Infection Management for Geriatrics in Long Term Care Facilities

34 Castle by level of impairment of specific components of immunity may help to advance our ability to improve host defense in an at-risk population (2). II. IMMUNOSENESCENCE: AGING CHANGES IN IMMUNITY EXCLUDING IMPACT OF DISEASE Immunosenescence is defined as the state of dysregulated immune function that contributes to the increased susceptibility of the elderly to infection and, possibly, to autoimmune disease and cancer (3). This perspective will focus on the rele- vance that age-related immune dysregulation has on susceptibility to infectious disease; however, there is growing interest in the pathogenetic role of a dysregu- lated immune system in common age-related illness such as atherosclerosis, Alzheimer’s dementia, diabetes mellitus, or osteoporosis. The immune system is arbitrarily divided into innate (natural) and adaptive (acquired) components, but recent advances in the field have focused attention on the interface between these two components. Extensive studies in inbred laboratory animals and in very healthy elderly humans have identified age-related changes in immunity, which have been essentially limited to phenotypic and functional changes in the T-cell component of adaptive immunity. In an attempt to standardize laboratory methods and isolate aging changes from external changes of disease and medications, studies over the past 15 years have included only the very healthy elderly. This has been accomplished by the exclusion of subjects with evidence of disease or use of medications, by applying rigorous criteria as defined by the SENIEUR Protocol (4). This concept of distin- guishing nature (genetic) versus nurture (environment) has long been debated and tends to distinguish the subtle differences in the interests of gerontologists (the study of aging) and geriatricians (the care of the aged). The SENIEUR Protocol criteria exclude subjects with unhealthy lifestyle choices; any clinical information that suggests the presence of infection, inflammation, malignancy, or other im- mune disorders; and any laboratory data that suggest abnormal organ function as well as anyone on medications for treatment of a defined disease. These stringent criteria exclude 90% of subjects aged 65 or older, 25% of younger subjects, and virtually 100% of the population in LTCFs (5–7). It would appear the original in- tent of the SENIEUR protocol was to develop a reference population, but it has been applied to exclude subjects with significant external/environmental expo- sure, which limits our understanding of mechanisms of vulnerability to infections in the at-risk population with underlying chronic diseases. Hence, despite exten- sive studies on possible mechanisms for age-related changes in T-cell phenotype and function in a very healthy population, no compelling scientific evidence has shown that these changes have direct relevance to the common infections seen in the aged population (4–7).

Age and Illness-Related Immune Dysfunction 35 III. IMPACT OF CHRONIC ILLNESS ON INFECTIONS SEEN IN THE ELDERLY Despite 90% complete involution of the thymus by age 40, true opportunistic in- fections are NOT seen among elderly patients, even those with significant chronic disease. This suggests that there is likely compensation for the lost activity of the thymus gland. Infections that are a problem in this population are well known to the clinician, that is, primarily bacterial infections (pneumonia, urinary tract, and skin and soft tissue) and some viral infections (reactivation of herpes zoster, and signif- icantly increased morbidity and mortality associated with influenza virus). In addi- tion, changes in immunity create difficulty in detecting both active (primary and re- activation) and inactive tuberculosis. Response to vaccination, which requires intact cell-mediated immunity to drive the humoral response, is clearly diminished in many different elderly populations as well as in laboratory animals (3,5,6). A. Impact of Age and Chronic Illness on Influenza Age-related changes in immunity likely have the most clinical relevance towards an impaired response to influenza infection and/or immunization to influenza. An estimated 90% of the 10,000 to 40,000 excess deaths attributed to influenza an- nually in the United States occur in persons 65 years of age or older. The Centers for Disease Control and Prevention Report on prevention and control of influenza states that when the antigenic match between vaccine and circulating virus is close, infection is prevented in 70% to 90% of subjects younger than 65, compared with only 30% to 40% in those 65 years of age or older (8). A past review on an- tibody response to influenza found that 10 studies identified a decline in antibody response in an aged population, 16 reported no change, and four showed an in- creased response. This variability is related to both differences between popula- tions and differences in defining a protective antibody response. Influenza vaccine efficacy in elderly persons is a complicated issue for a variety of reasons includ- ing the low attack rate and challenge in confirming actual influenza infection. There are also differences in defining an antibody response because older indi- viduals often have higher prevaccination antibody levels compared with younger individuals who have had less exposure to infection and fewer vaccinations. Even if antibody responses were intact, they may not provide the same level of protec- tion as in younger individuals. Thus, for example, in a study reporting on 72 vac- cinated elderly who later were confirmed to have influenza infection, 60% of these individuals had antibody titers of 1:40 or higher, and 31% had titers of 1:640 or higher 4 weeks after vaccination (9,10) (see Chapter 20). Not only is vaccine re- sponse less in this population, but even when vaccine response appears adequate, protection from infection is lower than in younger adults, which is likely related to the quality of the antibody produced in neutralizing viral pathogenesis. Never-

36 Castle theless, despite the low efficacy in prevention of infections, it needs to be empha- sized that vaccination in people aged 65 and older has been effective in reducing adverse events. In those 65 years of age or older, vaccination reduced the inci- dence of hospitalization or pneumonia 50% to 60%, and mortality was reduced by 80%. In a 3-year study on more than 75,000 community-dwelling el- derly, there was a 46% (range of 39% to 54%) reduction in all-cause mortality as- sociated with individuals who received influenza vaccination. Antibody response to vaccination (both magnitude and duration) is impaired in those aged 65 and older, but protective benefit to host defense likely occurs because cytotoxic T lym- phocyte activity towards both killing efficiency of viral infected cell and duration of activity has been reported to be intact in older subjects (8,9,11,12). Underlying chronic illness dramatically increases the risk of influenza in- fection and impairs the response to vaccination. The presence of one or two chronic illnesses (such as emphysema, diabetes mellitus, or chronic renal insuffi- ciency) is associated with a 40- to 150-fold increase in the basal incidence rate for influenza pneumonia (11). Whether chronic illnesses, medications, or other re- lated external conditions directly contribute to further compromise of immune competence has not been elucidated. One study on vaccine response in nursing home residents demonstrated only 50% of residents had an adequate response, based upon a definition of a fourfold increase in antibody titers. Furthermore, the response to vaccination did not correlate with nutritional status or dehy- droepiandosterone levels (13). Another study in a nursing home setting reported that only 36% of 137 vaccinated residents demonstrated a rise in antibody titer, and there was no correlation with age, body mass index, or functional status, as measured by the Barthel Index (14). B. Impact of Age and Chronic Illness on Pneumonia and Tuberculosis The risk and severity of pneumococcal pneumonia and tuberculosis increase with age. The incidence of pneumococcal infection is high in the first 2 years of life, then declines through adulthood, and finally increases dramatically in the geriatric population. Mortality is higher in elderly subjects and rises with advanced age, ap- proaching 80% in those older than age 85. Rates of bacteremia and meningitis from the pneumococcal infection are higher in the elderly. In fact, unlike all other age groups, mortality from pneumococcal pneumonia has actually increased in those older than age 75 since the antibiotic era (1950 vs 1985). Efficacy of the pneumococcal vaccine in preventing infection has been difficult to demonstrate in randomized control trials but has been reported to be 50% to 80% effective in case series. Five years after vaccination, the efficacy remains about 70% for those younger than age 75, but only 53% in subjects 75 to 85, and only 22% effective in those older than age 85 (15,16) (see Chapter 20).

Age and Illness-Related Immune Dysfunction 37 The overall case rate for tuberculosis declined 26% in the United States be- tween 1992 and 1997, with the highest number of cases reported in the 25 to 44 age group, which is a reflection of the human immunodeficiency virus (HIV) epidemic (see Chapter 15). Prior to this epidemic, tuberculosis case rates had an upward inflection point at age 75, due to both reactivation and primary cases of residents in institutional settings and community cases going undetected (11). The disease in the elderly remains largely distinct from tuberculosis in association with HIV infection, and the majority of the cases remain isoniazid sensitive. Differ- ences in presentation between young and older persons with tuberculosis include more subtle presentation (less pronounced cough, night sweats or X-ray findings), and skin testing is difficult to interpret due to both a waning of delayed hypersen- sitivity (false negative for inactive and active disease), but a more pronounced booster effect (false positive for “conversion”). Mouse studies on tuberculosis show increased susceptibility with minor shifts in the host response. Briefly, it ap- pears that in older animals there is a delayed recruitment of CD4ϩ T cells, with less interferon gamma production. Hence, the infection tends to disseminate more and eventual containment is less. Adoptive transfer studies show that transfer of young T cells into old animals reverses many of these changes (17). C. Changes in Immunity in Subjects with Herpes Zoster Infection The incidence of herpes zoster dramatically increases in individuals older than age 75 (see Chapter 17). Younger individuals who develop active zoster infections have an increased association of immunosuppressive illness, but outbreaks are not associated with occult malignancy in older individuals. Factors that control or pre- dict reactivation of latent infections are not known. Limited epidemiological stud- ies have shown that blacks have less risk of developing zoster than whites, but measures of stress were not significantly associated with herpes zoster. Risk fac- tors associated with the development of postherpetic neuralgia, such as the degree of immunological recall to the virus, have not been studied (18,19). No careful studies of immune changes in subjects who have had an outbreak of acute zoster or postherpetic neuralgia have been done; there are no studies to assess associa- tion of herpes zoster with subsequent development of bacterial, tubercular or in- fluenza infection, or response to vaccination. D. Risk Factors for Colonization with Resistant Bacteria Colonization of resistant bacteria in residents of LTCFs, including methicillin- resistant Staphylococcus aureus, vancomycin-resistant enterococci, aminoglyco- side-resistant enterococci, and multidrug-resistant gram-negative bacilli varies significantly from facility to facility (see Chapters 21–24). Colonization may be

38 Castle more common in LTCFs than in acute care settings, whereas infection from these organisms is less common. Epidemiological markers of risk of colonization and infection are generally any marker of end-stage illness and frailty, including prior acute hospitalization, length of stay in an LTCF, poor functional status, recurrent urinary bladder catheterization, urinary incontinence, pressure ulcers, and gas- trostomy tubes (20). Despite these markers of frailty, no studies have correlated colonization with resistant bacteria with specific changes in immunity, poor vaccine response, or risk of influenza or other bacterial infections. Hence, efforts toward prevention of colonization with resistant bacteria have included control/appropriate use of antibiotics and infection control measures of hand washing and appropriate isolation protocols. IV. AGE-RELATED CHANGES IN COMPONENTS OF IMMUNE RESPONSE A. Innate and Acquired Immunity The immune response consists of two interactive components, an innate (natural) and an acquired (adaptive) response. Innate immunity is less studied but has cel- lular components, that is, macrophages, polymorphonuclear (PMN) cells, natural killer (NK) cells, and dendritic cells (DCs); and noncellular components, which involve recognition molecules, such as C-reactive protein, serum amyloid protein, mannose-binding protein and the complement cascade. The noncellular compo- nents bind carbohydrate structures that do not occur in eukaryotes to help differ- entiate invading pathogens from self, which are then eradicated by the cellular component (3). Adaptive or acquired immunity has unique characteristics: (1) there is a specific response to a given antigen; (2) interaction of cells is required to activate either a cellular (cytotoxicity) and/or humoral (antibody) response; (3) memory is present, which enables more rapid response upon subsequent rechal- lenge of the same antigen; and (4) both the cell-mediated and humoral functions are dependent on T cells. To initiate an adaptive immune response, T cells must be activated by functional antigen-presenting cells (APC). The degree of interac- tion of both T cells and APC can influence the subsequent type, quality, and quan- tity of immune response. Hence, it is at this key interface between innate and ac- quired immunity that regulation of turning on or off of a response occurs (3,11,21,23). Interaction between the different immune cell types that constitute these components of host defense is carried out by the relative mix of cytokines, hor- mone-like proteins that act locally in directing the characteristics of an inflamma- tory response, and the ability of effector cells to differentiate or respond to specific signals, both of which are likely affected by aging and chronic illness. In general, activation of acquired immunity that involves a cell-mediated immune re-

Age and Illness-Related Immune Dysfunction 39 sponse and is protective against most infectious agents is described as a T helper 1 (Th1) response and is associated with high levels of the cytokines interleukin-2 (IL-2) and interferon-gamma (IFN-␥). In contrast, a T helper 2 (Th2) response, which is associated with allergic or parasitic infections but not associated with clearance of most bacterial or viral infections, is associated with high levels of IL- 10, IL-4, and IL-5 but low levels of IL-2 and IFN-␥. The relative concentrations of so-called proinflammatory cytokines, defined as those that upregulate a Th1 re- sponse (such as IL-12, IL-1, tumor necrosis factor alpha), or anti-inflammatory cytokines (important in turning off an inflammatory response such as IL-10, trans- forming growth factor beta, or IL-1 receptor antagonist) that are produced in local tissues, circulating APC including DCs, or by existing tissue macrophages, allow further refinement of the eventual outcome of an inflammatory response by influ- encing gene activation in effector immune cells (3,11,21–23). Mature DCs are required for efficient activation of influenza-specific cyto- toxic (memory CD8ϩ) T cells (24). Hence, the differentiation of regulatory APC at the site of inflammation is important in determining the quality of the subse- quent immune response, either due to the age-related changes in the cytokine mix at the site of inflammation or to cell-specific changes from aging or chronic dis- ease. Examples of differentiation-dependent measures of the efficiency of the in- teraction include the ability of T-cell binding to APC, IFN-␥ production, and the ability to generate influenza-specific cytotoxic T cells. Hence, adjuvants that may affect the ability of the DC to differentiate or mature should be considered to im- prove the suboptimal vaccine response seen in residents of LTCFs. B. Age-Related Changes in Acquired Immunity The overall impact of age on host immunity is thought to occur primarily along two mechanisms. The first is replicative senescence that may limit T-cell clonal expansion (the Hayflick phenomenon or loss of telomerase activity/telomere length, and may be more related to exposure to antigen than age). The second is developmental changes associated with involution of the thymus that precedes dysfunction of the T-cell component of adaptive immunity. Studies have shown a decrease in telomere length with age in T cells and B cells; recently, however, it was demonstrated that there was no significant change in telomerase activity upon stimulation of cells (25). This study may not have included individuals with re- peated exposure to antigen (perhaps in individuals with chronic illness), but it sug- gests that age-specific changes are more due to developmental changes in T cells. Several recent reviews have summarized extensive studies on changes in T-cell function with aging (26–29). The age-related decline in T-cell function is pre- ceded by involution of the thymus gland (cortex involutes much more than the medulla), with dramatic declines in thymic hormone levels, which has been de- scribed in both animals and humans. In addition, changes in bone marrow stem

