TJVM Vol. 48 No. 4 December 2018

658 Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. Figure 1 Kaplan-Meier survival curves for hypertrophic cardiomyopathy cats with arterial embolism (ATE) (regular line) and without ATE (bold line). The two curves differ significantly (p < 0.0001). Hypertrophic cardiomyopathy cats with ATE had a shorter survival time. Table 3 The percentage of hypertrophic cardiomyopathy cats categorized in comparison to survival time Parameters Category Numbers (%) P -value Age (years) Breed >7 and <7 27 (57.4) and 19 (40.4) 0.343 DSH and others 23 (48.9) and 24 (51.1) 0.401 Sex Persian and others 19 (40.4) and 28 (56.6) 0.944 DSH and Persian 23 (48.9) and 19 (40.4) 0.667 Male and Female 24 (51.1) and 23 (48.9) 0.163 Heart sound Normal and abnormal 17 (36.2) and 30 (63.8) 0.396 Blood pressure (mmHg) <160 and >160 27 (57.4) and 10 (21.3) 0.662 ATE Presence and absence 6 (12.8) and 41 (87.2) <0.0001 CKD Presence and absence 13 (27.7) and 34 (72.3) 0.657 Echocardiography Fractional shortening (%) <40 6 (12.8) 0.976 40-60 18 (38.3) >60 23 (48.9) Types of hypertrophy Asymmetry and symmetry 27 (57.4) and 20 (42.6) 0.736 LVIDd (mm) <10.8 and >10.8 7 (14.9) and 40 (85.1) 0.009 LA (mm) <16 and >16 23 (48.9) and 24 (51.1) 0.542 LA:Ao <2 and >2 27 (57.4) and 19 (40.4) 0.217 E:A <1 17 (36.2) 0.086 1-2 13 (27.7) >2 13 (27.7) Spontaneous contrast Presence and absence 13 (27.7) and 34 (72.3) 0.954 Medication ACEi and Furosemide With and without 23 (57.5) and 17 (42.5) 0.019 P<0.05 indicates statistically significant ACEi, angiotensin converting enzyme inhibitor; ATE, aortic thromboembolism; CKD, chronic kidney disease; DSH, domestic short hair; LVIDd, left ventricular internal dimension at end-diastole; LA, left atrium; LA:Ao, the ratio of left atrial dimension to the aortic annulus dimension; E:A, the ratio of peak velocity of early diastolic to late diastolic transmitral flow of mitral valve

Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. 659 Figure 2 Kaplan-Meier survival curves for the group of cats with normal left ventricular internal dimension at end-diastole (LVIDd) >10.8 mm (bold line) and <10.8 mm (regular line). The two curves differ significantly (p = 0.009). Cats with LVIDd >10.8 mm had a longer survival time. Figure 3 Kaplan-Meier survival curves for cats treated with angiotensin converting enzyme inhibitor (ACEi) and furosemide (regular line) and cats treated with other cardiovascular drugs (bold line). The two curves differ significantly (p = 0.019). Cats treated with ACEi and furosemide had a longer survival time.

65680 Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. Table 4 The percentage of hypertrophic cardiomyopathy cats without chronic kidney disease categorized in comparison to survival time Parameters Category Numbers (%) P -value Age (years) >7 and <7 14 (41.18) and 19 (55.88) 0.227 Breed DSH and others 17 (50) and 17 (50) 0.375 Persian and others 13 (38.24) and 21 (61.76) 0.906 Sex DSH and Persian 17 (50) and 13 (38.24) 0.652 Heart sound Blood pressure (mmHg) Male and female 18 (52.94) and 16 (47.06) 0.270 ATE Normal and abnormal 16 (47.06) and 18 (52.94) 0.757 <160 and >160 21 (61.76) and 3 (8.82) 0.746 Presence and absence 6 (17.65) and 28 (82.35) 0.001 Echocardiography <40 6 (17.65) 0.925 Fractional shortening (%) 40-60 12 (35.29) >60 16 (47.06) 0.562 Types of hypertrophy Asymmetry and symmetry 15 (44.12) and 19 (55.88) 0.001 LVIDd (mm) 0.195 LA (mm) <10.8 and >10.8 6 (17.65) and 28 (82.35) 0.506 LA:Ao 0.057 E:A <16 and >16 14 (41.18) and 20 (58.82) <2 and >2 17 (50) and 16 (47.06) 0.954 Spontaneous contrast <1 10 (29.41) Medication 1-2 12 (35.29) >2 10 (29.41) Presence and absence 12 (35.29) and 22 (64.71) ACEi and Furosemide With and without 5 (50) and 5 (50) 0.696 P<0.05 indicates statistically significant ACEi, angiotensin converting enzyme inhibitor; ATE, aortic thromboembolism; CKD, chronic kidney disease; DSH, domestic short hair; LVIDd, left ventricular internal dimension at end-diastole; LA, left atrium; LA:Ao, the ratio of left atrial dimension to the aortic annulus dimension; E:A, the ratio of peak velocity of early diastolic to late diastolic transmitral flow of mitral valve Table 5 Different median survival time in hypertrophic cardiomyopathy cats with different categories Parameters Category Median 95% CI All HCM cats ATE Presence 30 5-55 LVIDd (mm) Absence 505 125-885 <10.8 7 2-12 ACEi and Furosemide >10.8 339 14-664 With 505 156-854 Without 69 1-274 HCM cats without CKD ATE Presence 30 5-55 Absence 674 16-1332 LVIDd (mm) <10.8 5 1-10 ACEi and Furosemide >10.8 222 134-373 With 267 162-453 Without 63 1-127 ACEi, angiotensin converting enzyme inhibitor; ATE, aortic thromboembolism; CI, confident interval; CKD, chronic kidney disease; HCM, hypertrophic cardiomyopathy; LVIDd, left ventricular internal dimension at end-diastole Discussion and age did not affect the survival time based on the result of this study. A previous study demonstrated that Ragdoll and Main Coon cats have more chance to develop 63.8 % of cats in this study had abnormal HCM (Payne et al., 2010). Moreover, the survival time heart sound. Other studies found that 78-88.2% of cats of these breeds was shorter than other breeds. with HCM presented with abnormal heart sound However, Ragdoll and Main Coon cats are rare in (Payne et al., 2010; Rush et al., 2002). A previous study Thailand. The popular cat breeds in Thailand are DSH showed that the presence of a galloping heart sound and Persian, which are also the major population of was related to a worse survival time (Payne et al., HCM cats in this study. The median age of HCM cats 2013). However, the characteristics of heart sound did in this study was eight years (range 3.8-12.0 years). not affect the survival time in the group of cats in this This result is in agreement with a previous study study. suggesting that HCM usually develops in aging cats (Wess et al., 2010; Visser et al., 2017). However, breed There were variations in survival time in HCM cats reported previously including 1276 days in

Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. 661 England (Payne et al., 2010) and 732 days in the United Vachira Hunprasit for statistical analysis assistance. States (Atkins et al., 1992). However, the inclusion Also, we appreciate the cat owners for giving good criteria for cats in each study were different. The stage cooperation and providing information. of the disease, the treatment protocol and owner compliance might affect the survival time. Therefore, References the survival time of cats from each study may not be comparable. Atkins CE, Gallo AM, Kurzman ID and Cowen P 1992. Risk factors, clinical signs and survival in cats There was a study reporting that up to 21% of with a clinical diagnosis of idiopathic cats with HCM develop ATE (Fuentes, 2012). 12.8% of hypertrophic cardiomyopathy: 74 cases (1985- cats in the present study had ATE. These ATE cats had 1989). J Am Vet Med Assoc. 201: 613-618. a shorter survival time and a higher risk of death than those cats without ATE. This result suggests that signs Boon J 2017. Measurement and assessment of 2-D and of ATE are a good indicator for a poorer prognosis in M-mode images. In: Two-dimensional and M- cats affected with HCM. mode echocardiography for the small animal practitioner. 