Journal of Bodywork and Movement Therapies Volume 14 2010

Official journal of the: Journal of ® Association of Bodywork Neuromuscular Therapists, and Movement Ireland Therapies ® Australian Pilates Method Association ® National Association of Myofascial Trigger Point Therapists, USA ® Pilates Foundation, UK Volume 14 Issue 1 2010 EDITOR-IN-CHIEF Leon Chaitow ND, DO c/o School of Integrated Health, University of Westminster, 115 New Cavendish Street, London W1M 8JS, UK Preferred mailing address: P.O.Box 41, Corfu, Greece 49100 ([email protected]) ASSOCIATE EDITORS John Hannon DC Dimitrios Kostopoulos PhD, DSc, PT San Luis Obispo, CA, USA ( [email protected]) Hands-on Physical Therapy, New York, NY, USA ([email protected]) Glenn M. Hymel EdD, LMT Craig Liebenson DC Department of Psychology, Loyola University, New Orleans, LA, Los Angeles, CA, USA ([email protected]) USA ([email protected]) ASSOCIATE EDITORS: PREVENTION & REHABILITATION Warrick McNeill MCSP Matt Wallden MSc, Ost, Med, DO, ND London, UK ([email protected]) London, UK ([email protected]) International Advisory Board D. Beales MD (Cirencester, UK) S. Fritz LMT (Lapeer, MI, USA) J. M. McPartland DO (Middleburg, VT, USA) G. Bove DC, PhD (Kennebunkport, ME, USA) G. Fryer PhD. BSc., (Osteopath), ND C. Moyer PhD (Menomonie, WI, USA) C. Bron PT (Groningen, The Netherlands) D. R. Murphy DC (Providence, RI, USA) I. Burman LMT (Miami, FL, USA) (Melbourne City, Australia) T. Myers (Walpole, ME, USA) J. Carleton PhD (New York, USA) C. Gilbert PhD (San Francisco, USA) C. Norris MSc CBA MCSP SRP (Sale, UK) F. P. Carpes PhD (Uruguaiana, RS, Brazil) C. H. Goldsmith PhD (Hamilton, ON, Canada) N. Osborne BSc DC FCC (Orth.), FRSH, ILTM Z. Comeaux DO FAAO (Lewisburg, WV, USA) S. Goossen BA LMT CMTPT (Jacksonville, FL, USA) P. Davies PhD (London, UK) S. Gracovetsky PhD (Ocracoke, NC, USA) (Bournemouth, UK) J. P. (Walker) DeLany LMT (St Petersburg, FL, M. Hernandez-Reif PhD (Tuscaloosa, AL, USA) B. O’Neill MD (North Wales, PA, USA) P. Hodges BPhty, PhD, MedDr (Brisbane, Australia) J. L. Oschman PhD (Dover, NH, USA) USA) B. Ingram-Rice OTRLMT (Sarasota, FL, USA) D. Peters MB CHB DO (London, UK) M. Diego PhD (Florida, USA) J. Kahn PhD (Burlington, VT, USA) M. M. Reinold PT, DPT, ATC, CSCS (Boston, MA, J. Dommerholt PT, MS, DPT, DAAPM (Bethesda, R. Lardner PT (Chicago, IL, USA) P. J. M. Latey APMA (Sydney, Australia) MD, USA) MD, USA) E. Lederman DO PhD (London, UK) G. Rich PhD (Juneau, AK, USA) J. Downes DC (Marietta, GA, USA) D. Lee BSR, FCAMT, CGIMS (Canada) C. Rosenholtz MA, NCTMB (Boulder, CO, USA) C. Fernandez de las Peñas PT, DO, PhD (Madrid, D. Lewis ND (Seattle, WA, USA) R. Schleip MA, PT (Munich, Germany) W. W. Lowe LMT (Bend, OR, USA) J. Sharkey MSc, NMT (Dublin, Ireland) Spain) J. McEvoy PT MSC DPT MISCP MCSP (Limerick, D. G. Simons MD (Covington, GA, USA) T. M. Field PhD (Miami, FL, USA) D. Thompson LMP (Seattle, WA, USA) P. Finch PhD (Toronto, ON, Canada) Ireland) E. Wilson BA MCSP SRP (York, UK) T. Findley MD, PhD (New Jersey, USA) L. McLaughlin DSc PT (Ontario, Canada) A. Vleeming PhD (Rotterdam, D. D. FitzGerald DIP ENG, MISCP, MCSP (Dublin, C. McMakin MA DC (Portland, OR, USA) The Netherlands) Ireland) Visit the journal website at http://www.elsevier.com/jbmt Available online at www.sciencedirect.com Amsterdam • Boston • London • New York • Oxford • Paris • Philadelphia • San Diego • St. Louis Printed by Polestar Wheatons Ltd, Exeter, UK

Journal of Bodywork & Movement Therapies (2010) 14, 1e2 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt EDITORIAL Has osteopathy a role to play in treatment of flu? H1N1 influenza (also referred to as swine flu)e for most Osteopathy, now the Kirksville College of Osteopathic people who contract it e is no more aggressive or Medicine . contacted all their alumni. This effort culmi- dangerous than regular seasonal influenza. (CDC 2009) nated in 2445 osteopaths responding in treating 110,122 patients with influenza, with a resulting mortality of 0.25%. And yet for some the infection has proved fatal, with One of the few osteopathic medical hospitals, 400- bed reports suggesting that the majority of fatalities, related to Massachusetts Osteopathic Hospital, in Boston, also repor- H1N1 infection, have occurred in individuals who are ted a mortality of 0.25% for that period. (Walter 1992) immune compromised, or who have serious secondary pathologies, including diabetes, liver and/or heart disease. Building on this historical evidence, Hruby & Hoffman (2007) note that, although there were no controlled studies Most fatalities occur when the infection moves from the (and no descriptive comparisons between MD and DO standard influenza-like symptoms, to a severe acute patients), osteopaths achieved a high success rate perhaps respiratory syndrome (SARS), leading to pneumonia. This is due to osteopathic manipulative therapy a similar pattern to those who contracted Avian H5N1 influenza, several years ago. (MMWR 2003) What treatment did osteopaths use Osteopathic possibilities? OMT (osteopathic manipulative treatment) comprised a series of modalities that attempted to enhance thoracic The Spanish flu outbreak of 1918 was the first of its kind to mobility and lymphatic drainage, as well as liver, spleen have a variety of modern treatment approaches applied. and abdominal function. These included osteopathic, naturopathic and chiropractic care, in addition to standard medical care. The U.S. Dept Hruby & Hoffman have described the range of approaches Health & Human Services lists three reasons why, at that used e not as a specific protocol, but, ‘‘as a listing of OMT time, standard medical care was ineffective. First, physi- procedures as a resource for use in an overall treatment plan cians mistakenly believed Pfeiffer’s bacillus (rather than for a given patient ... These include thoracic, hepatic, a virus) was responsible, despite a lack of supportive splenic, abdominal and pedal lymphatic pump procedures, scientific evidence. Secondly, masks were relied upon as well as rib raising procedures. Also included are other OMT despite their ineffectiveness with viruses (masks CAN procedures that, although not thoroughly researched, have prevent bacterial spread). Lastly, although few physicians been clinically observed to provide similar effects. These believed in miasmas and imbalances in the humours, their procedures include soft tissue procedures, pectoral trac- remedies derived from these theories. (U.S. Dept Health & tion, mandibular drainage, frontal and maxillary lifts, and Human Services 2009) diaphragm doming .[as well as]., muscle energy tech- niques that can help to improve rib cage biomechanics.’’ Magoun (2004) has presented a well-documented approach to osteopathic care at that time e with implica- Most such approaches would be familiar to osteopathic tions for those who have influenza nowadays, whether practitioners. regular seasonal, Avian H5N1, or the current model, H1N1. Belief Magoun (2004) discusses osteopathic manipulative approaches: It may be useful to reflect on the effects of the strong and widespread conviction, held by many osteopaths (and In the United States, more than 28% of the population chiropractors) e that manipulative methods are capable of succumbed (Kolata, 2001) In US military hospitals, the encouraging the self-regulating functions of the body e and mortality rate averaged 36%, while the mortality rate in US how such convictions e(possibly more widely held in 1918 medical hospitals fell between 30% and 40%, with the exception of a rate of 68% in medical hospitals in New York City. (Patterson 2000) . the American School of 1360-8592/$36 ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2009.09.001

2 Editorial than 2010?) were able to translate to their flu-ridden The potential value of such methods, applied in appro- patients? priate situations, alongside standard medical care, by osteopaths, physiotherapists, chiropractors, or other suit- Paulus (2006) articulates this view when he says that the able trained therapists/practitioners, deserves further ‘‘quintessential goal’’ of the osteopath should be to: study, and not just in relation to H1N1. ‘‘diagnose the lack of motion and to help restore any quality of motion to the disordered region .. restoration References of motion, not alignment of the musculoskeletal system, activates the therapeutic process that bring about CDC, 2009. http://www.cdc.gov/h1n1flu/sick.htm (accessed 14. healing.’’ 09.09.). Current evidence? Hruby, R., Hoffman, K., 2007. Avian influenza: an osteopathic component to treatment. Osteopat Med Primary Care 1, 10. Interestingly, although PubMed lists 956 H1N1 citations during the past six months, not one includes these Jackson, K., et al., 1998. Effect of lymphatic and splenic pump keywords: manual therapy, lymphatic or even physical techniques on the antibody response to hepatitis B vaccine: therapy. (PubMed, 2009) However recent research supports a pilot study. J. Am. Osteopath. Assoc. 98, 155e160. the possibility that general OMT is beneficial in enhancing immune function, particularly with respect to upper Knott, M., et al., 2005. Lymphatic pump treatments increase respiratory infections. thoracic duct flow. J. Am. Osteopath. Assoc. 105, 447e456. For example: Kolata, G., 2001. Flu e the Story of the Great Influenza Pandemic of 1918 and the Search for the Virus that Caused It. Simon &  Sleszynski and Kelso (1993) demonstrated that preven- Schuster, New York. tion of post surgical atelectasis, using osteopathic thoracic manipulation was just as successful as incen- Magoun, H., 2004. More about the use of OMT during influenza tive spirometry epidemics. J Am Osteopathic Assoc 104 (10), 407.  Jackson et al. (1998) found lymphatic and splenic pump Morbidity & Mortality World Report, Mar 28, 2003. Outbreak of techniques enhanced the antibody response to hepa- Severe Acute Respiratory SyndromedWorldwide. Centers for titis B vaccination Disease Control. 52:241e246, 248.  Noll et al. (1999, 2000) provided clear evidence of the Nicholas, A., Oleski, S., 2002. J. Am. Osteopath. Assoc. 102 (Suppl. value of OMT in care of elderly hospitalized pneumonia 3), S5eS8. patients. Manual methods were applied to elderly hospitalized patients with pneumonia, Hospital time Noll, D., et al., 1999. Adjunctive osteopathic manipulative treat- was reduced from a mean of 8.6 days without OMT to ment in the elderly hospitalized with pneumonia: a pilot study. 6.6 days for those receiving OMT. Additionally OMT J. Am. Osteopath. Assoc. 99, 143e152. patients required less intravenous antibiotics Noll, D., et al., 2000. Benefits of osteopathic manipulative treat-  Nicholas & Oleski (2002) Described a four-step protocol, ments for hospitalized elderly patients with pneumonia. J Am composed of rib raising and treatment of the thoracic Osteopath Assoc. 100, 776e782. inlet, respiratory diaphragm, and pelvic diaphragm e for postoperative pain. ‘‘Patients who receive Patterson, M., 2000. Osteopathic methods and the great flu pandemic morphine preoperatively and OMT postoperatively tend of 1917e1918. J. Am. Osteopathic. Assoc. 100, 309e310. to have less postoperative pain and require less intra- venously administered morphine. In addition, OMT and Paulus, S., 2006. Concerning osteopathy: vital motions and material relief of pain lead to decreased postoperative forms 2006. http://www.interlinea.org/ (viewed September 18, morbidity and mortality and increased patient satis- 2009). faction. Also, soft tissue manipulative techniques and thoracic pump techniques help to promote early PubMed, 2009. http://www.ncbi.nlm.nih.gov/pubmed?termZ ambulation and body movement.’’ %28swineþORþH1N1%29þANDþ%28fluþORþinfluenzaþORþvirus þORþoutbreakþORþpandemic%29þANDþ%22lastþ6þmonths%22  Knott et al. (2005) demonstrated that osteopathic [edat] (viewed September 16, 2009). thoracic pump, and abdominal pump techniques, increased the flow of lymph through the thoracic ducts Sleszynski, S.L., Kelso, A.F., 1993. Comparison of thoracic manip- of mongrel dogs. ulation with incentive spirometry in preventing postoperative atelectasis. J. Am. Osteopathic. Assoc. 93, 834e838. 843e845. U.S. Dept Health & Human Services, 2009. http://1918.pandemicflu. gov/the_pandemic/03.htm (viewed September 16, 2009). Walter, G., 1992. The First School of Osteopathic Medicine. The Thomas Jefferson University Press at Northeast Missouri State University, Kirksville, Mo, p. 95. Leon Chaitow New Medicine Group, 144 Harley Street, London W1, United Kingdom E-mail address: [email protected]

Journal of Bodywork & Movement Therapies (2010) 14, 3e12 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt CLINICAL HYPOTHESIS Fascia: A missing link in our understanding of the pathology of fibromyalgia Ginevra L. Liptan a,b,* a Dept. of Medicine, Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239, United States b Legacy Good Samaritan Pain Management Center, 1130 NW 22nd Ave, Suite 345, Portland, OR 97210, United States Received 26 January 2009; received in revised form 7 August 2009; accepted 11 August 2009 KEYWORDS Summary Significant evidence exists for central sensitization in fibromyalgia, however the Central sensitization; cause of this process in fibromyalgiadand how it relates to other known abnormalities in fibro- Myofascial release; myalgiadremains unclear. Central sensitization occurs when persistent nociceptive input Manual therapy; leads to increased excitability in the dorsal horn neurons of the spinal cord. In this hyperex- Inflammation; cited state, spinal cord neurons produce an enhanced responsiveness to noxious stimulation, Connective tissue; and even to formerly innocuous stimulation. Growth hormone No definite evidence of muscle pathology in fibromyalgia has been found. However, there is some evidence for dysfunction of the intramuscular connective tissue, or fascia, in fibromyalgia. This paper proposes that inflammation of the fascia is the source of peripheral nociceptive input that leads to central sensitization in fibromyalgia. The fascial dysfunction is proposed to be due to inadequate growth hormone production and HPA axis dysfunction in fibromyalgia. Fascia is richly innervated, and the major cell of the fascia, the fibroblast, has been shown to secrete pro-inflammatory cytokines, particularly IL-6, in response to strain. Recent biopsy studies using immuno-histochemical staining techniques have found increased levels of collagen and inflammatory mediators in the connective tissue surrounding the muscle cells in fibromyalgia patients. The inflammation of the fascia is similar to that described in conditions such as plantar fasciitis and lateral epicondylitis, and may be better described as a dysfunctional healing response. This may explain why NSAIDs and oral steroids have not been found effective in fibromyalgia. * Legacy Good Samaritan Pain Management Center, 1130, NW 22nd Ave, Suite 345, Portland, OR 97210, United States. Tel.: þ1 503 413 7513; fax: þ1 503 413 7503. E-mail address: [email protected] 1360-8592/$ - see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2009.08.003

4 G.L. Liptan Inflammation and dysfunction of the fascia may lead to central sensitization in fibromyalgia. If this hypothesis is confirmed, it could significantly expand treatment options to include manual therapies directed at the fascia such as Rolfing and myofascial release, and direct further research on the peripheral pathology in fibromyalgia to the fascia. ª 2009 Elsevier Ltd. All rights reserved. Introduction osteoarthritis normalized following successful hip replace- ment surgery (Kosek and Ordeberg, 2000). The etiology of fibromyalgiada disorder characterized by widespread muscle pain and tenderness at specific soft-tissue What causes central sensitization in tender pointsdremains unclear. However, in the past decade fibromyalgia? evidence for abnormal pain processing in fibromyalgia has significantly advanced our understanding of this disorder. In Since patients with fibromyalgia complain of sore, painful 2002, a functional MRI study demonstrated that it took much muscles, investigators have long been searching for muscle less thumbnail pressure in fibromyalgia patients to activate pathology in fibromyalgia. These studies have included muscle the pain sensing areas of the brain compared to controls biopsies with structural and ultra-structural observation, (Gracely et al., 2002). Another study found that fibromyalgia magnetic resonance imaging and metabolic studies, electro- patients experienced stronger pain and larger areas of myography, and studies of blood flow and muscle strength. For referred pain after intramuscular injection of hypertonic the most part these studies have not shown consistent saline (Sorensen et al., 1998). Other research has shown differences between healthy and fibromyalgia muscles. In abnormal temporal summation and wind-up of pain in fibro- Simm’s rigorous review of 32 studies of muscle in fibromyalgia myalgia (Staud et al., 2004). These findings demonstrate that he states ‘Although controversy persist, the weight of in fibromyalgia the central nervous system has an exaggerated evidence from studies that are methodologically sound response to pain, a phenomenon called central sensitization. suggests that muscles are not abnormal’ (Simms, 1996). Central sensitization is caused by repeated or sustained Others still argue that peripheral pain mechanisms must noxious input to the dorsal horn neurons leading to play an important role in fibromyalgia pain. ‘Central increased neuronal responsiveness or central sensitization. sensitization has to have an initial genesis and nociceptive In fibromyalgia, however, no evidence of muscle pathology stimuli from painful foci in muscle are increasingly recog- has been described, leading to speculation that the central nized as being relevant to the development of fibromyalgia’ sensitization in fibromyalgia may occur spontaneously (Bennett, 2004). Supporting the idea that there are soft- though some as yet unknown mechanism (Ji et al., 2003). tissue abnormalities in fibromyalgia is the distribution of pain, which is not uniform as one would expect if the pain Others argue that myofascial trigger points cause the was generated solely from a spontaneous central nervous central sensitization in fibromyalgia (Staud, 2008). system hypersensitivity, but is most prevalent in certain areas of the soft-tissue, especially the shoulders, chest, However recent biopsy studies have found increased and lower back (Starz et al., 2008). The often observed levels of collagen and inflammatory mediators in the fascia worsening of fibromyalgia pain after an episode of muscle of fibromyalgia patients. This paper proposes that overuse also argues for a peripheral pathology in FM. dysfunction and inflammation of the intramuscular connective tissue, or fascia, leads to the central sensiti- Myofascial trigger points have been suggested as the zation seen in fibromyalgia. peripheral source of painful input leading to central sensi- tization in fibromyalgia. Myofascial trigger points are Central sensitization discrete painful spots located in a palpable taut band of skeletal muscle, classified as active if they cause pain at Central sensitization, a state of heightened sensitivity in the rest, and latent if they are painful only with palpation spinal cord, is thought to be a physiologic adaptation of the (Simons et al., 1999). However, attributing the central nervous system to sustained painful input. It is the end result sensitization seen in fibromyalgia solely to trigger points is of a complex neuronal response to peripheral nerve injury or problematic. Not all patients with fibromyalgia have trigger tissue inflammation. Recent studies support an important role points, and not all patients with trigger points have fibro- for dorsal horn glial cells (support cells for neurons) and NMDA myalgia. One study found 68% of fibromyalgia patients had receptors in producing abnormal pain sensitivity in the spinal identifiable trigger points (Granges and Littlejohn, 1993), cord (Watkins et al., 2001; Dickenson and Sullivan, 1987). and another found trigger points in only 38% of fibromyalgia subjects examined (Wolfe et al., 1992). Myofascial trigger In lab animals, central sensitization can be induced by points are also quite commond33e54% of completely injecting inflammatory chemicals into muscle, and by asymptomatic individuals have latent trigger points (Sola damaging peripheral nerves. Central sensitization has also et al., 1955; Schiffman et al., 1990). been described in many chronic pain conditions, including endometriosis (Bajaj et al., 2003), peripheral arterial Background disease (Lang et al., 2006), and chronic low back pain (O’Neill et al., 2007). In these conditions there is a known The symptoms of fibromyalgia have historically been source of persistent nociceptive input that keeps the CNS in described by many different terms, including ‘Chronic a continued state of sensitization. One group showed that central sensitization associated with painful hip

Fascia: a missing link 5 Rheumatism’ and ‘Muscular Rheumatism’. In a review article Figure 1 Published with kind permission of Ron Thompson. in 1904, Stockman described the symptoms of chronic rheu- to its roles in regulation of inflammation and wound repair. matism as ‘pain, aching, stiffness, a readiness to feel Fibroblast activation is induced by various stimuli that muscular fatigue, interference with free muscular move- occur with tissue injury. Activated fibroblasts isolated from ment, and very often a want of energy and vigour’ the site of a healing wound will continue to secrete higher (Stockman, 1904). Chronic rheumatism was not thought by levels of ECM and proliferate more rapidly than fibroblasts Stockman to affect the joints themselves, but rather the obtained from normal tissue. Fibroblasts are also an fibrous tissues structures of the muscles. He attributed important source of ECM degrading proteases, and have chronic rheumatism primarily to infectious causes, particu- larly rheumatic fever and influenza, but also noted some cases with no infectious etiology. Stockman notes the work of Balfour and Scudamore, two British physicians who separately in the early 19th century put forward the idea that the pain of muscular rheumatism occurs as a result of thickenings developing in the fibrous connective tissue of muscle. Sir William Gowers attributed symptoms of muscular rheumatism to the ‘inflammation of fibrous tissue’, and proposed that the condition should be called ‘fibrositis’ (Gowers, 1904). Due to lack of evidence of peripheral inflammation, in 1976 the term ‘fibromyalgia’ was proposed, and ultimately adopted by the American College of Rheumatology when they released formal diagnostic criteria for the condition in 1990 (Wolfe et al., 1990). However the current terminology still reflects the concept of connective tissue abnormality in fibromyalgia, as the name is composed of the Latin words for fiber, muscle, and pain. Allopathic medicine has historically regarded fascia as relatively inert. According to a recent article in Science magazine ‘medical books barely mention fascia and anatomical displays remove it’ (Grimm, 2007). However in osteopathic medicine, the fascia has long been recognized as a potential cause of pain and soft-tissue dysfunction. As one osteopath writes ‘The whole of OMT [osteopathic manipulative treatment] has been concerned, purposefully or not, with manipulation of the fascia’ (Danto, 2003). Fascia Fascia is the dense connective tissue that envelopes Figure 2 Structure of skeletal muscle, illustrating the layers muscles grossly, and also surrounds every bundle of muscle of surrounding connective tissue known as the fascia, which fibers and each individual muscle cell. This connective includes the deep fascia, epimysium, perimysium and tissue is inextricably linked with the muscle, and is contin- endomysium. uous with the tendons and periosteum (Figures 1 and 2). The fascia is composed of cellsdincluding fibroblasts, macrophages and mast cellsdand extracellular matrix. The extracellular matrix (ECM) is composed of ground substance and collagen and elastin fibers. Fascia is essentially a dense gel (the ground substance) in which cells and fibers are suspended, giving it colloidal properties. Fascia is richly innervatedda histological study found nerve fibers in all specimens of the deep fascia, including a variety of both free and encapsulated nerve endings, especially Ruffini and Pacini corpuscles (Stecco et al., 2006). In fact muscle innervation is primarily located in the fascia: consisting of 25 percent stretch receptors of muscle cells, and 75 percent free nerve endings in intramuscular fascia, and in the walls of blood vessels and tendons (Bonica, 1990). The principal cell of the connective tissue is the fibro- blast, which produces the extracellular matrix, in addition