40 Castle cells have also been described that are distinct from thymic changes. These changes are thought to result in a shift in the phenotype of circulating T cells, with a decrease in the number of naïve T cells (CD45RAϩ CD4ϩ), and a relative ac- cumulation of memory T cells (CD45ROϩ CD4ϩ). In addition, the memory cells that remain include a spectrum of normal functioning and hypofunctioning T cells in comparison with younger adult controls. The decrease in functioning cells re- sults in impaired proliferative capacity and IL-2 and IL-2 receptor expression. These changes can be traced to defective upstream postreceptor signaling at mul- tiple steps, including the phosphorylation of mitogen-activated protein kinases (29). At the same time, there appears to be a propensity towards a shift to a Th2 anti-inflammatory response, as evidenced by an increase in IL-10 production (3,11,30). A wide spectrum of findings have been reported, but the healthier the population, the less impact on changes in T-cell function can be identified. Age- related changes in T-cell cytokines other than IL-2 and IL-10 have demonstrated a much more varied response, especially IFN-␥ and IL-4, which may relate as much to different species (mouse studies differ markedly from humans) and the type of stimulation. Finally, despite the rather universal changes in T-cell response with age, the relevance is unclear. In vitro support of antibody response of T cells has been shown to be impaired with age (9), but impaired proliferative response, even to specific antigen, was not able to predict impaired antibody response to in- fluenza immunization (9). Studies on changes in CD8ϩ cells are much less nu- merous and have described some age changes, with impaired binding to targets, but once bound, killing capacity appears intact with aging (11,23). Changes in B cells are much less clear but appear to have some similarities to T-cell age-related changes. B cells from older individuals show impaired acti- vation and proliferation that also may be related to changes in costimulatory molecule expression (11,23). Both the primary and secondary antibody responses to vaccination have been impaired, with the degree of impairment being greater when T-cell involvement is required to drive the antibody response (usually re- lated to the complexity of the antigen). The specificity and efficacy of antibodies produced in older individuals is lower than in younger populations (11,27). C. Interaction Between Innate and Acquired Immunity with Aging Age-specific changes in immunity have been largely limited to the T-cell com- partment of acquired-immunity. Antigen-presenting cell function in healthy el- derly is intact, but infection rates are increased in chronically ill elderly. Thus, it is surmised that the impact of chronic illness on host immunity may be manifested by impaired efficiency of interaction of innate and acquired immunity. Because the innate component of immunity is critical to both the number of immunocompetent units, as well as the magnitude of the immunological burst upon

Age and Illness-Related Immune Dysfunction 41 activation, it very well could be the target of chronic illness in reducing immune competence beyond normal age-related changes (2). Evidence, for the most part, has suggested innate immunity remains intact or is upregulated with aging. The fre- quently reported nonspecific increase of proinflammatory substances produced by the innate immune system and downregulation of specific immunity may reflect a compensatory event by either component, with causality unclear (2,11,17). In SENIEUR Protocol Healthy elderly, larger numbers of DCs were gener- ated from circulating immune cells in comparison with younger adults, and DCs from elderly were effective in restoring proliferative capacity and in preventing the development of apoptosis (programmed cell death) in T cells grown to senescence (no longer able to proliferate) in culture (31,32). Likewise, antigen presentation ca- pacity of circulating APCs has actually been shown to be higher in community- dwelling elderly in comparison with younger adult controls, and was associated with higher IL-12 and IL-10 levels (30). Preliminary studies in a nursing home pop- ulation with chronic illness suggests a reversal in APC function, with impaired anti- gen presentation, impaired differentiation as manifested by reduced surface marker expression of major histocompatibility complex (MHC) class II and CD40 and no increased levels of the proinflammatory cytokine IL-12 (33, unpublished data). Hence, whether APC function, and DC in particular, is the specific target or the fi- nal common pathway of lost immune competence from chronic illness is unclear. The differentiation of DC has been identified as a key variable in the stimulation of effector T-cell function (IFN-␥ production and cytotoxic T-cell function), and will be considered an important target for immunotherapeutic adjuvants to improve antigen delivery and boost immunity in general (24). The production and interaction of cytokines produced by cells of innate im- munity are very complex. The timing and relative signal strength of these cy- tokines are crucial to the overall priming of the acquired immune response. Mul- tiple studies suggest that there is a nonspecific increase in production of proinflammatory proteins in the aged population. Animal studies show exquisite sensitivity in the old animals to bacterial endotoxins, with significantly more end- organ inflammation (15). Low-level, nonspecific autoimmunity throughout dif- ferent tissues may play a role in gradual loss of reserve capacity of a given organ system, a hallmark of aging, as well as a subsequent nonresponsiveness of immu- nity to infectious pathogens. Studies on age-related changes in proinflammatory cytokines show varied findings, most likely related to the very complex nature of response to cytokine networks, but most have shown an increase in stimulated production of IL-6, IL-8, and TNF-alpha, and a decrease in IL-1 (3,11,23,34). A recent review on IL-6 and aging describes 14 studies that report increases in IL-6 (34). Interleukin-6 itself has been shown to be inhibitory to TNF-mediated my- cobacteriostatic activities in macrophages (35). Chronic illness likely contributes to further dysregulation of control of im- mune response. A recent study compared IL-2 and IL-6 levels in young and el-

42 Castle derly healthy and “almost healthy” populations by including individuals who did not meet the SENIEUR protocol because of a lack of regular exercise and the use of medications for conditions such as hypertension or osteoarthritis. The almost healthy group demonstrated lower levels of IL-2 and higher levels of IL-6 in both young and old age groups, with the most pronounced changes in the elderly almost healthy population (4). Table 1 summarizes the components of the immune sys- tem, the impact of age and chronic illness on immunity, and the interaction be- tween innate and acquired immunity. Table 1 Summary of Immunity, Aging, Chronic Diseases, and Their Interactions Innate immunity Interface Acquired immunity Key elements NK cells- Surveillance & infected NK- Produce T cells- Provide long-term cell killing proinflammatory memory, direct lysis of Aging effects cytokines IL-12 infected cells, or rapidly PMN- Migration to sites of amplify response Impact of chronic infection; phagocytosis, local PMN- Cytokines from other production of disease inflammatory cells rescue PMNs; cells (GM-CSF, IL-2, proinflammatory production of ROS for killing LPS) prevent PMNs from cytokines (IL-2, IFN-␥). Abbreviations rapid cell death M/APC- Phagocytosis, killing of B cells- Production of organisms, regulation of M- Antigen presentation antibodies to antigen; cytokines, wound healing requires Ms, and signals require help from T cells, to T cells determine stimulation of M/APC’s NK cells- Incr cell number quality of immune compensates for decr in response T cells- Memory cells efficiency; low count is accumulated are much associated with 3ϫ incr mortality M-T cell: Many studies slower to divide, and show impared response produce less IL-2, more PMN- Tissue migration is intact in to PHA, likely due to IL-10. healthy elders decr in costimulatory molecule function; incr B cells- Produce more auto M- Less efficient tumor lysis in IL-10 (anti- Ab (to self); levels of Ab stimulation response to IFN-␥ inflammatory) occurs remain stable, but instead and is either a specificity for antigen, NK- Little studied cause or consequence of pathogen decr PMN- Tissue migration impaired in incr nonspecific inflammation with aging Further impairment due to chronic bronchitis, brittle DM; loss of stimulation of very high oxidants produced in Suggestion of more APC association with atherosclerosis impairment of interaction M- Shown to be inhibitory to killing with coexisting disease response to tuberculosis in lungs and little or enhanced NK- Natural killer cells response in very PMN- Neutrophil, healthy/elderly polymorphonuclear cell M- Macrophage/monocyte IFN- ␥ Interferon-gamma Ag- Antigen APC- Antigen presenting cells IL- Interleukin Ab- Antibody GM-CSF- Granulocyte-macrophage Decr/incr- ROS- Reactive oxygen colony-stimulating factor LPS- Lipopolysaccharide Decrease/increase species PHA- Phytohemagglutinin DM- Diabetes mellitus

Age and Illness-Related Immune Dysfunction 43 V. IMMUNE POTENTIATING EFFECTS OF MEDICATIONS AND DIETARY SUPPLEMENTS A. Modification of Immunity with Nutritional Treatments in Elderly with Nutritional Deficiencies Persons with protein energy undernutrition generally have associated micronutri- ent deficits. Specific nutrient deficiencies are also possible in individuals without gross undernutrition, but who may not have adequate resources in preparation of balanced meals. Elderly subjects who all met the SENIEUR Protocol but differed in having slightly low serum albumin values (3.0 to 3.5 g/dL) underwent a comparison of immune response. In comparison with those subjects with normal albumin values, the group with the lower albumin had significant decrease in pro- liferation to phytohemagglutinin, lower IL-2, and decrease in delayed-type hy- persensitivity testing. However, there was less of a difference to antibody re- sponse to influenza vaccine (36,37). Micronutrient deficits are associated with alteration in immune parameters, especially deficits in zinc, selenium, folic acid, vitamin B-6, and vitamin E. The impact of zinc deficiency on immunity has been extensively studied, as zinc is a cofactor in many postreceptor activation steps that are essential to cell prolifera- tion (see Chapter 5). Zinc deficiency has been found to be associated with im- paired peripheral blood T-cell counts, impaired T-cell proliferation, decreased IL- 2 production, and diminished CD8ϩ T-cell cytotoxicity. Zinc deficiency has been associated with progression of some disease that is thought to be due to a shift from a Th1 to a Th2 response, such as is found in leprosy, leishmania, schistoso- miasis, and acquired immunodeficiency syndrome (38–40). In vitro models have reproduced this clinical observation, as zinc-deficient subjects have reduced pro- duction of IL-2 and IFN-␥ but no change in Th2 cytokines of IL-4, IL-6, and IL- 10 (41). Of note, children with protein energy malnutrition have evidence of thymic atrophy and impaired T-cell differentiation, which is reversed by zinc sup- plementation and correlated with plasma zinc levels. In addition, it has been shown in several tissue culture conditions that zinc deficiency is associated with increased activation of apoptosis of both endothelial cells as well as T cells, which is reversed by zinc supplementation (42). Hence, the ability of local tissues and APCs in stimulating an appropriate immune response may be impaired in the pres- ence of zinc deficiency due to induction of apoptosis of APCs and/or effector T cells upon activation and less support of a subsequent Th1 immune response. B. Effect on Immunity of Dietary Supplements in Elderly Without Known Nutrient Deficiency Other studies have investigated the immunopotentiation of dietary supplements, even if no deficiency is identified. One study has reported a significant increase in

44 Castle CD4ϩDRϩ cells in a randomized placebo design with zinc supplementation, but a decline in total CD3ϩ and CD4ϩ cells with vitamin A supplementation (43). An- other study randomly assigned physiologic supplementation of vitamins (B6, A, C, E) and trace elements in 96 independently living elderly and evaluated changes in immune parameters and the incidence of infections over a 12-month period. In the intervention group, there were significant increases in percent CD3ϩ, CD4ϩ, and NK cells, with no change in CD8ϩ or B cells. There was a significant increase in proliferative response of T cells to phytohemagglutinin with increases in IL-2 and IL-2 receptor expression, as well as NK activity. There was no significant change in the response to influenza vaccine (39). The supplemented group had less num- ber of days of infection and less duration of use of antibiotics. Likewise, a study randomizing hemodialysis patients to 120 mg of zinc sulfate supplementation af- ter each dialysis showed a significant increase in the serum zinc levels, as well as an increase in the percentage of B cells and the antibody response to influenza vac- cination (44). Another group has reported an increase in delayed-type hypersensi- tivity testing at 12 months after supplementation with ascorbate, beta carotene, al- pha tocopherol, folate, retinol, riboflavin, copper, and zinc. There was a significant increase in serum levels of all but retinol, riboflavin, copper, and zinc (38). Topi- cal zinc has been associated with a boost in delayed-type hypersensitivity skin test- ing in elderly hospitalized patients (45). Finally, a recent randomized control trial comparing zinc acetate lozenges (a 12.8 mg zinc acetate lozenge every 2 to 3 hours while awake) with placebo in 50 ambulatory volunteers within 24 hours of devel- oping cold symptoms showed a significant reduction in duration and severity of cold symptoms but did not demonstrate any change in proinflammatory cytokine levels while on treatment with zinc (46). Vitamin E, an antioxidant, has been found to be deficient in 50% of the pop- ulation studies of healthy elderly in New Mexico (47–50). There is evidence that vitamin E inhibits the formation of prostaglandin E2 (PGE2). Prostaglandin E2 has been described to be elevated with age in humans and animal models and is asso- ciated with decreases in proliferative capacity and IL-2 production, and an increase in IL-10 production, all of which are thought to be associated with aging. Dose re- sponse improvements in immune function of old mice when given vitamin E sup- plementation, including delayed-type hypersensitivity, proliferative response and IL-2 with a decrease in PGE2 have been reported (51)). Similar results were ob- tained in healthy elderly humans supplemented with 800 IU of vitamin E for 30 days, with significant increases in delayed-type hypersensitivity and IL-2 and a trend toward increased proliferation of lymphocytes (48). Further studies on hu- mans demonstrated that low doses of vitamin E (60 mg/day) were able to increase delayed-type hypersensitivity response, but higher doses were needed (200–800 mg/day) to demonstrate an increase in antibody response to several vaccines (hep- atitis B, tetanus, and diphtheria) (48,49) and a 30% lower incidence in self-reported infections for a period of 235 days. In a murine influenza model, vitamin E sup-