2nd ed. Iowa: John Wiley and Sons, A decrease of LVIDd was significantly Inc. 95. associated with decreased survival time in the group of cats in this study. This result is in agreement with the Fuentes VL 2012. Arterial thromboembolism risks, result of a prior study (Sugimoto et al., 2015). The realities and a rational. J Feline Med Sur. 14(7): smaller the left ventricular chamber, the less blood 459-470. volume fills up the ventricle resulting in a decrease in cardiac output. Eventually, the left atrium will become Gordon SG and Coˆt e E 2015. Pharmacotherapy of enlarged and left-sided congestive heart failure will feline cardiomyopathy: chronic management of develop. From these consequences, a decrease in heart failure. J Vet Cardiol. 17: 159-172. LVIDd can increase the risk of cardiac death in HCM cats. This presumption supports the result of this study Payne JR, Borgeat K, Connolly DJ, Boswood A, Dennis that found a higher risk of death of approximately S, Wagner T, Menaut P, Maerz I, Evans D, Simons three times in cats with a small ventricular chamber VE, Brodbelt DC and Fuentes VL 2013. Prognostic size than those with a normal ventricular chamber size. indicators in cats with hypertrophic cardiomyopathy. J Vet Intern Med. 27: 1427-1436. Hypertrophic cardiomyopathy cats treated with ACEi and furosemide stayed longer than those Payne J, Fuentes VL, Boswood A, Connolly D, Koffas treated with other cardiovascular drugs besides ACEi H and Brodbelt D 2010. Population characteristics and furosemide. Furosemide helps HCM cats by and survival in 127 referred cats with reducing excess extravasation fluid in peripheral hypertrophic cardiomyopathy (1997 to 2005). J tissues. Angiotensin converting enzyme inhibitor Small Anim Pract. 51: 540-547. inhibits renin angiotensin aldosterone system (RAAS) and has benefits to HCM cats by reducing preload and Rush JE, Freeman LM, Fenollosa NK and Brown JD afterload increasing the amount of blood being 2002. Population and survival characteristics of pumped by the heart (Gordon and Coˆt e, 2015). For cats with hypertrophic cardiomyopathy: 260 these reasons, the combination use of these two drugs, cases (1990-1999). J Am Vet Med Assoc. 220(2): furosemide and ACEi would provide benefits by 202-207. increasing survival time particularly in HCM cats with CHF. Sugimoto K, Fuji Y, Sunahara H and Aoki T 2015. Assessment of left ventricular longitudinal Although the study was carefully prepared, function in cats with subclinical hypertrophic data from some cats was missed due to the limitation cardiomyopathy using tissue Doppler imaging in gathering the information of cats from the owners. and speckle tracking echocardiography. J Vet This limitation resulted from the retrospective design Med Sci. 77(9): 1101-1108. of the study. Because the missing data was varied in each parameter, the multivariate regression analysis Visser LC, Sloan CQ, and Stern JA 2017. could not be performed. Echocardiographic assessment of right ventricular size and function in cats with In conclusion, this study showed that the hypertrophic cardiomyopathy. J Vet Intern Med. presence of ATE and the size of the left ventricular 31: 668-677. chamber during diastole are related to shorter survival time in HCM cats with CHF. These parameters should Wess G, Sarkar R and Hartmann K 2010. Assessment be included in the diagnostic investigation and the of left ventricular systolic function by strain prognosis of the disease. Treating HCM cats with CHF echocardiography in various stages of feline by ACEi and furosemide helps cats stay longer than hypertrophic cardiomyopathy. J Vet Intern Med. treating by other cardiovascular drugs besides ACEi 24: 1375-1382. and furosemide. Lastly, this study reports an estimated survival time of HCM cats with CHF in Thailand. White AJM 2015. End-stage hypertrophic cardiomyopathy in a cat. Can Vet J. 56(5): 509-511. Acknowledgements We would like to express gratitude for the grant from academic affair, Faculty of Veterinary Science Chulalongkorn University and thank Dr.