6 G.L. Liptan a crucial role in maintaining homeostasis and repair in the thoracolumbar fascia than healthy controls (Langevin ECM (Kalluri and Zeisberg, 2006). et al., 2009). Fascia has been demonstrated in vitro to have some Eosinophilic fasciitis, a rare condition resulting in contractile behavior. Some fibroblasts, called myofibro- widespread eosinophilic infiltration and inflammation of blasts, express alpha-smooth-muscle actin and are able to the fascia results in significant fibrosis of the fascia. contract (Schleip et al., 2005, 2006). Increased expression ‘Adhesions seen in eosinophilic fasciitis, which develops of smooth-muscle actin is thought to be triggered by grossly thickened fascia and fibrosis are indicative of the mechanical stimulation and inflammation in order to potential for fascial inflammation to cause adhesions’ promote wound healing and tissue repair. (Franklyn-Miller et al., 2009). Fibroblasts also respond to mechanical stretch with Evidence for fascial dysfunction in hyperplasia and secretion of inflammatory cytokines (Skutek fibromyalgia et al., 2001). Using in vitro models, Dodd et al. demon- strated that fibroblasts respond to acyclic mechanical strain When Stockman examined muscle biopsy studies of patients by altering shape and alignment, undergoing hyperplasia with ‘chronic rheumatism’ in 1904, he found inflammatory and secreting inflammatory cytokines, including IL-6 (Dodd hyperplasia of the connective tissue. Specifically he et al., 2006). described a section of inflamed perimysium which on light microscopic evaluation consisted of a ‘proliferated and Fibroblasts have a vital role in the regulation of inflam- oedematous fibrous tissue with an amorphous matrix’, mation. Dysregulation of fibroblasts has been implicated in leading him to conclude that ‘the essential pathological the chronic inflammation seen in rheumatoid arthritis. changes in chronic rheumatism are confined to white Synovial fibroblasts isolated from rheumatoid arthritis fibrous tissue’ (Stockman, 1904). However, Collins later joints were found to secrete increased amounts of NF-kB, examined Stockman’s published illustrations and noted a transcription factor that ‘appears to play a critical role in ‘scarcely more variation in fibrous tissue structure than can perpetuating both tissue hyperplasia and the inflammatory be encountered normally in different situations in the response at sites of chronic inflammation’ (Miagkov et al., human body’. Collins also examined 7 ‘typical fibrositic’ 1998; Buckley et al., 2001). nodules under light microscopy and found no evidence of inflammation (Collins, 1940). Both of these early studies Fibrosis and adhesions suffered from methodological flaws including lack of controls groups and poorly defined diagnostic criteria. One of the hallmarks of connective tissue, including fascia, is its mutability and remodeling in response to mechanical More recent studies of FM muscle using standard histo- stress. However, under certain conditionsdexcess pathology techniques under light microscopy have not mechanical stress, inflammation or immobilitydthis shown any consistent pathology (Lindh et al., 1995; Drewes process can result in excessive and disorganized collagen et al., 1993). However one group described a ‘network of and matrix deposition resulting in fibrosis and adhesions reticular fibers connecting muscle fibers’ causing a ‘rubber- (Langevin, 2008). band like’ constriction of muscle fibers seen under light microscopy (Bartels and Danneskold-Samsoe, 1986). In plantar fasciitis and tendinitis of the elbow these types of changes have been reported. Two series of surgical Electron microscopic studiesdwhich examine the biopsies in patients with plantar fasciitis reported fascial myofibrils and sarcomeres that make up individual muscle thickening, collagen disorganization and increased fibro- cellsdhave also not shown any differences between fibro- blasts. Jarde et al. (2003) in a report on 38 cases of plantar myalgia muscles and controls (Yunus et al., 1989a). fasciitis noted ‘collagen degeneration with fibers losing their longitudinal arrangement and presenting with While no consistent abnormalities have been found at a haphazard orientation, with an increase in fibroblastic either the ultrastructural or structural level of muscle cells cellular density’. They also noted microcalcifications in the using standard techniques, two recent studies using fascia of a few of the surgical specimens. The authors found specialized immuno-histochemical staining techniques that these lesions were similar to those found in cases of focused on the intramuscular connective tissue have tendon injury. discovered some intriguing abnormalities. Tendons are essentially a denser version of intramus- Spaeth et al. describe an increase in collagen IV cular fascia with the same components of fibroblasts, surrounding the muscles of fibromyalgia patients. collagen and extracellular matrix. According to a review of Comparing immuno-stained muscle biopsies from 25 fibro- the histopathological changes found in lateral epicondylitis, myalgia patients to 26 healthy controls, they described the most common findings were hypertrophy of fibroblasts a ‘slight, but significant increase in collagen surrounding and abundant disorganized collagen (Kraushaar and Nirschl, the muscle cells of the fibromyalgia patients’ (Spaeth 1999). et al., 2005). A biopsy study of the thoracolumbar fascia in chronic Ruster et al. also found increased levels of collagen in mechanical low back pain found evidence suggestive of the endomysium in fibromyalgia muscles, and in addition fascial inflammation, in particular degenerative changes in describe evidence for endomysial inflammation and tissue the collagen fibers and microcalcifications in the fascia damage. Specifically, they note elevated levels of N-car- (Bednar et al., 1995). In an ultrasound-based comparison, boxymethyllsine (CML), an advanced glycation end-product chronic low back pain patients had approximately (AGE) that is considered to be a marker of oxidative stress 25% thicker perimuscular connective tissue in the and tissue damage, in the fascia of fibromyalgia patients.

Fascia: a missing link 7 ‘CML staining was stronger in the fibromyalgia patients, and depression. Unlike FM, pain is not a major feature was detected primarily in the interstitial tissue between described in adult GH deficiency syndrome. However the muscle fibers’ (emphasis added). They reported treatment with GH has been reported to improve pain increased staining of collagen types I, II, and VI in the levels in adult GH deficient patients (Cuneo et al., 1998). interstitial tissue compared to healthy subjects and found These conditions may not be directly comparable, however, ‘the collagens and CML were co-localized, suggesting because true adult GH deficiency is usually acquired due to that the AGE modifications were related to collagen’. In pituitary damage and is generally accompanied by multiple addition, they found increased levels of CD-68 positive other pituitary hormone deficiencies. In contrast, fibro- macrophages and activated NF-kB in the interstitial tissue myalgia patients have normal pituitary responses but have of fibromyalgia muscles (Ruster et al., 2005). As described subtle alterations in hypothalamic control of growth earlier, NF-kB is a transcription factor that plays an hormone release (Leal-Cerro et al., 1999). important role in the regulation of fibroblast hyperplasia and cytokine release, and high levels of NF-kB have also Fibroblasts have growth hormone receptors, and in been reported in synovial fibroblasts from rheumatoid response to growth hormone secrete many important locally joints (Miagkov et al., 1998). acting growth factors, such as IGF-1 (Murphy et al., 1983; Oakes et al., 1992). Fibroblasts play a central role in wound This immuno-histochemical evidence is suggestive of healing, and IGF-1 is a major physiological mediator of fascial inflammation in fibromyalgia. As described earlier, normal wound healing (Suh et al., 1992) A study of wound focal fascial inflammation has been described in other healing in rats revealed increased IGF-1 immunoreactivity in conditions as plantar fasciitis and low back pain. Giesecke fibroblasts, epidermal cells and macrophages in the inci- et al. found evidence for central sensitization in idiopathic sional area (Todorovic et al., 2008). Improved wound heal- chronic low back pain patients (Giesecke et al. 2004). Since ing and increased staining for IGF-1 in healing tissue have local myofascial inflammation as described in chronic low been reported after administration of recombinant human back pain could be a trigger of central sensitization, it is growth hormone (Gilpin et al., 1994; Herndon et al., 1995). possible that a more generalized fascial inflammation could Local IGF-1 administration has also been found to improve lead to central sensitization as well. In fact peripheral wound healing (Suh et al., 1992; Beckert et al., 2007). afferent nociceptors of muscle, the majority of which reside in the fascia, have been shown to be highly effective An intriguing study of gamma-hydroxybutyrate, a medi- at causing central sensitization (Wall and Woolf, 1984). cation known to increase slow wave sleep, was found to both increase growth hormone levels and improve wound healing Growth hormone and sleep abnormalities in rats (Murphy et al., 2007). This medication has also shown benefit in recent human studies of patients with FM as well, Moldofsky was able to cause symptoms of fibro- and the improvements in sleep significantly correlated with myalgiadwidespread muscle pain and fatiguedin healthy improvements in pain scores (Russell et al., 2009). patients by depriving them of deep (slow-wave) sleep experimentally (Moldofsky and Scarisbrick, 1976). These Hypothesis symptoms resolved once subject were again allowed deep sleep. Sleep studies have demonstrated that fibromyalgia Fascial dysfunction and inflammation may lead to the wide- patients experience reduced deep sleep that is frequently spread pain and central sensitization seen in fibromyalgia. interrupted with alpha-waves which are normally associ- This paper proposes that the fascial dysfunction in fibro- ated with states of wakefulness (Moldofsky et al., 1975). myalgia could be caused by chronic tension in the fascia and an impaired fascial healing response due to inadequate growth Growth hormone is primarily secreted during deep sleep hormone stimulation. In genetically prone individuals, and after exercise, and is responsible for regulating the a trauma may trigger prolonged dysfunction of the stress healing and maintenance of tissues. Nearly 70% of total GH response. This chronic autonomic arousal and hypervigilance secretion occurs at night, and GH secretion ‘will not occur may cause excess fascial tension, interfere with deep sleep if sleep stage III or IV is prevented by awakening the and impair growth hormone release (Figure 3). subject’ (Felig et al., 1995). There seems to be a genetic component to fibro- Reduced 24 h secretion of GH in FM has been reported, myalgiadfirst-degree relatives of patients with fibro- with the decrease most noticeable during the night when myalgia are 8.5 times more likely to have fibromyalgia than GH secreted in the patients was much lower than in relatives of patients with rheumatoid arthritis (Arnold controls (Leal-Cerro et al., 1999). Another group also found et al., 2004). An association between trauma and fibro- reduced GH secretion during sleep compared to controls myalgia has also been reported, with one study finding that (Landis et al., 2001). More than 90% of fibromyalgia patients ‘physical trauma in the preceding 6 months is significantly have inadequate growth hormone response to exercise associated with the onset of FM’ (Al-Allaf et al., 2002). (Paiva et al., 2002) and one third have significantly low circulating IGF-1 levels (Bennett et al., 1992). Human Hyperactivity of the stress response has also been growth hormone replacement in FM patients resulted in described in fibromyalgia, with dysfunction of both hypo- significant improvement of symptoms and reduction in thalamicepituitaryeadrenal axis and of the autonomic tender points in one study (Bennett et al., 1998). nervous system (Adler et al., 1999; Cohen et al., 2000). Hyperactivity of the HPA axis can also cause a blunted Some of the clinical features of FM are similar to those growth hormone response (Jones et al., 2007). described in adult GH deficiency syndrome including fatigue, muscle weakness, impaired exercise tolerance and Chronic sympathetic dominance of the nervous system may also promote chronic tension in the fascial system.

8 G.L. Liptan mechanical stress from daily activities, and thus have higher levels of fascial inflammation. The areas near muscle/tendon junctions are particularly susceptible to microinjuries from mechanical forces. In fact, six of the 18 tender points used to define the condition occur in or near areas of tendinous insertions, namely those at the sub- occipital muscle insertions, near the epicondyles and at the medial fat pad of the knee (Figure 4a and b). Anti-inflammatories in fibromyalgia If fascial inflammation exists in FM, why are non-steroidal anti-inflammatory medications (NSAID) and corticosteroids ineffective? No improvement in fibromyalgia symptoms was reported with prednisone 15 mg per day for two weeks, or with the NSAID medications ibuprofen and naproxen (Clark et al., 1985; Goldenberg et al., 1986; Yunus et al., 1989b). This paper argues that there is indeed fascial inflam- mation in fibromyalgia, but that it is a type of inflammation that is not responsive to oral NSAIDs or corticosteroids. The fascial inflammation proposed to exist in fibromyalgia is similar to that described in chronic overuse injuries such as lateral epicondylitis and plantar fasciitis. This inflammation is attributed to cumulative microtrauma that overwhelms the tissue’s ability to repair itself, resulting in a chronic inflammatory reaction that may be more appropriately termed a ‘dysfunctional healing response’. The response to injury of connective tissue, including fascia, ligaments and tendons, occurs in three phases (Kumar, 1999). Figure 3 Proposed etiology of central sensitization in 1) Inflammatory phasedinvasion of polymorphonuclear fibromyalgia. cells and monocytes/macrophages, and release of prostaglandin and cytokines 2) Proliferative phasedfibroblasts activated to produce collagen and extracellular matrix that is laid down in disorganized fashion 3) Remodeling phasedprogressive maturation and align- ment of collagen fibers and remodeling of extracellular matrix Fascia has recently been shown to be able to have signifi- The anti-inflammatory effect of NSAIDs is due to their cant contractile force in vitro, and this fascial contractility interference with prostaglandin production, thus they are is though to contribute to the incredible feats of strength effective in the initial inflammatory phase of injury repair. humans can perform in emergenciesdsituations in which NSAIDs have been shown to be helpful in decreasing pain the sympathetic nervous system is also dominant (Schleip and swelling in acute soft-tissue injuries, but not in chronic et al., 2005; Schleip et al., 2006). soft-tissue inflammation (Heere, 1987). A randomized controlled trial of NSAIDs in plantar fasciitis found that both In response to chronic excess fascial tension, fibroblasts placebo and NSAID group improved over time, and there would likely overproduce collagen and extracellular matrix was no statistical difference between the groups at 1, 2 or in a continuous attempt to respond to the increased 6 months (Donley et al., 2007). Another randomized mechanical stress. However due to inadequate growth controlled study found no difference between placebo and hormone stimulation of fibroblast there may be an impaired NSAID treatment in chronic achilles tendinopathy (Astrom fascial healing response resulting in chronic fascial inflam- and Westlin, 1992). mation; there is ‘a critical role for fibroblasts in regulating the switch from acute to chronic inflammation in tissues’ Local corticosteroid injections have shown effectiveness (Buckley et al., 2001). in overuse injuries but this effect tends to be short-lived. A randomized controlled trial of steroid injections in This widespread dysfunctional fascial healing response plantar fasciitis found a statistically significant pain could be considered a ‘bodywide fasciitis’ as compared to reduction at 1 month in the treatment group that had dis- the more focal fasciitis seen in other conditions such as appeared by 3 months post treatment (Crawford et al., plantar fasciitis. The tender points of fibromyalgia may 1999). In lateral epicondylitis steroid injections also provide reflect areas that suffer the greatest microtrauma and

Fascia: a missing link 9 Figure 4 a and b: 18 tender points of fibromyalgia as established by 1990 ACR criteria (Wolfe et al., 1990). only temporary improvement, and ‘the significant short- however only be removed by local and well-directed term benefits of corticosteroid injections are paradoxically manipulations’ (Stockman, 1904). This idea was reiterated reversed after six weeks with high recurrence rates’ recently by a leading fascia researcher, ‘Treatments (Bissett et al., 2006). involving direct mechanical stimulation of connective tissue can potentially reverse connective tissue fibrosis’ In an animal model of chronic muscle inflammation (Langevin, 2008). Myofascial fibrotic changes can theoreti- created by injecting inflammatory stimulants into the cally be treated by breaking up excessive collagen adhe- hamstrings of mice, neither NSAIDs nor high-dose oral sions through soft-tissue and myofascial release techniques corticosteroids were effective in reducing inflammation. (Ward, 2003). If there is excess tension in the fascial system The inflammation could only be reduced by local cortico- in fibromyalgia due to chronic sympathetic nervous domi- steroid injection directly into the muscle (Green and nance, manual therapy may also help reduce that tension. Mangan, 1980). Notably, while NSAIDs and oral steroids have been tested in FM, the effectiveness of local steroid A randomized controlled pilot study demonstrated that injections in FM has not been assessed. osteopathic manipulative treatment (OMT), in conjunction with medication, was more effective in relieving symptoms NSAIDs and corticosteroids are not only ineffective in of fibromyalgia than medication alone (Gamber et al., relieving chronic soft-tissue inflammation but may actually 2002). A total of 24 patients were included in the study, and hinder the healing process. Two studies reported slowed the treatment group received once weekly OMT sessions for muscle repair in animals treated with an NSAID (Obremsky 23 weeks. The control group received either moist heat et al., 1994; Almekinders and Gilbert, 1986). Indomethacin packs at each visit or no additional treatment beyond their added to repetitively stretched fibroblasts in vitro reduced usual medications. The osteopathic manipulative tech- the secretion of prostaglandins but also inhibited the niques used in this study were individualized for each synthesis of DNA, an effect that may be detrimental in the patient, so it is difficult to assess how much treatment remodeling phase of repair (Almekinders et al., 1995). directed specifically at the fascia that each patient Corticosteroids are also notorious for impairing surgical received. Each patient received Jones strain/counterstrain wound healing (Suh et al., 1992). Thus NSAIDs and corti- techniques and other modalities per provider dis- costeroids may actually worsen an already dysfunctional cretiondincluding myofascial release, muscle energy, soft- tissue repair response in fibromyalgia. tissue treatment and craniosacral manipulation. Manual therapy in fibromyalgia treatment A Swedish study on connective tissue massage in fibro- myalgia found a pain-relieving benefit of 37% in addition to In 1904 Stockman recognized the potential of manual reduced use of analgesics and positive effects on quality of therapy in treating chronic rheumatism (what is now called life. The treatment group consisted of 23 patients who fibromyalgia) and noted that ‘indurated fibrous tissue can received 15 treatments over 10 weeks, while the control

10 G.L. Liptan group participated in weekly discussion groups. The Al-Allaf, A.W., Dunbar, K.L., Hallum, N.S., 2002. A case-control connective tissue massage is described as a ‘manual tech- study examining the role of physical trauma in the onset of niques to detach dense connective tissue’, but no further fibromyalgia syndrome. Rheumatology (Oxford) 41, 450e453. description of the technique is provided. Interestingly, this treatment was chosen for the study because ‘experienced Almekinders, L.C., Gilbert, J.A., 1986. Healing of experimental massage therapists who were surveyed prefer connective muscle strains and the effects of nonsteroidal antiinflammatory tissue massage for the treatment of individuals with fibro- medication. American Journal of Sports Medicine 14 (4), myalgia’ (Brattberg, 1999). 303e308. However, for manual therapies to be effective in fibro- Almekinders, L.C., Baynes, A.J., Bracey, L.W., 1995. An in vitro myalgia, they must take into account the colloidal prop- investigation into the effects of repetitive motion and nonste- erties of fascia, and according to Chaitow and DeLany ‘the roidal antiinflammatory medication on human tendon fibro- amount of resistance colloids offer increases proportionally blasts. American Journal of Sports Medicine 23 (1), 119e124. to the velocity of force applied to them. This makes a gentle touch a fundamental requirement . when Arnold, L.M., Hudson, J.J., Hess, E.V., et al., 2004. Family study of attempting to produce a change in, or release of restricted fibromyalgia. Arthritis & Rheumatism 50, 944e952. fascial structures which are all colloidal in their behavior’ (Chaitow and DeLany, 2000). Therefore, only slow and Astrom, M., Westlin, N., 1992. No effect of piroxicam on achilles sustained pressure will effect changes in the fascial tissue. tendinopathy. A randomized study of 70 patients. Acta Ortho- paedica Scandinavica 63 (6), 631e634. Appropriate manual therapy must allow for the state of reduced growth hormone and thus reduced capacity for Bajaj, P., Bajaj, P., Madsen, H., Arendt-Nielson, L., 2003. Endo- tissue repair in fibromyalgia by allowing for sufficient rest metriosis is associated with central sensitization: a psycho- between sessions. Utilizing the growing body of knowledge physical controlled study. The Journal of Pain 4 (7), 372e380. on the properties of fascia can help manual therapists treat fibromyalgia patients with techniques that don’t cause Bartels, E.M., Danneskold-Samsoe, B., 1986. Histological abnor- further injury and inflammation, but rather gently break malities in muscle from patients with certain types of fibrositis. apart existing fascial restrictions and adhesions and Lancet 1 (8484), 755e757. promote tissue healing. Beckert, S., Haack, S., Hierlemann, H., et al., 2007. Stimulation of Conclusion steroid-suppressed cutaneous healing by repeated topical appli- cation of IGF-1: different mechanisms of action based on mode of This paper presents the hypothesis that fascial dysfunction IGF-1 delivery. Journal of Surgical Research 139 (2), 217e221. in fibromyalgia leads to widespread pain and central sensitization. Using other known abnormalities in fibro- Bednar, D.A., Orr, F.W., Simon, G.T., 1995. Observations on the myalgia, a proposed mechanism leading to fascial pathomorphology of the thoracolumbar fascia in chronic dysfunction in fibromyalgia is described. mechanical back pain. Spine 20 (10), 1161e1164. The in vivo microdialysis techniques developed by Shah’s Bennett, R.M., Clark, S.R., Campbell, S.M., Burkhardt, C.S., 1992. group to assess myofascial trigger points could also be used Low levels of somatomedin-C in patients with the fibromyalgia to evaluate the chemical composition of fascial interstitial syndrome: a possible link between sleep and muscle pain. fluid for evidence of inflammation (Shah et al., 2005). In Arthritis & Rheumatism 35 (10), 1113e1116. vitro examination of fibroblasts removed from fascial tissues in fibromyalgia could look for evidence of activa- Bennett, R.M., Clark, S.C., Walczyk, J., 1998. A randomized, tion, such as excess secretion of extracellular matrix and double-blind, placebo-controlled study of growth hormone in inflammatory mediators. Comparing fascial IGF-1 levels in the treatment of fibromyalgia. American Journal of Medicine fibromyalgia to controls may also be useful. 104 (3), 227e231. Finally, clinical studies of manual therapies that target Bennett, R.M., 2004. Fibromyalgia: present to future. Current Pain the fascia, like Rolfing and myofascial release, could help and Headache Reports 8, 379e384. define the role of fascia in producing fibromyalgia pain. Directly comparing a therapy aimed at releasing fascial Bissett, L., Beller, E., Jull, G., et al., 2006. Mobilisation with restriction such as myofascial release to a massage therapy movement and exercise, corticosteroid injection, or wait and that focuses primarily on muscle relaxation would be see for tennis elbow: randomized trial. British Journal of informative. If there is clinical improvement with manual Medicine 333 (7575), 939. therapies targeting the fascia, this could significantly improve our ability to treat fibromyalgia, and guide further Bonica, J.J., 1990. The Management of Pain. Lea & Febinger, research on the peripheral pathology of fibromyalgia Philadelphia, p. 34. towards the fascia. Brattberg, G., 1999. Connective tissue massage in the treatment of References fibromyalgia. European Journal of Pain 3, 235e245. Adler, G.K., Kinsley, B.T., Hurwitz, S., Mossey, C.J., Goldenberg, D.L., Buckley, C.D., Pilling, D., Lord, J.M., et al., 2001. Fibroblasts 1999. Reduced hypothalamic-pituitary and sympathoadrenal regulate the switch from acute resolving to chronic persistent responses to hypoglycemia in women with fibromyalgia syndrome. inflammation. Trends in Immunology 22 (4), 199e204. The American Journal of Medicine 106, 534e543. Chaitow, L., DeLany, J.W., 2000. Clinical Applications of Neuro- muscular Techniques. Churchill Livingstone, London, p. 3. Clark, S., Tindall, E., Bennett, R.M., 1985. A double blind crossover trial of prednisone versus placebo in the treatment of fibrositis. Journal of Rheumatology 12 (5), 980e983. Cohen, H., Neumann, L., Shore, M., Amir, M., Cassuto, Y., Buskila, D., 2000. Autonomic dysfunction in fibromyalgia: application of a power spectral analysis of heart rate variability. Seminars in Arthritis and Rheumatism 29, 217e227. Collins, D., 1940. Fibrositis and infection. Annals of the Rheumatic Diseases 2, 114e126. Crawford, F., Atkins, D., Young, P., et al., 1999. Steroid injection for heel pain: evidence for short-term effectiveness. A randomized controlled trial. Rheumatology (Oxford) 38 (10), 974e977. Cuneo, R.C., Judd, S., Wallace, J.D., et al., 1998. The Australian multicenter trial of growth hormone (GH) treatment in GH-deficient adults. Journal of Clinical Endocrinology & Metabolism 83 (1), 107e116.