Age and Illness-Related Immune Dysfunction 45 plementation showed significant reduction in lung influenza viral titers in old mice, with only modest improvement in young mice (50,51). However, epidemiological studies in elderly subjects have not found any correlation between immune pa- rameters (lymphocyte response, delayed-type hypersensitivity, serum antibody levels) with vitamin E intake (52). In human studies, supplementation of vitamin E has not shown a shift from Th2 to Th1 cytokine profiles after supplementation, although delayed-type hypersensitivity has repeatedly been demonstrated to be boosted with vitamin E, especially in the lowest responders (53). The effect of supplementation of vitamins A, C and E on cell-mediated im- munity in long-stay nursing home residents did demonstrate improved T-cell number, CD4/CD8 ratio, and T-cell proliferative response to phytohemagglutinin (54). Another study investigated the individual and combined effects of vitamin C (1 g) and E (400 mg alpha-tocopheryl acetate) on APC cytokines, proinflamma- tory cytokines, IL-1, IL-6 TNF-alpha, and PGE2 in response to lipopolysaccha- ride stimulation of peripheral blood mononuclear cells. The combination resulted in the higher increases in cytokines than either vitamin given alone and was also associated with a reduced PGE2 production (55). Other studies using 100 mg of alpha-tocopherol in 52 subjects aged 65 years and older did not demonstrate sig- nificant changes in antibody production or T-cell proliferative response (53,56). However, there may be subsets of elders who are more likely to benefit from vi- tamin E supplementation, especially low immune responders (53,55,57). C. Common Medications That May Have Significant Immunopotentiation In addition to obvious immunosuppressive medications, that is, corticosteroids, nonsteroidal anti-inflammatory agents (including cyclooxygenase type 2 in- hibitors), and antineoplastic agents, there are several common medications that demonstrate some evidence of immune potentiation. Studies involving chroni- cally ill individuals who are on multiple medications need to control for these medications in particular. The other approach would be to directly test if these medications have potential impact on boosting impaired immune competence in crossover designs. One class of agents that should be considered include beta adrenergic receptor antagonists (beta blockers) (58–60). Beta blockers have been shown to block the immunosuppressive effects of acute stress (58). One study on patients with dilated cardiomyopathy showed a decrease in the rate of anergy (from 70% down to 20%) with an increase in percentage of T cells and NK cells, and an increase in concanavalin A stimulation of IL-2 receptors, in comparison with those randomized to no beta blockers (61). Histamine has been found to af- fect immune response in a complex manner. Histamine antagonists, especially type 2 inhibitors, also modulate immune response and have been shown to boost proliferative response of T cells to IL-2, improve healing to herpes zoster, and

46 Castle boost delayed-type hypersensitivity response (62). Hormonal agents that appear to have an impact on immunity include sex hormones, growth hormone, and mege- strol acetate (63–65). Estrogens are likely to have a complex effect, as women tend to have more autoimmune disorders and diminished skin testing responses (64). Growth hormone levels diminish with aging, whereas supplementation of growth hormone was found to reverse involution of the thymus and was associ- ated with enhanced activation of immune systems, including a 50% increase in T- cell proliferative response and increases in IL2 receptor on T cells (65). Other con- founding medical conditions that may alter immunity include stress and depression, as well as any primary disease that affects primary organ function, such as failure of the heart, kidney, or liver. These conditions make it difficult to identify causation of impaired host defense; however, they likely manifest by a limited number of impaired potential T cell-antigen presenting cell functions. REFERENCES 1. Stevenson KB. Regional data set of infection rates for long-term care facilities: De- scription of a valuable benchmarking tool. Am J Infect Control 1999; 27:20–26. 2. Castle SC, Uyemura K, Makinodan T. The SENIEUR Protocol after 16 years: A need for a paradigm shift? Mech Ageing Dev 2001; 122:127–140. 3. Pawelec G. Immunosenescence: Impact in the young as well as the old? Mech Age- ing Dev 1999; 108:1–7. 4. Ligthart GJ, Corberand JX, Fournier C, Meinders AE, Knook DL, Hijmans W. Ad- mission criteria for immunogerontological studies in man: The SENIEUR Protocol. Mech Ageing Dev 1984; 28:47–55. 5. Mysliwski J, Bryl E, Foerster J, Mysliwski A. The upregulation of TNFa production is not a generalised phenomenon in the elderly between their sixth and seventh decades of life. Mech Ageing Dev 1999; 107:1–14. 6. Wick G, Grubeck-Loebenstein B. The aging immune system: Primary and secondary alterations of immune reactivity in the elderly. Exp Gerontol 1997; 32:401–413. 7. Rowe JW, Kahn RL. Successful Aging. New York, Pantheon, 1998. 8. Fukuda F, Bridges CB, Brammer TL, Izurieta HS, Cox NJ. Prevention and control of influenza: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1999; 48:1–23. 9. Bernstein E, Kaye D, Abrutyn E. Immune response to influenza vaccination in a large elderly population. Vaccine 1999; 17:82–94. 10. Gravenstein S, Drinka PJ, Duthie EH, Miller BA, Brown CS, Hensley M, Circo R, Langer E, Ershler WB. Efficacy of an influenza hemagglutinin-diptheria toxoid con- jugate vaccine in elderly nursing home subjects during an influenza outbreak. J Am Geriatr Soc 1994; 42:245–251. 11. Burns EA, Goodwin JS. Immunodeficiency of aging. Drugs Aging 1997; 11: 374–397. 12. Mullooly JP, Bennett MD, Hornbrook MC, Barker WH, Williams WW, Patriarca PA,

Age and Illness-Related Immune Dysfunction 47 Rhodes PH. Influenza vaccination programs for elderly persons: Cost-effectiveness in a health maintenance organization. Ann Intern Med 1994; 121:947–952. 13. Fulop T, Wagner JR, Khalil A, Weber J, Trottier L, Payette H. Relationship between response to influenza vaccination and the nutritional status in institutionalized elderly subjects. J Gerontol A Biol Sci Med Sci 1999; 54A:M59–M64. 14. Potter JM, O’Donnel B, Carman WF, Roberts MA, Stott DJ. Serological response to influenza vaccination and nutritional and functional status of patients in geriatric medical long term care. Age Ageing 1999; 28:141–145. 15. Butler JC, Spika JS, Nichol KL, Shapiro ED, Breinan RF. Effectiveness of pneumo- coccal vaccine. Lancet 1998; 351:1961–1962. 16. Hedlund JU, Kalin ME, Ortqvist A, Henrichsen J. Antibody response to pneumococ- cal vaccine in middle-aged and elderly patients recently treated for pneumonia. Arch Intern Med 1994; 154:1961–1965. 17. Orme I. Mechanisms underlying the increased susceptibility of aged mice to tuber- culosis. Nutr Rev 1995; 53(S4):S35–S40. 18. Schmader K. Postherpetic neuralgia in immunocompetent elderly people. Vaccine 1998; 16:1768–1770. 19. Schmader K, George LK, Burchett BM, Pieper CF. Racial and psychosocial risk fac- tors for herpes zoster in the elderly. J Infect Dis 1998; 178 Suppl 1:S67–S70. 20. Kaufman CA, Hedderwick SA, Bradley SF. Antibiotic resistance: Issues in long-term care. Infect Med 1999; 16:122–128. 21. Horan MA, Ashcroft GS. Ageing, defence mechanisms and the immune system. Age Ageing 1997; 26-S4:15–19. 22. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998; 392:245–252. 23. Hirokawa K. In: Pathy MSJ (ed). Principles and Practice of Geriatric Medicine, 3rd edition. New York, John Wiley & Sons Ltd., 1998:35–47. 24. Larsson M, Messmer D, Somersan S, Fonteneau JF, Donahoe SM, Lee M, Dunbar PR, Cerundolo V, Julkunen I, Nixon DF, Bhardwaj N. Requirement of mature den- dritic cells for efficient activation of influenza A-specific memory CD8ϩ T cells. J Immunol 2000; 165:1182–1190. 25. Son NH, Murray S, Yanovski J, Hodes RJ, Weng N. Lineage-specific telomere short- ening and unaltered capacity for telomerase expression in human T and B lympho- cytes with age. J Immunol 2000; 165:1191–1196. 26. Chakravarti B, Abraham GN. Review: Aging and T-cell-mediated immunity. Mech Ageing Dev 1999; 108:183–206. 27. Hodes RJ, Fauci AS (eds). Report of Task Force on Immunology and Aging. National Institutes of Aging and Allergy and Infectious Disease. US Department of Health and Human Services, March 1996. 28. Cossarizza A, Ortolani C, Monti D, Franceschi C. Cytometric analysis of immunose- nescence. Cytometry 1997; 297–313. 29. Miller RA. Cellular and biochemical changes in the aging mouse immune system. Nutr Rev 1995; 53:S8–S17. 30. Castle S, Uyemura K, Crawford W, Wong W, Klaustermeyer WB, Makinodan T. Age-related impaired proliferation of peripheral blood mononuclear cells is associ- ated with an increase in both IL-10 and IL-12. Exp Gerontol 1999; 34:243–252.

48 Castle 31. Steger MM, Maczek C, Grubeck-Loebenstein B. Morphologically intact dendritic cells can be derived from the peripheral blood of aged individuals. Clin Exp Immunol 1996; 105:544–550. 32. Steger MM, Maczek C, Grubeck-Loebenstein B. Peripheral blood dendritic cells reinduce proliferation in in vitro aged T cell populations. Mech Ageing Dev 1997; 93:125–130. 33. Castle S, Uyemura K, Wong W, Makinodan T. Evidence of enhanced type 2 response and impaired upregulation of type 1 response in frail elderly nursing home residents. Mech Ageing Dev 1997; 94:7–16. 34. Ershler WB, Keller ET. Age-associated increased interleukin-6 gene expression, late- life diseases, frailty. Annu Rev Med 2000; 51:245–270. 35. Bermudez LE, Wu M, Petrofsky M. Interleukin-6 antagonizes tumor necrosis factor- mediated mycobacteriostatic and mycobactericidal activities in macrophages. Infect Immun 1992; 60:4245–4252. 36. Lesourd B. Protein undernutrition as the major cause of decreased immune function in the elderly: Clinical and functional implications. Nutr Rev 1995; 53:S86–S94. 37. Lesourd B. Nutrition and immunity in the elderly: Modification of immune responses with nutritional treatments. Am J Clin Nutr 1997; 66:478S–484S. 38. Bogden JD. Studies on micronutrient supplements and immunity in older people. Nutr Rev 1995; 53(II):S59–S65. 39. Chandra R. Nutrition and immunity in the elderly: Clinical significance. Nutr Rev 1995; 53(II):S80–S85. 40. Sprietsma JE. Zinc-controlled Th1/Th2 switch significantly determines development of disease. Med Hypoth 1997; 49:1–14. 41. Prasad AS. Effects of zinc deficiency on Th1 and Th2 cytokine shifts. J Infect Dis 2000; 182:(Suppl) S62–S68. 42. Lepag LM, Giesbrecht JA, Taylor CG. Expression of T lymphocyte p56(lck), a zinc- finger signal transduction protein, is elevated by dietary zinc deficiency and diet re- striction in mice. J Nutr 1999; 129:620–627. 43. Fortes C, Forastiere F, Agabiti N, Fano V, Pacifici R, Virgili F, Piras G, Guidi L, Bar- toloni C, Tricerri A, Zuccaro P, Ebrahim S, Perucci CA. The effect of zinc and vita- min A supplementation on immune response in an older population. J Am Geriatr Soc 1998; 46:19–26. 44. Turk S, Bozfakioglu S, Ecder ST, Kahraman T, Gurel N, Erkoc R, Aysuna N, Turk- men A, Bekiroglu N, Ark E. Effects of zinc supplementation on the immune system and on antibody response to multivalent influenza vaccine in hemodialysis patients. Int J Artif Organs 1998; 21:274–278. 45. Kwok T, Fotherby MD, Cookson J, Potter JF, Castleden CM. Can topical zinc ac- centuate tuberculin reactivity in the elderly? Respir Med 1994; 88:47–48. 46. Prasad AS, Fitzgerald JT, Bao B, Beck FWJ, Chandrasekar PH. Duration of symp- toms and plasma cytokine levels in patients with the common cold treated with zinc acetate: A randomized, double-blind, placebo-controlled trial. Ann Intern Med 2000; 133:245–252. 47. Meydani SN. Vitamin E enhancement of T cell-mediated function in healthy elderly: Mechanisms of action. Nutr Rev 1995; 53:S52–S58.

Age and Illness-Related Immune Dysfunction 49 48. Meyandi SN, Beharka AA. Recent developments in vitamin E and immune response. Nutr Rev 1996; 56:S49–S58. 49. Meyandi SN, Meyandi M, Blumberg JB, Leka LS, Siber G, Loszewski R, Thompson C, Pedrosa MC, Diamond RD, Stollar BD. Vitamin E supplementation and in vivo immune response in healthy elderly subjects: A randomized controlled trial. JAMA 1997; 277:1380–1386. 50. Han SN, Meydani SN. Vitamin E and infectious diseases in the aged. Proc Nutr Soc 1999; 58:697–705. 51. Han SN, Meydani SN. Antioxidants, cytokines and influenza in aged mice and el- derly humans. J Infect Dis 2000; 182 Suppl 1:S74–S80. 52. Goodwin JS, Garry TJ. Relationship between megadoses of vitamin supplementation and immunological function in a healthy elderly population. J Clin Exper Immunol 1983; 51:647–653. 53. Pallast EG, Shouten EG, de Waart FG, Fonk HC, Doekes G, von Blomberg BM, Kok FJ. Effect of 50- and 100-mg vitamin E supplements on cellular immune function in noninstitutionalized elderly persons. Am J Clin Nutr 1999; 69:1273–1281. 54. Penn ND, Purkins L, Kelleher J, Heatley RV, Mascie-Taylor BH, Belfield PW. The effect of dietary supplementation with vitamins A, C and E on cell-mediated immune function in elderly long-stay patients: A randomized control trial. Age Ageing 1991; 20:169–174. 55. Jeng KC, Yang CS, Siu WY, Tsai YS, Liao WJ, Juo JS. Supplementation with vita- mins C and E enhances cytokine production by peripheral blood mononuclear cells in healthy adults. Am J Clin Nutr 1996; 64:960–965. 56. De Waart FG, Portengen L, Doekes G, Verwaal CJ, Kok FJ. Effect of 3 months vita- min E supplementation on indices of the cellular and humoral immune response in el- derly subjects. Br J Nutr 1997; 78:761–774. 57. Beharka AA, Wu D, Han SN, Meydani SN. Macrophage prostaglandin production contributes to the age-associated decrease in T cell function which is reversed by the dietary antioxidant vitamin E. Mech Ageing Dev 1997; 93:59–77. 58. Bachen EA, Manuck SB, Cohen S, Muldoon MF, Raible R, Herbert TB, Rabin BS. Adrenergic blockade ameliorates cellular immune responses to mental stress in hu- mans. Psychosom Med 1995; 57:366–372. 59. Hedberg A, Gerber JG, Nies AS, Wolfe BB, Molinoff PB. Effects of pindolol and propranolol on beta adrenergic receptors on human lymphocytes. J Pharmacol Exp Ther 1986; 239:117–123. 60. Feldman RD, Hunningshake GW, McArdle WL. Beta-adrenergic-receptor-mediated suppression of interleukin-2 receptors in human lymphocytes. J Immunol 1987; 139:3355–3359. 61. Maisel AS. Congestive heart failure/LVH: Beneficial effects of metoprolol treatment in congestive heart failure: Reversal of sympathetic-induced alterations in immune function. Circulation 1994; 90:1774–1780. 62. Komlos L, Notmann J, Arieli J, Hart J, Levinsky H, Halbrecht I, Sendovsky U. In vitro cell-mediated reaction in herpes zoster patients treated with cimetidine. Asian Pacif J Allergy Immunol 1994; 12:51–58. 63. Mantonvani G, Maccio A, Lai P, Massa E, Ghiani M, Santona MC. Cytokine in-

50 Castle volvement in cancer anorexia/cachexia: Role of megesterol acetate and medroxy progesterone acetate on cytokine down regulation and improvement in clinical symp- toms. Crit Rev Oncogenes 1998; 9:99–106. 64. Rook GAW, Hernandez-Pando R, Lightman SL. Hormones, peripherally activated prohormones and regulation of the Th1/Th2 balance. Immunol Today 1994; 15: 301–303. 65. Khorram O, Yeung M, Vu L, Yen SSC. Effects of growth hormone-releasing hor- mone (GHRH) (1-29)-NH2 administration on the immune system of aging men and women. J Clin Endocrinol Metab 1997; 82:3590–3596.