662 Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. บทคัดย่อ ระยะเวลาการอยรู่ อด และปัจจยั ในการพยากรณ์โรคกลา้ มเนอื้ หวั ใจ ผดิ ปกติในแมวทีม่ ภี าวะหวั ใจลม้ เหลว ลลดิ า ตันตสิ ุวัฒน์1 ฐิตินันท์ ต้ังอาษาศลิ ป์1 ปานดวงใจ พวงอาไพ1 พรชนก ปานภกั ดี1 สริ ิลักษณ์ สุรเชษฐพงษ์2* โรคกลา้ มเนื้อหัวใจหนาผิดปกตเิ ป็นโรคกล้ามเนื้อหัวใจทพ่ี บไดบ้ อ่ ยในแมว ปัจจบุ ันยงั ไม่มีการศกึ ษาถงึ ระยะเวลาการอยรู่ อดของ แมวทเ่ี ปน็ โรคกลา้ มเนื้อหวั ใจหนาผิดปกติในประเทศไทย การศกึ ษาน้มี ีวัตถุประสงคเ์ พ่อื หาระยะเวลาการรอดชวี ติ และปัจจยั ทม่ี ผี ลต่อ ระยะเวลาการรอดชีวิตในแมวทเ่ี ปน็ โรคกล้ามเน้อื หัวใจหนาผดิ ปกติรว่ มกับภาวะหัวใจล้มเหลว การศึกษาน้เี ปน็ การศกึ ษาขอ้ มลู ยอ้ นหลงั ใน แมวท้งั หมด 47 ตัว เพ่ือวเิ คราะห์หาระยะเวลาการรอดชีวติ และความสัมพนั ธข์ องระยะเวลาการรอดชวี ิตกบั ปัจจัยต่างๆ ไดแ้ ก่ เพศ อายุ พนั ธ์ุ น้าหนัก ผลการตรวจรา่ งกาย การปรากฏของโรคภาวะแทรกซอ้ น การรกั ษาทางยา และผลการตรวจหวั ใจดว้ ยคลน่ื สยี งสะทอ้ นความถสี่ ูง น้า ข้อมูลมาวิเคราะหท์ างสถิติ โดยใช้ Log-rank test และแสดงผลในรปู แบบของกราฟ Kaplan-Meier โดยผลการศึกษาพบว่าแมวท่เี ปน็ โรค กล้ามเนือ้ หวั ใจหนาผิดปกติและมภี าวะหวั ใจล้มเหลวมรี ะยะเวลาการรอดชีวติ 283 วัน และปจั จัยทท่ี า้ ให้ระยะเวลาการรอดชวี ติ สั้นลงคือ การ ปรากฏของภาวะลมิ่ เลือดอดุ ตนั ในหลอดเลอื ดแดงเอออรต์ า ขนาดของหัวใจหอ้ งลา่ งซ้ายขณะคลายตัวสดุ ที่น้อยกวา่ 10.8 มลิ ลเิ มตร และการ ใช้ยารกั ษาโรคหวั ใจชนิดอน่ื นอกจากการใช้ยากลมุ่ ยับยั้งเอนไซมแ์ องจิโอเทนซินคอนเวอร์ตต้งิ รว่ มกบั ยาฟโู รซไี มด์ ขอ้ มูลจากการศกึ ษานมี้ ี ประโยชน์ตอ่ การพยากรณ์โรคและการวางแผนการจัดการในแมวทีเ่ ปน็ โรคกล้ามเน้ือหัวใจหนาผดิ ปกติ คาสาคญั : แมว ภาวะหัวใจล้มเหลว โรคกล้ามเนอ้ื หวั ใจหนาตัวผดิ ปกติ ระยะเวลาการรอดชีวติ 1นสิ ติ ช้ันปที ่ี 6 ปกี ารศกึ ษา 2560 คณะสัตวแพทยศาสตร์ จฬุ าลงกรณม์ หาวทิ ยาลัย 39 อังรีดนู ังต์ เขตปทุมวัน กรงุ เทพฯ 10330 ประเทศไทย 2ภาควิชาอายุรศาสตร์ คณะสัตวแพทยศาสตร์ จฬุ าลงกรณ์มหาวิทยาลยั 39 องั รดี ูนังต์ เขตปทุมวัน กรุงเทพฯ 10330 ประเทศไทย *ผรู้ ับผิดชอบบทความ E-mail: [email protected]

Original Article Emergence of a classical variant of porcine epidemic diarrhea virus novel to Thailand responsible for the milder clinical disease in a herd previously infected with a pandemic variant Gun Temeeyasen1 Anchalee Srijangwad1 Thitima Tripipat1 Angkana Tantituvanont2 Dachrit Nilubol1* Abstract In 2014, porcine epidemic diarrhea (PED) outbreaks characterized by mild clinical disease and low pre- weaning mortality increased in frequency in Thailand. An outbreak investigation, based on the detection of the classical variant and the allowance of the herd owners, was conducted in 5 swine herds and 25 intestinal samples were submitted from herds with outbreaks of diarrhea for sequencing analysis. PEDV isolation and complete nucleotide sequences of the spike (S) glycoprotein were determined. The PEDV isolate EAS1 was identified. Phylogenetic analysis based on the full-length spike gene demonstrated that EAS1 belonged to the classical variant group with high nucleotide and amino acid identity (98.4-99.0% and 96.6-97.9%) with LZC and SM98, which are the Chinese and Korean classical variant PEDVs, respectively. The results suggest the emergence of a classical variant PEDV responsible for the mild clinical disease in Thailand. The introduction could have been due to the use of illegally imported modified live vaccines from other countries. Keywords: Emergence, Novel, Porcine epidemic diarrhea virus, Spike gene, Thailand 1Department of Veterinary Microbiology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand 2Department of Pharmaceutics and Industrial Pharmacy, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok 10330, Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 663-669.