Fascia: a missing link 11 Danto, J.B., 2003. Review of integrated neuromusculoskeletal Jones, K.D., Deodhar, P., Lorentzen, A., et al., 2007. Growth release and the novel application of a segmental ante- hormone perturbations in fibromyalgia: a review. Seminars in rior/posterior approach in the thoracic, lumbar, and sacral Arthritis and Rheumatism 36 (6), 357e379. regions. Journal of the American Osteopathic Association 103 (12), 583e596. Kalluri, R., Zeisberg, M., 2006. Fibroblasts in cancer. Nature Reviews Cancer 6 (5), 392e401. Dickenson, A.H., Sullivan, A.F., 1987. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of Kosek, E., Ordeberg, G., 2000. Abnormalities of somatosensory deep rat dorsal horn nociceptive neurons following C fibre perception in patients with painful osteoarthritis normalize stimulation. Neuropharmacology 26, 1235e1238. following successful treatment. European Journal of Pain 4 (3), 229e238. Dodd, J.G., Good, M.M., Nguyen, T.L., Grigg, A.I., Batia, L.M., Standley, P.R., 2006. In vitro biophysical strain model for under- Kraushaar, B.S., Nirschl, R.P., 1999. Tendinosis of the elbow (tennis standing mechanisms of osteopathic manipulative treatment. elbow): clinical features and findings of histological, immuno- Journal of the American Osteopathic Association 106 (3), 157e166. histochemical and electron micrsocopy studies. The Journal of Bone and Joint Surgery 81 (2), 259e278. Donley, B.G., Moore, T., Sferra, S., et al., 2007. The efficacy of oral nonsteroidal anti-inflammatory medication (NSAID) in the treat- Kumar, S., 1999. Biomechanics in Ergonomics. Taylor & Francis, ment of plantar fasciitis: a randomized, prospective, placebo- Philadelphia, p. 36. controlled study. Foot & Ankle International 28 (1), 20e23. Landis, C., Lentz, M., Rothermel, J., et al., 2001. Decreased Drewes, A.M., Andreasen, A., Schroder, H.D., et al., 1993. nocturnal levels of prolactin and growth hormone in women Pathology of skeletal muscle in fibromyalgia: a histo-immuno- with fibromyalgia. The Journal of Clinical Endocrinology & chemical and ultrastructural study. British Journal of Rheuma- Metabolism 86 (4), 1672e1678. tology 32 (6), 479e483. Lang, P.M., Schober, G.M., Rolke, R., et al., 2006. Sensory Felig, P., Baxter, J.D., Frohman, L.A., 1995. Endocrinology and neuropathy and signs of central sensitization in patients with Metabolism. McGraw-Hill, New York, p. 307. peripheral arterial disease. Pain 124 (1e2), 190e200. Franklyn-Miller, A., Falvey, E., McCrory, P., 2009. Fasciitis first Langevin, H.M., 2008. Potential role of fascia in chronic musculo- before tendinopathy; does the anatomy hold the key? British skeletal pain. In: Audette, Bailey (Eds.), Integrative Pain Journal of Sports Medicine (epub ahead of print). Medicine. Humana Press, p. 125. Gamber, R.G., Shores, J.H., Russo, D.P., Jimenez, C., Rubin, B.R., Langevin, H.M., Stevens-Tuttle, D., Fox, J.R., et al., 2009. Ultra- 2002. Osteopathic manipulative treatment in conjunction with sound evidence of altered lumbar connective tissue structure in medication relieves pain associated with fibromyalgia human subjects with chronic low back pain. Abstract only. syndrome: results of a randomized clinical pilot project. Fascia Research Congress. Journal of the American Osteopathic Association 102 (6), 321e325. Leal-Cerro, A., Povedano, J., Astorga, R., et al., 1999. The growth hormone (GH)-releasing hormone-GH-insulin-like growth factor- Giesecke, T., Gracely, R.H., Grant, M.A.B., et al., 2004. Evidence 1 axis in patients with fibromyalgia syndrome. The Journal of of augmented central pain processing in idiopathic chronic low Clinical Endocrinology & Metabolism 84 (9), 3378e3381. back pain. Arthritis & Rheumatism 50 (2), 613e623. Lindh, M., Johansson, G., Hedberg, M., et al., 1995. Muscle fiber Gilpin, D.A., Barrow, R.E., Rutan, R.L., et al., 1994. Recombinant characteristics, capillaries and enzymes in patients with human growth hormone accelerates wound healing in children fibromyalgia and controls. Scandinavian Journal of Rheuma- with large cutaneous burns. Annals of Surgery 220 (1), 19e24. tology 24 (1), 34e37. Goldenberg, D.L., Felson, D.T., Dinerman, H., 1986. A randomized Miagkov, A.V., Kovalenko, D.V., Brown, C.E., et al., 1998. NF-kB controlled trial of amitriptyline and naproxen in the treatment activation provides the potential link between inflammation of patients with fibromyalgia. 29, 1371e1377. and hyperplasia in the arthritic joint. Proceedings of the National Academy of Sciences of the United States of America Gowers, W., 1904. Lumbago: its lessons and analogues. British 95, 13859e13864. Medical Journal 1, 117e121. Moldofsky, H., Scarisbrick, P., England, R., Smythe, H., 1975. Gracely, R., Petzke, F., Wolf, J., Clauw, D., 2002. Functional magnetic Musculoskeletal symptoms and non-REM sleep disturbance in resonance imaging evidence of augmented pain processing in patients with fibrositis and healthy subjects. Psychosomatic fibromyalgia. Arthritis & Rheumatism 46 (5), 1333e1343. Medicine 37, 341e351. Granges, G., Littlejohn, G., 1993. Prevalence of myofascial pain Moldofsky, H., Scarisbrick, P., 1976. Induction of neurasthenic syndrome in fibromyalgia syndrome and regional pain musculoskeletal pain syndrome by selective sleep stage depri- syndrome: a comparative study. Journal of Musculoskeletal Pain vation. Psychosomatic Medicine 38 (1), 35e44. 1, 19e35. Murphy, K.D., Rose, M.W., Chinkes, D.L., et al., 2007. The effects Grimm, D., 2007. Cell biology meets Rolfing. Science 318, 1234e1235. of gamma-hydroxybutyrate on hypermetabolism and wound Green, A.P., Mangan, F.R., 1980. The effect of steroidal and non- healing in a rat model of large thermal injury. Journal of Trauma 63 (5), 1099e1107. steroidal anti-inflammatory drugs on chronic muscle inflamma- tion. Journal of Pharmacy and Pharmacology 32, 319e322. Murphy, L.J., Vrhovsek, E., Lazarus, L., 1983. Identification and Heere, L.P., 1987. Piroxicam in acute musculoskeletal disorders characterization of specific growth hormone receptors in and sports injuries. European Journal of Rheumatology and cultured human fibroblasts. Journal of Clinical Endocrinology & Inflammation 8 (1), 60e64. Metabolism 57 (6), 1117e1124. Herndon, D.N., Hawkins, H.K., Nguyen, T.T., et al., 1995. Charac- terization of growth hormone enhanced donor site healing in Oakes, S.R., Haynes, K.M., Waters, M.J., et al., 1992. Demonstra- patients with large cutaneous burns. Annals of Surgery 221 (6), tion and localization of growth hormone receptors in human 649e659. skin and skin fibroblasts. Journal of Clinical Endocrinology & Jarde, O., Diebold, P., Havet, E., et al., 2003. Degenerative lesions Metabolism 75 (5), 1368e1373. of the plantar fascia: surgical treatment by fasciectomy and excision of the heel spur. A report on 38 cases. Acta Ortho- Obremsky, W.T., Seaber, A.V., Ribbeck, B.M., et al., 1994. paedica Belgica 69 (3), 267e274. Biomechanical and histologic assessment of a controlled muscle Ji, R.R., Kohno, T., Moore, K.A., Woolf, C.J., 2003. Central sensi- strain injury treated with piroxicam. American Journal of Sports tization and LTP: do pain and memory share similar mecha- Medicine 22 (4), 558e561. nisms? Trends in Neurosciences 26 (12), 696e705. O’Neill, S., Manniche, C., Graven-Nielsen, T., Arendt-Nielsen, L., 2007. Generalized deep-tissue hyperalgesia in patients with chronic low-back pain. European Journal of Pain 11 (4), 415e420.

12 G.L. Liptan Paiva, E.S., Deodhar, A., Jones, K.D., Bennett, R., 2002. Impaired Starz, T.W., Vogt, M., Gold, K., 2008. Fibromyalgia: what’s tender, growth hormone secretion in fibromyalgia patients: evidence what’s not. In: American College Rheumatology 2008 Annual for augmented hypothalamic somatostatin tone. Arthritis & Scientific Meeting, abstract 1405. Rheumatism 46 (5), 1344e1350. Staud, R., Price, D.D., Robinson, M.E., et al., 2004. Maintenance of Russell, I.J., Perkins, A.T., Michalek, J.E., 2009. Sodium oxybate windup of second pain requires less frequent stimulation in relieves pain and improves function in fibromyalgia syndrome. fibromyalgia patients compared to normal controls. Pain 110 Arthritis & Rheumatism 60 (1), 299e309. (3), 689e696. Ruster, M., Franke, S., Spaeth, M., et al., 2005. Detection of Staud, R., 2008. The role of peripheral input for chronic pain elevated N-carboxymethyllysine levels in muscular tissue and syndromes like fibromyalgia syndrome. Journal of Musculoskel- serum of patients with fibromyalgia. Scandinavian Journal of etal Pain 16 (1 & 2), 67e74. Rheumatology 34 (6), 460e463. Stecco, C., Porzionaato, A., Macchi, V., Tiengo, C., Parenti, A., Schiffman, E.L., Fricton, J.R., Haley, D.P., 1990. The prevalence Aldegheri, R., Delmas, V., De Caro, R., 2006. A histological and treatment needs of subjects with temporomandibular study of the deep fascia of the upper limb. Italian Journal of disorders. Journal of the American Dental Association 120, Anatomy and Embryology 11 (2), 1e6. 295e303. Stockman, R., 1904. The causes, pathology and treatment of chronic Schleip, R., Klingler, W., Lehmann-Horn, F., 2005. Active fascial rheumatism. Edinburgh Medical Journal 15 (107e116), 223e235. contractility: fascia may be able to actively contract in a smooth muscle-like manner and thereby influence musculo- Suh, D.Y., Hunt, T.K., Spencer, E.M., 1992. Insulin-like growth skeletal dynamics. Medical Hypotheses 65, 273e277. factor-1 reverses the impairment of wound healing induced by corticosteroids in rats. Endocrinology 131 (5), 2399e2403. Schleip, R., Lehmann-Horn, F., Klingler, W., 2006. Fascia is able to contract in a smooth muscle-like manner and thereby influence Todorovic, V., Pesko, P., Micev, M., 2008. Insulin-like growth factor-1 musculoskeletal mechanics. In: Proceedings of the 5th World in wound healing of rat skin. Regulatory Peptides 150 (1e3), 7e13. Congress of Biomechanics, pp. 51e54. Munich, Germany. Wall, P.D., Woolf, C.J., 1984. Muscle but not cutaneous C-afferent Shah, J., Phillips, T., Danoff, J., Gerber, L., 2005. An in vivo input produces prolonged increases in the excitability of the microanalytical technique for measuring the local biochemical flexion reflex in the rat. The Journal of Physiology (London) 356, milieu of human skeletal muscle. Journal of Applied Physiology 453e458. 99, 1977e1984. Ward, R.C., 2003. Foundations for Osteopathic Medicine. Lippin- Simms, R., 1996. Is there muscle pathology in fibromyalgia cott, Williams & Wilkins, Philadelphia, p. 1158. syndrome? Rheumatic Disease Clinics of North America 22 (2), 245e266. Watkins, L.R., Milligan, E.D., Maier, S.F., 2001. Glial activation: a driving force for pathological pain. Trends in Neurosciences 24 Simons, D.G., Travell, J.G., Simons, L.S., 1999. Myofascial Pain and (8), 450e455. Dysfunction: The Trigger Point Manual, vol. 1. Lippincott Wil- liams & Wilkins, Baltimore, p. 19. Wolfe, F., Smythe, H.A., Yunus, M.B., et al., 1990. The American College of Rheumatology 1990 criteria for the classification of Skutek, M., van Griensven, M., Zeichen, J., et al., 2001. Cyclic fibromyalgia. Arthritis & Rheumatism 33 (2), 160e172. mechanical stretching enhances secretion of interleukin 6 in human tendon fibroblasts. Knee Surgery, Sports Traumatology, Wolfe, F., Simons, D.G., Fricton, J., et al., 1992. The fibromyalgia Arthroscopy 9 (5), 322e326. and myofascial pain syndrome: a preliminary study of tender points and trigger points in persons with fibromyalgia, myofas- Sola, A.E., Rodenberger, M.L., Gettys, B.B., 1955. Incidence of cial pain syndrome and no disease. Journal of Rheumatology 19 hypersensitive areas in posterior shoulder muscles. American (6), 944e951. Journal of Physical Medicine 34, 585e590. Yunus, M.B., Kalyan-Raman, U.P., Masi, A.T., Aldag, J.C., 1989a. Sorensen, J., Graven-Nieslen, T., Henriksson, K.G., et al., 1998. Electron microscopic studies of muscle biopsy in primary fibro- Hyperexcitability in fibromyalgia. Journal of Rheumatology 25 myalgia syndrome: a controlled and blinded study. Journal of (1), 152e155. Rheumatology 16, 97e101. Spaeth, M., Fischer, P., Lagner, C., Pongratz, D., 2005. Increase of Yunus, M.B., Masi, A.T., Aldaq, J.C., 1989b. Short term effects of collagen IV in skeletal muscle of fibromyalgia patients. Journal ibuprofen in primary fibromyalgia syndrome: a double blind, of Musculoskeletal Pain 12 (9S), 67. placebo controlled trial. Journal of Rheumatology 16 (4), 527e532.

Journal of Bodywork & Movement Therapies (2010) 14, 13e18 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt APPLIED PHYSIOLOGY Skin distraction at select landmarks on the spine midline in the upright and fully flexed postures Paul J. Moga, D.O., Ph.D Post Office Box 4088, Ann Arbor, Michigan 48106e4088, USA Received 31 January 2008; received in revised form 13 March 2008; accepted 28 April 2008 KEYWORDS Summary Background: This study was aimed at quantifying superoinferior and mediolateral Skin distraction; skin distraction over the spine’s midline for the purpose of designing a unique surface marker Spinal motion; for use in a study originally proposed by Wojtys and Ashton-Miller. It was also aimed at testing Spinal angles; the null hypotheses H01: There is no difference in the amount of skin distraction between Hamstring tightness; hamstring normal and hamstring tight subjects and H02: There are no age or gender differ- Fascia; ences of skin distraction. Manual medicine; Methods: Nine male and twelve female volunteers served as the convenience subjects for this Spine IRB-approved study. Eight subjects were classified as hamstring ‘‘tight’’ (short) using the Finger-to-Floor Reach Test. Skin distraction was measured at five spine midline landmarks palpated on the subjects’ bared backs: T1, T10, L3, S1, and the posterioresuperior iliac spine (PSIS). A pattern of four dots was placed at each landmark using a rectangular template and non-toxic, water-soluble ink. Measurements were taken between superoinferior and mediolat- eral pairs of template points with subjects in both upright (‘‘initial’’) and fully flexed (‘‘final’’) postures. Between-measurement differences were then calculated, expressed as percent strain, and pooled for mean percent strain values. Repeated measures produced a maximum strain error of about 1.7%. Results: With the exception of the skin over the T10 landmark, distraction in the superoinferior direction was greater than that in the mediolateral direction. There were no significant differ- ences in skin distraction between age or gender groups. However, hamstring short males had significantly smaller superoinferior skin distraction at L3 than their hamstring normal counterparts [35% (Æ5.2) vs. 46% (Æ4.6), pZ0.049), while hamstring short females had a smaller mean medio- lateral distraction at the same level that approached significance [2.5% (Æ2.5) vs. 7.6% (Æ5.4), pZ0.080). At this landmark, there was a moderately strong, inverse correlation (rZÀ0.720) between mediolateral percent strain and reach distance in hamstring tight subjects. Conclusion: In general, superoinferior percent strain increased and mediolateral percent strain decreased from thoracic to sacral regions, likely reflecting the relative increase in spine segment motion from thoracic to lumbar region. The larger mean mediolateral E-mail address: [email protected] 1360-8592/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2008.04.037

14 P.J. Moga distraction at the T10 level was probably the result of traction on the skin at that level by the dependence of appendicular structures in forward flexion. Finally, the negative value at the T1 landmark was probably the result of the cervical spine extension that occurred during flexion as the subjects lifted their heads to look up. ª 2008 Elsevier Ltd. All rights reserved. Introduction methods measure the angle between two lines that demarcate the spine segment of interest. In the Cobb The accuracy of the spine angle measurements based on technique (1948), the most cephalad vertebra of the spine skin surface markers depends on the marker’s ability to segment of interest is demarcated by a line tangential to its withstand skin distraction during subject movement, as this superior endplate. The most caudad vertebra of the same movement may introduce error (e.g., Vanneuville et al., spine segment is similarly marked by a tangent to its infe- 1994). Quantification of such movement is therefore rior endplate. The angle between the two tangents at their essential to surface marker design. Because such informa- intersection represents the spine segment angle (Figure 1). tion is lacking in the literature, a small study was performed to answer two questions: first, how much does In Ferguson’s method (1949), the intersecting lines of the skin distract over the midline of the spine at select the spine segment are based on three centroids, rather landmarks and second, are there significant differences in than two vertebral endplates. Centroids are points ‘‘at the the amount of distraction between landmarks? The answers geometric center of a polygon’’, which ‘‘.can be used to to these questions would be incorporated into the design of represent a polygon as a point’’ (Gregory, 2002). In this a unique surface marker for use in an optoelectronic study case, the polygon is the vertebral body or centrum. The (Moga, 2002) that compared the spine angles of subjects vertebrae selected are at the cephalad aspect, the apex, who were hamstring normal (able to touch the ground in and the caudad aspect of the spine segment of interest sagittal plane flexion) and hamstring short (unable to touch (Figure 2). Centroids are determined as the intersection of the ground in sagittal plane flexion). two lines drawn on each vertebral body from a corner to the opposite corner. Then two lines are drawn, one from In keeping with the hamstring category comparison, the the cephalad vertebral centroid to that of the apical following null hypotheses were advanced: vertebra, and the other from the caudad vertebral centroid to that of the apical vertebra. As in the Cobb method, the Primary null hypothesis angle between the intersection of two lines represents the spine segment angle. H01: There is no difference in the amount of skin distrac- tion between hamstring normal and hamstring short subjects. It was expected that hamstring short subjects would have less skin distraction than hamstring normal subjects, which would reflect a reduced flexion range of motion. Secondary null hypothesis H02: There are no age or gender differences of skin distraction. Background There are primarily two approaches used to determine Figure 1 Cobb method (after Riseborough and Herndon, 1975) spine segment angles: contact and non-contact. The angle q, representing the angle of the spine segment, is contact approach requires the application of goniometric measured as the angle between tangents to vertebral end- measuring devices directly to the skin over palpated land- plates (image ª Paul John Moga, 2002). marks. These devices include Loebl’s pendulum goniometer (1967); Debrunner’s kyphometer (1972), and Willner’s pantograph (1981). The non-contact approach may be subdivided into two basic categories: radiographic and non-radiographic. The radiographic approach utilizes X-ray exposure to identify specific vertebral landmarks. The angles between these landmarks are then determined generally by using either the Cobb or, less commonly, the Ferguson method. Both

Skin distraction at select landmarks on the spine midline 15 Figure 2 Ferguson method angle q, representing the angle of Figure 3 Balsa template horizontal base: 2.8 cmÂ1.8 cm the spine segment, is measured as the angle between centroid- 0.6 cm; vertical stanchion: 4.4 cmÂ1.2 cmÂ0.6 cm; image ª based lines (image ª Paul John Moga, 2002). Paul John Moga, 2002. Non-radiographic, non-contact techniques used to Harris (1983) or Stokes et al. (1987) have shown that the measure spine segment angles include photogrammetric error between skin mounted targets and true skeletal methods that mitigate the need for X-ray exposure. Like motion ranges from 8% to 10%. Leroux, et al. (2000) contact goniometric methods, non-radiographic techniques determined correlation coefficients of 0.94 and 0.91 for generally require both demarcation and labeling of spine kyphosis and lordosis, respectively, between radiographic segment landmarks. However, rather than relying on Cobb measurements and ‘‘spatial localization’’ of markers radiographic landmarks, surface-mounted markers are placed superficial to spinous processes. Troup et al. (1968) placed at manually palpated structures that define spine obtained correlation coefficients of 0.91 between surface segment endpoints. marker and X-ray spine angles for erect and fully flexed postures for 14 observations, with no significant differences Once the spine segment is defined, techniques similar to between the methods (Student’s t-test, p<0.001). Despite those of Cobb and Ferguson may be used to determine the the strong correlation and similar marker system, they spine angle. A modification of the Ferguson method was observed that ‘‘displacements of the skin in the long axis of used in the photogrammetric method as described by the spine. proved to be considerable’’, and added that Wojtys et al. (2000). In that study, flat, adhesive markers such displacement was a particular source of errors, espe- were placed on the skin’s surface at either end of the spine cially in the lumbar region. segment of interest. A test platform having cameras with photographic axes perpendicular to the plane of the Skin markers utilized at the lumbar area are less reliable, subjects’ backs was used. A custom software program was primarily because of soft tissue thickness (Willner, 1981; developed to measure spine angles from the photographs as Bryant et al., 1989). When using non-radiographic methods, the angle of intersection between lines tangential to the special care must be taken when placing surface markers at skin’s surface landmarks. this region, as the thickness of the superficial tissues can affect marker position relative to the bony reference point. A modification of the Cobb method was utilized in the optoelectronic method as described by Moga (2002). In that So, skin surface motion must be incorporated into study, specially designed platform markers were positioned a marker’s design. If not, the marker’s base may be dislodged on the skin over select spine landmarks at either end of the during torso movement. Also, the protrusion of vertebral spine segment of interest. A commercially available, elec- spinous processes during torso flexion must be considered, tronic videographic system having camera axes parallel to the skin surface was used to record marker position. Using Table 1 Mean percent strain at various spine landmarks. the system’s software, spine angles were then calculated as the intersection of the angles between lines perpendicular Landmark Superoinferior Mediolateral to the skin’s surface, as demarcated by the skin markers’ perpendicular arms. Mean s.d. Mean s.d. At least one problem exists with the use of surface- T1 À3.7 13.1 0.0 6.1 mounted markersdDo the markers accurately reflect the position of bony landmarks? Authors such as Thurston and T10 10.7 10.2 20.5 10.2 L3 41.7 7.1 6.1 6.5 S1 49.7 9.7 1.6 6.3 PSIS 53.9 11.8 0.1 4.9 s.d.Zstandard deviation. Note: mean length % strain increases from T1 to S1.