5 Nutrition and Infection Kevin P. High Wake Forest University School of Medicine, Winston-Salem, North Carolina I. INTRODUCTION Aging is associated with a decline in immune competence and an increased risk of infection (see Chapter 4). Nutritional factors have been shown to play a signifi- cant role in age-associated immune dysfunction (1–4), and the prevalence of mal- nutrition among older adults is greatest in residents of long-term care facilities (LTCFs) (5,6). Reversal of underlying nutritional deficits is an attractive and in- expensive option for reducing morbidity and mortality in elderly residents of LTCFs; however, there are few randomized, controlled trials of sufficient power to clearly define the utility of such interventions for clinical endpoints. Most frequently, surrogate markers of nutrition or immune function (i.e., reversal of previously documented vitamin deficiency, increases in serum albumin, vaccine responses, or delayed-type hypersensitivity reactions) have been the outcomes measured in clinical trials. With this limitation clearly stated, this chapter will re- view the prevalence, causes, methods of detection, and clinical relevance of mal- nutrition in residents of LTCFs, and provide evidence-based suggestions for prac- tical interventions to boost immune response and reduce the risk of infection in this at-risk population. II. PREVALENCE AND CAUSES OF MALNUTRITION IN OLDER RESIDENTS OF LTCFs Global malnutrition, a deficiency of protein and calories, is the most common form of malnutrition in older adults. Estimates of the prevalence of protein and 51

52 High caloric malnutrition is dependent on the variable used to define malnutrition. If one considers reduced daily intake to reflect malnutrition, the proportion of el- derly adults who are malnourished ranges from 2% to 33% in both healthy, free- living elderly adults and residents of LTCFs. However, using nutritional variables such as anthropometric measures or laboratory determinations (i.e., serum albu- min, total lymphocyte count), the estimated incidence of global malnutrition in healthy older adults is 3% versus 15% to 66% in institutionalized populations (6). Residents of LTCFs are at greater risk for global malnutrition for two basic reasons: reduced nutritional intake and increased metabolic demands. Reduced intake in LTCF residents is rarely due to poverty, in contrast to poor nutritional in- take in the community, because regulations require that meals meet specific nu- tritional standards. However, serving the meal does not guarantee that it will be consumed (Table 1). Obviously, many LTCF residents have significant disabili- ties that reduce their capacity to feed themselves or properly chew and swallow. Many residents may be depressed, have anorexia consequent to comorbid condi- tions or drugs, or other conditions that reduce the desire for food. Furthermore, the environment of LTCFs may not be conducive to caloric intake for some residents. Many elderly in the community maintain caloric intake through “grazing,” that is, eating small amounts throughout the day. Scheduled times for meals, a short du- ration of time to complete the meal, a lack of adequate staffing to feed all residents at meal time, and reduced preferences for “institutional” ways of preparing food all contribute to reduced caloric intake in residents of LTCFs (6,7). Finally, cog- Table 1 Barriers to Voluntary Nutrient Consumption in Older Residents of Long-Term Care Facilities Physical conditions Cachexia/anorexia of underlying disease Disability (inability to feed oneself) (e.g., cancer, infection) Medications (see Table 6) Poor dentition/swallowing Increased metabolic demands (e.g., Gastrointestinal disorders (e.g., peptic wound healing, renal disease) ulcer disease, gastroesophageal reflux, constipation) Metabolic disorders (poorly controlled Restrictive diets diabetes mellitus, thyroid disease) Cultural/psychosocial Bereavement Food preferences (based on religious Depression or cultural norms) Social isolation Restrictive meal times System barriers Inadequate staffing Lack of food between meals (inability of elderly to “graze”)

Nutrition and Infection 53 Table 2 Common Nutritional Deficiencies in Older, Long-Term Care Facility Residents Nutrient Prevalence Comment Protein/calories 17%–65% Manifested by wasting, low BMI, low Vitamin A 2%–20% serum albumin, lymphopenia Vitamin B6 (pyridoxine) 28%–49% Deficiency more common if measured by dietary intake or corneal cytology than Vitamin B12 0%–20% by serum levels Vitamin D 20%–48% Vitamin E 5%–40% Particularly important when LTCF Zinc 0%–21% residents placed on isoniazid for tuberculosis prophylaxis/therapy Atrophic gastritis is common in elderly Decreased sunlight exposure and dairy product intake Supplementation documented to improve some vaccine responses in older adults Zinc supplementation to speed wound healing probably only helpful in residents who are zinc deficient BMI, Body mass index; LTCF, Long-term care facility. Source: Refs. 1, 6. nitively impaired patients may not perceive hunger and thirst in the same way, thus further limiting their internal drive to consume protein and calories. For many of these same reasons, specific nutritional deficiencies also are more common in residents of LTCFs (Table 2). Vitamins A, B6, B12, D, E, and the trace elements zinc and selenium are most often found to be deficient in residents of LTCFs, with prevalence estimates of 40% to 50% for some micronutrients. Specific risk factors for micronutrient deficiencies in elderly LTCF residents in- clude reduced oral intake (vitamins A, B6, D, E, and zinc), increased metabolic re- quirements (e.g., zinc for wound healing), decreased exposure to sunlight (vitamin D), and a high prevalence of atrophic gastritis (vitamin B12). Like protein/calorie malnutrition, the prevalence of micronutrient deficiencies again varies widely with the techniques used to measure deficiency. For example, vitamin A, a fat-sol- uble vitamin that is stored in the liver, is essential for proper immune function and the integrity of skin and mucous membranes. A French study (8) showed that the prevalence of vitamin A deficiency in LTCF residents was 2%, 6%, 21%, or 55% depending on whether serum levels, corneal cytology, urinary excretion after a given oral load of vitamin A, or evaluation of oral intake, respectively, was used as the determinate of “deficiency.” Furthermore, there is significant debate as to what the “recommended” daily amount of vitamins or minerals should be in older

54 High adults (9). Recommended intakes have tended to focus on the average intake in large population studies rather than the optimal level of intake for the majority of the population. Obviously, these recommendations are usually handicapped by a lack of data as to what is optimal. Finally, recent data suggest that energy re- quirements predicted by frequently cited methods (such as the Harris-Benedict [H-B] equation) do not accurately reflect the metabolic needs of elderly subjects, overestimating metabolic needs in 20%, and underestimating metabolic need in 35% of LTCF residents (10). The accuracy of the H-B equation-predicted metabolic need is not improved by adding a commonly used “stress factor.” Thus, estimating metabolic or specific micronutrient needs in LTCF residents is difficult based on current techniques. Two excellent recent reviews (11,12) have examined the anorexia of aging, outlining the physiological and pathophysiological causes for reduced caloric intake in older adults. There are many instances in which undernutrition is a con- sequence of physiology, and thus cannot be modified by earlier recognition or in- tervention. Specific examples include incurable cancer, a competent patient’s re- fusal to eat or take supplements, endstage disease such as severe congestive heart failure, or chronic obstructive pulmonary disease. However, undernutrition is of- ten the result of a reversible cause, if recognized (Table 3). Specifically address- ing this problem in LTCFs, one study (7) identified 15 modifiable causes for un- dernutrition in LTCF residents (Table 4). Most of the suggested remedies could be set in place at minimal or no additional cost; others, like increasing or retraining Table 3 Mnemonic for Identifying Causes of Weight Loss in Older Long-Term Care Facility Residents Medications Emotional problems (depression) Anorexia tardive (nervosa); alcoholism Late-life paranoia Swallowing disorders Oral factors No money (insufficient funds in Medicaid facilities for palatable, individualized diets and consultant dietitian) Wandering and other dementia-related behavior Hyperthyroidism, hyperparathyroidism, hypoadrenalism Enteric problems (malabsorption) Eating problems (inability to feed oneself) Low-salt, low-cholesterol diets Social problems (ethnic food preferences, isolation, “disgusting” food habits of other residents) Source: Ref. 12, with permission.

Nutrition and Infection 55 Table 4 Reversible Causes of Malnutrition and Suggested Remedies Cause Method of identification Corrective action Staff unawareness Lack of documentation in Staff education chart by MD, RN, or RD Inappropriate use of Patient receiving a Replace by ad lib diet restricted diets restricted diet no longer indicated Discontinue or replace Use of drugs which impair offending drug desire or ability to eat Review of medications Provide assistance or Unmet need for eating Observation and calorie devices assistance or self-help count eating devices Retrain the nursing aide Observation Suboptimal technique of Improve the environment eating assistance Observation Increase prescription to 1.5 Suboptimal dining Ͻ 1.5 ϫ RDA of calories ϫ RDA calories and environment and protein prescribed protein Prescription of Weight and/or albumin Project MD will consult on maintenance instead of decline during illness; each NHCU patient repletion dietary intakes inadequate nutrition during intercurrent (oral or enteral) support illness Inadequate nutritional Daily temperatures reveals Identify and treat support during elevations infections intercurrent illness Clinical review Prescribed indicated Unrecognized febrile modified diet illness Prescribed tube-feeding volume not being Correct management of Unmet need for modified administered or complication diet absorbed Prompt dental care Inadequate management of Oral examination Consult speech pathology tube-feeding Clinical signs suggest complications for swallowing dysphagia; evaluation evaluation Poor dental status not requested Speech pathologist retrains Unmet need for dysphagia Recommendations of ward staff speech pathology not evaluation being followed Suboptimal treatment of dysphagia MD, Medical doctor; RN, Registered nurse; RD, Registered dietitian; RDA, Recommended daily allowance; NHCU, Nursing home care unit. Source: Ref. 7, with permission.

56 High staff, would require significant resources. However, as outlined in the following sections, there is potential for significantly better outcomes for LTCF residents if malnutrition is recognized. One important and often underrecognized cause of malnutrition in older LTCF residents deserves special focus: depression (and other psychiatric disor- ders). These disorders account for 22% to 32% of cases of significant weight loss in older LTCF residents (13,14) and community-dwelling elderly (15). Given the prevalence and frequently reversible nature of depression, it is incumbent upon all physicians and physician extenders who care for residents of LTCFs to have a heightened awareness of this disorder. III. ASSESSMENT OF NUTRITIONAL STATUS AND CONSEQUENCES OF MALNUTRITION IN LTCF RESIDENTS Considerable research has been aimed at identifying at-risk or malnourished LTCF residents. Many identification methods include complex measures not available in most LTCFs. However, a review of the relevant literature suggests a number of indicators, readily available from common data sets, that correlate with more sophisticated measures of nutritional status to help identify those at risk. Re- cently, one study (16) confirmed that the weight and body mass index (BMI; weight divided by height in kg/m2) measures available in the Minimum Data Set (MDS) closely correlates with more sophisticated anthropometric measures and bioelectrical impedance analysis. In another study, BMI closely correlated with a widely used 30-point scale (Mini-Nutritional Assessment [MNA]) predictive of clinical outcomes in many studies of the elderly (Fig. 1) (17). In that Swedish study, malnutrition (an MNA Ͻ17) was present in 33% to 71% of LTCF residents depending on the type of facility. A BMI less than 24 kg/m2 correlated with an MNA score of less than 17, identifying the majority of malnourished elderly. Sev- eral studies have documented that a recent loss of more than 5% of body weight, a weight less than 90% ideal body weight for age/gender and complaints of anorexia by patients correlate with malnutrition (6,7,18). It may be obvious, but there is a strong correlation between physical im- pairment and risk for malnutrition. In a recent study (19), it was reported that mal- nutrition in elderly subjects (from the community or nursing home), as determined at the time of hospitalization, was much more likely in subjects dependent in at least one activity of daily living (ADL), and dependence for any one of the ADLs investigated (bathing, dressing, transfer, toileting, and eating) was independently predictive of malnourishment. Interestingly, of the comorbid conditions exam- ined, congestive heart failure, dementia, chronic obstructive pulmonary disease, cancer, and diabetes mellitus, only diabetes mellitus was related to nutritional sta-

Nutrition and Infection 57 Figure 1 Correlation of body mass index (BMI) with a 30-point comprehensive nutri- tional assessment, the Mini-nutritional Assessment (MNA). Reproduced from Ref. 17 with permission. tus and was negatively associated with malnourishment. Several laboratory pa- rameters also indicate a likelihood for malnutrition; low serum albumin (Ͻ4.0 g/dL), total cholesterol (Ͻ160 mg/dL), total lymphocyte count (Ͻ1500/mm3), and hemoglobin (Ͻ13 g/dL) all should raise the possibility of malnutrition in an LTCF resident (6,20,21). Nutritional factors are strongly predictive of subsequent hospitalization, disability, and mortality (6,7,18,21–23). One recent study specifically focusing on LTCF residents (23) evaluated 350 randomly selected residents and determined the value of 96 medical, functional, socioeconomic, and nutritional variables for predicting severe (life-threatening) complications. Only five of the 96 variables were predictive, of which three were nutritional (serum albumin, BMI, and amount of weight loss in the prior year), with the other two being renal function and functional status (ADLs). In a subsequent cohort of 110 residents, the authors found that these five variables could predict life-threatening events with a sensi- tivity of 88% and a specificity of 65% (23). The impact of any one indicator of malnutrition cannot predict with cer- tainty which patients will live and which will die, nor are there specific data that reversing any one variable (e.g., serum albumin) will improve outcomes. How- ever, the magnitude of the association for some nutritional variables with mortal- ity is strong in elderly populations (reviewed in 20). For every decrease in serum