664 Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. Introduction have PEDV sequencing data from previous PED outbreak to compare the genetic difference between Porcine epidemic diarrhea (PED), an enteric the PEDV responsible for previous and current disease characterized by acute severe watery diarrhea outbreaks. With these three criteria, only five herds and high mortality (Pensaert and De Bouck, 1978), is were selected for further investigation. Face-to-face caused by the PED virus (PEDV), a virus in the family interviews were conducted at the farm. In the Coronaviridae. Two PEDV variants are recognized: interview, information based on herd size, production classical and pandemic variants (Sun et al., 2015). The type, PEDV vaccination use, killed (KV) or modified spike gene is a distinguishing feature. The pandemic live virus (MLV) vaccine, how long had PEDV vaccines variant contains two insertions of 4 (56GENQ59) and 1 been used, frequency of repeated PEDV outbreaks and (140N) amino acids at positions 55-60 and 140, pre-weaning mortality following the current outbreak respectively, and a deletion of 2 amino acids (160DG161) was collected. at positions 160-161 (Li et al., 2012) and has been reported as an emerging variant worldwide (Li et al., Source of specimens and PEDV isolation: Twenty-five 2012; Chen et al., 2014; Lee and Lee, 2014; Lin et al., intestinal samples were collected from 3-4-day-old pigs 2014; Hanke et al., 2015; Masuda et al., 2015; Ojkic et displaying severe watery diarrhea from 5 affected pig al., 2015; Sun et al., 2015; Theuns et al., 2015; Vui et al., farms. The intestinal samples were submitted to the 2015). laboratory of Dr.Dachrit Nilubol from December 2013 to December 2014. The samples were assayed for the PED first emerged in Thailand in late 2007 presence of swine coronaviruses including PEDV, (Puranaveja et al., 2009). A herd in Nakhon Pathom, a transmissible gastroenteritis virus (TGEV) and porcine province in the Western region of Thailand, reported a deltacoronavirus (PDCoV). Detection for PEDV and disease outbreak characterized by severe diarrhea and TGEV was performed following the previously high mortality. The pandemic outbreak was then described protocols using specific primers for spike (S) reported across Thailand. PEDV was detected and the gene of PEDV (Park et al., 2007), and specific primers variant responsible for the outbreaks and circulating in for N gene of TGEV (Kim et al., 2000). To screen for the Thailand, was a pandemic variant closely related to presence of PDCoV, PCR amplification was performed Chinese and Korean isolates (Temeeyasen et al., 2014). on cDNA using specific primers for the N gene of Since then, PED has become endemic and causes PDCoV as previously described (Wang et al., 2014). sporadic outbreaks in which outbreaks of one or two times a year have been reported. Normally, repeated All intestinal samples positive for PEDV by outbreaks of 4-6 months apart have been observed. PCR were subjected to PEDV isolation using the same However, in late 2013, repeated PED outbreaks with a protocol as the previous study (Temeeyasen et al., shorter duration and a milder clinical disease were 2014). In brief, intestinal samples were minced into reported in several herds. It is worthy of note that small pieces and suspended in 10 mL of phosphate outbreaks reoccurred within 2-3 months after a buffer saline (PBS; 0.1 M, pH 7.2). Suspended samples previous outbreak and displayed milder clinical were centrifuged at 4,800xg for 10 min. Supernatant disease compared to the earlier outbreaks. was filtered through 0.22 m filters and inoculated into a 25 cm2 flask that previously estimated the fully It is unknown whether a novel variant of confluent Vero cells. The inoculated flask was PEDV with milder clinical severity was introduced into incubated at 37 C in 5% CO2 for 1 hour, then replaced Thailand or if previous isolates evolved until they were by 10 ml MEM 2% FBS and incubated again for a no longer recognized by the immune system. In order period of 2-3 days and daily observed for CPE. The to identify the PEDV variants responsible for the recent CPE positive flask was frozen and thawed 2 times outbreaks, intestinal samples from herds experiencing before being centrifuged at 2,500 rpm for 10 min to PED outbreaks with mild clinical disease were collect the supernatant and store at -80 C. collected. PEDV was isolated and the full-length spike gene obtained from intestinal samples was Reverse transcription-polymerase chain reaction: characterized and compared to spike gene sequencing Supernatant from PEDV isolation was subjected to data previously reported. RNA extraction using the Nucleospin®RNA Virus (Macherey-Nagel Inc., PA, USA) in accordance with Materials and Methods the manufacturer’s instructions and synthesized cDNA from the extracted RNA using M-MuLV Herd selection and investigation: A disease Reverse Transcriptase (New England BioLabs Inc., investigation was conducted in swine herds that MA, USA). Complete S genes PCR amplification was experienced PED outbreaks in Eastern and Western performed on the cDNA using previously described regions of Thailand from July 2014 to December 2014. primers (Temeeyasen et al., 2014). PCR products were Herd selection was based on three criteria. Firstly, the visualized by agarose gel electrophoresis. permission of the herd owners. Secondly, emphasis was placed on investigating swine herds that Cloning, plasmid purification and sequence experienced a reoccurrent outbreak of PED with mild determination: PCR products were cloned into clinical disease. Mild clinical disease characterized by plasmid vectors for the subsequent transformation of PED associated clinical signs included vomiting, Escherichia coli cells using a commercial kit (pGEM-T® diarrhea and dehydration but mortality was low Easy Vector System I (Promega, WI USA), and the compared to PED previously experienced. Low controls were included at all stages of cloning and mortality was approximately 30-50%. Finally, the transformation. Two to three bacterial transformant availability of PEDV sequencing data from previous outbreaks. The investigated herds were required to

Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. 665 colonies were randomly selected from each sample for outbreak were observed according to the PEDV plasmid purification using the Nucleospin® Plasmid variants that were successfully isolated during the kit (Macherey-Nagel Inc., PA, USA) and the selected recurrent outbreaks. In two of the five herds, the colonies were grown in LB broth for 24 h and subjected recurrent PEDV outbreak occurred 16-24 weeks after to plasmid isolation and sequencing. Sequencing was the previous outbreaks and the causes of repeated PED performed by First BASE Laboratories Inc. (Selangor, outbreaks were pandemic variant. The mortality rates Malaysia) using an ABI Prism 3730XL DNA sequencer. in these herds were 56.9 and 52.1% that were higher compared to those observed in herds in which the Sequence analyses: Nucleotide and deduced amino recurrent PEDV outbreaks had occurred within 3 acid sequences were aligned using the CLUSTALW months. Three of five herds had a recurrent outbreak program (Thompson et al., 1994). A phylogenetic within 12 week of the previous outbreak and the analysis was constructed based on complete S genes classical variant of PEDV were the cause of repeated from this study and as previously reported. A outbreaks. The mortality rates in these 3 herds were midpoint-rooted neighbor-joining tree was generated 32.5%, 33.1% and 28.7%, respectively (Table 1). based on the Kimura 2-parameter model using MEGA4 (Tamura et al., 2007). The robustness of the Phylogenetic analysis of complete spike gene: All 25 phylogenetic analysis and significance of the branch intestinal samples were PEDV-positive. The full-length order were determined by bootstrap analysis with spike gene of 25 PEDV isolates was sequenced. To 1,000 replicates. The percentage of homology between investigate the heterogeneity, the 25 full-length spike the isolates at the nucleotide and amino acid levels was genes (accession number KR610991-KR610993 and calculated. KR941552-KR941557) were aligned with 81 previously published PEDV spike genes, including the previously Results reported Thai PEDV. The phylogenetic analysis revealed that all PEDV isolates were grouped into 3 Herd investigation: Five herds experiencing PEDV clusters: G1, G2 and G3 (Figure 1). outbreaks were investigated. Two patterns of recurrent Table 1 Radiographic scoring system for evaluation of bone healing (Johnson et al., 1996b). Herds Geographic Herd size Use of PEDV variants PEDV variants Duration from a Pre-weaning responsible for previous mortality (%) during region (sows) modified live responsible for recurrent outbreak outbreak recurrent outbreak PEDV vaccine previous outbreak DT Central 1,300 Yes Pandemic Classical 8 weeks 32.5 JK East 2,000 Yes Pandemic Classical 9 weeks 33.1 ET East 2,000 No Pandemic Classical 12 weeks 28.7 SA West 2,000 No Pandemic Pandemic 16 weeks 56.9 PR West 1,200 No Pandemic Pandemic 24 weeks 52.1 The phylogenetic analysis demonstrated that performed. The sequencing result demonstrated the Thai PEDV is presently categorized into two vaccine isolate was similar to the classical variant genetically distinct clusters: G1 and G3. Four Thai PEDV isolates in this study (Figure 1). Nucleotide and PEDVs isolated in 2014 included in G1 are novel to amino acid identity between the vaccine isolate and the Thailand. Comparative analyses of the full-length classical isolates in this study were 99.8% and 99.4%, spike genes demonstrated that the difference between respectively. both clusters lies in the characteristics of the spike gene. The PEDV isolates in G3 contained two insertions of 4 Discussion (56GENQ59) and 1 (140N) amino acids at positions 55-60 and 140 of the S gene, respectively, and a deletion of 2 Since its first emergence in Thailand in 2007, amino acids (160DG161) at positions 160-161. A previous PED has become endemic in Thai swine farms and the study reported that Thai PEDV is a pandemic variant PEDV responsible for disease outbreaks in Thailand belonging only to G3. The emergence of PEDV in G1 has been a pandemic variant. A classical variant of suggests that this is a novel isolate to Thailand. The PEDV had not been identified (Temeeyasen et al., spike gene sequencing results showed that the newly 2014). However, in late 2013, several swine farms isolated PEDV, Thai PEDV in G1, shared very high reported PED outbreaks with mild clinical disease and nucleotide and amino acid identity (98.4-99.0% and mainly low pre-weaning mortality. Intestinal samples 96.6-97.9%, respectively) with LZC and SM98, which were then submitted for further genetic are the Chinese and Korean classical variant PEDVs, characterization. The presence of other coronaviruses respectively. Vaccine virus was isolated from the including TGE and PCDoV was ruled out due to the vaccine vial and the sequence of the vaccine isolate was negative results of PCR. A disease investigation was conducted following the analysis of the sequencing