16 P.J. Moga Table 2 Mean differences of percent strain between groups. Repetition of the measurements produced an error landmark points. which, with the exception of two outliers, ranged from an underestimation of 1.5 mm to an overestimation of the Superoinferior Mediolateral initial value by 1.5 mm. The mean error was an underesti- mation of 0.07 mm. This measurement error, similar to that T1eT10 14.4 20.5 found by van Weeren and Barneveld (1986), was propagated T10eL3 31 À14.4 as a maximum strain error of approximately 1.7%. L3eS1 S1ePSIS 8 À4.5 Results 4.2 À1.5 lest the perpendicular orientation of the markers’ vertical Mean superoinferior and mediolateral percent strains for arm relative to the skin’s surface be affected. In this event, each spine landmark are listed in Table 1. With the the result would be inaccurate spine angle measurements. exception of the skin over the T10 landmark, distraction in the superoinferior direction was greater than that in the Methods mediolateral direction. In the superoinferior direction, percent strain ranged from À3.7% at T1 to 53.9% at the PSIS Nine male and twelve female (nZ21) volunteers served as landmark. The negative value at the T1 landmark was likely the subjects for this Human Use Committee-approved the result of cervical spine extension that may have study. They had no gross kyphoscoliosis, a mean age of 26.9 occurred during flexion as the subjects lifted their heads to (Æ10.5) years, a mean height of 1.73(Æ0.09) m, and a mean look up. weight of 67.68(Æ9.71) kg. Approximately one-half exhibi- ted hamstring tightness (shortness) by the Finger-to-Floor Mean percent strain differences between adjacent Reach Test. landmarks are displayed in Table 2. The greatest change in superoinferior strain was between the T10 and L3 levels, Skin distraction was measured at five landmark points on and was more than twice that of the T1ÀT10 difference. each subject’s bared back. These points were the first and The greatest difference in mediolateral strain was between the tenth thoracic vertebrae (T1, T10), the third lumbar T1 and T10, almost twice that of T10ÀL3. These findings vertebra (L3), the first sacral segment (S1), and the were exhibited by both gender groups (Figures 4 and 5). posterior-superior iliac spine (PSIS). The skin over these landmark points was marked using non-toxic, water-soluble Independent groups’ t-tests were then used to compare ink. Four dots were placed in a rectangular pattern at each the means of the skin distraction percent strains at the corner of a balsa wood template (Figure 3). various landmarks between hamstring normal and hamstring short subjects. There were no significant Distraction over the each landmark site was determined differences of percent strains between the two groups by comparing the distances between two pairs of points (calculated t-test statistic valueZ0.1140; pZ0.88, using measured with subjects in two positions. The four skin two-tailed distribution for heteroscedastic samples with surface dots were paired as superoinferior and mediolateral a set at the 0.05 level of significance). However, after template points. The two postures were upright and fully dividing the subjects according to gender, t-tests revealed flexed positions. Each participant was asked to flex (bend a significant difference of superoinferior percent strain at forward) in the sagittal plane to the best of his or her the L3 landmark between hamstring groups for males ability. Each set of measurements were then repeated and (calculated t-test statistic valueZ2.8000; pZ0.049) and averaged. a difference that approached significance (calculated t-test statistic valueZ1.9519; pZ0.080) of the mediolateral The differences between the initial and final measure- percent strain at the L3 landmark between hamstring ments at all landmark points for each subject were calcu- groups for females (Table 3). Although no inferential anal- lated and expressed as percent strain, with mean strains ysis was done to determine its statistical significance, there calculated for both hamstring normal and hamstring short was a moderately strong, inverse correlation (Pearson Figure 4 Superoinferior skin surface strain by gender (image Figure 5 Mediolateral skin surface strain by gender (image courtesy of L.J. Huston, MSE). courtesy of L.J. Huston, MSE).

Skin distraction at select landmarks on the spine midline 17 Table 3 Skin distraction (mean percent strain) at l3 for hamstring normal and Hamstring short subjects, along with calculated independent groups t-test statistic values and p-values. Hamstring normal Hamstring Tight Statistical value p-value Mean s.d. Mean s.d. Females Mediolateral 7.62 5.42 2.49 2.51 1.9519 0.080 0.049 n7 5 Males Superoinferior 46.33 4.61 35.12 5.18 2.8000 n3 3 s.d.Zstandard deviation. rZÀ0.720) between mediolateral percent strain and reach The increase of the skin’s percent strain from thoracic distance in hamstring short subjects at this landmark. spine (about 11% over T10) to lumbar spine (about 42% at These findings led to the failure to reject the first null L3) was in keeping with each segment’s relative contribu- hypothesis. tion to overall flexion range of motion (6 of flexion at T9À10 vs. about 16 at L3À4 (White and Panjabi, 1978 in The data were then examined for significant differences Nordin and Frankel, 1989). of skin distraction by age and between genders. Subjects were divided into two age groupsdunder 30 years old and Mean mediolateral strains diminished from T10 to PSIS. A over 30 years old (Table 4). Contrary to what was expected, larger mean mediolateral distraction at the T10 level may there were no significant differences of percent strains have been the result of traction on the skin by the depen- between age groups (calculated t-test statistic val- dence of osseous and soft tissue structures in forward ueZ0.0531; pZ0.94). Nor were there significant differ- flexion, such as the upper extremities, shoulder girdle, and ences of percent strains between genders (calculated t-test breast tissue. Tissue dependence was likely also the statistic valueZ0.1609; pZ0.84), even though males were explanation for the large difference in mediolateral strain significantly taller and heavier than females (Table 5). between T1 and T10. These findings led to the failure to reject the second null hypothesis. How did the differences of percent strain from landmark to landmark compare? There was a pronounced difference Discussion in mean superoinferior strain between the T10 and L3 landmarks. This likely reflected the greater flexion range of How did skin distraction in the two directions compare? At motion of the superior lumbar spine as compared to the every spine landmark level, with the exception of T10, inferior thoracic spine, which, of course, is relatively fixed mean skin distraction in the superoinferior direction was due to costal attachments. The greater flexion range of greater than mean distraction in the mediolateral direc- motion may also be the explanation for the moderately tion. The superoinferior strains increased consistently from strong inverse correlation between reach distance and cephalad to caudad along the spine, with the greatest mean mediolateral distraction at L3 in hamstring short subjects. distraction over the PSIS and S1 (53.9% and 49.7% strain, respectively). These results have some clinical implications. As distraction was determined over boney prominences and Although it is speculation, the 4% difference in strain not muscle, it is likely that distraction occurred in the skin over PSIS and S1 may reflect an anatomic relationship of and subdermal tissues. The increase of superoinferior these landmarks and the lumbodorsal fascia. Grossly, distraction in a cephalocaudad manner corresponds to the fascial fiber direction at PSIS appears to be more parallel increase in sagittal plane, flexion range of motion of the with that of the Latissimus dorsi, and, because of that spine from the thoracic to the lumbar region. Skin relation, may be more susceptible to the distraction when distraction was greatest at the lumbosacral regiondwhere the back is stretched and the upper limbs are dependent in the distal spine range of motion is at its peak. the flexed posture. In contrast, load over the pre-S1 fascia may be more symmetrical (along the midline), and the The study design prohibits any cause-and-effect conclu- fascia might be more fixed over that landmark point. sions. However, it may be possible that the converse of these findings is truedthat motion restrictions in these same soft Table 5 Mean weight and height for females and males. Weight (kg) Height (m) n Table 4 Mean ages of subjects grouped as over 30 years Mean s.d. Mean s.d. and under 30 years of age. Mean years s.d. n Females 61.51 4.70 1.67 0.06 12 Males 75.91 8.50 1.81 0.06 9 <30 years 23.72 2.82 18 p-value 0.000 0.000 >30 years 46.00 19.92 3 s.d.Zstandard deviation. Males were significantly larger than s.d.Zstandard deviation. females.

18 P.J. Moga tissues may impair overall flexion range. The findings of Cobb, J.R., 1948. Outline for the study of scoliosis. AAOS Instruc- a significantly smaller distraction in hamstring-short subjects tional Course Lecture 5, 261e275. lend support to this notion. In addition, it has already been demonstrated that tissues such as the lumbodorsal fascia Debrunner, H.U., 1972. The kyphometer (German). Zeitschrift fuer contribute to load transfer in the lower trunk (Vleeming Orthopaedic und Ihre Grenzgebiete 110 (3), 389e392. et al., 1995). Because of the functional importance of these soft tissues, manual treatment of the low back and pelvis Ferguson, A.B., 1949. Roentgen diagnosis of the extremities and spine. must go beyond boney and articular somatic dysfunction or In: Annals of Roentgenology: A Series of Monographic Atlases, vol. subluxation. It must address not only muscle tightness issues XVII, second ed. Paul B. Hoeber, Inc., NY, pp. 364e365. (e.g., Hamstrings, Latissimus, Gluteal group), but should include treatment of the more superficial, soft tissue struc- Gajdosik, R.L., Albert, C.R., Mitman, J.J., 1994. Influence of hamstring tures, such as the lumbodorsal fascia as well. length on the standing position and flexion range of motion of the pelvic angle, lumbar angle, and thoracic angle. Journal of Ortho- Conclusions paedic and Sports Physical Therapy 20 (4), 213e219. Several important findings resulted from this small study. Gregory, I., 2002. A place in history: a guide to using geographical Quantification of the relative amounts of skin distraction at information systems in historical research. In: Arts and several landmark points along the human spine’s midline in Humanities Data Service Guides to Good Practice. University of sagittal plane flexion helped to augment the current Essex, Colchester, UK. knowledge base. From thoracic to sacral regions, super- oinferior percent strain was found to increase in sagittal Leroux, M.A., Zabjek, K., Simard, G., Badeux, J., Coillard, C., plane flexion, in keeping with the spine’s ‘‘stretching’’ Rivard, C.H., 2000. A noninvasive anthropometric technique for while bending forward. Mediolateral percent strain was measuring kyphosis and lordosis: an application for idiopathic found to decrease, which was thought to be the result of an scoliosis. Spine 25 (13), 1689e1694. approximation of paravertebral tissue that became prom- inent during flexion. Loebl, W.Y., 1967. Measurements of spinal posture and range of spinal movements. Annals of Physical Medicine 9, Another finding was that hamstring normal subjects who 103e110. were able to touch the floor in a Finger-to-Floor Reach Test had significantly larger superoinferior and mediolateral skin Moga, P.J., 2002. On the relation between thoracic kyphosis, distraction changes than hamstring short individuals, or athletic training, hamstring shortness, and anthropometry in those unable to touch the floor in sagittal plane flexion. the developing spine. Doctoral Dissertation, University of This finding was believed to reflect the restricted lumbo- Michigan, Ann Arbor, MI. pelvic flexion that has been shown to occur in hamstring short subjects (Gajdosik et al., 1994). Nordin, N., Frankel, V.H., 1989. Basic Biomechanics of the Muscu- loskeletal System. Lea and Febiger, Philadelphia. Disclosures Riseborough, E.J., Herndon, J.H., 1975. Scoliosis and other This project was part of a larger study sponsored in part by Deformities of the Axial Skeleton. Little, Brown, Boston. a Grant from the Orthopedic Research and Education Foundation, Rosemont, IL, USA. Stokes, I.A., Bevins, T.M., Lunn, R.A., 1987. Back surface curvature and measurement of lumbar spinal motion. Spine 12 (4), Acknowledgments 355e361. The author would like to thank University of Michigan Thurston, A.J., Harris, J.D., 1983. Normal kinematics of the lumbar faculty Drs. E.W. Wojtys and J.A. Ashton-Miller for their spine and pelvis. Spine 8 (2), 199e205. support during the data collection phase of this work, and Laura Huston, M.S.E., for her assistance. Troup, J.D.G., Hood, C.A., Chapman, A.E., 1968. Measurement of the sagittal mobility of the lumbar spine and hips. Annals of References Physical Medicine 9, 308e321. Bryant, J.T., Reid, J.G., Smith, B.L., Stevensen, J.M., 1989. van Weeren, P.R., Barneveld, A., 1986. A technique to quantify skin Method for determining vertebral body positions in the sagittal displacement in the walking horse. Journal of Biomechanics 19 plane using skin markers. Spine 14 (3), 258e265. (10), 879e883. Vanneuville, G., Kyndt, T., Massaux, M., Harmand, Y., Garcier, J.M., Monnet, J.P., Guillot, M., Cluzel, P., Escande, G., Poumarat, G., 1994. Preliminary opto-electronic study on vertebral movement. Surgical and Radiologic Anatomy 16 (4), 385e391. Vleeming, A., Pool-Goudzwaard, A.L., Stoeckart, R., van Wingerden, J.P., Snijders, C.J., 1995. The posterior layer of the thoracolumbar fascia. Its function in load transfer from spine to legs. Spine 20 (7), 753e758. White, A.A., Panjabi, M.M., 1978. Clinical Biomechanics of the Spine. J.B. Lippincott, Philadelphia. Willner, S., 1981. Spinal pantographda noninvasive technique for describing kyphosis and lordosis in the thoracolumbar spine. Acta Orthopedica Scandinavica 52, 525e529. Wojtys, E.M., Ashton-Miller, J.A., Huston, L.J., Moga, P.J., 2000. The association between athletic training time and the sagittal curvature of the immature spine. American Journal of Sports Medicine 28 (4), 490e498.

Journal of Bodywork & Movement Therapies (2010) 14, 19e26 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt INVESTIGATIVE STUDY: HUMAN PHYSIOLOGY Relationship between hand-grip isometric strength and isokinetic moment data of the shoulder stabilisers Dimitris Mandalidis, PT, Ph.D.*, Moira O’Brien, FRCPI Department of Anatomy, Trinity College Dublin, Dublin 2, Ireland Received 4 March 2008; received in revised form 1 May 2008; accepted 2 May 2008 KEYWORDS Summary The purpose of this study was to examine the relationship between hand-grip (HG) Hand grip; isometric strength and isokinetic moment of the shoulder musculature in 18 healthy male Isokinetic strength; volunteers. HG isometric strength at 0, 90 and 180 of shoulder flexion and isokinetic peak Shoulder musculature and average concentric moments of the shoulder rotators and abductors and the elbow flexors at 60 sÀ1 were measured on both the dominant and non-dominant sides. Pearson correlation coefficients revealed statistically significant positive relationships between HG isometric strength and isokinetic moments of the shoulder external rotators (rZ0.40e0.54), the shoulder abductors (rZ0.42e0.71) and the elbow flexors (rZ0.45e0.66) regardless of hand dominance. The positive relationships between HG isometric strength and isokinetic strength of the shoulder stabilisers was probably attributed to mechanisms providing stability to the elbow and shoulder joints either by force transmission via myotendinous and myofascial path- ways or by ‘‘overflow’’ of muscular activity via neural circuits. The results of the present find- ings suggested that HG isometric strength can be used to monitor isokinetic strength of certain muscle groups contributing to the stability of the shoulder joint; however, HG strength may account only for approximately 16e50% of the variability in isokinetic strength of these muscle groups. ª 2008 Elsevier Ltd. All rights reserved. Introduction * Corresponding author at: Avlonos 41-104 43, Athens, Hellas, Strength improvement of the shoulder’s muscles is a major Greece. Tel.: þ30 3201 51122344. goal of every rehabilitation exercise programme concerning functional restoration of the injured shoulder as it plays E-mail address: [email protected] (D. Mandalidis). a significant role in the stability of the shoulder joint. Constant monitoring of such changes enables clinicians to 1360-8592/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2008.05.001

20 D. Mandalidis, M. O’Brien modify the preventive/rehabilitation exercise programme informed consent form. Subjects with shoulder pathology whenever it is required and assist them, at least in part, in or present shoulder/neck pain were excluded from the decision making regarding patients’ return to previous study. The mean and standard deviation of age, body mass, athletic or working activities. The increased use of iso- and height were 25.7Æ4.8 years, 78.4Æ10.0 kg and kinetic dynamometers over the last few years has provided 1.8Æ0.1 m, respectively. Fifteen volunteers were right and more accurate and reliable information regarding the three were left handed. Appropriate ethical approval functional status of the shoulder musculature than other granted prior to the study. tools, such as manual muscle testing, strain gauges, and hand held dynamometers (Mandalidis et al., 2001a, b; HG strength measurements Mandalidis and O’Brien, 2001). However, factors such as (i) the low availability of isokinetic dynamometers probably HG strength measurements of both dominant and non- due to the high cost of these devices, (ii) the high level of dominant side were performed prior to isokinetic assess- proficiency, as far as the testing procedure is concerned, ment with each subject standing, the shoulder at 0, 90, which is required by the examiner, (iii) the high motivation and 180 of flexion (see Figure 1aec), and the elbow in full which is necessary by the examinee, and (iv) the fact that extension using a standard HG dynamometer (Lafayette isokinetic testing procedures, particularly those concerning Instruments, Lafayette IN, US). The arm was halfway the shoulder musculature, are time consuming, make iso- between internal and external rotation of the shoulder (the kinetic strength assessments of the shoulder impractical if palm was facing towards the body) and the elbow was not difficult. Furthermore, in order to ensure adequate extended. The opening of the handle was adjusted prior to stabilisation of the joint and to avoid articular range limi- testing so that the crease of the proximal interphalangeal tations and pain inhibition of the shoulder musculature, joints of the subject rested on top of the adjustable handle many authors have performed isokinetic assessment of the allowing 90 of flexion. shoulder several months after surgical repair (Walker et al., 1987; Walmsley and Hartsell, 1992; Leroux et al., 1994). Each volunteer was instructed to perform one sub- These restrictions prevent postoperative evaluation of the maximum and one near maximum effort attempt for warm shoulder’s isokinetic strength for several months after up following by three attempts of maximum efforts, surgery. squeezing the handle of the instrument as hard as possible. Before each attempt the indicator needle of the instrument Hand-grip (HG) dynamometry, a relatively inexpensive, was set at the zero mark. Fifteen to twenty seconds rest simple and clinically useful method that has been mainly was allowed between each attempt. One-minute break was used in the evaluation of hand’s functional capacity, has the interval between testing of the dominant and non- also been utilised by several authors as a valid estimate of dominant hand. The best of the three maximum efforts was total upper limb strength (Bohannon, 1998; Adams et al., used in the statistical analysis. In order to avoid bias, hand 2004), as well as total body’s (Niebuhr and Marion, 1990), dominance and arm position (0, 90, and 180 of flexion) back extensors (Sinaki, 1989) and pulmonary muscles were randomly assigned. strength (Sahin et al., 2004). Isokinetic strength measurements Its validity in relation to the total upper limb strength has been determined based on the significant correlations Gravity corrected peak and average concentric moments of revealed between HG strength measurements and manual the dominant and non-dominant internal and external muscle testing of specific shoulder, elbow and hand muscles rotators and abductors of the shoulder as well as the flexors (Bohannon, 1998) and self-report of upper limb functional of the dominant and non-dominant elbow were measured assessments (Adams et al., 2004), in elderly people using a Kin Com II dynamometer (Chattecx Corporation receiving home care (Bohannon, 1998) and in patients Chattanooga, TN, USA) at 60 sÀ1. The internal and external suffering injuries (Goldman et al., 1991), chronic stroke rotators of the shoulder were measured between 0 (hori- (Boissy et al., 1999) and early rheumatoid arthritis (Adams zontal level) and 60 of external rotation with each subject et al., 2004). No such correlation has been established seated upright with the arm at 45 of abduction and 30 of between HG strength and isokinetic moment data of the horizontal adduction (scapular plane). The elbow was set at shoulder stabilisers. The purpose of this study was to 90 of flexion and the forearm was pronated (see Figure 2). investigate the relationship between HG isometric strength Isokinetic testing of the shoulder abductors was performed and isokinetic moment of the shoulder rotators, shoulder between 30 and 90 of shoulder elevation with each abductors and elbow flexors in a healthy male population. volunteer seated and set at an angle of 30 anterior to the frontal plane (scapular plane), so that the scapular plane Methods was parallel to the plane of rotation of the dynamometer’s actuator arm. The elbow was in extension and the forearm Subjects in pronation (see Figure 3). Isokinetic strength of elbow flexion was tested over a range of 60 (20e80 of elbow HG isometric strength and isokinetic concentric moments of flexion) with each volunteer seated and strapped onto the shoulder rotators, shoulder abductors and elbow flexors a chair with the arm placed at 90 of forward flexion and of both dominant and non-dominant upper extremity were the forearm midway between supination and pronation (see measured in a convenience sample of 18 collegiate-level Figure 4). Stabilisation of the subject’s trunk and body part male athletes. All individuals were invited and agreed to under investigation as well as proper alignment between participate in the study after they read and signed an

Relationship between hand-grip isometric strength and isokinetic moment data of the shoulder stabilisers 21 Figure 1 Subject’s positioning for hand-grip isometric strength assessment at 0 (a), 90 (b), and 180 (c) of shoulder flexion. the biological and motor centres of rotation was performed additional attempt was allowed. Five minutes break was according to the isokinetic testing procedures described in allowed between each side tested. To minimise learning previous studies (Mandalidis et al., 2001a, b; Mandalidis and effects and bias due to fatigue, the order of limb domi- O’Brien, 2001). nance (dominant, non-dominant side) and muscle group tested (shoulder internal and external rotators, shoulder A 10-min warm up which was consisted of 5-min arm abductors and elbow flexors) was randomly assigned. No cranking on an arm cycle ergometer at a low work load and verbal encouragement was given and no visual feedback 5-min of specific mobility and stretching exercises of the from the screen was allowed during isokinetic testing. All shoulders, performed by each subject before each testing isokinetic testing procedures were considered reliable session. Familiarisation with the isokinetic device was based on the ICC values calculated in previous studies achieved by allowing each subject to perform three sub- (Mandalidis et al., 2001a, b; Mandalidis and O’Brien, 2001). maximum concentric actions, with at least one of the action at/or near maximum effort. After 1-min rest, each Statistical analysis volunteer performed three maximum intermittent concen- tric actions (20-s interval between actions). An attempt was The relationship between HG tests and isokinetic deleted if the volunteer reported a lack of effort or if measurements was examined using the one-tailed Pearson a significant difference was observed on the screen between two consecutive torque curves. In this case, an

22 D. Mandalidis, M. O’Brien productemoment correlation coefficients (r). The coeffi- cient of determination was presented as a percentage of r2. Results Figure 2 Subject’s positioning for isokinetic strength Grip strength of the dominant hand was not statistically assessment of the internal and external rotators of the significant greater than grip strength of the non-dominant shoulder. hand measured at 0 (46.3Æ7.0 kg vs. 45.5Æ7.5 kg), 90 (42.7Æ6.3 kg vs. 41.7Æ6.6 kg), and 180 of shoulder flexion (47.0Æ6.5 kg vs. 42.9Æ6.6 kg). HG strength development was also greater when the shoulder was at 180 compared with 0 and 90 of shoulder flexion; however, the differ- ences were also not statistically significant either for the dominant or the non-dominant side. The mean and stan- dard deviation of isokinetic moment data of the shoulder internal and external rotators, shoulder abductors and elbow flexors of both dominant and non-dominant side are listed in Table 1. Pearson correlation coefficients between HG strength and isokinetic average and peak moments of the shoulder internal and external rotators, shoulder abductors and elbow flexors ranged between 0.06 and 0.62 (Table 2) for the dominant and between 0.28 and 0.71 for the non- dominant upper extremity (Table 3). However, HG strength was statistically significant related only with isokinetic moments of the shoulder external rotators, the shoulder abductors and the elbow flexors (see Tables 2 and 3 for levels of statistically significant relationships). In general, better relationships were obtained between HG strength and average moments compared with peak moments and between isometric HG strength and isokinetic strength of the shoulder abductors compared with shoulder external rotators and elbow flexors. The majority of the correlation coefficients were also greater when isokinetic moments were related with HG measured at 90 and 180 of shoulder flexion compared with 0, for both the dominant and non- dominant side. Discussion The findings of our study revealed a statistically significant positive correlation between HG force measured at 0, 90, and 180 of shoulder flexion and isokinetic average and Figure 3 Subject’s positioning for isokinetic strength Figure 4 Subject’s positioning for isokinetic strength assessment of the abductors of the shoulder. assessment of the elbow flexors.