58 High albumin of 1.0 g/dL, mortality increases 10% to 22%. Furthermore, there is a four- fold risk of mortality when the total lymphocyte count is less than 1500/mm3, and a tenfold risk of mortality in subjects in whom the total cholesterol is less than 120 mg/dL, even after controlling for the presence of malignancy. Importantly, even well-nourished elderly residents of LTCFs often become malnourished during an acute hospitalization (24) and are more likely to become so than their community-dwelling counterparts (21). These elderly are much more likely to require discharge to a nursing or rehabilitation facility (relative risk [RR] 2.3; 95% confidence interval [1.1–4.6]), experience inhospital death (RR 8.0 [2.8–22.6] and death outside the hospital within 90 days (RR 2.9 [1.4–6.1]). Im- portant and, at times, unavoidable interventions such as “NPO” (nothing per oral) orders without adequate replacement nutrition contribute heavily to the number of subjects malnourished during hospitalization. IV. NUTRITIONAL INTERVENTIONS TO REDUCE INFECTION AND IMPROVE OUTCOMES IN LTCF RESIDENTS It is evident from the data outlined above that elderly LTCF residents are fre- quently malnourished, and poor nutritional status is associated with an increased risk of adverse outcomes; however, there are relatively few supplementation tri- als in LTCF residents (Tables 5 and 6). In most instances, the trials that have been published suffer from a lack of clinically relevant endpoints with regard to infec- tion risk and have been significantly underpowered to detect such benefits. Most studies demonstrate increased caloric intake, increased weight, improved MNA scores, or higher serum levels of micronutrients. A few demonstrate trends toward reduced infection or improved vaccine responses, but frequently, other small stud- ies contradict these findings. To the knowledge of the author, no trial of nutritional intervention in elderly LTCF residents has ever shown improved survival. A. Commercial Formulas/Protein-Energy Supplements Early data (reviewed in 7) suggested that commercially available nutritional sup- plements sometimes enhanced caloric intake and increased serum albumin and transferrin, but the effect on physical function or infection risk was not assessed. Several studies of oral supplementation of commercially available formulas have been performed in LTCF residents in the last 10 years (25–29), but unfortunately, still leave many questions unanswered. An early retrospective, case-control study (25) demonstrated that oral supplements are usually instituted for weight loss or poor appetite, and that oral supplementation resulted in weight gain back to base- line in the majority of residents. Serum albumin, lymphocyte counts, cholesterol,

Nutrition and Infection 59 Table 5 Trials of Commercial Formula Supplements in Long-Term Care Facility Residents Total N Time/ Author (reference) (n on suppl) study type Comment Johnson (25) 109(56) Variable/C-C Supplements started primarily due to Kayser-Jones (26) 40(29) Variable/O weight loss (71%) and poor appetite (16%); weight regained, but no effect on Turic (27) 58(28) 6 weeks/R,P infection or hospitalization rate; Lauque (28) mortality higher in supplemented group Fiatarone Singh (29) 88(37) 2 months/ (significantly more ill at baseline) mixed* Supplements rarely administered or 50(24) 10 weeks/R,P consumed as ordered by the physician (overall mean percentage consumed 55% of that prescribed); supplements often prescribed without adequate evaluation of cause for weight loss; calorie intake of supplement offset by reduced meal caloric intake Compared commercial formula vs a snack three times a day; increased energy/protein/nutrient intake in formula group Good compliance, average energy intake increased 400 kcal/day; weight and mini nutritional assessment scores improved in supplemented residents vs control groups Several parameters (total caloric intake, serum folate/vitamin D) approached significance, but only serum transferrin was significantly improved by supplementation; meal caloric intake declined in supplemented group; no change in performance status vs control * C-C, Case-control; O, Observational; R, Randomized; P, Placebo-controlled. This study used four groups, one well nourished that did not receive supplements (n ϭ 19), two groups of “at risk” subjects randomly assigned to supplement (n ϭ 19; provided one of four different supplement types) or no supplement (n ϭ 22), and a fourth group of malnourished elderly all provided a supplement (n ϭ 24).

60 High Table 6 Selected Micronutrient Supplementation Trials in Older, Long-Term Care Facility Residents Author Total N Time/ (reference) (n on suppl) trial type Nutrient(s) Comment Penn (31) 30 1 month/R,P Vitamins ⇑ T cells, CD4 cells, A,C,E lymphocyte responses. Murphy (36) 109(53) Single, high dose/R Vitamin A No change in infection rate or Van der Wielen 33(15) antibiotic use (32) 12 weeks/R,P kcal, thiamine, B6, B12, Significantly increased weight, Girodon (33) 81(61) 2 years/R,P,F folate serum thiamine, vitamin B6, decreased serum Znϩϩ ϩ homocysteine; no selenium or immunologic outcomes Vitamins A, measured C, E or both ⇑ serum selenium levels in Znϩϩ ϩ selenium group and BOTH group; ⇓ infectious episodes in the groups receiving Znϩϩ ϩ selenium, but not vitamins alone group Fortes (35) 118(88) 3 months/R,P,F Vitamin A, ⇓ CD3ϩ and CD4ϩ in vitamin Provinciali (37) Znϩϩ, both A group, Girodon (34) or neither ⇑ CD3ϩ, CD4ϩ, CD16ϩ and 384(194) 2 months/R Znϩϩ or Znϩϩ CD56ϩ lymphocytes ϩ arginine Influenza vaccine administered 725(543) 2 years/R,P,F Znϩϩ ϩ in 3 years; no difference in % selenium or responders or mean antibody Vitamins A, titer post-vaccination C, E or both ⇑ serum micronutrient levels, but no effect on DTH responses; improved responses to influenza vaccine in Znϩϩ ϩ selenium groups, borderline reduction in respiratory infection in Znϩϩ ϩ selenium groups (P ϭ 0.06), no effect of vitamins alone R, Randomized; P, Placebo-controlled; F, Factorial; DTH, Delayed-type hypersensitivity; ⇑, increased; ⇓ decreased.

Nutrition and Infection 61 and hemoglobin tended to improve, but there were too few residents to allow pur- poseful conclusions. A trend toward higher mortality was apparent in the supple- mented group (8/56 vs. 2/53; P ϭ 0.057), but these findings are subject to the significant bias of nonrandomized trials in which the intervention is often per- formed in a population that is more ill. In fact, in the indicated study, the supple- mented group did have an older mean age and were more likely to have a history of stroke or pressure sores. There was no difference with regard to infection or hospitalization. However, a nonrandomized observational study (26) suggested that only slightly more than 50% of the volume of supplement ordered is actually consumed by LTCF residents, and that up to half of residents placed on oral supplements will continue to lose weight. Although not quantified, this study suggested that oral supplement use between meals “destroyed the residents’ appetites” in some cases, reducing the total caloric intake. This issue was also raised in a subsequent, ran- domized trial discussed below (29). Three recent randomized trials of commercial formula supplementation in older LTCF residents have been published. A study of 53 LTCF residents in four LTCFs in Ohio (27) compared an 8-oz serving of a commercially available sup- plement with a “snack” at 10 AM, 2 PM, and before bed, and assessed several nu- tritional variables at 3 and 6 weeks. Significantly greater intake of protein, many vitamins and trace elements, and caloric intake were documented, and the study found no decrease in the energy intake from meals alone. No clinical outcomes of health, infection, or functional status were measured in that study. In a second French study, 88 residents in an LTCF were assessed via the MNA (28). Those with a score of less than 17 (n ϭ 24) received supplementation. Those with a score of 24 or higher (n ϭ 19) received no supplementation. Those residents with scores of 17 to 23.5 (n ϭ 41) were considered nutritionally at risk and randomized in an unblinded fashion to receive oral supplements or not (no placebo provided). Twenty-two were randomized to no supplements. All of the 22 completed the observation period and were therefore included in the analysis. However, of the 19 randomized to receive supplements, six withdrew consent or were admitted to the hospital and were excluded from the analysis. The groups were evaluated at baseline and on day 60. Mean caloric (~25%; ~400 kcal) and protein (~30%; 25–30 g) intake improved in the supplemented groups and did not change in the groups that did not receive supplements. There were too few sub- jects to determine any differences in other outcomes measured between the ran- domized groups, even with regard to MNA score, and no information was pro- vided regarding infection risk. The third study randomized and examined 50 LTCF residents in a placebo controlled, blinded trial of an oral liquid supplement and determined the impact of supplementation on caloric intake, serum measures of nutritional status, body composition, and health/physical status (29). There was no effect of supplemen-

62 High tation in any of the outcomes measured, even caloric intake. This was due to the fact that nonsupplemental calories (i.e., intake during meals) fell in the interven- tion group. Clearly there are issues of power in a study of only 25 subjects per arm, but these are the data available regarding oral supplementation in LTCF residents. In summary, protein-calorie supplements are frequently prescribed in LTCF residents, but compliance is often poor. Furthermore, such supplements inconsis- tently raise caloric intake due to reduced meal-time caloric intake by some resi- dents. No impact on infection risk or survival has been demonstrated for these supplements in LTCF residents. B. Multivitamin/Mineral Supplements There have been many studies of micronutrient supplementation in elderly sub- jects. Most have been performed in free-living elderly rather than residents of LTCFs (reviewed in 1,30). In LTCF residents, there are four studies of multivita- min/mineral supplements (31–35), and a few studies of individual micronutrients (35–37) (Table 6). These studies have often shown contradictory results; however, some unifying principles can be gleaned from review of the interventions reported to date. Clearly, micronutrient supplements can enhance vitamin and mineral in- take in LTCF residents and increase serum levels of many micronutrients. Fur- thermore, compliance with such supplementation is excellent and inexpensive. Trace elements, primarily zinc and selenium, may increase post-vaccine antibody titers, raise CD4 cell numbers, and reduce the risk of respiratory infection (33–35,37), whereas vitamin supplementation with vitamins A, C, E or ␤-carotene has little or no effect in the studies outlined to date. One large, well-designed study of vitamin/mineral supplementation high- lights the potential benefits and limitations of current data. This study (34) ran- domized 725 LTCF residents in 25 facilities in a factorial design to receive trace elements (zinc 20 mg ϩ selenium 100 mg), three vitamins (C 120 mg, E 15 mg, and ␤-carotene 6 mg [ϭ 1000 retinol equivalents]), both or neither for 2 years. Mortality was expectedly high in all four groups (~30%) and not different between groups. There was no effect on delayed-type hypersensitivity (DTH) responses, but a greater proportion of subjects in the trace element groups (vitamins ϩ trace elements or trace elements alone) had protective antibody titers after influenza vaccination (P Ͻ 0.05). Surprisingly, vitamins alone appeared to have a negative effect on antibody titers (P Ͻ 0.05). There was no effect on urogential tract infec- tions, but a trend toward reduced incidence of respiratory tract infections (P ϭ 0.06) occurred in both trace element groups but not in those receiving vita- mins alone. These data confirm the findings of a smaller, prior study by the same investigators (33), but in the earlier study, the reduction in respiratory infections in the Znϩϩ ϩ selenium groups reached statistical significance. The dose of vita- min E used in this and the previously mentioned study (33,34) was modest, 15

Nutrition and Infection 63 mg/d. It should be noted that higher doses of vitamin E (200 or 800 mg/d) have been found to effectively enhance immune responses (DTH responses and T-cell dependent vaccine responses) in free-living elderly (38,39). Reportedly, a ran- domized trial of vitamin E supplementation in higher doses is underway in LTCF residents (40). Vitamin A (or ␤-carotene, a vitamin A precursor) supplementation has been extensively studied. The rationale for vitamin A supplementation is strong be- cause it is an essential nutrient for immune function and regeneration of epithelial surfaces in the gastrointestinal tract and respiratory tree, and vitamin A deficiency is relatively common in LTCF residents (8). However, no study has shown bene- fit from vitamin A supplementation in LTCF residents, and, somewhat surpris- ingly, some studies have found potential harm (33–36). Thus, based on current data, specific vitamin A supplementation should probably be avoided. V. SPECIFIC SYNDROMES WHERE NUTRITIONAL SUPPLEMENTATION MAY BE OF BENEFIT A. Pneumonia There have been no specific studies of nutritional supplementation in LTCF resi- dents with pneumonia. However, a recent randomized, single-blind study of a commercially available protein-calorie supplement for 1 month after hospitaliza- tion for community-acquired pneumonia in elderly subjects did include LTCF res- idents (41). The supplement improved nutritional status for a variety of variables measured but, most importantly, functional status was improved at a follow-up visit 3 months later when compared with the placebo group. The study was not powered to detect differences in rehospitalization or mortality and did not perform any measures of immune competence. Nevertheless, short-term (1 month) nutri- tional supplementation after an episode of community-acquired pneumonia may be of benefit in elderly LTCF residents. B. Pressure Ulcers Debate is considerable over whether nutritional supplementation can prevent or speed the healing of pressure ulcers (42,43). A recent multicenter trial (44) demonstrated a slightly reduced risk of pressure ulcers in a group receiving pro- tein-calorie supplements, but several issues have been raised about this study (45). The caloric intake did not increase in supplemented subjects and the difference in incidence of pressure ulcers was relatively small: 41% in the supplemented group versus 47% in the control group. More widely accepted but based on no more convincing data is the practice of micronutrient (e.g., Znϩϩ) supplementation to assist in healing of pressure ul-

64 High cers. Most data suggest that if there is a benefit of vitamin/mineral supplementa- tion for healing skin/soft tissue wounds, that it is likely limited to individuals de- ficient at baseline. Current recommendations are to provide adequate calories (30–35 kcal/kg) and protein (1–1.25 g/kg) to avoid negative nitrogen balance, and zinc at a dose of 220 mg/day (42). Vitamins A, C, and several B complex vitamins are necessary for wound healing, but there are no specific recommendations re- garding dosing. C. Urinary Tract Infections One nutritional intervention has been reasonably well studied for the prevention of urinary tract infection (UTI) in elderly subjects: cranberry juice consumption (reviewed in 46). There has been only one study in LTCF residents (47); however, there has been one reasonably sized, randomized trial (48), another small crossover study that demonstrated benefit in the elderly (49), and a variety of stud- ies in younger patients (46). However, there are valid criticisms against all these studies. In the randomized trial (48), the endpoint was reduction of bacteriuria with pyuria (15% in the 300 mL/d cranberry juice group vs. 28% in the control group), not symptomatic UTI. As outlined in Chapter 12, asymptomatic bacteri- uria in the elderly does not require therapy. There was a trend toward reduced an- tibiotic use in the treatment group (1.7 vs. 3.2 antibiotics per 100 patient months) which, if confirmed in a larger study, would be of great value in and of itself. In the LTCF study (47), reported only in abstract form, both cranberry juice (220 mL/d) and capsules of cranberry juice extract were used, and the control group was historical. Five hundred thirty-eight LTCF residents were studied and UTIs reduced from 27/month in the historical control group to 20/month in the treat- ment group (P ϭ 0.01). There may be other benefits of cranberry juice (46). One possible cause for reducing antibiotic use could be a reduction by cranberry juice in malodorous urine, a common trigger for urinalysis and urine culture for institutionalized el- derly. In addition, a small study of patients with urostomies who consumed 160 to 330 mL/d of cranberry juice experienced improvement in the skin surrounding the stoma. This could be of benefit in LTCF residents with incontinence and immo- bility who are at risk for skin breakdown, but no substantive trial testing this hy- pothesis has been performed. VI. APPETITE STIMULANTS Appetite stimulants are poorly studied in LTCF residents, but a recent random- ized, double-blind trial was reported that indicated megestrol acetate (MA) may be of some value (50). In that study, 800 mg/d of MA or placebo was provided for