Relationship between hand-grip isometric strength and isokinetic moment data of the shoulder stabilisers 23 Table 1 Means and standard deviations (in brackets) of order probably to position and stabilise the carpal, mid- isokinetic average and peak concentric moment (N m) of carpal and metacarpophalangial joints of the hand. dominant (D) and non-dominant (ND) shoulder internal and external rotators (SIR, SER), shoulder abductors (SAB), and The forces developed by these muscles during HG may elbow flexors (EF), nZ18. further be transmitted to the more proximally located areas both via myotendinous and myofascial pathways Average moment Peak moment providing stability to the elbow and shoulder joints. Force transmission via the myotendinous pathway is enabled by D ND D ND the proximal tendinous attachments of the long extensors and flexors of the wrist onto the humerus. It may also be SIR 35.1 (7.4) 33.2 (9.1) 39.2 (8.2) 36.8 (4.5) possible via myofascial sequences formed by muscle fibres SER 24.0 (4.9) 23.1 (4.3) 27.8 (5.8) 27.2 (10.2) that are inserted into the overlying fascia and septa, and SAB 54.2 (7.9) 52.2 (8.4) 59.3 (9.4) 57.2 (10.2) fascial components. According to the Fascia Manipulation EF 44.6 (10.1) 43.0 (6.9) 51.1 (11.9) 48.6 (7.5) model that proposed by Stecco (2004), there are six such sequences that are arranged in the upper limb according to peak moments of the shoulder external rotators (rZ0.40e the three spatial planes. Three of these myofascial 0.54), shoulder abductors (rZ0.42e0.71), and elbow sequences, the antemotion, the mediomotion and the flexors (rZ0.45e0.66) for the dominant and non-dominant intra-rotation sequence are located in the anterior portion side. Similar relationships have been reported in a previous of the upper limb and literally connect the palmar muscles study between dynamometer-measured HG and manually and fascia of the hand with those of the shoulder. The tested strength of the shoulder abductors (rZ0.57 and 0.72 antemotion myofascial sequence distally begins with the for the right- and left-hand side, respectively) and elbow muscles and fascia of the thenar eminence and continues flexors (rZ0.58 for the right- and rZ0.54 for the left-hand with the flexor carpi ulnaris muscle and the biceps brachii, side) in a group of 37 home-care patients with a mean age via the bicipital aponeurosis, which provides continuity of 78 years (Bohannon, 1998). between the anterior antebrachial and brachial fasciae. The brachial fascia, in turn, continues proximally with the The relationship between HG strength and isokinetic fascia of the pectoralis major. The mediomotion sequence moment of the shoulder stabilisers can partly be explained connects the hypothenar muscles, the flexor carpi ulnaris based on the mechanism by which an efficient action of the muscle, the medial intermuscular septum and coraco- muscles that act at a distal joint can be performed only brachialis muscle. The intra-rotation sequence begins with when the proximal joint or joints to it are also efficiently the lumbricals, which are inserted into the tendons of the stabilised by the surrounding musculature. HG is a dynamic flexor digitorum profundus, and continues with the task which requires a delicate and efficient action of the pronator teres, medial intermuscular septa and sub- majority of the extrinsic muscles of the hand and wrist scapularis. Recent anatomical studies, which revealed aided by some intrinsic muscles of the hand (Long et al., numerous tendinous insertions from the muscles of the arm 1970). The closure of the fingers around an object, such as and forearm onto the deep fascia of the upper limb, sup- the handle of the dynamometer that was used in our study, ported Stecco’s model (Stecco et al., 2007, 2009). Based on is performed by the long flexors of the fingers in association these findings the bicipital aponeurosis is merged with the with the extensors of the wrist (Smith et al., 1996; Johanson antebrachial fascia and the palmaris longus is inserted into et al., 1998). The contribution of the wrist extensors to the the palmar fascia and, via tendinous expansions, is con- development of isometric strength during HG, however, is of nected with the fascia overlying the thenar eminence great importance as these muscles stabilise the wrist in muscles. The pectoralis major fascia also is connected with extension opposing the action of the long finger flexors to the anterior brachial fascia (via a clavicular expansion), and move the joint into flexion (Smith et al., 1996). Johanson the medial brachial fascia and the medial intermascular et al. (1998) have reported that apart from a high EMG septum (via a costal expansion). Stecco et al. (2007, 2009) activity the long flexors of the fingers (flexor digitorum have also shown that at the posterior aspect of the upper supeficialis and profundus), the majority of the forearm limb the fascia of the latissimus dorsi is connected with the muscles (extensor carpi radialis longus and brevis, the triceps brachii fascia via a fibrous lamina. The triceps extensor carpi ulnaris, the extensor digitorum communis and tendon is inserted partially into the antebrachial fascia and the brachioradialis) were also active during power grip in the extensor carpi ulnaris sent a tendinous expansion to the Table 2 Pearson correlation coefficients (r) and percentages of the coefficients of determination (r2) between hand-grip strength (HG) at 0, 90, and 180 of shoulder flexion and concentric average and peak moment (in brackets) of the shoulder internal and external rotators (SIR, SER), shoulder abductors (SAB), and elbow flexors (EF) of the dominant side. SIR SER SAB EF R %r2 r %r2 r %r2 r %r2 HG-0 0.27 (0.21) 7.3 (4.4) 0.53b (0.43b) 28.1 (18.5) 0.59a (0.58b) 34.8 (33.6) 0.45b (0.47b) 20.3 (22.1) HG-90 0.29 (0.23) 8.4 (5.3) 0.54b (0.45b) 29.2 (20.3) 0.62a (0.62a) 38.4 (38.4) 0.51b (0.52b) 26.0 (27.0) HG-180 0.12 (0.06) 1.4 (0.4) 0.41b (0.36) 16.8 (13.0) 0.56a (0.55a) 31.4 (30.3) 13.7 (14.4) 0.37 (0.38) a p<0.01. b p< 0.05 (nZ18).

24 D. Mandalidis, M. O’Brien Table 3 Pearson correlation coefficients (r) and percentages of the coefficients of determination (r2) between hand-grip strength (HG) at 0, 90, and 180 of shoulder flexion and concentric average and peak moment (in brackets) of the shoulder internal and external rotators (SIR, SER), shoulder abductors (SAB), and elbow flexors (EF) of the non-dominant side. SIR SER SAB EF r %r2 r %r2 r %r2 r %r2 HG-0 0.29 (0.28) 8.4 (7.8) 0.45c (0.39) 20.3 (15.2) 0.50c (0.42c) 25.0 (17.6) 0.52c (0.56b) 27.0 (31.4) HG-90 0.34 (0.34) 11.6 (11.6) 10.9 (7.8) 0.61b (0.52c) 37.2 (27.0) 0.52c (0.54c) 27.0 (29.2) HG-180 0.37 (0.36) 13.7 (13.0) 0.33 (0.28) 16.0 (18.5) 0.71a (0.69a) 50.4 (47.6) 0.66b (0.64b) 43.6 (41.0) 0.40c (0.43c) ap<0.001. bp<0.01. cp<0.05 (nZ18). fascia of the hypothenar muscles). These anatomical find- during light manual precision work particularly when the ings confirmed also, at least in part, the Anatomy Trains shoulder and the elbow joints were in various degrees of Model introduced by Myers (1997b), by which force trans- flexion. This mechanism is probably mediated by neural mission may be possible via two myofascial chainsd‘‘train circuits located in the spinal cord and in cortical and lines’’dthat run in the front and the back of the upper limb subcortical areas of the central nervous system which connecting the proximal with its more distal regions. Myers’ increase the neural drive to agonists and synergists muscle concept (1997a) was based on the ontogeny and disposition in order to execute a specific task more effectively (Carroll of the fascial net during its embryonic development by et al., 2006). which is constituted one large and unified organ that allows constant transmission of tension and compression forces The better relationship between HG strength obtained throughout the human structure. at 90 and 180 compared with 0 of shoulder flexion and isokinetic moment of the shoulder external rotators, the Force may also be transmitted towards the proximally shoulder abductors and the elbow flexors was probably located regions of the upper limb, along a myofascial attributed to the dynamic state of the entire upper limb pathway that permits force transmission between the when HG was performed with the shoulder in elevation. sarcomeres and the endomysium (Huijing, 2003). Experi- Shoulder flexion requires the action of the primary movers mental animal studies have shown that force that originates (anterior part of the deltoid) as well as the contribution of (at least in part) from the endomysium, not from the the rotator cuff muscles (infraspinatus and teres minor) sarcomeres in series, as a result of its counteraction to the and the long head of the biceps brachii (shoulder abductor shortening of the active or passive sarcomeres, may be and elbow flexor) in order to provide dynamic stability to transferred within a muscle, either longitudinally or onto the joint (Sigholm et al., 1984; Kronberg et al., 1990; neighbouring fascicles and then to the epimysium (intra- Sakurai et al., 1998). The muscles around the elbow muscular force transmission) (Huijing, 2003). Force may should also contract in order to maintain the joint fully further be transmitted between the intramuscular extended. It has been proposed that muscle contraction connective tissue of a muscle and the extramuscular during flexion of the upper limb may stretch the continuity connective tissues, such as the fascia that constitutes the of the myofascial structures located anterior to it, boundaries within a compartment (extramuscular force allowing more effective force transmission via myofascial transmission), and between intramuscular layers of pathways (Stecco et al., 2007a, b), resulting probably in connective tissue of two adjacent muscles within a muscle a better relationship between HG and shoulder muscle’s group (intermuscular force transmission) (Huijing, 2003). strength. However, it should be noted that myofascial force trans- mission was explored in these experimental laboratory Correlation coefficients in this study were calculated studies under contractions of extremely lengthened between HG strength and both average and peak isokinetic muscles (Huijing and Baan, 2003; Maas et al., 2001) and moment of the shoulder stabilisers. Average moment in therefore should be distinguished from the force that may isokinetic dynamometry is a measure of muscle strength be transmitted along a myofascial pathway under maximum that is equal to the sum of moment values obtained at muscular contractions that performed by a living individual. every sampling point of a specific range of motion divided by the number of points. Peak moment, on the other hand Furthermore, the relationship between HG and iso- is the highest moment that is produced by a muscle group kinetic moment of the shoulder and elbow musculature may at a single point in this range. Many authors have recom- be explained based on the mechanism by which muscular mended the use of average instead of peak moment activity is ‘‘overflowed’’ to muscles beyond the essential because it is more reliable (Mandalidis et al., 2001a, b) and prime movers or synergists that take part in a particular it is less affected by artefacts usually occurring during movement. Carey et al. (1983) using the term muscular or isokinetic assessment of muscles operating on heavy excitation overflow to describe this mechanism reported segments like the shoulder joint (Dvir, 1995). The present that, among other muscles, the EMG activity of the biceps findings suggested that the average isokinetic moment of brachii was increased during precision HG. Sporrong et al. the shoulder stabilisers, not the peak, may be reflected (1995, 1996, 1998) have also shown an increased EMG more accurately with HG as the majority of correlation activity of the supraspinatus and infraspinatus during coefficients were greater when HG was related with intermittent 30e50% of maximum HG force production and average instead of peak isokinetic moment.

Relationship between hand-grip isometric strength and isokinetic moment data of the shoulder stabilisers 25 The lack of consistency regarding the positions used for However, HG isometric strength should be treated with HG and shoulder isokinetic testing was one of the limita- caution as an indicator of shoulder muscles’ strength as it tions of our study. HG strength has been measured by accounts only for a 16.0e29.2% of the variability in iso- several authors in a wide variety of positions in order to kinetic strength of the shoulder external rotators, a 17.6e assess whether its level can be influenced by the position of 50.4% of the variability in isokinetic strength of shoulder the individual under investigation and the position of the abductors and a 20.3e43.6% of the variability in isokinetic shoulder and elbow joints (Richards, 1997; Su et al., 1994; strength of the elbow flexors regardless of hand dominance. Balogun et al., 1991). In our study, HG was assessed in the standing position with the elbow fully extended as previous Conflict of interest studies revealed that isometric strength of the HG in this position was greater compared with either standing or All authors deny any conflicts of interest including personal, sitting position with the elbow at 90 of flexion (Balogun financial, or other related to the submitted manuscript et al., 1991). Isokinetic testing, particularly of the shoulder entitled ‘‘Relationship between HG isometric strength and rotators and abductors, has also been carried out in several isokinetic moment data of the shoulder stabilizers’’. different positions, revealing different moment outputs (Soderberg and Blaschak, 1987). Internal and external References rotation of the shoulder in the present study was assessed at 45 of abduction in the scapular plane. Shoulder Adams, J., Burridge, J., Mullee, M., Hammond, A., Cooper, C., abductors were also assessed in the scapular plane. Rath- 2004. Correlation between upper limb functional ability and bun and Macnab (1970) in a previous study have shown that structural hand impairment in an early rheumatoid population. when the shoulder is abducted the vascularity of the Clinical Rehabilitation 18, 405e413. supraspinatus tendon is enhanced. Anatomical and biome- chanical data have also revealed that when arm movements Balogun, J., Akomolafe, C., Amusa, L., 1991. Grip strength: effects are performed in the scapular plane the stresses placed on of testing posture and elbow position. Archives of Physical the anterior capsuloligamentous structures are lower, the Medicine and Rehabilitation 72, 280e283. impingement of the greater tuberosity under the acromion is prevented, better congruency between the articular Bohannon, R., 1998. Hand-grip dynamometry provides a valid surfaces of the glenohumeral joint is provided and optimum indication of upper extremity strength impairment in home care lengthetension relationship of the rotator cuff muscles is patients. Journal of Hand Therapy 11, 258e260. achieved (Johnston, 1937; Greenfield et al., 1990; Kuhlman et al., 1992). Furthermore, previous studies have shown Boissy, P., Bourbonnais, D., Carlotti, M., Gravel, D., Arsenault, B., good reliability of the isokinetic testing procedures of all 1999. Maximal grip force in chronic stroke subjects and its the muscle groups tested (Mandalidis et al., 2001a, b; relationship to global upper extremity function. Clinical Reha- Mandalidis and O’Brien, 2001). bilitation 13, 354e362. The range of HG isometric strength or isokinetic strength Carey, J., Allison, J., Mundale, M., 1983. Electromyographic study of the shoulder stabilisers may also have affected the of muscular overflow during precision handgrip. Physical correlation coefficients calculated in the present study and Therapy 63, 505e511. therefore the obtained relationships were limited to the study group. It is known that if a correlation exists between Carroll, T., Herbert, R., Munn, J., Lee, M., Gandevia, S., 2006. two variables, restricting the range of either of the vari- Contralateral effects of unilateral strength training: evidence ables will reduce the correlation (Pagano, 1994). The and possible mechanisms. Journal of Applied Physiology 101, participants in the present study were healthy males 1514e1522. between 17 and 35 years of age with a range of HG strength between 29.0 and 60.5 kg. This range was narrower Dvir, Z., 1995. Isokinetics. Muscle Testing Interpretations and compared with the range reported by Mathiowetz et al. Clinical Applications. Churchill Livingstone, Edinburgh. (1985) for a healthy male population 20e94 years of age (12.2e79.8 kg). Therefore, the participants in our study Goldman, S., Cahalan, T.D., An, K.N., 1991. The injured upper should be considered as a subgroup and although the extremity and the JAMAR five-handle position grip test. Amer- correlation coefficients calculated were considered ican Journal of Physical Medicine and Rehabilitation 70, 306e acceptable for biological studies (Pagano, 1994) they could 308. be greater if the study group was more representative of the general healthy male population. Greenfield, B., Donatelli, R., Wooden, M., Wilkes, J., 1990. Iso- kinetic evaluation of shoulder rotational strength between the Conclusions plane of scapula and the frontal plane. The American Journal of Sports Medicine 18, 124e128. The findings of the present study revealed a statistical positive correlation between HG isometric strength and Huijing, P., 2003. Muscular force transmission necessitates isokinetic moment produced by the shoulder external a multilevel integrative approach to the analysis of function of rotators and abductors and the elbow flexors. It is sug- skeletal muscle. Exercise and Sport Sciences Review 31, 167e gested that HG isometric strength measurements could be 175. used to monitor strength levels of the shoulder stabilisers. Huijing, P., Baan, G., 2003. Myofascial force transmission: muscle relative position and length determine agonist and synergist muscle force. Journal of Applied Physiology 94, 1092e1107. Johanson, M., James, M., Skinner, S., 1998. Forearm muscle acti- vation during power grip and release. The Journal of Hand Surgery 23A, 938e944. Johnston, T., 1937. The movements of the shoulder-joint. A plea for the use of the plane of the scapula as the plane of reference for movements occurring at the humero-scapular joint. The British Journal of Surgery 25, 252e260. Kronberg, M., Nemeth, G., Brostro¨m, L.-A., 1990. Muscle activity and co-ordination in the normal shoulder. An electromyographic study. Clinical Orthopaedics 257, 76e85.

26 D. Mandalidis, M. O’Brien Kuhlman, J., Iannotti, J., Kelly, M., Riegler, F., Gevaert, M., Sakurai, G., Ozaki, J., Tomita, Y., Nishimoto, K., Tamai, S., 1998. Ergin, T., 1992. Isokinetic and isometric measurement of Electromyographic analysis of shoulder joint function of the strength of external rotation and abduction of the shoulder. biceps brachii muscle during isometric contraction. Clinical Journal of Bone and Joint Surgery 749, 1320e1333. Orthopaedics 354, 123e131. Leroux, J., Codine, P., Thomas, E., Pocholle, M., Mailhe, D., Sigholm, G., Herberts, P., Almstro¨m, C., Kadefors, R., 1984. Elec- Blotman, F., 1994. Isokinetic evaluation of rotational strength in tromyographic analysis of shoulder muscle load. Journal of normal shoulder and shoulders with impingement syndrome. Orthopaedic Research 1, 379e386. Clinical Orthopaedics 304, 108e115. Sinaki, M., 1989. Relationship of muscle strength of back and upper Long II, C., Conrad, P., Hall, E., Furler, S., 1970. Intrinsiceextrinsic extremity with level of physical activity in healthy women. muscle control of the hand in power grip and precision handling: American Journal of Physical Medicine and Rehabilitation 68, an electromyographic study. Journal of Bone and Joint Surgery 134e138. 52, 853e867. Smith, L., Weiss, E., Lehmkuhl, L.D., 1996. Brunnstrom’s Clinical Maas, H., Baan, G., Huijing, P., 2001. Intermascular interaction via Kinesiology, fifth ed. F.A. Davis Company, Philadelphia. myofascial force transmission: effects of tibialis anterior and extensor digitorum longus length on force transmission from rat Soderberg, G., Blaschak, M., 1987. Shoulder internal and external extensor digitorum longus muscle. Journal of Biomechanics 34, rotation peak moment production through a velocity spectrum 927e940. in differing positions. Journal of Orthopaedic and Sports Phys- ical Therapy 8, 518e524. Mandalidis, D.G., O’ Brien, M., 2001. Isokinetic strength of the elbow flexors with the forearm in supination and in the neutral Sporrong, H., Palmerud, G., Herberts, P., 1995. Influences of position. Isokinetics and Exercise Science 9, 111e117. handgrip on shoulder muscle activity. European Journal of Applied Physiology 71, 485e492. Mandalidis, D., O’ Reagan, M., Donne, B., O’ Brien, M., 2001. Reliability of isokinetic shoulder rotation in the scapular plane. Sporrong, H., Palmerud, G., Herberts, P., 1996. Hand group Isokinetics and Exercise Science 9, 65e72. increases shoulder muscle activity. An EMG analysis with static handcontractions in 9 subjects. Acta Orthopaedica Scandinavica Mandalidis, D., O’ Reagan, M., Donne, B., O’ Brien, M., 2001. Effect 67, 485e490. of transient moment-oscillations on the reliability of isokinetic shoulder elevation in the scapular plane. Isokinetics and Exer- Sporrong, H., Palmerud, G., Kadefors, R., Herberts, P., 1998. The cise Science 9, 101e109. effect of light manual precision work on shoulder muscles e an EMG analysis. Journal of Electromyography and Kinesiology 8, Mathiowetz, V., Kashman, N., Volland, G., Weber, K., Dowe, M., 177e184. Rogers, S., 1985. Grip and pinch strength: normative data for adults. Archives of Physical Medicine and Rehabilitation 66, 69e72. Stecco, L., 2004. Fascial Manipulation for Musculoskeletal Pain. Piccin, Padova, pp. 123e130. Myers, T., 1997a. The anatomy trains. Journal of Bodywork and Movement Therapies 1, 134e145. Stecco, A., Macchi, V., Stecco, C., Porzionato, A., Day, J., Delmas, V., De Carlo, R., 2009. Anatomical study of myofascial Myers, T., 1997b. The anatomy trains: part 2. Journal of Bodywork continuity in the anterior region of the upper limb. Journal of and Movement Therapies 1, 91e101. Bodywork and Movement Therapies 13 (1), 53e62. Niebuhr, B., Marion, R., 1990. Voluntary control of submaximal grip Stecco, C., Gagey, O., Macchi, V., Porzionato, A., De Carlo, R., strength. American Journal of Physical Medicine and Rehabili- Aldegheri, R., Delmas, V., 2007. Tendinous muscular insertions tation 69, 96e101. onto the deep fascia of the upper limb. First part: anatomical study. Morphologie 91, 29e37. Pagano, R.R., 1994. Understanding Statistics in the Behavioural Sciences. West Publishing Company, Minneapolis, St. Paul. 149pp. Su, C.-Y., Lin, J.-H., Chien, T.-H., Cheng, K.-F., Sung, Y.-T., 1994. Grip strength in different positions of the elbow and shoulder. Rathbun, J., Macnab, I., 1970. The microvascular pattern of the Archives of Physical Medicine and Rehabilitation 75, 812e815. rotator cuff. The Journal of Bone and Joint Surgery 52B, 540e 553. Walker, S., Couch, W., Boester, G., Sprowl, D., 1987. Isokinetic strength of the shoulder after repair of a torn rotator cuff. Richards, L., 1997. Posture effects on grip strength. Archives of Journal of Bone and Joint Surgery 69A, 1041e1044. Physical Medicine and Rehabilitation 78, 1154e1156. Walmsley, R., Hartsell, H., 1992. Shoulder strength following Sahin, G., Ulubas, B., Calikoglu, M., Erdogan, C., 2004. Handgrip surgical rotator cuff repair: a comparative analysis using iso- strength, pulmonary function tests, and pulmonary muscle kinetic testing. Journal of Orthopaedic and Sports Physical strength in fibromyalgia syndrome: is there any relationship? Therapy 15, 215e222. South Medical Journal 97, 25e29.

Journal of Bodywork & Movement Therapies (2010) 14, 27e34 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt CLINICAL FASCIA RESEARCH Changes in the structure of collagen distribution in the skin caused by a manual technique Helga Pohl, Ph.D.* Centre for Bodytherapy, Tassiloweg 2, D 82319 Starnberg, Germany Received 15 March 2008; received in revised form 1 June 2008; accepted 3 June 2008 KEYWORDS Summary Objective: When treating patients with functional disorders using a special Functional disorders; manual technique, tissue changes can be felt by the therapist and the patient. This study Connective tissue; was conducted to objectively document these changes. Collagen dermis; Method: In the author’s practice for body therapy, 30 patients were measured with high- Manual treatment; frequency ultrasound (22 MHz) immediately before and after their first treatment in the area Ultrasound study; where they experienced pain or other discomfort and/or movement restriction. Treatment effects Results: Highly significant differences can be seen in the structure of the collagen matrix in the dermis before and after treatment. These changes reflect the differences in tension, softness and regularity, which can be palpated before and after treatment and are thought to be caused by changes in the mechanical forces of fibroblasts and increased microcirculation. ª 2008 Elsevier Ltd. All rights reserved. Background patients suffer from chronic pain and movement restriction and have several complaints. Their disturbances are often In the author’s practice for body therapy, patients who called functional, psychosomatic, neurogenic, or age- mainly have chronic complaints that are referred to by the related disorders. It is usually assumed that there is no author as sensory motor disorders are treated. This means bodily source for these complaints or it exists only in the patients have unpleasant sensations such as chronic pain, central nervous system, or that it involves degeneration of nausea, dizziness, anxiety, depression, numbness, tingling, the bones. globus pharynges, burning feet, etc. At the same time they have movement disorders such as restriction, stiffness, From the clinical experience of the author, these instability, cramps, sudden weakness, voice disorders, disorders originate from the surface of the body. There is speech disorders, restless legs, etc. The majority of these always a bodily basis of these disorders that can be found in the muscles, muscle fasciae (e.g. trigger points) or in * Tel.: þ49 8151 78171; fax: þ49 8151 3743. the connective tissue of the skin. These changes can be E-mail address: [email protected] palpated, and observed. Patients can pinpoint the site of their discomfort (for example anxiety at the front of their ribcage) although they may feel this discomfort as coming 1360-8592/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2008.06.001

28 H. Pohl from within. A movement restriction (in breathing, for Figure 1 Treatment of the connective tissue of the skin. example) can be seen where patients show their complaints, and more often than not also a dimpling in the Subjects skin. Structural changes in the muscles, in muscle fasciae, Thirty patients who came for body therapy, 16 females, 14 or in the skin can commonly be palpated at these males, 23e80 years old. The mean age was 52 years. locations. Criteria for inclusion The afflicted area was easy to measure and had no trigger Besides the changes in the area of their complaints, points or muscle tenderness under pressure. patients may have patterns of heightened tension in other Treatment surface areas throughout their body, of which they may be Patients were treated in the area of one of their completely unaware prior to treatment. complaints. Treatment lasted for 4e29 min with an average of 14 min. For treatment of all these bodily changes, Sensory Motor Method Bodytherapy was used, which consists of several hands-on A Minhorst Collagenosonª was used for measurement. This techniques, movement, and body awareness train- is a high-resolution high-frequency ultrasound (22 MHz), ingdcombined with treating heightened tonus in muscles and connective tissue, thereby changing sensory impres- sions, movement, and posture. One of the manual techniques employed is a treatment of the connective tissue of the skin using rolling and pressing movements of the fingers (Figure 1). This is not the common technique of skin rolling over large areas of skin. Rather, treatment is carried out in a very localised area, going back and forth in all directions with a pressing and squeezing of the connective tissue between the thumb and the index finger and then moving to the next small area only a few millimetres distant from the first one. After a short while, treatment is repeated at the first location. Via this method, the tissue is thoroughly worked through, 1 mm after the other (Figure 2). At the beginning of the treatment, the afflicted area of the skin feels hard, firm, and cold to the therapist. The skin is not easy to move in this area and it has an irreg- ular structure. Skin fold thickness is greater. For the patient it feels (very) painful when treating the afflicted area and the pressure used during treatment is perceived to be very strong. Patients regularly overestimate the power of the pressure. The same pressure used in a healthy area is estimated as being much less. In addi- tion, treatment is not painful in non-afflicted area- sdthere seems to be oversensitivity to pressure in the affected area. After a few minutes of treatment, the therapist feels the tissue becoming softer, warmer, and easier to move. A skin fold becomes thinner. This relaxation response is often very abrupt, from one moment to the next. At the moment of relaxation, patients report treatment to be less painful and sense the pressure as markedly reduced. Oversensitivity seems to have disappeared and this is when treatment at this site is considered complete. Immediately after treatment, the patient feels the treated area as warm and more alive and as having lost the previous pain or discomfort. In the treated area, a reddening can be seen and sometimes a little swelling without other signs of lasting oedema. There is also a larger range of movement along with more stability in movement in this area. Method Figure 2 Ultrasound measurement with Collagenosonâ. In a first step to objectively document the immediate changes caused by the treatment of the connective tissue of the skin, we conducted an ultrasound study.

Changes in the structure of collagen structure 29 Figure 3 Distribution of collagen in afflicted areas before treatment. reaching 4 mm beneath the surface. It makes visible objects a diagram AZamplitude scan. The latter shows the curve of the size of micrometres, that are dense enough to give an summed collagen density in the various layers of the skin. echo, in this case, bundles of collagen. For measurement, a probe is moved along a 2.5 cm path of the skin. The following characteristics of normal skin can be seen in all our subjects: Measurement On the left side of the pictures (in the B Scan), there is Patients were measured immediately before and after their always a thin dark band showing the high echo coming from first treatment in one of their afflicted areas. the epidermis. This is regularly the highest peak of the entire curve in the A Scan. On the right is the dermis, which Results always contains a lot of collagen, which is seen in the B pictures as the high density of dark points and in the A scans Figure 3 shows four examples of the measurements as an elevated curve with several peaks. The second high- obtained before treatment. est peaks of the graphs are often seen at the transition from the dermis to the subcutis. In the subcutis on the right side In general, the distribution of bundles of collagen fibrils of the pictures, the density of dark points is reduced; can always be seen as a picture BZbrightness scan and consequently, the curve is flatter and the peaks are less high. This reflects subcutis containing markedly less collagen. Figure 4 32-year-old female, pain felt in the inferior left Figure 5 56-year-old male, pain felt at the front of the right abdomen. shoulder.

30 H. Pohl Figure 6 64-year-old female, pain felt on top of the left Figure 8 57-year-old male, instability felt in the left calf. foot. For a summary of factors influencing thickness of the Besides these general traits, large differences between skin, see Krackowizer (2007). the different specimens can be seen. Due to these differencesdboth by individual and Thickness of the skin and quantity of collagen are known betweendthere were no baseline data available, no to depend on the following: matched control group could be established, and only pre/ post comparisons were possible.  area of the body (e.g. Smalls et al., 2006)  gender (e.g. Roberts et al., 1975) Individual pre/post comparisons  age (e.g. Roberts et al., 1975; Hall et al., 1981)  hormonal status (e.g. Quatresooz et al., 2006) Qualitative description of differences before and  diseases (e.g. Pitt et al., 1986; Collier et al., 1986) after treatment  medication (e.g. Cossmann and Welzel, 2006)  mechanical forces (e.g. Kligman and Takase, 1988) At first glance there is a high congruence of what can be  exposure to sun (e.g. Kligman and Takase, 1988) palpated and what the ultrasound pictures show. It looks as  dominance of an extremity (e.g. Smalls et al., 2006) it feels.  and many other factors. Figure 7 41-year-old female, nausea felt in superior Figure 9 60-year-old female, pain felt on the left tibialis abdomen. anterior.

Changes in the structure of collagen structure 31 Figure 10 37-year-old female, breathing restriction upper right abdomen. Epidermis does not seem to change with this sort of In the last example (Figure 10), an extreme densification treatment (B scan and A scan). in the papillary layer of the dermis (next to the epidermis) can be seen, which disappears over the course of Dermis: After treatment the dermis looks broader. In the treatment. B pictures, there seems to be more fluid within the dermis than before and collagen is more distributed in the Statistical analysis and results surrounding fluid. There is a more homogeneous structure with less high densities and a less-marked border to the For measuring the differences in collagen distribution subcutis. These differences are reflected in the A scans: before and after treatment, two measurements of the A after treatment the highest peaks in the dermis are lower scans (curves) were taken: height of the highest peak in the than before and the curves are smoother. dermis (second highest peak of the entire curve) and thickness of the skin. Both were measured before treat- Subcutis: Since the dermis seems to have broadened and ment (at T1) and after treatment (at T2). Measurement was the Collagenosonâ measures only at a depth of 4 mm, there done by an independent observer. is less of the subcutis seen after the treatment. Hence, direct comparisons are not possible. For the variable height of the highest peak in the dermis, the distance of the highest point in the dermis from the x- For a qualitative description of pre/post differences, see axis was measured. The difference of the mean height of Figures 4e9. the highest peak in the dermis before (MZ3.72, SDZ.68) and after (MZ2.59, SDZ.55) treatment was highly 5Mean 4,5 Figure 11 Example for the measurement of the height of highest peak in the dermis. 4 3,5 3 2,5 2 1,5 1 0,5 0 Highest Peak T1 Highest Peak T2 Figure 12 Mean difference in the height of the highest peak in the dermis before and after treatment.

32 H. Pohl Table 1 Differences in thickness of skin by gender and results of t-test for matched pairs. T1 T2 N t-ratio p< M SD M SD Female 2.01 .67 2.08 .66 16 .90 n.s. .01 Male 1.63 .79 1.99 .70 14 3.52 Table 2 Differences in thickness of skin by age separated for younger and older patients and results of t-test for matched pairs. T1 T2 N t-ratio p< Figure 13 Example of the measurement of the thickness of M SD M SD the skin. Younger significant (t-test for matched pairs, t-ratioZÀ8.03, DFZ29, p<.0001) (see Figures 11 and 12). Female 1.94 .62 2.20 .61 6 6.68 .01 No gender or age influence could be found for this Male 1.59 .74 1.93 .48 9 2.21 .10 variable. Older The variable thickness of the skin (epidermis and dermis, without subcutis) was determined as length of the x-axis, Female 2.05 .73 2.00 .70 10 À.50 n.s. where the first time the curve from the y-axis cuts the line .05 by 15% in height. The difference of the mean height of the Male 1.71 .94 2.40 1.06 5 4.14 highest peak in the dermis before (MZ1.83, SDZ.74) and after (MZ2.04, SDZ.67) treatment was highly significant Unfortunately, up to now we do not have a measure for (t-test for matched pairs, t-ratioZ3.06, DFZ29, p<.01) homogeneity to support this observation statistically. (see Figures 13 and 14). Treatment in non-afflicted areas There was a gender and age influence on the change of thickness of the skin. It was not significant for women, As mentioned earlier, we have no real control group, but in whereas the increase of thickness for men was highly 10 cases (six females, four males, mean age 49 years), the significant (see Table 1). same treatment was administered to patients in non- afflicted areas of their skin. There were no complaints in In a more fine-grained analysis, it turns out that only these areas, specimen tissue felt soft from the very younger female patients (up to 55 years) showed a highly beginning and the treatment was not painful. After treat- significant change (see Table 2 and e.g. Figures 4 and 7) ment, neither reddening nor swelling could be seen. No whereas older female patients remained unchanged. Male pre/post differences could be found when measuring patients showed the opposite finding with a highly signifi- cant change for older men and a statistical trend for younger men. Differences in homogeneity before and after treatment are likely, as can be seen from the graphs (B pictures). 5 4,5 4 3,5 Mean 3 2,5 2 1,5 1 0,5 0 Thickness T1 Thickness T2 Figure 14 Mean difference in the thickness of the skin Figure 15 43-year-old male, non-afflicted area (forearm). before and after treatment.

Changes in the structure of collagen structure 33 Figure 16 21-year-old female, non-afflicted area (neck). and Gnoth, 1991) that exerts active mechanical forces in patients in non-afflicted areasdeither in terms of height of many fascial tissues like subcutis (Langevin and Cornbrooks, the highest peak in the dermis nor in the thickness of the 2004; Langevin et al., 2005), muscle fascia (Schleip, 2006), skin (see Figures 15 and 16). and dermis (Fray et al., 1998; Grinnell et al., 2003). It has also been shown that fibroblasts of the skin are able to Conclusions contract collagen lattices and that their ability to contract can be changed by various substances (Coulomb et al., 1984; The changes seen in ultrasound scans seem to mirror the Adams and Priestley, 1988). differences in tension, density, and homogeneity in the skin, palpable immediately before and after treatment. It Before treatment there may be a contracture of the looks as it feels. fibroblasts in the skin, which can be felt as tension in the tissue and may be seen in the echoes of the ultrasound as The main differences in the distribution of collagen that somewhat higher densities of dermal collagen fibres. This were measured, showed: contraction may also exert a pressure on receptors in the skin, causing feelings of discomfort.  a reduction of the highest densifications in the dermis (found mainly in the transition zone with the subcutis); The manual treatment of the skin may influence mech- anoreceptors, causing processes leading to relaxation of  an increase in thickness of the dermis. the fibroblasts. To date, we can only speculate about the causes of the changes we could observe. It may be an interaction of two In the transition zone of the dermis to the subcutis and overlapping processes: relaxation of fibroblasts and in the papillary dermis, where we found the highest increase in microcirculation. densities of collagen (see Figure 17, right side), there are Fibroblasts are very numerous in the dermis. A count of also the plexus of blood vessels of the dermis (see dermal fibroblasts produced numbers between 2100 and Figure 17, left side, and Braverman, 2000)daround blood 4100 mmÀ3 in the midst of the dermis (Miller et al., 2003) and vessels there is the highest concentration of fibroblasts 1Â106 cells per cm2 in a 100 mm thickness of papillary dermis, (Randolph and Simon, 1998). Therefore, contracture of respectively (Randolph and Simon, 1998). Fibroblasts and fibroblasts may restrict microcirculation. Pericytes may their dendrites have been shown to form a network (Novotny also be involved in this process. Thus relaxation of fibroblasts may lead to:  widening of the skin,  reduction of the highest collagen densities,  facilitation of movement of interstitial fluid,  mechanical and chemical changes influencing blood and lymph vessels,  increased microcirculation and a thickening of the skin. Relaxation of fibroblasts and increased microcirculation may also lead to the clinical phenomena of:  less-restricted movement,  reddening and swelling,  sensations of softness and warmth,  feelings of wellbeing. Further research is needed to find out how long-lasting the immediate changes are and what their contributions to healing processes may be. Figure 17 Blood vessels in the skin (left) and collagen distribution (right).

34 H. Pohl Since every manual treatment directly or indirectly also Kligman, A.M., Takase, Y. (Eds.), 1988. Cutaneous Aging. University treats the skin, the connective tissue of the skin should of Tokyo Press, pp. 47e60. become a focus of scientific attention, especially in func- tional sensory motor disorders. Krackowizer, P., 2007. Sonographische Dickenmessung der Cutis. Grundlage fu¨r die sonographische Lympho¨demdiagnostik. Acknowledgments Dissertation Universita¨t Innsbruck. The author expresses her appreciation to all her patients Langevin, H.M., Cornbrooks, C.J., 2004. Fibroblasts form a body- making this work possible, to Prof. Dr. Erhard Mergenthaler wide cellular network. Histochemistry and Cell Biology 122, and Dr. Robert Schleip, both with the University of Ulm, 7e15. Germany, for their helpful comments and suggestions. Langevin, H.M., Bouffard, N.A., Badger, G.J., Iatridis, J.C., References Howe, A.K., 2005. Dynamic fibroblast cytoskeletal response to subcutaneous tissue stretch ex vivo and in vivo. Amer- Adams, L.W., Priestley, G.C., 1988. Contraction of collagen lattices ican Journal of PhysiologydCell Physiology 288, by skin fibroblasts: drug induced changes. Archives of Derma- C747eC756. tological Research 280, 114e118. Miller, C.C., Godeau, G., Lebreton-DeCoster, C., Desmoulie`re, A., Braverman, I.M., 2000. The cutaneous microcirculation. Journal of Pellat, B., Dubertret, L., Coulomb, B., 2003. Validation of Investigative Dermatology Symposium Proceedings 5, 3e9. a morphometric method for evaluating fibroblast numbers in normal and pathologic tissues. Experimental Dermatology 12 Collier, A., Matthews, D.M., Kellett, H.A., Clarke, B.F., (4), 403e411. Hunter, J.A., 1986. Change in skin thickness associated with cheiroarthropathy in insulin-dependent diabetes-mellitus. Novotny, G.E.K., Gnoth, C., 1991. Variability of fibroblast British Medical Journal 292 (6525), 936. morphology in vivo: a silver impregnation study on human digital dermis and subcutis. Journal of Anatomy 177, Cossmann, M., Welzel, J., 2006. Evaluation of the atrophogenic 195e207. potential of different glucocorticoids using optical coherence tomography, 20-MHz ultrasound and profilometry; a double- Pitt, P., O’Dowd, T.M., Brincat, M., Moniz, C.J., Studd, J.W.W., blind, placebo-controlled trial. British Journal of Dermatology Berry, H., 1986. Reduction of skin collagen with increased skin 155 (4), 700e706. thickness in postmenopausal women with rheumatoid arthritis. Rheumatology 25, 263e265. Coulomb, B., Dubertret, L., Bell, E., Touraine, R., 1984. The contractility of fibroblasts in a collagen lattice is reduced by Quatresooz, P., Pie´rard-Franchimont, C., Gaspard, U., corticosteroids. Journal of Investigative Dermatology 82, 341e344. Pie´rard, G.E., 2006. Skin climacteric aging and hormone replacement therapy. Journal of Cosmetic Dermatology 5 (1), Fray, T.R., Molloy, J.E., Armitage, M.P., Sparrow, J.C., 1998. 3e8. doi:10.1111/j.1473-2165.2006.00215.x. Quantification of single dermal fibroblast contraction. Tissue Engineering 4 (3), 281e291. Randolph, R.K., Simon, M., 1998. Dermal fibroblasts actively metabolize retinoic acid but not retinol. Journal of Investigative Grinnell, F., Ho, C.H., Tamariz, E., Lee, D.J., Skuta, G., 2003. Dermatology 111, 478e484. Dendritic fibroblasts in three-dimensional collagen matrices. MBC Online 14 (2), 384e395. Roberts, M.A., Andrews, G.R., Caird, F.I., 1975. Skinfold thickness on the dorsum of the hand in the elderly. Age and Ageing 4 (1), Hall, D.A., Blacket, A.D., Zajac, A.R., Switala, S., Airey, C.M., 8e15. 1981. Changes in skinfold thickness with increasing age. Age and Ageing 10 (1), 19e23. Schleip, R., 2006. Active fascial contractility. Implications for muscoskeletal mechanics. Dissertation, Faculty of Medicine, Ulm University, Ulm. Smalls, L.K., Wickett, R.R., Visscher, M.O., 2006. Effect of dermal thickness, tissue composition, and body site on skin biome- chanical properties. Skin Research and Technology 12 (1), 43e49.

Journal of Bodywork & Movement Therapies (2010) 14, 35e39 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt PATIENT CARE: CASE STUDY Glove use and the HIV positive massage therapy client Sarah Elizabeth Welch, HBOR/BA, RMT a,*, Judah Bunin, B.Sc.H., M.Sc., N.D. b a Time Out Massage Therapy, 1714 Rothesay Road, Saint John, New Brunswick E2H 2J4, Canada b Research Coordinator and instructor, Atlantic College of Therapeutic Massage, 440 King Street, Fredericton, New Brunswick E3B 5H8, Canada Received 29 July 2008; received in revised form 22 January 2009; accepted 27 January 2009 KEYWORDS Summary Massage therapy is often used to treat stress, and other symptoms of HIV/AIDS. HIV/AIDS; Massage therapy standards of practice require the use of gloves only when contact with blood Gloves; and bodily fluids is expected. Health care professionals often mistrust universal precautions Skin-to-skin; and use gloves when their use is not indicated, especially when dealing with HIV positive Touch; clients. This case report explored the effects of un-indicated glove use on stress levels, satis- Stress; faction with treatment, perception of the therapist, and perceived stigma during a massage Stigma; therapy treatment. In this case, gloved treatments were only 80% as effective at reducing Universal precautions stress as ungloved treatments. No difference was found in sense of stigma, perception of the therapist, or overall satisfaction in ungloved compared to gloved treatments. Suggestions for future considerations and additional research are made. ª 2009 Elsevier Ltd. All rights reserved. Introduction alternative health care (CAHC), and massage therapy in particular, are treatment modalities that individuals living UNAIDS (2007) currently estimates that there are more than with the HIV/AIDS often seek. Gore-Felton et al. (2003) 1.3 million people living with the human immunodeficiency found that 67% of people living with HIV/AIDS supplement virus/acquired immunodeficiency syndrome (HIV/AIDS) in their HIV-related medication use with CAHC. It was North America. Studies show that complementary and estimated in Gore-Felton et al’s (2003) study that 23% of those HIV positive individuals using CAHC chose to use * Corresponding author at: 145-3 Orange Street, Saint John, New massage therapy out of the many other CAHC modalities. Brunswick E2L 1M7, Canada. Tel.: þ1 506 635 7883; fax: þ1 506 635 With the increasing interest in, and availability of massage 0206. therapy, Baskwill et al. (2003) note that, ‘the likelihood, as massage therapists, that we will treat a client living with E-mail address: [email protected] (S.E. Welch). HIV (whether the client is aware, or not) is almost assured’ 1360-8592/$ - see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2009.01.008