Nutrition and Infection 65 12 weeks to LTCF residents with weight loss or low body weight, and then the res- idents were followed up for an additional 13 weeks for subsequent health out- comes. At the conclusion of the 12-week supplementation period, there was no difference in weight, but appetite and sense of well-being were significantly bet- ter in the MA group. However, by the end of the 13-week follow-up period, the MA recipients were more likely to have gained 4 lb or more over their baseline. This approach appears to deserve further study. VII. DRUG-NUTRIENT INTERACTIONS The elderly LTCF resident is likely to be receiving multiple prescription drugs. In- creasingly, there is recognition that nutrient-drug interactions can cause serious adverse effects. In a recent study of residents in three LTCFs in New York (51), residents consumed approximately five drugs per patient and, on average, were at risk for 1.4 to 2.7 drug-nutrient interactions per month. With specific regard to in- fection, this is most likely to cause difficulty with antibiotic administration. Tetra- cyclines and fluoroquinolones may be poorly absorbed when antacids, multivalent cations (e.g., calcium), or tube feedings are provided. Certain antifungal com- pounds, particularly itraconazole, may be poorly absorbed with concomitant ad- ministration of antacids or hydrogen ion (H2) antagonists/proton pump inhibitors. More likely than nutrients influencing drug metabolism, drugs are likely to influence nutrient intake (Table 7). The most commonly prescribed drugs that are likely to induce anorexia and decrease nutrient intake are antibiotics, antidepres- sants, and other psychiatric drugs, digoxin, and anti-inflammatory agents. A crit- ical part of nutritional care for elderly LTCF residents is frequent, thorough review of all medications with discontinuation of nonessential therapies. VIII. CONCLUSIONS Residents of LTCFs are often at risk for malnutrition, and reversible causes of malnutrition are common. Most at-risk residents can be initially identified by in- formation available in the MDS (weight and BMI) and initial screening laborato- ries (serum albumin, total lymphocyte count). More sophisticated assessments, such as the MNA, have been shown to be valid in LTCF residents. Once identi- fied, data support the correction of underlying medical causes, particularly de- pression, and the use of nutritional supplements or appetite stimulants to increase caloric and protein intake in LTCF residents to reverse weight loss. However, the role of such supplements for preventing infection is less well defined by currently available data. Current data support the use of trace mineral supplements (20 mg/d Znϩϩ-sulfate and 100 ␮g/d selenium sulfide) in most LTCF elderly, as the ex-

66 High Table 7 Drugs that Cause Anorexia in Older Adults Anorectic agents Cardiovascular drugs Digoxin Amiodarone Procainamide Quinidine Spironolactone Gastrointestinal drugs Cimetidine Interferon Psychiatric drugs Phenothiazines Butyrphenones Lithium Amitriptyline Impramine Fluoxetine and other selective serotonin-reuptake inhibitors Anti-infective drugs Most antibiotics Metronidazole Griseofulvin Nutrient supplements Iron sulfate Potassium salts Vitamin D (in excess) Antineoplastics Cyclophosphamide and most others Antirheumatic drugs Nonsteroidal anti-inflammatory agents Colchicine Penicillamine Pulmonary drugs Theophylline Malabsorptive agents Laxatives Cholestyramine Methotrexate Colchicine Neomycin Ganglionic blockers Agents that increase metabolism Theophylline L-Thyroxine (in excess) Thyroid extract Triiodotyrosine (in excess) D-Pseudoephedrine Source: Ref. 11, with permission.

Nutrition and Infection 67 pense and risk of adverse effects is small, and there appears to be a reduced risk of respiratory infection. Vitamin supplements have variable effects. Vitamin E at 200 mg/d improves immune responses in healthy elderly and may be of value in LTCF residents, but vitamin A supplementation should probably be avoided. Spe- cific nutritional supplementation may be of value in certain infectious diseases such as recovery from community-acquired pneumonia (protein-calorie supple- ments) and prevention of UTIs (cranberry juice). Finally physicians should be aware of the potential for antibiotic-nutrient interactions and the effect of anorec- tic medications on nutrient intake. REFERENCES 1. High KP. Nutrition and infection. In: Yoshikawa TT, Norman DC (eds). Infectious Disease in the Aging: A Clinical Handbook. Totowa, NJ: Humana Press, 2001: 299–312. 2. Wick G, Grubeck-Loebenstein B. Primary and secondary alterations of immune re- activity in the elderly: Impact of dietary factors and disease. Immunol Rev 1997; 160:171–184. 3. Heuser MD, Adler WH. Immunological aspects of aging and malnutrition: Conse- quences and intervention with nutritional immunomodulators. Clin Geriatr Med 1997; 13:697–715. 4. Lesourd B, Mazari L, Ferry M. The role of nutrition in immunity in the aged. Nutr Rev 1998; 56:S113–S125. 5. Kerstetter JE, Holthausen BA, Fitz PA. Malnutrition in the institutionalized older adult. J Am Diet Assoc 1992; 92:1109–1116. 6. Rudman D, Feller AG. Protein-calorie undernutrition in the nursing home. J Am Geriatr Soc 1989; 37:173–183. 7. Abbasi AA, Rudman D. Undernutrition in the nursing home: Prevalence, conse- quences, causes and prevention. Nutr Rev 1994; 52:113–122. 8. Azais-Braesco V, Moriniere C, Guesne B, Partier A, Bellenand P, Baguelin D, Grolier P, Alix E. Vitamin A status in the institutionalized elderly. Critical analysis of four evaluation criteria: Dietary vitamin A intake, serum retinol, relative dose-re- sponse test (RDR) and impression cytology with transfer (ICT). Int J Vitam Nutr Res 1995; 65:151–161. 9. Bendich A. Criteria for determining recommended dietary allowances for healthy older adults. Nutr Rev 1995; 53:S105–S110. 10. Roubenoff R, Giacoppe J, Richardson S, Hoffman PJ. Nutrition assessment in long- term care facilities. Nutr Rev 1996; 54:S40–S42. 11. Morley JE. Anorexia of aging: Physiologic and pathologic. Am J Clin Nutr 1997; 66:760–773. 12. Morley JE, Silver AJ. Nutritional issues in nursing home care. Ann Intern Med 1995; 123:850–859. 13. Morley JE, Kraenzle D. Causes of weight loss in a community nursing home. J Am Geriatr Soc 1994; 42:583–585.

68 High 14. Wright BA. Weight loss and weight gain in a nursing home: A prospective study. Geriatr Nurs 1993; 14:156–159. 15. Wilson MG, Caswani S, Liu D, Morley JE, Miller DK. Prevalence and causes of un- dernutrition in medical outpatients. Am J Med 1998; 104:56–63. 16. Blaum CS, O’Neill EF, Clements KM, Fries BE, Fiatarone MA. Validity of the min- imum data set for assessing nutritional status in nursing home residents. Am J Clin Nutr 1997; 66:787–794. 17. Saletti A, Lindgren EY, Johansson L, Cederholm T. Nutritional status according to mini nutritional assessment in an institutionalized elderly population in Sweden. Gerontology 2000; 46:139–145. 18. Sullivan DH. Undernutrition in older adults. Ann Long-Term Care 2000; 8:41–46. 19. Covinsky KE, Martin GE, Beyth RJ, Justice AC, Sehgal AR, Landefeld CS. The re- lationship between clinical assessments of nutritional status and adverse outcomes in older hospitalized medical patients. J Am Geriatr Soc 1999; 47:532–538. 20. Omran ML, Morley JE. Assessment of protein energy malnutrition in older persons, part II: Laboratory evaluation. Nutrition 2000; 16:131–140. 21. Kamel HK, Karcic E, Karcic A, Barghouthi H. Nutritional status of hospitalized el- derly: Differences between nursing home patients and community-dwelling patients. Ann Long-Term Care 2000; 8:33–38. 22. Volkert D, Kruse W, Oster P, Schlierf G. Malnutrition in geriatric patients: Diagnos- tic and prognostic significance of nutritional parameters. Ann Nutr Metab 1992; 36:97–112. 23. Sullivan DH, Walls RC. The risk of life-threatening complications in a select popu- lation of geriatric patients: The impact of nutritional status. J Am Coll Nutr 1995; 14:29–36. 24. Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospi- talized patients. A prospective study. JAMA 1999; 281:2013–2019. 25. Johnson LE, Dooley PA, Gleick JB. Oral nutritional supplement use in elderly nurs- ing home patients. J Am Geriatr Soc 1993; 41:947–952. 26. Kayser-Jones J, Schell ES, Porter C, Barbaccia JC, Steinbach C, Bird WF, Redford M, Pengilly K. A prospective study of the use of liquid oral dietary supplements in nursing homes. J Am Geriatr Soc 1998; 46:1378–1386. 27. Turic A, Gordon KL, Craig LD, Ataya DG, Voss AC. Nutrition supplementation en- ables elderly residents of long-term-care facilities to meet or exceed RDAs without displacing energy or nutrient intakes from meals. J Am Diet Assoc 1998; 98: 1457–1459. 28. Lauque S, Arnaud-Battandier F, Mansourian R, Guigoz Y, Paintin M, Nourhashemi F, Vellas B. Protein-energy oral supplementation in malnourished nursing-home res- idents. A controlled trial. Age Ageing 2000; 29:51–56. 29. Fiatarone Singh MA, Bernstein MA, Ryan AD, O’Neill EF, Clements KM, Evans WJ. The effect of oral nutritional supplements on habitual dietary quality and quan- tity in frail elders. J Nutr Health Aging 2000; 4:5–12. 30. High KP. Micronutrient supplementation and immune function in the elderly. Clin In- fect Dis 1999; 28:717–722. 31. Penn ND, Purkins L, Kelleher J, Heatley RV, Mascie-Taylor BH, Belfield PW. The effect of dietary supplementation with vitamins A, C and E on cell-mediated immune

Nutrition and Infection 69 function in elderly long-stay patients: A randomized controlled trial. Age Ageing 1991; 20:169–174. 32. van der Wielen RP, van Heereveld HA, de Groot CP, van Staveren WA. Nutritional status of elderly female nursing home residents: The effect of supplementation with a physiological dose of water-soluble vitamins. Eur J Clin Nutr 1995; 49:665–674. 33. Girodon F, Lombard M, Galan P, Brunet-Lecomte P, Monget A-L, Arnaud J, Preziosi P, Hercberg S. Effect of micronutrient supplementation on infection in institutional- ized elderly subjects: A controlled trial. Ann Nutr Metab 1997; 41:98–107. 34. Girodon F, Galan P, Monget AL, Boutron-Ruault MC, Brunet-Lecomte P, Preziosi P, Arnaud J, Manuguerra JC, Herchberg S. Impact of trace elements and vitamin sup- plementation on immunity and infections in institutionalized elderly patients: A ran- domized controlled trial. MIN. VIT. AOX. geriatric network. Arch Intern Med 1999; 159:748–754. 35. Fortes C, Forastiere F, Agabiti N, Fano V, Pacifici R, Virgili F, Piras, Guidi L, Bar- toloni C, Tricerri A, Zuccaro P, Ebrahim S, Perucci CA. The effect of zinc and vita- min A supplementation on immune response in an older population. J Am Geriatr Soc 1998; 46:19–26. 36. Murphy S, West KPJ, Greenough WB, Cherot E, Katz J, Clement L. Impact of vita- min A supplementation on the incidence of infection in elderly nursing-home resi- dents: A randomized controlled trial. Age Ageing 1992; 21:435–439. 37. Provinciali M, Montenovo A, Di Stefano G, Colombo M, Daghetta L, Cairati M, Veroni C, Cassino R, Della TF, Fabris N. Effect of zinc or zinc plus arginine supple- mentation on antibody titre and lymphocyte subsets after influenza vaccination in el- derly subjects: A randomized controlled trial. Age Ageing 1998; 27:715–722. 38. Meydani SN, Leka L, Loszewski R. Long term vitamin E supplementation enhances immune response in healthy elderly. FASEB J 1994; 8:A272. 39. Meydani SN, Meydani M, Blumberg JB, Leka LS, Siber G, Loszewski R, Thompson C, Pedrosa MC, Diamond RD, Stollar BD. Vitamin E supplementation and in vivo immune response in healthy elderly subjects. A randomized controlled trial. JAMA 1997; 277:1380–1386. 40. Han SN, Meydani SN. Vitamin E and infectious diseases in the aged. Proc Nutr Soc 1999; 58:697–705. 41. Woo J, Ho SC, Mak YT, Law LK, Cheung A. Nutritional status of elderly patients during recovery from chest infection and the role of nutritional supplementation as- sessed by a prospective randomized single-blind trial. Age Ageing 1994; 23:40–48. 42. Chernoff R. Policy: Nutrition standards for treatment of pressure ulcers. Nutr Rev 1996; 54:S43–S44. 43. Finucane TE. Malnutrition, tube feeding and pressure sores: Data are incomplete. J Am Geriatr Soc 1995; 43:447–451. 44. Bourdel-Marchasson I, Barateau M, Rondeau V, Dequae-Merchadou L, Salles-Mon- taudon N, Emeriau JP, Manciet G, Dartigues JF. A multi-center trial of the effects of oral nutritional supplementation in critically ill older inpatients. GAGE Group. Groupe Aquitain Geriatrique d’Evaluation. Nutrition 2000; 16:1–5. 45. Hessov I. Can nutritional intervention reduce the incidence of pressure sores? Nutri- tion 2000; 16:141. 46. Harkins KJ. What’s the use of cranberry juice? Age Ageing 2000; 29:9–12.