36 S.E. Welch, J. Bunin (p. 13). Treating a client who is infected with HIV/AIDS is an importance of skin-to-skin contact in the treatment of HIV area in which all Registered Massage Therapists (RMTs) positive individuals, as described below. should be trained and comfortable. Low self-esteem is often attributed to feelings of guilt The issue of HIV/AIDS and stigma is not new. Numerous and shame, and a sense of being punished for past research studies have been conducted in an effort to behaviours. In fact, health care workers’ attitudes may understand the extreme sense of stigma that surrounds even reinforce the sense of worthlessness in patients this condition. Of particular interest are the studies that with AIDS. As a result, these patients often see them- highlight the perceived stigma HIV positive patients feel selves as repulsive to other people, which may interfere from health care professionals (HCPs). This sense of with successful coping. stigma is often associated with the practice of double gloving, lack of touch, and overuse of universal precau- Using latex gloves when dealing with blood and bodily tions. Fourn and Duci (1993) found that although nursing fluids is a necessity in caring for all patients. In situations students have an excellent understanding of the routes of not requiring protective clothing, however, patients HIV transmission, and use of universal precautions, 77.4% with AIDS should have as much skin-to-skin contact with were fearful of contracting the disease during their nurses as other patients. hospital training. That study also notes that the fear of HIV infection decreased as knowledge levels rose. McCann Patients with AIDS also have a special need for comfort and Sharkey (1998) note in a similar study that following touch in light of their many losses (p. 63). an educational intervention with nurses regarding HIV transmission routes, participants were more willing to The lack of cutaneous stimulation derived from skin-to- work with a colleague who was HIV positive, and to treat skin comfort touching compounds sensory deprivation patients who are HIV positive. These same participants, and magnifies the complexity of psychosocial problems however, would, ‘take additional precautions for fear of experienced by these patients (p. 64). contracting HIV in the workplace’ (p. 267). This brings to light the fact that HCPs are not confident that universal Chapman (2000) also highlights the importance of touch precautions will protect them from becoming infected and satisfaction of tactile hunger for HIV positive individ- with HIV. Chapman (2000) reports that the HIV positive uals. Rattray and Ludwig (2000) note that, ‘As HIV/AIDS patient may internalize the actions of HCPs who are not may be seen as an ‘‘untouchable’’ disease, touch itself can comfortable working with an HIV positive patient as be healing’ (p. 1037). It seems that research has completely stigma. One participant in Chapman’s (2000) study stated, neglected the patient’s feelings about HCP glove use, and ‘even though she’s a nurse.that’s with people who the effects of glove use on treatments. To the researchers’ understand all the facts e they still are frightened of it knowledge there has been no study conducted regarding even though they know’ (p. 846). the effects of glove use on massage therapy treatment outcomes. Current Canadian massage therapy standards of practice in Ontario indicate adherence to universal precautions to Exploration as to the potential negative effects of un- manage the transmission of infectious disease through indicated glove use is necessary. This case report was blood and bodily fluids (CMTO, 2006). The massage therapy designed to determine if un-indicated glove use during standards of practice follow the Infection Control Guide- a massage therapy treatment on an HIV positive participant lines for Regulated Professionals (COTO, 2008). These has a negative effect on stress levels, satisfaction with guidelines recommend that gloves are only to be used if treatment, client perception of the RMT, or perceived stigma. clients or practitioners have open wounds that will be contacted during the treatment. The use of universal Methods precautions assumes that every client or practitioner could potentially be carrying a blood-borne pathogen. For this Profile of participant reason, in theory, all clients are treated equally, regardless of actual health status. The following information was gathered from the client through an intake survey, a health history form, and Trends have been found in RMT student opinions through verbal communication with the therapist regarding the use of gloves with HIV positive clients. Ivany throughout the massage treatments. The participant is (2008) reports that some RMT students express strong a 45-year-old female who was diagnosed with HIV more attitudes regarding the use of gloves, even when they are than 25 years ago by a family physician via blood test. The not indicated, if a client has disclosed positive HIV status. medical prognosis after diagnosis was not disclosed. The Again, this highlights the fact that RMTs, like nurses, are client did not reveal any health history issues other than not confident that only using universal precautions when being HIV positive, and being under significant amounts of indicated by current standards of practice will protect stress due to family dynamics, employment issues, and them from HIV infection. financial constraints. In the years following the diagnosis, the client has never required the use of medication to Present research in the area of glove use is focused on manage the condition. The client holds strong belief that tactile discrepancy, frequency of puncture/tearing, and adverse dermatological effects on HCPs who use gloves on a regular basis (Tiefenthaler et al., 2006; Noonan and Moyle, 2004; Nayer and Gottrup, 2000). Some research supports the notion that touch is a particularly important aspect in HIV/AIDS care. Butts (1995) indicates the

Glove use and the HIV positive massage therapy client 37 CAHC is an effective way to manage HIV. Due to financial treatments to ungloved treatments, and the final survey restrictions, and frequency of relocation, the participant was used to derive subjective information from the has been unable to receive massage therapy treatments participant after completion of all six treatments. regularly in recent years. In the past, the participant had used massage therapy on a regular basis for stress Six 45-minute relaxation massages were performed to management. The client currently lacks a family doctor, the back, arms and neck of this HIV positive participant. and receives blood tests infrequently due to lack of symp- The massage treatments were consistent in type, duration toms, difficulty in finding a family physician, and lack of and series of techniques used. Massage techniques used interest in participating in conventional health care. included effleurage, muscle squeezing, prolonged tapote- ment, scooping, wringings, fist compressions, palm Massage therapy precautions for treating a client who is spreading, light circular kneading and occipital origin- HIV positive include, as with any client, the use of insertion technique. All techniques were performed in universal precautions. This recommends that the therapist a slow, predictable manner as to increase relaxation. These use protective barriers (i.e., gloves, masks), if either the relaxation techniques were chosen due to their effects on client or the therapist has an open wound, or there is stimulating the parasympathetic nervous system, and potential for the exchange of blood/HIV carrying bodily decreasing sympathetic nervous system firing, thus leading fluids between the therapist and the client. Other to increased levels of relaxation and decreased levels of precautions include not treating the client if they have stress (Rattray and Ludwig, 2000). The first treatment was a cold or flu, or if the therapist has a cold, flu, or other ungloved, with each additional treatment alternating infection that could have an effect on the client’s between being an ungloved or gloved treatment. Before immune-status (Rattray and Ludwig, 2000). The client’s each treatment the participant was asked to determine desired outcomes from the massage included reduction of their current stress level on a five point Likert scale. stress, time for self, opportunity for mindebody connec- Following each massage the client was asked to complete tion, and general enjoyment. a post-treatment survey that evaluated her satisfaction with the treatment, feelings towards the therapist, views Treatment plan on stigmatization, and stress level. The client was also invited to add comments at the end of the survey for any The case report was designed to treat an HIV positive client additional feedback. Following the series of massages the for stress using relaxation massages, and to evaluate the client completed the final survey. outcome of each massage to establish if there was any change in stress level, satisfaction with treatment, percep- Results tion of the therapist, or perceived stigma between the gloved and ungloved treatments. In order to derive measurable The final survey revealed that overall, the participant felt results from the study, treatments, outcome measurement less stressed than when she was not receiving the treat- and location were kept consistent. An intake survey, post- ments. She was aware of the treatments where the thera- treatment survey and final survey were used to gather data. pist wore gloves, and felt that the quality of the touch The intake survey gathered demographic information from changed with glove use. She reported that when HCPs use the participant, and evaluated the participant’s experience gloves to treat, there is a feeling of ‘coldness’. The with: massage therapy; glove use in massage therapy; and participant was satisfied and grateful for the massages, experiences of perceived stigmatization from health care regardless of glove use. The participant stated that she professionals. The post-treatment survey was derived from could have let the use of gloves during the treatments the Attitudes towards HIV Health Care Providers (AHHCP) affect her in a negative way, but she chose not to do so. She scale, with some additional questions regarding current stated that, ‘people who are still attached to their status stress levels and satisfaction with treatments. The questions would feel the ‘‘dirty syndrome’’. I don’t have the capacity used both positive and negative wording, and required the to allow myself that ugliness’. The participant also com- participant to answer on a five point Likert-style scale mented that, ‘if I had come in here in the first year of my ranging from do not agree to agree strongly. The AHHCP scale diagnosis, I would have felt it (the shame of being dirty due was created to evaluate an HIV positive person’s attitude to the therapist wearing gloves)’. She explained that the towards their medical team, and takes into account HIV- participant’s reaction to the therapist wearing gloves is related stigma. It has been reported to have an excellent dependant on their progression through accepting their HIV internal consistency of .92 (Bodenlos et al., 2004). status, and how they choose to deal with stigmatization. The final survey was subjectively structured and allowed Table 1 outlines descriptive data related to the changes the participant to respond to open-ended questions in stress level before and after each treatment. All treat- regarding their satisfaction with the series of massage ments yielded a decrease in stress level, with ungloved treatments, feelings related to gloves being used during treatment number 3 showing the most difference at 2.5 treatments, sense of stigmatization during the treatments, marks on the Likert scale. Changes in stress level with each and feelings towards the massage therapist, rather than the treatment ranged from 1 to 2.5 marks on the Likert scale. more quantitative structure of the post-treatment, Likert- When pre-treatment stress levels were 3.5, the ungloved style survey. In summary, the intake survey was used to treatment was more effective in decreasing the stress collect data on demographics and past experiences, the level. Effectiveness of the ungloved treatments ranged post-treatment survey was used to evaluate any change in from decreasing stress level by 1e2.5 marks. Gloved quantitative effectiveness of treatments from gloved treatment effects on stress ranged from 1 to 1.5 marks.

38 S.E. Welch, J. Bunin Table 1 Changes in stress level by treatment session. Change in stress level 1 Treatment number Pre-treatment stress level Post-treatment stress level 1.5 2.5 1 Ungloved 3 2 1 2 Gloved 3 1.5 1.5 3 Ungloved 3.5 1 1.5 4 Gloved 2.5 1.5 5 Ungloved 2.5 1 6 Gloved 3.5 2 The mean pre-treatment stress level for both gloved and positive clients, and to explore the notion that the ungloved treatments was 3. The mean post-treatment improper use of universal precautions may potentially stress level for gloved treatments was 1.67, while the mean have a negative effect on several different dependant post-treatment stress level for ungloved treatments was variables or outcomes. 1.33. Ungloved treatments, therefore, lead to a mean decrease in stress of 1.67. Gloved treatments resulted in One shortcoming in the design of this case report is that a mean decrease in stress level of 1.33. Figure 1 below the client turned in the completed surveys and reports to reflects these data. the same RMT who performed the massage, which might have biased the client’s responses. Thus, it is possible that Discussion the overall benefits of the treatments on stress levels may be somewhat exaggerated, but this does not explain the Outcomes of this case report show that relaxation apparent differences observed between the gloved and massage may be an effective way to lower stress levels in ungloved treatments. individuals living with HIV/AIDS. In this case report, gloved treatments were only 80% as effective as ungloved Future work on this subject should include a third party treatment in reducing one participant’s stress levels. For researcher collecting the data, with the RMT being kept some individuals, the use of gloves when not indicated unaware of their client’s remarks. Having this design, and may lead to negative internalization of stigma, and advising the client that the therapist will not know his/her feelings of contamination. These findings coincide with responses, is one method of reducing such demand the material previously published by Butts (1995). Some- characteristics. what unexpectedly, there was no change in overall satisfaction with treatments related to glove use in this Further research with larger participant numbers at study. various stages of acceptance of HIV status seems prudent. Research on the effectiveness of gloved treatments among It is important to note that, as a case study, the participants with other infectious diseases (i.e., Hepatitis applicability of these results to the population as C), and without infectious diseases will lead to a greater a whole is limited. The purpose of this case study is to understanding of the effects of un-indicated glove use on bring to light the topic of improper use of universal treatment outcomes and perceptions, satisfaction with precautions in massage therapy sessions with HIV treatments, and perceived stigma in massage therapy sessions. It is time for health care research to focus on the Mean Pre Treatment Stress Level effects of treatment-related variables on client perception Mean Post Treatment Stress Level and treatment outcomes. 3 Not only does un-indicated glove use have the potential to have negative effects on a client emotionally, and 2.5 psychosocially, but it appears that it may decrease the effectiveness of treatments. It is hoped that the next time 2 an RMT or HCP reaches for a pair of gloves when they are not indicated, that they will reconsider, and choose skin-to- 1.5 skin contact. 1 Acknowledgments 0.5 The authors wishe to extend thanks to instructors Nadine Currie-Jackson, Lisa Ivany, and Candace Gilmore for their 0 Gloved Treatment contribution to the study when it was in its early stages. Ungloved Treatment Special thanks to Dr. Julie McKeen for her enthusiasm, encouragement, and assistance in finding a participant. Figure 1 Mean pre- and post-treatment stress levels for Also special thanks to Dr. Janie Butts at the University of ungloved vs gloved treatments. Southern Mississippi for sharing her knowledge of gloves and touch as related to the HIV positive client. Dr. Matthew Alexander, thank you for your tireless editing skills, and expertise in English grammar.

Glove use and the HIV positive massage therapy client 39 References Gore-Felton, C., Vosvick, M., Power, R., Koopman, C., Ashton, E., Bachmann, M.H., Israelski, D., Spiegel, D., 2003. Alternative Baskwill, A., Kilty, M., Hamilton, A., Atkinson, H., 2003. Massage therapy therapies: a common practice among men and women and immune measures in HIVþ clients: a critique of the current living with HIV. Journal of the Association of Nurses in AIDS Care research. The Journal of Soft Tissue Manipulation 10 (4), 3e5. 14 (3), 17e27. Bodenlos, J.S., Grothe, K.B., Kendra, K., Whitehead, D., Ivany, L., 2008. Personal Communications. RMT Student Attitudes Copeland, A.L., Brantley, P.J., 2004. Attitudes towards HIV Towards Treating HIV Positive Clients. January 16, 2008. health care providers scale: development and validation. AIDS Patient Care and STDs 18 (12), 714e720. McCann, T.V., Sharkey, R.J., 1998. Educational intervention with international nurses and changes in knowledge, attitudes, and Butts, J.B., 1995. Transcending the latex barrier: the therapeutics willingness to provide care to patients with HIV/AIDS. Journal of of comfort touch in patients with acquired immunodeficiency Advanced Nursing 27 (2), 267e273. syndrome. Holistic Nursing Practice 10 (1), 61e67. Nayer, L., Gottrup, F., 2000. Incidence of glove perforations in Chapman, E., 2000. Conceptualisation of the body for people living gastrointestinal surgery and the protective effect of double with HIV: issues of touch and contamination. Sociology of gloves: a prospective, randomized controlled study. European Health & Illness 22 (6), 840e857. Journal of Surgery 166 (4), 293e296. CMTO College of Massage Therapists of Ontario, 2006. Communi- Noonan, M., Moyle, M., 2004. Latex glove allergy in health care. cation/Public Health Standard 5: Risk Identification and Australian Nursing Journal 12 (3), 1e3. Management for an Outbreak of Infectious Diseases Available online: http://www.cmto.com/pdfs/CPH%205.pdf Retrieved Rattray, F., Ludwig, L., 2000. Clinical Massage Therapy: Under- June 2, 2008. standing, Assessing, and Treating over 70 Conditions. Talus Incorporated, Elora, 1178 pp. COTO College of Occupational Therapists of Ontario, 2008. Infection Control for Regulated Health Professionals Available online: Tiefenthaler, W., Gimpl, S., Wechselberger, G., Benzer, A., http://www.coto.org/pdf/InfectionControlforRegulatedHealth- 2006. Touch sensitivity with sterile standard surgical gloves ProfessionalsFederationGuide.pdf Retrieved June 2, 2008. and single-use protective gloves. Anesthesia 61 (10), 959e961. Fourn, L., Duci, S., 1993. AIDS: knowledge and fear of contagion in nursing students during their hospital training. Medecine UNAIDS, 2007. 2007 AIDS Epidemic Update Available online: Tropicale 53 (3), 315e319. http://www.unaids.org/en/KnowledgeCentre/HIVData/ EpiUpdate/EpiUpdArchive/2007/ Retrieved June 2, 2008.

Journal of Bodywork & Movement Therapies (2010) 14, 40e49 available at www.sciencedirect.com journal homepage: www.elsevier.com/jbmt STATISTICAL MEASUREMENT OF FUNCTIONAL PATHOLOGY Range of normality versus range of motion: A functional measure for the prevention and management of low back injury Serge A. Gracovetsky Concordia University, Montreal, Q.C, Canada Received 20 September 2008; received in revised form 26 January 2009; accepted 1 February 2009 KEYWORDS Summary The Range of Motion (ROM) is a popular measurement used in the determination of Low back pain; disability for low back pain subjects in spite of serious objections to its clinical usefulness. It is Range of motion; proposed to consider a different index called the Range of Normality (RON) which is defined to Range of normality; be the portion of the ROM that an injured subject is able to do quasi-normally. This permits Lordosis; a direct assessment of the return to work parameters and the restrictions that ought to be Intersegmental placed on activities. It also allows follow up since the RON is expected to fill up the ROM as mobility; the injured subject recovers from his injury. Skin markers; ª 2009 Elsevier Ltd. All rights reserved. Z score; Safe zone; Return to work Introduction outright malingerers. The pertinent medical issue in eval- uating dysfunction of the spine is to objectively quantify Back pain is the most costly ailment of working-age adults, the severity of the disability (Gracovetsky et al., 1997). with economic consequences exceeding $50 billion per annum in the USA alone (Deyo et al., 1991; Frymoyer and The Range of Motion (ROM) of the trunk is one of the Cats-Baril, 1991). The most prevalent of these are low back variables used by the clinician to rate the disability of the injuries incurred in the workplace (Fathallah et al., 1998); back-injured patient (Fitzgerald et al., 1983). Many approximately 2% of all workers sustain back injuries each believed that a correlation between the ROM for a subject year (Sontag, 1991). Snook (1978) has shown that only could be related to pathology (Mayer and Gatchel, 1988) a very small percentage of the workforce is classified as because the average ROM of a large population correlates with impairment of the spine. Subjects diagnosed with E-mail address: [email protected] stenosis, degenerative disk disorders or disk prolapse tend to reduce their ROM, whereas those with spondylolisthesis or non-specific low back pain show a trend of greater ROM than the normal population. 1360-8592/$ - see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbmt.2009.02.001

Range of normality versus range of motion 41 The ROM depends on such a large number of factors (Anderson et al., 1985). Thus, when the same patient repeats introducing variability and limiting reproducibility, that many the flexion while lifting a load, the point at which the insta- investigators have discounted its diagnostic importance (Lea bility appears on the radiographs changes. Segmental and Gerhardt, 1995; Samo et al., 1997; Mayer et al., 1997). stability is responsible for maintaining the equilibrium of ROM is under voluntary control and dependent upon the dynamic motion parameters within the ROM (Ogon et al., cooperation of the patient. It is not entirely objective. 1997). The measurement of segmental function is more informative than measuring ROM, which correlates poorly Nevertheless, ROM is readily observable and continues to with the radiographic ‘‘gold standard’’ (Samo et al., 1997). be a popular measurement used by the American Medical Association and the American Academy of Orthopedic Radiography is traditionally used as the ‘‘gold standard’’ Surgeons in determining impairment ratings (Venditti, for evaluating low back intersegmental function (Gian- 1991). ROM is a parameter recognized as being indicative of turco, 1944; Pearcy, 1985) even if low back dysfunction functional pathology in spite of the fact that the existing does not correlate well with either pain or radiological models of trunk motion characteristics and musculoskeletal imaging (Waddell, 1987) and that diagnostic radiology is of capacity still require formal validation (Marras et al., 1995) limited value in the first evaluation of the majority of spinal and that ROM has yet to be shown to be a good predictor of disorders (Spitzer et al., 1987). Since function cannot be job performance or injury occurrence (Venditti, 1991). deduced from radiology or pain, it must be directly measured. Yet, reported pain is still the single most The ROM of the trunk does not take into account the important factor in the clinical diagnosis of LBP (Graco- allocation of work between spine and pelvis and cannot vetsky et al., 1998). account for the distribution of motion between the different levels of the lumbar spine. For example, a patient During dynamic lifting, there are specific stresses such as suffering from facet joint stiffness at the lower lumbar torque, compression, and shear forces corresponding to the spine may produce a normal ROM by either increasing pelvic trunk flexion and load weight (Chaffin and Andersson, 1984; mobility (Mayer et al., 1984) or by increasing the mobility of Davis et al., 1998). Minimizing or limiting these stresses the thoraco lumbar junction. Thus, in spite of the presence requires finding that portion of the ROM within which it is of spinal dysfunction, the ROM appears normal. According safe to operate. to Horn et al. (1991), the reality is that ‘‘ROM in the lumbar spine cannot distinguish between normal individuals and By definition, movements executed outside of the indi- those with a true impairment due to pathological condi- vidual’s safe functional arc of motion increase the risk of tions.’’ There is not a single blind study with controls back injury. The part of the ROM during which spinal proving that the spinal impairment or disability of a subject coordination remains close to the normal average repre- can be derived with reasonable certainty from his/her ROM. sents a more appropriate measure defining safety and disability parameters. This region of ‘‘safe’’ functionality is ROM provides a non-specific index of spinal function and termed the ‘‘Range of Normality’’ (RON). measurements of dynamic motion are more informative than static ROM (McGregor et al., 1997). For example, consider This section introduces the concept of RON as a statis- the unstable patient who, in the erect stance, has a normal tical measure of normal function and defines its differences ROM and a near normal radiograph. As the patient flexes, with ROM. The practical applications of RON to the occu- there is a moment at which the radiograph reveals an pational risk management of low back disorders will now be abnormal spinal motion. The dynamic changes in the motion investigated. betray the presence of pathology and influence the clin- ician’s diagnosis far more than the magnitude of the ROM. Method There is a deeper mathematical reason for the failure of Subject selection ROM to account for a specific diagnostic. It is rooted in the statistical distribution of the ROM of the so-called normal and Subjects with mechanical low back dysfunction were abnormal populations. The variance of these populations is selected according to a very rigid inclusion/exclusion so large that it does overlap both averages (see Figure 1a, b). criteria (Gracovetsky et al., 1998). Even if the average normal is distinct from the average Subject evaluation abnormal, that finding is clinically useless because the clinician is not dealing with averages but with a single Each subjects’ performance was evaluated according to patient. Indeed, the clinical diagnosis is a reverse identifi- previously published methods (Gracovetsky et al., 1989; cation problem: given that the ROM of my patient is 63, Gracovetsky et al., 1995). In brief, kinematic data were what is the probability that this is due to an injury at L4/L5 obtained by tracking the motion of light-emitting diode There is no known answer to that question. markers placed on the skin over spinous processes and iliac crests of the lumbar spine, while the activity of the mul- This difficulty is exacerbated by the fact that 90% of LBP tifidus over the L5 level was recorded by surface electro- is non-specific (Spratt et al., 1990) and the argument myography (EMG). Patients underwent the standard becomes circular. If the clinician is unable to reliably evaluation protocol, performing sagittal flexion, recovery- distinguish between the normal and the abnormal, what is from-flexion, and lateral bending movements with incre- the credibility of any statistics collected of these pop- mental load increases. ulations? Regrettably, in many studies, a normal subject is arbitrarily defined to be an asymptomatic subject, thereby The kinematic data were used to estimate the spinal introducing a bias from the start. curve and spinal levels and several spinal coordination In addition, load places local strain on the ligaments, muscles and disc tissues of intervertebral segments

42 S.A. Gracovetsky Figure 1 a. The Gaussian approximation. The ROM of a given population can be approximated by a Gaussian distribution. The standard deviations away from the average are indicated. b. Normal and abnormal populations. The ROM distributions of the normal and abnormal populations overlap significantly. Hence, given the ROM of a patient, it is not possible to determine with a reasonable certainty whether that ROM belongs to the normal or to the abnormal population. The false positives AND negatives are simultaneously unacceptably high. This is why ROM alone cannot be useful in diagnosing LBP. parameters such as trunk flexion and the intersegmental amplitude of the total movement, regardless of the quality motion (EISM), in subjects with spinal pathology. Each of the movement, whereas the portion of the ROM that is subject’s parameters were compared to those of a matched executed close enough from the normal average corresponds normal population (age, sex, occupation) extracted from to the RON. a robust normative database. Consider the problem of determining the range of normality The normative database of the lordosis’ variation of a 30 years old male lifting 22 kg. The term ‘‘lordosis’’ used in the text is the estimated lumbo- A validated normative database of 40 normal subjects sacral angle derived from the correlation between radio- selected according to stringent inclusion/exclusion graphic studies and skin markers’ kinematics (Figure 3a). The criteria was previously generated (Gracovetsky et al., normative database will generate the reference against which 1995). This database is representative of sex, age (20e30, the subject performance will be assessed (Figure 3b). 30e40, 40e50 and 50þ) and load lifted. Using various parameters derived from external measurements of spine Although the normalized estimated lordosis’ variation as motion (including estimated intersegmental mobility the loaded trunk flexes is the physiological parameter (EISM), lumbar lordosis/pelvic range of motion ratio and chosen to illustrate the definition of the RON, there are EMG pattern), normal function was characterized with many other relevant parameters (Gracovetsky, 2009). The appropriate standard deviations. These particular physio- normal zone (Æ2 SD) around the average is the shaded area. logical parameters were chosen because their normal behavior turned out to be highly predictable and corre- Figure 3c demonstrates that the measured lordosis lated with pathology. remains within the shaded area for the first arc of motion of the trunk (zero to 50). Beyond 50, the normal unfolding of Statistical analysis the lordosis slows down for unknown reasons, but the restricted motion can be clearly detected. The conclusion from this experiment is that although the subject has a functional anomaly, his response is essentially normal It was assumed that the distribution of the population can be approximated by a Gaussian. Even if the distribution of the physiological parameters of the normal population is not truly Gaussian, over 95% of the subjects fall within the interval of Æ2 SD around the average (Newman et al., 1996). For every physiological parameter measured, there was a corresponding normative value with its variance and the corresponding probabilities of normality were calculated using the Z-descriptor (Newman et al., 1996). Definition of range of normality As illustrated in Figure 2, the RON is defined to be the part of Figure 2 Range of Normality versus Range of Motion. the ROM during which the patient’s spinal coordination remains close enough (Æ2 SD) from the normal average. By definition, the RON equals the ROM for the normal. The situation is quite different for the abnormal in which the ROM is greater than the RON. The ROM characterizes the