70 High 47. Dignam R, Ahmen M, Denman S, Zayon RN, Wilks T, Shipman C, Wolfert R, Kle- ban M. The effect of cranberry juice on UTI rates in a long term care facility. J Am Geriatr Soc 1997; 45(Supplement):S53 (p. 169). 48. Avorn J, Monane M, Gurwitz JH, Glynn RJ, Choodnovskiy I, Lipsitz LA. Reduction of bacteriuria and pyuria after ingestion of cranberry juice. JAMA 1994; 271:751–754. 49. Haverkorn MJ, Mandigers J. Reduction of bacteriuria and pyuria using cranberry juice. JAMA 1994; 272:590. 50. Yeh SS, Wu SY, Lee TP, Olson JS, Stevens MR, Dixon T, Porcelli RJ, Schuster MW. Improvement in quality-of-life measures and stimulation of weight gain after treat- ment with megestrol acetate oral suspension in geriatric cachexia: Results of a dou- ble-blind, placebo-controlled study. J Am Geriatr Soc 2000; 48:485–492. 51. Lewis CW, Frongillo EA, Jr, Roe DA. Drug-nutrient interactions in three long-term- care facilities. J Am Diet Assoc 1995; 95:309–315.

6 Clinical Manifestations of Infections Dean C. Norman VA Greater Los Angeles Healthcare System, and UCLA School of Medicine, Los Angeles, California I. INTRODUCTION Infectious diseases are a leading cause of morbidity and mortality in the frail nurs- ing home population and a leading cause of transfer of residents from a long-term care facility (LTCF) to an acute care facility. Higher morbidity and mortality rates in older patients in general result partly because of diminished physiological re- serves and altered host defenses brought on by aging and comorbidities. This problem is magnified in residents of LTCFs because of debility caused by chronic disease. Elderly residents of long-term care institutions are typically taking mul- tiple medications. This practice, coupled with age and morbidity-related changes in pharmacology of drugs, including antibiotics, increases the risk for adverse drug interactions in nursing home residents. Infected nursing home residents frequently are transferred to acute care hospitals. Unfortunately, hospitalization may be complicated by nosocomial in- fection and iatrogenic illness. Furthermore, once hospitalized, the elderly are more likely to undergo invasive procedures and to suffer complications. Vigorous pre- vention measures to control infectious diseases plus rapid diagnosis and timely initiation of appropriate empiric antimicrobial therapy will reduce the impact of infectious diseases in the long-term care population. However, atypical presenta- tion and variability of diagnostic testing and other factors may delay diagnosis and therapy in a population that can least afford such delays. Fortunately, the differential diagnosis of important infectious diseases in the elderly is somewhat limited and dependent on the clinical setting and the patient’s functional status. Respiratory infections, including pneumonia, and urinary tract and soft tissue infections (acronym “PUS”), as well as gastrointestinal infections 71

72 Norman comprise the majority of acute infections in residents of long-term care institu- tions (1). The types of infections may be limited, but the microbial etiology of infections is more diverse in the elderly when compared with the younger popu- lation. In general, a variety of pathogens may account for a given infection. For example, pneumonia in the young is usually caused by relatively few pathogens, such as Streptococcus pneumoniae and Mycoplasma pneumoniae. Urinary tract infection in the young adult is usually caused by Escherichia coli. However, in the older adult a variety of pathogens may cause either of these common infections. A small but significant number of cases of community-acquired pneumonia in the elderly are caused by gram-negative bacilli, and a higher percentage of lower res- piratory tract infections in nursing homes is caused by gram-negative bacilli and mixed flora (see Chapter 14). Similarly, urinary tract infection in the elderly in both the community and long-term care setting may be caused by any one of sev- eral species of gram-negative and gram-positive bacteria. Chronic indwelling bladder catheter-associated infections are typically polymicrobial. The diverse microbial etiology of urinary tract infections in residents of LTCFs requires ob- taining urine cultures before initiation of antibiotic therapy for symptomatic uri- nary tract infection (see Chapter 12). II. ATYPICAL PRESENTATIONS OF INFECTION Once an infection develops, the cornerstone of successful treatment is timely di- agnosis and the rapid initiation of empiric antimicrobial therapy. Nonclassic pre- sentations of acute illnesses occur often in the frail elderly, and acute infections are no exception. In the nursing home, infectious diseases provide unique diag- nostic challenges because of atypical presentations—especially diminished fever responses (described in the next section), the frequent presence of cognitive im- pairment in nursing home residents, and the lack of availability of timely radio- graphs or laboratory data. The clinician caring for residents of LTCFs should be aware that virtually any acute change in functional status may herald the onset of a serious infectious disease. These manifestations include, but are not limited to lethargy, anorexia, falls, focal neurologic signs and delirium (Table 1). Changes in mental status from baseline are seen even when the infection does not involve the central nervous system. Common infections may present without classic symptoms. For example, pneumonia may be present without cough, purulent sputum, fever, or chest pain, and the only sign alerting the clinician may be tachypnea; meningitis may occur without a stiff neck; and symptomatic urinary tract infection may at times present solely as a decline in cognitive function with- out dysuria, urgency, or frequency. Moreover, the presentation of illness may not be in proportion to the severity of the underlying infection, a large percentage of

Clinical Manifestations of Infections 73 Table 1 Nonspecific Presentations of Acute Infection in Nursing Home Patients Diagnostic clues • Anorexia • Any unexplained change in functional status from baseline • Change in baseline body temperature (a decrease may be caused by sepsis) • Decline in cognition • Failure to thrive • Focal neurologic signs (may be a clue to presence of meningitis, endocarditis) • Lethargy • Hypotension (may be a clue to presence of sepsis) • Tachypnea (may be initial finding in pneumonia, sepsis) elderly women who present with symptoms and signs of pyelonephritis, which are indistinguishable from those of younger women, will be bacteremic, which is in contrast to younger women (2). Localizing peritoneal findings may be delayed in cases of severe intra-abdominal infection (3,4). This is especially problematic in LTCFs, where physicians are rarely present to perform physical examinations at the time of onset of symptoms. III. FEVER A. Significance of Diminished Fever Response Studies of specific infections in older adults, including pneumonia (5,6), bac- teremia (7,8), endocarditis (9,10), nosocomial febrile illness (11), meningitis (12), and intra-abdominal infection (13) confirm that fever, the hallmark of invasive mi- crobial infection, may be blunted or absent in up to one-third of infected elderly pa- tients (14–16). The absence of this cardinal sign of infection has implications be- yond confounding clinical diagnosis. First, an absent or blunted fever response to infection is a poor prognostic sign, as demonstrated by a study of several hundred patients with bacteremia and fungemia (17). The results of this study confirmed that those patients responding to bacteremia or fungemia with a robust fever were more likely to survive. This classic study’s conclusion is now well established for many infectious diseases and applies to both young and older adults. Second, al- though the prognostic significance of the febrile response to infection is clear, it is less well established that fever is an essential adaptive mechanism that is an im- portant host defense in humans. However, there is strong evidence that fever is an important host defense for a variety of other species (18). Cold-blooded animals such as certain types of lizards and fish seek warmer environments in order to raise body temperatures in response to infection. Laboratory experiments confirm that

74 Norman fever is an important host defense in these poikilothermic animals (19,20). In fact, enhanced resistance to infection appears to occur with increased body temperature (fever) in a variety of mammalian animal models (18). Therefore, based on these animal data, it can be considered that fever is potentially an important host defense in humans. The effect of fever on host defenses is independent of a direct effect of elevated body temperature on bacterial replication. The exceptions are that physi- ologically achievable temperature elevations may inhibit bacterial growth directly of Treponema pallidum, Neisseria gonorrhoeae, and certain strains of pneumo- cocci. The mechanism by which fever augments host defenses and improves natu- ral and cellular immune responses appears to be multifactorial; it minimally in- volves elevating certain monokine and cytokine production and enhancing cytokine activity. These cytokines include interleukins 1 and 6, tumor necrosis fac- tor, alpha interferon, and others. These particular cytokines are also endogenous pyrogens and have many effects on cellular components of the immune response. One effect appears to be to facilitate adherence of leukocytes to endothelial cells and leukocyte migration to extravascular areas of infection. The mechanism by which a significant number of infected older adults fail to mount a febrile response is not known. Potential mechanism(s) have been pro- posed based on the current understanding of the pathogenesis of fever. The role of cytokines as endogenous mediators of fever has been recently reviewed (21,22). Bacterial products such as lipopolysaccharide (endotoxin) induce macrophages and other cells to produce cytokines that act as endogenous pyrogens. These py- rogens are produced either locally at the site of infection and enter the circulation or by macrophages adhering to endothelium in circumventricular organs of the brain. The pyrogens act on the anterior hypothalamus resulting in a biochemical cascade including the release of prostaglandin E2. This cascade raises the central nervous system “thermostat” (22). These changes result in shivering, vasocon- striction, and various behavioral responses, all of which elevate core body tem- perature, which then becomes the new baseline. When the infection is over, the thermostat is reset to the previous baseline; sweating and temperature-lowering behaviors occur, thus returning body temperature to normal. These pathways could be affected by aging, and thermoregulation in the elderly appears to be im- paired to some degree. This is evidenced by the increased morbidity and mortal- ity in older persons from heat stroke and hypothermia. A variety of endogenous pyrogens results in a lower fever response in older mice compared with younger mice (23–25). Other data in rats demon- strated that intracerebroventricular injection of interleukin 1 results in similar immediate fever responses between young and old animals. This suggests that blunted febrile responses observed in older mammals may result from an inabil- ity of peripheral endogenous mediators to reach the central nervous system rather than an unresponsiveness of the central nervous system (26). Other stud- ies have demonstrated diminished production of endogenous pyrogens with age

Clinical Manifestations of Infections 75 in various rodent models (27). Select rodent experiments have yielded evidence that changes in thermogenic brown fat may play a role in the blunted fever re- sponse of aging (28). Thus, reduced production and response to endogenous py- rogens may be important in the pathogenesis of the blunted febrile response to infection observed in the elderly. There is no evidence in humans that reducing a fever with antipyretic drugs increases the risk of morbidity and mortality if appropriate antimicrobial therapy is initiated. Fever in the elderly can result in discomfort, tachycardia, and other physiologic stresses that may be harmful, and these symptoms provide a rationale for antipyretic use (18). B. Baseline Temperatures and Definition of Fever in the Nursing Home The normal and febrile body temperature for older adults has been thoroughly re- viewed recently (16). Baseline temperature and diurnal variation of temperature as measured by electronic thermometry was decreased in a nursing home popula- tion (29,30). Mean baseline morning rectal temperature was established to be 98.6 degrees Fahrenheit (F) (37 degrees centigrade [C]) in 22 residents in whom oral temperatures could not be easily obtained. The mean oral temperature of 85 addi- tional residents was 97.4 degrees F (37.3 degrees C). Diurnal variation was only 0.6 degrees F (0.3 degrees C) for rectal temperatures and 0.4 degrees F (0.2 de- grees C) for oral temperatures. In another study, 50 randomly selected nursing home residents had mean oral baseline temperature of 97.4 degrees F (36.3 de- grees C) (29). A retrospective review found 69 infections in 26 of these residents, with the mean maximum temperature reaching 101.3 degrees F (38.5 degrees C). In nearly half these infections, the temperature did not reach 101 degrees F. Yet, a majority of these patients significantly increased their temperature over baseline by at least 2.4 degrees F (1.3 degrees C). Lowering the criterion for fever to 100 degrees F (37.8 degrees C) or higher raised the sensitivity to 70% for predicting infection with a specificity of 90%. These findings led to the recommendation by the Practice Guidelines Committee of the Infectious Diseases Society of America that a clinical evaluation be done for nursing home residents with a single oral temperature higher than 100 degrees F (37.8 degrees C) or persistent oral or rec- tal temperature higher than 99 degrees F (37.2 degrees C) or greater than 99.5 degrees F (37.5 degrees C), respectively. Two or more readings of more than 2 de- grees F (1.1 degrees C) over baseline regardless of site of measurement should also stimulate an evaluation for infection (31). The chance of an infection is fur- ther increased if obvious symptoms and signs of infection exist or if there is any change in functional status accompanying the temperature changes (see Table 1). In some cases a significant decrease in temperature might also indicate a serious infection complicated by sepsis.

76 Norman C. Significance of Robust Fever Response and Fever of Unknown Origin (FUO) Infected elderly residents of LTCFs who mount a vigorous febrile response simi- lar to younger adults, as defined by 101 degrees F orally (38.3 degrees C), can be expected to have a serious or life-threatening infection. This conclusion is extrap- olated from a classic study of 1,200 ambulatory care patients (32) and two other confirmatory studies (33,34). In contrast to the young in whom these fevers were usually the result of benign viral infections, the elderly, especially the very old, usually suffered from a serious or life-threatening infection. Finally, infection is the leading etiology of FUO in the elderly, followed by connective tissue diseases such as temporal arteritis. A lesser number of cases are the result of malignancy (35–37). Many of these underlying conditions are treatable and, unless advanced directives preclude an extensive evaluation, an underlying cause for FUO in the older person should be sought. IV. CONCLUSION The clinician must be familiar with all the manifestations of infections in the long- term care setting to minimize the impact of infectious diseases in this population. Preventive measures, rapid, aggressive diagnosis, and empiric therapy are essen- tial to further reduce morbidity and mortality from infection. Fever, the hallmark of infection, may be absent or blunted in 20% to 30% of infections in the frail el- derly. Alternatively, the presence of a vigorous fever in the older LTCF resident is more likely to be associated with a serious bacterial infection compared with a younger population, and requires a thorough and prompt evaluation. REFERENCES 1. Bradley SF. Infections and infection control in the long-term care setting. In: Yoshikawa TT, Norman DC (eds). Infectious Disease in the Aging. A Clinical Hand- book. Totowa, Humana Press, 2001:245–256. 2. Gleckman RA, Bradley PJ, Roth RM, Hibert DM. Bacteremic urosepsis: A phe- nomenon unique to elderly women. J Urol 1985; 133:174–175. 3. Norman DC, Yoshikawa TT. Intraabdominal infection: Diagnosis and treatment in the elderly patient. Gerontology 1984; 30:327–338. 4. Campbell BS, Wilson SE. Intraabdominal Infections. In: Yoshikawa TT, Norman DC (eds). Infectious Disease in the Aging. A Clinical Handbook. Totowa: Humana Press, 2001:91–98. 5. Bentley DW. Bacterial pneumonia in the elderly: Clinical features, diagnosis, etiol- ogy and treatment. Gerontology 1984; 30:297–307.