Range of normality versus range of motion 43 Figure 3 a. Definition of the estimated lordosis J and the true lumbosacral angle J*. The angle J is calculated from the skin markers while the angle J* is calculated from radiographs. The relationship J/J* is described in Gracovetsky (2009). The angle J is termed ‘‘lordosis’’ in the text. b. From the normative database. The statistics of the variation of the normalized estimated lordosis for a 22 kg load lift for a population of 30 years old male subjects is represented by its average and the Æ2 SD spread. By definition, to have a normal lordosis for a given angle of trunk flexion, a 30 years old male subject lifting 22 kg subject must control his lordosis to remain within the shaded area. c. Definition of RON. The subject measured response remains within the shaded area only when the trunk angle remains smaller than 50. That is the RON. For trunk flexion angles greater than 50, the spine is restricted and the response escapes the shaded normal zone. The maximum flexion angle is 70, which is the subject’s ROM. d. Definition of RON. The same data of Figure 3b is represented as function of the distance to normality expressed in term of Z score. The normal range is now a rectangle with a height of Æ2 SD. The lordosis remains within the normal range for up to 50 of trunk flexion and is restricted. The advantage of this representation is that the range of normality is always a rectangle of fixed size. from zero to 50 when lifting 22 kg. That is a very specify pathology of various degrees, manifested as either a hypo- information with direct application in the workplace. mobile or a hypermobile joint. Only the pathologies cor- responding to a distance greater than Æ2 SD are relevant. Figure 3d contains the same data as Figure 3b but rep- resented as a function of the distance to normality The mechanics of assigning a probability of normality to expressed in Z score. a given response, depicted in Figure 4b, are straightfor- ward, requiring only a basic statistical manipulation of the The same analysis is repeated for each physiological trunk flexion data. For example, when the trunk flexes by parameter detected by the system (see Figure 10). 80, the distance between the patient’s response and the average normal was 1.5 times the standard deviation. This Results corresponded to a probability of abnormality of 6.7%. Accordingly, the patient’s probability of normality is: Recall that a variation of a physiological parameter is considered normal as long as its distance to the average (1 À 0.067) Z 93.3% remains within Æ2 SD, corresponding to a 95% probability of normality. The corresponding range of trunk flexion was By repeating this evaluation for each angle of trunk termed the Range of Normality (RON). flexion, the probability of normality for the entire ROM was determined. Figure 4a depicts a Gaussian interpretation of ROM of the spine, showing the probability of exceeding a distance Figure 5a further demonstrates this statistical treatment of one or two standard deviations from the mean or average of ROM data as a Gaussian distribution. Again, the proba- value of the distribution. In this theoretical model, a bility of normality is a function of the trunk flexion angle deviation from the mean ROM value is interpreted as

44 S.A. Gracovetsky Figure 4 a. ROM distribution. The Gaussian distribution approximates the distribution of ROM of the normal population. Pathology manifested as a hypomobile or a hypermobile joint causes a deviation from the mean value. b. Deviation from normality. To transform the patient’s response into a probability of normality, the distance between the response and the average normal must be determined. and the direction of the deviation from normal distin- the subject’s response and the mean value was considered guished between hypomobile and hypermobile joints. without distinguishing between hypermobility and hypo- mobility, the probabilities on each side of the mean were The important issue for return to work decision is not doubled. The probability of normality shows that the quality whether the patient is hypermobile or hypomobile but of motion was not consistent throughout the entire ROM, whether the patient is normal for the task. Hence, the factor supporting the tenet that spinal dysfunction is far more of interest is the absolute value of the distance to the mean complex than a simple deviation from the mean ROM value of and not its sign. When considering the distance from the the normal population. mean rather than specifying a hyper or hypomobile condition, the sum of the probabilities must be added. In this example, The relationship between RON and ROM is further high- for a distance from normality of 1.5 times the standard lighted in Figure 7a. Here, the subject’s probability of deviation, the probability that the subject was normal was normality for EISM of the L5/S1 joint began to decrease at 20 the sum of the areas on each side of the bell curve. In other of trunk flexion. It became unmistakably suggestive of words, the corresponding probability was doubled to account abnormal spinal coordination by 40 of flexion, where it fell for the symmetry of the Gaussian curve (Figure 5b). below 25% probability of normality. Furthermore, as shown in Figure 7b, the performance was affected by load. As the Figure 6a shows the normal variation of the estimated load lifted increases, the RON decreases, as demonstrated intersegmental mobility (EISM) of the lumbosacral (L5/S1) by the leftward shift in the probability function. This shows spinal articulation as the trunk flexes. This particular that spinal dysfunction reveals itself more clearly at higher parameter is clinically noteworthy since the L5/S1 joint is levels of biomechanical stress, and that a proper assessment subjected to a substantial amount of biomechanical stress of spinal function requires putting the spine under stress. during flexion and lifting (Anderson et al., 1985; Thompson, 1991) and is thus a common site of injury. Discussion When the response of an injured patient was compared to The Range of Motion (ROM) characterizes the amplitude of the normal (Figure 6b), it was apparent that the ‘‘quality’’ of the total movement and depends on the collaboration of the the motion throughout the ROM deviated substantially from subject. Although low back pain sufferers demonstrate that of the normal average. This injured response was a reduction in ROM in all planes of motion and often exhibit characteristic of a hypomobile joint, as the function fell below that of normal levels. Because the distance between Figure 5 a. Probability of normality. A distance of greater than À1.5 times the standard deviation occurs with a probability of 6.7%. The minus sign signifies that the motion is hypomobile. b. Gaussian symmetry. The probability that an abnormal response is at a distance of 1.5 times the standard deviation is twice the initial hypomobile value of 6.7, which is 13.4%.

Range of normality versus range of motion 45 Figure 6 a. Changes in the L5/S1 estimated intersegmental angle during trunk flexion. Average normal (n Z 40) unloaded measured effective intersegmental angle at L5/S1 during trunk flexion. The probability of normality is plotted against the angle of trunk flexion. b. Comparing the unloaded injured with the normal. The injured response is near normal up to 40 of trunk flexion, at which point it deviates from normality. slow, guarded motion (McGregor et al., 1997), very few patterns are useful physiological parameters in the objec- individuals are actually impaired to such an extent that their tive evaluation of spinal pathology (Newman et al., 1996). entire ROM is abnormal. An impaired subject is not neces- sarily a functionally disabled subject (Marriott et al., 1998). This study examined spinal coordination patterns to evaluate the quality of motion over the entire ROM (Gra- An injured subject may not be able to do heavy manual covetsky et al., 1989). This technology provided an accu- labor; however, the same subject could handle less rate fingerprint of the coordination of the unconstrained demanding tasks. The question that arises is ‘‘What is the spine during loaded or unloaded movements. The biome- appropriate level of disability for that person?’’ If the indi- chanical parameters of subjects with spinal pathology were vidual can find a desk job suited to his educational level, then measured and compared to those of a normative database he is clearly NOT disabled with respect to that particular (Gracovetsky et al., 1995). occupation. If the person does not have the skills to handle such a job, then he will be classified as disabled since he is The main feature of this normative database is the unable to perform in an occupation matched to his skills. surprisingly small variance for the physiological parameters Conversely, some individuals may have an abnormally small measured, which include EISM, spinal and pelvic contribu- ROM. Yet, within that restricted ROM, they may nevertheless tions to the task, and percentage elongation of the poste- be capable of performing their required duties at a near rior ligamentous system. This small variance indicates that normal level. This is why the pertinence of ROM measure- important physiological parameters, such as the rate of ments for the assessment of levels of dysfunction and decrease of lordosis during flexion, deviate little from one disability in low back injury remains controversial and normal individual to the next. Thus, coordination between limited in clinical practice (McGregor et al., 1997). the spinal segments to execute a motion is relatively independent of voluntary control and represents a desir- Dynamic motion characteristics differ between healthy able objective measure of spinal function. Pathology individuals and those with low back impairment (Ogon displays abnormal coordination patterns (Newman et al., et al., 1997). However, unlike ROM, the coordination 1996). Assessing the degree of a disability requires the between spine, pelvis and vertebrae is a characteristic of determination of the portion of the ROM that is executed at the human species that is relatively independent of volun- a normal or near normal level of spinal coordination for all tary controls (Gracovetsky et al., 1995). The inter-joint measured physiological parameters. That is the purpose of synergy of the spine is systematically coordinated by the Range of Normality (RON). a complex neural proprioceptive system (Ogon et al., 1997; Mitnitski et al., 1998). Lumbar and pelvic coordination In this context, the definition of normality is primarily functional; that is a subject is classified as normal if able to Figure 7 a. RON versus ROM. The quality of the motion during trunk flexion is not the same throughout the entire ROM. b. Effect of load. As the load lifted increases, the ROM and RON decrease.

46 S.A. Gracovetsky Figure 8 RON and safety. For a given level of normality, the 40. At the reduced level of normality of 10%, the trunk will relationship between load (arbitrary units) and trunk flexion be allowed to flex up to 70. The issue of determining what (angle) divides the plane into safe and unsafe zones. level of normality is desirable depends upon the acceptable level of risk for the specific worker and workplace setting. function within Æ2 SD of the matched control asymptom- To estimate risk levels, the occupational biomechanist must atic, even if he is clinically injured (Newman et al., 1996). assess both the requirements of the job and the capacity of the individual (Chaffin and Andersson, 1984). The relationship between the load and the RON permits to specify the maximum safe range of activity for an injured An industrial surveillance study demonstrated that worker. Figure 8 shows that the relationship between load and ‘‘there seems to be a threshold of sagittal flexion at which, if the RON for a given probability of normality divides the plane surpassed, simultaneous complex dynamic motions could be into two zones. Using a 25% level of normality as the criteria, it distinguished among risk groups’’ (Fathallah et al., 1998). is possible to measure the RON for a given load for a specific The RON measure is one tool to evaluate the complex kine- individual. The area below the 25% level of normality proba- matics of the spine in occupational risk reduction. bility curve corresponds to the safe zone within which the subject can assume any combination of loads and postures. Large compressive loads during the lifting of weight The performance of tasks outside of the functionally safe increase the probability of work-related injuries. Numerous range incurs an increased risk of low back injury. guidelines have been suggested in an attempt to diminish lumbar spine injuries (NIOSH, 1981). However, due to the For ergonomists and risk managers, the logical imple- number of variables involved, the rules regarding back mentation of this relationship between load and RON is the safety and lifting have been of a speculative nature and determination of safe loading configurations for workers may be overly conservative. The factors influencing safe injured or not. The load lifted may be small enough to lifting include load, lifting technique (Anderson and Chaf- maintain the worker in the safe zone. An injured worker may fin, 1984), lordosis and intersegmental mobility (Mitnitski still be employable for tasks involving low loads that he can et al., 1998), speed and acceleration, and the interplay of lift normally in spite of his injury (Gracovetsky et al., 1998). muscles and ligaments (Anderson et al., 1985). Instead of limiting the back safety decision-making process to From the measured characteristic response of a subject, isometric, isokinetic, and ROM measures, it is suggested it is possible to solve specific ergonomic problems. For that the ‘‘quality’’ of the motion also be considered is instance, Figure 9a shows the characteristic response of assessing work-related risk. a subject in a two dimensional plane with the load on the vertical axis and the RON on the horizontal axis. Suppose The ability to measure the RON under natural loaded a particular job requires a trunk flexion of 40 to bend over conditions permits a more precise approximation of work- a conveyor belt while holding a given load. For a level of place lifting requirements, and is relevant for those normality set at 25%, the maximum load that can be safely formulating back safety guidelines for occupational risk manipulated is ‘‘3’’ (arbitrary units). reduction. Indeed, since dysfunction may reveal itself only at higher levels of spinal stress, evaluating the range of The selection of a different level of normality will result normal function of the spine under conditions of loading in a different loading configuration (Figure 9b). For ought to be a better indicator of normality than that of example, lifting a load of ‘‘3’’ at a 25% level of normality unloaded function (Gracovetsky et al., 1998). demands that the subject avoids flexing his trunk more than Functional capacity testing and outcome measures for musculoskeletal disorders are assessed when returning the injured worker to the workplace or when determining their level of disability (Liebenson and Yoemans, 1997). RON may readily be employed to determine the type of movements that a worker is capable of performing quasi-normally, in spite of a pre-existing spinal dysfunction. By defining the functionally safe postural and loading parameters for a given individual and task, RON may be used to delineate Figure 9 a. Determining safety parameters. The characteristic curve of the individual determines the trunk flexion angle and maximum load parameters for a given probability of normality. b. Safety and risk levels. The safe zone within which a given task can be accomplished is larger as the probability of normality decreases however greater risk is incurred.

Range of normality versus range of motion 47 Figure 10 The variations in estimated intersegmental mobility of the five lumbar segments and T12/L1 when the trunk flexes under load is represented by their distances to normality in the Z score diagram (see the definition of Figure 3). In this example, the ROM is about 70. Note that the RON varies for each level. The minimum RON is obtained for L4/L5. By definition, the RON of that subject is the smallest RON that is about 15. From a clinical perspective, the L4/L5 joint is the limiting factor in the movement. safe work practice guidelines and reduce the incidence and other work-specific functions are also desirable parameters cost of low back injuries. to be included into RON estimates to enhance its applica- bility to occupational risk management. There is nothing The RON is a statistical measure used to ascertain the preventing the extension of the RON concept to models ‘‘quality’’ of a motion, with or without load, and is a func- other than that of low back dysfunction. tion of trunk flexion angle. Endurance, lateral bending and Figure 11 a. The RON of each level in Figure 10 is regrouped to show the minimum RON occurring at L4/L5 for a value of 15. That is the RON of the most restricted level and it sets the pattern of return to work restrictions. b. When the ROM is compared with each RON, the loss in normality becomes apparent (shaded area). This illustrates the direct impact of the pathology on the function of the spine.

48 S.A. Gracovetsky Each physiological parameter measured has its own local Frymoyer, J.W., Cats-Baril, W.L., 1991. An overview of the inci- RON (that is a RON for lordosis, for pelvic motion, for each dence and costs of low back pain. Orthop. Clin. North Am. 22, EISM etc.) (Figure 10). 263e271. The overall RON of the subject is defined to be the Gianturco, C., 1944. A roentgen analysis of the motion of lumbar smallest of all the measured RONs since it is the most vertebrae in normal individuals and in patients with low back restricted parameter that will set the upper limit of the pain. Am. J. Roentgenol. Rad. Ther. 52, 261e268. safe load (Figure 11). In particular, the actual reduction of normality for each level can be measured objectively. Gracovetsky, S., Kary, M., Levy, S., Ben Said, R., Pitchen, I., Knowing which physiological parameter is more restricted Helie, J., 1989. Analysis of spinal and muscular activity during represents a clue as to the underlying pathology. flexion/extension and free lifts. Spine 14, 327e331. Note that the system interprets all deviations from the Gracovetsky, S., Marriott, A., Richards, M., Newman, N., Asselin, S., normative data as being mechanical disorders. Since about 1997. The impact of inefficient clinical diagnosis on the cost of 10% of the LBP is non-mechanical, it is imperative that this managing low back pain. J. Healthc. Risk Manag. 17 (3), 21e31. test be complemented by a full clinical examination to rule out non-mechanical causes of low back pain. Gracovetsky, S., Newman, N., Pawlowsky, M., Lanzo, V., Davey, B., Robinson, L., 1995. A database for estimating normal spinal Conclusion motion derived from non-invasive measurements. Spine 20 (9), 1036e1046. The concept of Range of Normality and a methodology to measure it was compared to the familiar Range of Motion. Gracovetsky, S.A., Newman, N.M., Richards, M.P., Asselin, S., This simple statistical measure permits the quantitative Lanzo, V.F., Marriott, A., 1998. Evaluation of clinician and evaluation of the normal functional limits of the lumbar machine performance in the assessment of low back pain. Spine spine, with applicability to safe work practice guidelines, 23 (5), 568e575. conservative care and disability measurement. Gracovetsky, S.A., 2009. The Spinal Engine, pp. 103e239-ISBN: 978 Acknowledgements 1 4276 2997 5. The design of the machine used in the study was made Horn, T.J., Lowery, W.D., Jr., Wiesel, S.D., 1991. Impairment possible by the support of the National Research Council evaluation based on spinal range of motion. In: Presented at the of Canada (NRCC-IRAP program) and the Agence Que´- Annual Meeting of the International Society for the Study of the becoise de la Valorisation Industrielle de la Recherche Lumbar Spine, Heidelberg, Germany, pp. 73. (AQVIR). The normative database was calculated from the original data collected on 40 normal individuals as part of Lea, R.D., Gerhardt, J.J., 1995. Range of motion measurements. J. a research project supported by the Institut de Recherche Bone Joint Surg. 77A (5), 784e798. en Sante´ et Se´curite´ du Travail (IRSST), the research division of the Quebec Workers’ Compensation Board. The Liebenson, C., Yoemans, S., 1997. Outcomes assessment in work of Martha Cox is herby acknowledged together with musculoskeletal medicine. Man. Ther. 2 (2), 67e74. the collaboration of W. McIlwain M.D., E. Shapter M.D., K. Swan D.O. Marras, W.S., Parnianpour, M., Ferguson, S.A., Kim, J.-Y., Crowell, R.R., Bose, S., Simon, S.R., 1995. The classification of References anatomic- and symptom-based low back disorders using motion measure models. Spine 20 (23), 2531e2546. Anderson, C.K., Chaffin, D.B., Herrin, G.D., Matthews, L.S., 1985. A biomechanical model of the lumbosacral joint during lifting Marriott, A., Gracovetsky, S.A., Newman, N.M., Richards, M.P., activities.’’. J. Biomech. 18 (8), 571e584. Asselin, S., 1998. Improving clinical examinations by incorpo- ration objective measurements of functional limitations. Anderson, C.K., Chaffin, D.B., 1984. A biomechanical evaluation of Rehab. Manag. Dec/Jan, 70e74. five lifting techniques. In: Proceedings of the 1984 International Conference on Occupational Biomechanics, pp. 313e317. Mayer, T., Gatchel, R., 1988. Functional Restoration for Spinal Disorders: the Sports Medicine Approach. Lea & Febiger, Phila- Chaffin, D.B., Andersson, G.B.J., 1984. Mechanical work capacity delphia, PA. evaluation. In: Occupational Biomechanics. Wiley Interscience Publications, John Wiley & Sons, Toronto, pp. 78e110. Mayer, T.G., Kondraske, G., Brady Beals, S., Gatchel, R.J., 1997. Spinal range of motion: accuracy and sources of error with Davis, K.G., Marras, W.S., Waters, T.R., 1998. Evaluation of spinal inclinometric measurement. Spine 22 (17), 1976e1984. loading during lowering and lifting processes. Clin. Biomech. 13 (3), 141e152. Mayer, T.G., Tencer, A.F., Kristoferson, S., Mooney, V., 1984. Use of non-invasive techniques for quantification of spinal range-of- Deyo, R.A., Cherkin, D., Conrad, D., Volinn, E., 1991. Cost, motion in normal subjects and chronic low-back dysfunction controversy, crisis: low back pain and the health of the public. patients. Spine 9 (6), 588e595. Annu. Rev. Public Health 12, 141e156. McGregor, A.H., McCarthy, I.D., Dore, C.J., Hughes, S.P., Fathallah, F.A., Marras, W.S., Parniapour, M., 1998. The role of 1997. Quantitative assessment of the motion of the lumbar complex, simultaneous trunk motions in the risk of occupation- spine in the low back pain population and the effect of related low back disorders. Spine 23 (9), 1035e1042. different spinal pathologies on this motion. Eur. Spine J. 6, 308e315. Fitzgerald, G.K., Wynveen, K.J., Rheault, W., Rothschild, B., 1983. Objective assessment with establishment of normal values for Mitnitski, A.B., Yahia, L.H., Newman, N.M., Gracovetsky, S.A., lumbar spinal range of motion. Phys. Ther. 63 (11), 1776e1781. Feldman, A.G., 1998. Coordination between the lumbar spine lordosis and trunk angle during weight lifting. Clin. Biomech. 13 (2), 121e127. Newman, N., Gracovetsky, S.A., Itoi, M., Zucherman, J., Richards, M., Durand, P., Xeller, C., Carr, D., 1996. Can the computerized physical examination differentiate normal subjects from abnormal subjects with benign mechanical low back pain? Clin. Biomech. 11 (8), 466e473. NIOSH, 1981. Work Practice Guide for Manual Lifting. Department of Health and Human Services, National Institute for Occupational Health and Safety, Cincinnati, Ohio. Publication no. 81e122. Ogon, M., Bender, B.R., Hooper, D.M., Spratt, K.F., Goel, V.K., Wilder, D.G., Pope, M.H., 1997. A dynamic approach to spinal instability: part I: sensitization of intersegmental motion

Range of normality versus range of motion 49 profiles to motion direction and load condition by instability. Behavioral assessment of mechanical signs. Spine 15 (2), Spine 22 (24), 2841e2858. 96e102. Pearcy, M.J., 1985. Stereo radiography of lumbar spine motion. Thompson, D.A., 1991. Ergonomics. In: White, A.H., Anderson., R. Acta Orthop. Scand. 56 (212(Suppl.)). (Eds.), Conservative Care of Low Back Pain. Williams and Wil- Samo, D.G., Chen, S.-P.C., Crampton, A.R., Chen, E.H., kins, Baltimore, MD. Conrad, K.M., Egan, L., Mitton, J., 1997. Validity of three Venditti, P.P., 1991. Functional and work capacity evaluation. In: lumbar sagittal motion measurement methods: surface incli- White, A.H., Anderson, R. (Eds.), Conservative Care of Low Back nometers compared with radiographs. J. Occup. Environ. Med. Pain. Williams and Wilkins, Baltimore, MD. 39 (3), 209e216. Waddell, G., 1987. Clinical assessment of lumbar impairment. Clin Snook, S.H., 1978. The design of manual handling tasks. Ergonomics Orthop Relat. Res. 221, 110e120. 21 (12), 963e985. Sontag, M.J., 1991. Scientific basis of functional assessment of the Abbreviations lumbar spine patient. In: White, A.H., Anderson., R. (Eds.), Conservative Care of Low Back Pain. Williams and Wilkins, EMG: Electromyography; Baltimore, MD. EISM: Estimated Intersegmental mobility; Spitzer, W.O., Leblanc, F.E., Dupuis, M., 1987. Scientific approach ROM: Range of Motion; to the assessment and management of activity-related spinal RON: Range of Normality; disorders. Spine 12 (7 Suppl.), 1e59. SD: Standard deviation Spratt, K.F., Lehmann, T.R., Weinstein, J.N., Sayre, H.A., 1990 Feb. A new approach to the low-back physical examination.