Clinical Manifestations of Infections 77 6. Marrie TJ, Haldane EV, Faulkner RS, Durant H, Kwan C. Community-acquired pneumonia requiring hospitalization: Is it different in the elderly? J Am Geriatr Soc 1985; 33:671–680. 7. Gleckman R, Hibert D: Afebrile bacteremia. A phenomenon in geriatric patients. JAMA 1982; 248:1478–1481. 8. Finkelstein M, Petkun WM, Freedman ML, Antopol SC. Pneumococcal bacteremia in adults: Age-dependent differences in presentation and outcome. J Am Geriatr Soc 1983; 31:19–27. 9. Terpenning MS, Buggy BO, Kauffman CA. Infective endocarditis: Clinical features in young and elderly patients. Am J Med 1987; 83:626–634. 10. Werner GS, Schulz R, Fuchs JB, Andreas S, Prange H, Ruschewski W, Kreuzer H. Infective endocarditis in the elderly in the era of transesophageal echocardiography: Clinical features and prognosis compared with younger patients. Am J Med 1996; 100:90–97. 11. Trivalle C, Chassagne P, Bouaniche M, Landrin I, Marie I, Kadri N, Menard JF, Lemeland JF, Doucet J, Bercoff E. Nosocomial febrile illness in the elderly: Fre- quency, causes, and risk factors. Arch Intern Med 1998; 158(14):1560–1565. 12. Gorse GJ, Thrupp LD, Nudleman KL, Wyle FA, Hawkins B, Cesario TC. Bacterial meningitis in the elderly. Arch Intern Med 1984; 144:1603–1607. 13. Potts FE, IV, Vukov LF: Utility of fever and leukocytosis in acute surgical abdomens in octogenarians and beyond. J Geront A Biol Sci Med Sci 1999; 54A(2):M55–M58. 14. Yoshikawa TT, Norman DC: Fever in the elderly. Infect Med 1998; 15(10):704–706. 15. Norman DC. Fever and aging. Infect Dis Clin Pract 1998; 7(8):387–390. 16. Norman DC. Fever in the elderly. Clin Infect Dis 2000; 31:148–151. 17. Weinstein MP, Murphy JR, Reller RB, Lichenstein KA. The clinical significance of positive blood cultures: A comprehensive analysis of 500 episodes of bacteremia and fungemia II: Clinical observations with special reference to factors influencing prog- nosis. Rev Infect Dis 1983; 5:54–70. 18. Mackowiak, PA. Physiological rationale for suppression of fever. Clin Infect Dis 2000; 31(suppl 5):S185–S189. 19. Kluger MJ, Ringler DM, Anver MR: Fever and survival. Science 1975; 188:166–168. 20. Covert JB, Reynolds WM: Survival value of fever in fish. Nature 1977; 267:43–45. 21. Netea MG, Kullberg BJ, Van der Meer JWM. Circulating cytokines as mediators of fever. Clin Infect Dis 2000; 31:S178–S184. 22. Dinarello CA. Cytokines as endogenous pyrogens. J Infect Dis 1999; 179(suppl 2): S294–S304. 23. Norman DC, Yamamura RH, Yoshikawa TT. Fever response in old and young mice after injection of interleukin. J Gerontol 1988; 43:M80–M85. 24. Miller D, Yoshikawa TT, Castle SC, Norman DC. Effect of age in fever response to recombinant tumor necrosis factor alpha in a murine model. J Gerontol 1991; 46: M176–M179. 25. Miller DJ, Yoshikawa TT, Norman DC. Effect of age on fever response to recombi- nant interleukin-6 in a murine model. J Gerontol A Biol Sci Med Sci 1995; 50A: M276–M279. 26. Plata-Salamán CR, Peloso E, Satinoff E. Interleukin-1-induced fever in young and old Long-Evans rats. Am J Physiol 1998; 275:R1633–R1638.

78 Norman 27. Bradley SF, Vibhagool A, Kunkel SL, Kauffman CA. Monokine secretion in aging and protein malnutrition. J Leukocyte Biol 1989; 45:510–514. 28. Scarpace PJ, Bender BS, Burst SE. The febrile response of E. coli peritonitis in senes- cent rats. Gerontologist 1990; 30:215A. 29. Castle SC, Norman DC, Yeh M, Miller D, Yoshikawa TT. Fever response in elderly nursing home residents: Are the older truly colder? J Am Geriatr Soc 1991; 39: 853–857. 30. Castle SC, Yeh M, Toledo S, Yoshikawa TT, Norman DC. Lowering the temperature criterion improves detection of infections in nursing home residents. Aging Immunol Infect Dis 1993; 4:67–76. 31. Bentley DV, Bradley S, High K, Schoenbaum S, Taler, G, Yoshikawa TT. Practice guideline for evaluation of fever and infection in long-term care facilities. Clin Infect Dis 2000; 31:640–653. 32. Keating MJ III, Klimek JJ, Levine DS, Kiernan FJ. Effect of aging on the clinical sig- nificance of fever in ambulatory adult patients. J Am Geriat Soc 1984; 32:282–287. 33. Wasserman M, Levinstein M, Keller E, Lee S, Yoshikawa TT. Utility of fever, white blood cells, and differential count in predicting bacterial infections in the elderly. J Am Geriat Soc 1989; 37:534–547. 34. Schoeinfeld CN, Hansen KN, Hexter DA, Stearns DA, Kelen GD. Fever in geriatric emergency patients: Clinical features associated with serious illness. Ann Emerg Med 1995; 26(1):18–24. 35. Esposito AL, Gleckman RA. Fever of unknown origin in the elderly. J Am Geriatr Soc 1978; 26:498–505. 36. Berland B, Gleckman RA. Fever of unknown origin in the elderly: A sequential ap- proach to diagnosis. Postgrad Med 1992; 92:197–210. 37. Knockaert DC, Vanneste LJ, Bobbaers HJ. Fever of unknown origin in elderly pa- tients. J Am Geriatr Soc 1993; 41:1187–1192.

7 Ethical Issues of Infectious Disease Interventions Elizabeth L. Cobbs Washington D.C. VA Medical Center, and George Washington University, Washington, D.C. I. INTRODUCTION Long-term care facilities (LTCFs) provide care to dependent persons with a vari- ety of needs and expectations and, in a shifting medical marketplace, ethical issues are part of the daily routine. In the wake of shortened hospital stays, the use of nurs- ing homes for subacute post-hospital stays has surged. Short-term residents may receive rehabilitation services or continuing medical treatment for serious illnesses such as osteomyelitis. Improvement in function and health is the usual goal, and discharge to the community is often expected. The other larger group of residents living in LTCFs is composed of frailer individuals who are likely to reside in the LTCF through the end of life. The nursing home is their home. Their expectations for medical care are varied, as are their abilities to make decisions and express treat- ment preferences. A subset of those residents are near the end of life. In addition to their medical complexity, LTCF residents are at increased risk for infectious diseases because of physiological changes associated with aging, the impact of chronic conditions, and the effects of institutional living (1). Infec- tion has been cited as the most frequent cause of transfer to the hospital (2), and hospital transfer is a frequent response to the medically ill resident in the LTCF, although practice varies. To serve this diverse group of residents, the LTCF is expected to provide timely and appropriate medical care, while at the same time offering a comfort- able, personalized residence. The dual task of meeting the medical needs of many residents and providing a homelike environment that delivers a pleasing quality of life for dependent frail persons creates the setting for a number of ethical dilem- 79

80 Cobbs mas faced by LTCF practitioners. “The ethical issues in any care system reflect the nature of the care provided, the setting in which the care takes place, the ca- pabilities of the care recipients, the commitments of the care providers, and the so- cial and financial arrangements that society has created to structure and reimburse activities of caring” (3). II. THE PLACE OF ETHICS IN LTCFs Knowledge of medical ethics helps physicians and other practitioners to do the right thing in the long-term care environment, where competition between medi- cal and humanistic agendas are typical. Several ethical themes are integral to life in LTCFs. A. Autonomy Autonomy refers to self-determination without overbearing external influence, a prized attribute in American society. Autonomy becomes a ubiquitous ethical con- cern in the LTCF because all those who live in this setting do so because their abil- ity to function independently has been compromised. Serving the needs of depen- dent residents within a medical model requires a reworking of the definition of autonomy. Perception and experience of autonomy are influenced by many fac- tors, including culture. The Milwaukee Hmong community, for example, per- ceives dementia not as a chronic disease that robs a person of autonomy but as a natural part of the life cycle. Individuals suffering from dementia are cared for in their sons’ homes and rarely display difficult behaviors such as combativeness and wandering (4). Autonomy in an LTCF is most often expressed through a pattern of living rather than through discrete decisions (3). B. Beneficience and Nonmaleficience Beneficience refers to doing good, and in the practice of medicine this translates to doing the right thing for the patient. Defining the “right thing” has undergone a shift in recent years with increased understanding of the needs of the growing pop- ulation of older adults living with chronic conditions (causing both physical and mental deficits) and significant self-care needs. Weighing the burdens and bene- fits of possible interventions has become standard practice in making treatment decisions for LTCF residents, because of the risks associated with virtually all treatment options. Closely related to beneficience is the admonition to do no harm, known as nonmaleficence. Nonmaleficence has taken on greater importance as the burdens of common treatment options, such as hospitalization, are recognized, and options for prompt and effective out-of-hospital treatments increase. The era

Ethical Issues 81 recognizing residents’ rights in LTCFs came about after the 1987 passage of the Omnibus Budget Reconciliation Act, with perhaps the most noticeable result be- ing a change of practice in the use of physical and chemical restraints. C. Fidelity Fidelity embraces trust and confidentially and is a fundamental principle under- lying the doctor-patient relationship. The doctor-patient relationship remains fun- damental to the care of residents in LTCFs; however, certain compromises to this relationship are inevitable because of the high frequency of cognitive impairment found among residents. Issues of trust pertain also to relationships with others on the interdisciplinary team (IDT) and often are important for residents to achieve a sense of control. D. Justice Justice refers to the equitable distribution of resources and treatment and is espe- cially relevant when the interests of residents, staff, the institution, and families come into conflict with each other. Competing demands for staff attention and re- sources create the need for individuals and systems to negotiate settlements to conflicts so that the needs of residents and others in the LTCF community are most equitably served. E. Everyday Ethics The need for everyday ethical principles to guide practice in LTCFs is derived from the complexity of the organizations, their objectives, and their many partic- ipants. The nursing home environment seeks to blend two very different cultures: the autonomous, individually controlled home with the externally regulated, physician-directed structure of a medical institution. The requirement for physi- cian orders to permit basic elements such as diet, activity level, or permission to self medicate sets an overarching framework of paternalism. Affirmation of self occurs with new expressions of autonomy manifested by the residents’ activities being consistent with their personal values and preferences. The LTCF must cre- ate systems that encourage consistent decisions by residents to maximize auton- omy, despite disability and the institutional setting (3). At times competition exists between interests. The residents are the primary customers of the facility, yet LTCFs have been criticized for a lack of attention to the values and preferences of the individual resident. With limited staffing, resi- dents may compete for staff time and attention. An acutely ill resident becomes the focus of attention, sometimes creating a shortage of staff time to attend to the needs of other residents. Decisions about whether to hospitalize a resident are in-

82 Cobbs fluenced by competing interests, including institutional financial incentives and staffing capacity, in addition to patient preferences. Conflicts may also occur in the interface between the optimum well-being of the residents and the needs and preferences of the staff. Conflict can also be found between staff and facility, and the facility and outside organizations. Families and significant others are important members of the LTCF community and contribute to the caregiving process. Families, how- ever, may add to the conflict between competing interests. The facility needs a system of regular conflict resolution that effectively and consistently resolves conflicts between competing interests and values (3). The resident faces many obstacles to maintaining a self that is capable of au- tonomous action. Providing opportunities for choice and exertion of control over the environment and participation in the decision-making process have been shown to positively influence resident life. Communication and negotiation are means to achieve the best possible outcomes for the resident as well as the staff and the institution. Many daily infection control decisions have ethical dimensions that require choices between competing concerns or values. Common issues include whether to isolate residents colonized with resistant organisms, whether an ill healthcare worker should be allowed to work, and whether to investigate clusters of infec- tions (5). Additional issues have to do with when to treat, what to treat, whether to hospitalize, how to communicate with residents and families, what to do when treatment attempts become burdensome and residents refuse, how to improve staff behaviors that protect resident safety and health, and when not to treat. III. GOALS OF CARE The development of individualized goals of care for each resident is a process that creates the best mechanism to maximize autonomy, quality of life, and desirable medical outcomes. In the process of developing goals of care, a comprehensive biopsychosocial assessment is performed. This assessment provides a framework for the integration of disease factors with psychosocial factors and other resident characteristics. This assessment yields measures of resident functional capacity and points out where interventions to improve function and enhance independence might be placed. The values’ history, including information about the resident’s preferences, hopes, fears, basis for meaning, spirituality, and personal goals, is integrated with the comprehensive assessment. The physician and resident (or surrogate) aim to reach a shared understanding of the resident’s health status, care needs and pref- erences, options for future treatment interventions, and likely outcomes. From this assessment, a blueprint of goals and plans for care can be developed. In this way,

Ethical Issues 83 Table 1 An Example of a Scheme to Prioritize Goals of Care Intensive Comprehensive Basic Palliative Comfort care care care care care only First Prolong life Maintain Maintain Maximize Maximize priority physical physical comfort comfort Maintain and and Second physical and cognitive cognitive Maintain priority cognitive function function physical function and Third Prolong life Maximize cognitive priority comfort function Maximize Maximize comfort comfort Prolong life Prolong life Source: Adapted from Ref. 8. residents of widely differing decisional capacity, health status, and personal pref- erences have deliberately articulated, personally generated (to the extent possible) plans for care to guide treatment decisions. The health values of the seriously ill vary considerably from person to per- son, and they cannot be easily predicted from the person’s current state of health. Mental health is an important factor in determining how patients evaluate their health (6). There is no substitute for involving the resident (or surrogate in the event of resident incapacity) in developing goals for care. Practitioners can expect considerable variation in preferences for care based on a variety of factors, in- cluding ethnicity (7). Many institutions are developing structured approaches to advance care planning, including prioritization of goals of care. One scheme is shown in Table 1 (8). IV. THE DOCTOR-PATIENT RELATIONSHIP Over the last several decades, the paternalistic model of medical care in the United States has given way to an increased emphasis on patient autonomy. Residents of LTCFs typically have significant cognitive impairment, physical frailty, and chronic disease that will be with them for the rest of their lives. These residents are likely to benefit from the “enhanced autonomy” model in which an active ex- change of ideas and negotiation takes place with the goal of achieving the best possible decision for the resident (9). In many cases, goals of care will be dis-