666 Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. data. To compare the effect of repeated outbreaks and amino acid, respectively, with a vaccine variant caused by the classical variant and pandemic variant, from Korea (SM98). The emergence of a classical the disease severity among these five investigated variant of PEDV related to modified live vaccine herds was determined using pre-weaning mortality. A (MLV) was not surprising. Following PEDV outbreak, previous study demonstrated the different several swine herders have utilized the illegally pathogenicity of the classical variant and pandemic imported MLV to control PEDV. The vaccination variant PEDV in the experimental challenge model in program includes pre-farrow vaccination with two which piglets orally challenge with a S-INDEL variant doses of MLV. The vaccine virus in MLV may not be strain, the classical variant PEDV, exhibited milder properly attenuated and could become infectious in clinical outcomes when compared to that of a non-S- vaccinated sows with the potential to shed the vaccine INDEL, the US-PEDV prototype that genetically virus in the environment. The spill over from gestation related to the pandemic variant PEDV (Chen et al., unit to the farrowing unit might potentially occur 2016). The findings from the previous study are similar through the worker and fomites which could carry the to the results of the pre-weaning mortality observed in virus from unit to unit (Lowe et al., 2014). In addition, the present investigation in which the mortality rate in sows could potentially shed the vaccine virus to their herds repeated outbreaks with pandemic variant piglets, which could lead to a severe disease outbreak. PEDV was high. In contrast, the herds with repeated The route of excretion requires further investigation. outbreaks with classical variant PEDV showed milder clinical disease and a lower mortality rate. However, The detection of a vaccine-like variant in the one might speculate that the milder clinical disease and herd ET is of interest. Even through, the MLV was not lower mortality rate of the repeat outbreaks by classical used in the herd the herd was located in a one of the variant PEDV could be due to the immune status of the highest densities of the swine producing areas in the herds. Due to one of three criteria used to recruit PED Eastern region of Thailand. The virus could have been outbreak herds into the present investigation showed spread from some farms through airborne the genetic sequencing data of PEDV from previous transmission which have evidence of 16 kilometers outbreaks in which the PED-positive herds where the (Alonso et al., 2014), or from other route including information from previous study indicated the transport trailer (Lowe et al., 2014) or dead hauling pandemic variant PEDV was responsible for the truck. outbreaks. The herd immunity was supposed to be specific immunity against pandemic variant PEDV. A One controversy would be that the vaccine previous study demonstrated the decline of PEDV- virus might have been in this region for quite some specific antibody level after 6 months post infection time. However, based on previous reports, this (Ouyang et al., 2015). It is suggested that the duration hypothesis is still debatable. The first report of PEDV of the protection immunity should potentially last for emerging in Thailand suggested that, based on the 6 months. The findings are similar to the results of the partial spike sequencing data, the PEDV variant present study in which the duration between the responsible for the outbreaks in Thailand was Chinese- previous outbreak and the repeated outbreak by the like (Puranaveja et al., 2009). However, another pandemic variant PEDV varies between 4 to 6 months. investigation, that analyzed the nucleotide and amino In contrast, herds in this present study with repeated acid sequence based on complete spike and ORF3 outbreaks by the classical variant PEDV, the duration genes, found that the variants responsible in Thailand between the previous outbreak and repeated outbreak were all pandemic variants. In addition, the sequence was relatively shorter than 6 months. These results analysis of ORF3 gene indicated none of the PEDV might be associated with the cross-protection activity isolates was genetically related to MLV vaccines at that between classical and pandemic variants of PEDV time (Temeeyasen et al., 2014). In addition, further which can vary from partial to none. Therefore, the evolution analyses including the identification of potential cross protection between the classical and natural selection in different gene regions involving in pandemic variants should be further investigated. the evolutionary process of PEDV suggesting that the emergence of classical PEDV variant in Thailand was The introduction of classical PEDV was due to the external introduction, rather than the investigated in these herds. The source of the continuous evolution in the population (Cheun- Arom introduction could be an illegally smuggled modified et al., 2016). live PEDV vaccine (MLV) recently used as a pre-farrow vaccine in sow herds. In herds DT and JK, PED In summary, the results of the study outbreaks were observed a few days following the demonstrated the emergence of a novel PEDV in administration of the MLV. A few bottles of the MLV Thailand that is genetically related to the classical were collected and the genetic characterization of the variant. The introduction, molecular epidemiology and vaccine virus suggests that the vaccine virus and the cross protective immunity against the pandemic virus causing the outbreak were genetically related. An variant previously existing are of interest and require additional investigation of the genetic sequence of this further investigation. vaccine demonstrated that the full-length spike gene of the vaccine virus shares 99.8% and 99.4% similarity Acknowledgements with the classical PEDV variant responsible for the outbreak in this study at the nucleotide and amino acid We gratefully acknowledge the graduate levels, respectively. In addition, the genetic scholarship program of Chulalongkorn University. characterization of the classical PEDV variant in this This study was supported by the National Research study shared 99.0% and 97.9% similarity at nucleotide Council of Thailand and the Ratchadaphiseksomphot Endowment Fund 2013 of Chulalongkorn University (CU-56-527-HR).

Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. 667 Figure 1 Phylogenetic analysis of porcine epidemic diarrhea virus (PEDV) isolates based on the nucleotide sequences of the complete S glycoprotein genes. The tree was constructed using the neighbor-joining method with the Kimura 2-parameter and bootstrap re-sampling (1,000 replications). Blue filled triangles indicate isolation before 2013. Red filled dots indicate isolation in 2014. Unfilled squares indicate isolation from vaccine.

668 Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. References immune status in porcine epidemic diarrhea virus (PEDV) infected sows under field Alonso C, Goede DP, Morrison RB, Davies PR, Rovira conditions. Vet Res. 46: 140. A, Marthaler DG and Torremorell M 2014. Park SJ, Moon HJ, Yang JS, Lee CS, Song DS, Kang BK Evidence of infectivity of airborne porcine and Park BK 2007. Sequence analysis of the epidemic diarrhea virus and detection of airborne partial spike glycoprotein gene of porcine viral RNA at long distances from infected herds. epidemic diarrhea viruses isolated in Korea. Vet Res. 45: 73. Virus Genes. 35: 321-332. Pensaert M and De Bouck P 1978. A new coronavirus- Chen Q, Li G, Stasko J, Thomas JT, Stensland WR, like particle associated with diarrhea in swine. Pillatzki AE, Gauger PC, Schwartz KJ, Madson D, Archives of virology. 58: 243-247. Yoon KJ, Stevenson GW, Burrough ER, Harmon Puranaveja S, Poolperm P, Lertwatcharasarakul P, KM, Main RG and Zhang J 2014. Isolation and Kesdaengsakonwut S, Boonsoongnern A, characterization of porcine epidemic diarrhea Urairong K, Kitikoon P, Choojai P, Kedkovid R, viruses associated with the 2013 disease outbreak Teankum K and Thanawongnuwech R 2009. among swine in the United States. J Clin Chinese-like strain of porcine epidemic diarrhea Microbiol. 52: 234-243. virus, Thailand. Emerg Infect Dis. 15: 1112-1115. Sun M, Ma J, Wang Y, Wang M, Song W, Zhang W, Lu Cheun-Arom T, Temeeyasen G, Tripipat T, C and Yao H 2015. Genomic and epidemiological Kaewprommal P, Piriyapongsa J, Sukrong S, characteristics provide new insights into the Chongcharoen W, Tantituvanont A and Nilubol phylogeographical and spatiotemporal spread of D 2016. Full-length genome analysis of two porcine epidemic diarrhea virus in Asia. J Clin genetically distinct variants of porcine epidemic Microbiol. 53: 1484-1492. diarrhea virus in Thailand. Infect Genet Evol. 44: Tamura K, Dudley J, Nei M and Kumar S 2007. 114-121. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Hanke D, Jenckel M, Petrov A, Ritzmann M, Stadler J, Evol. 24: 1596-1599. Akimkin V, Blome S, Pohlmann A, Schirrmeier H, Temeeyasen G, Srijangwad A, Tripipat T, Beer M and Hoper D 2015. Comparison of porcine Tipsombatboon P, Piriyapongsa J, Phoolcharoen epidemic diarrhea viruses from Germany and the W, Chuanasa T, Tantituvanont A and Nilubol D United States, 2014. Emerg Infect Dis. 21: 493-496. 2014. Genetic diversity of ORF3 and spike genes of porcine epidemic diarrhea virus in Thailand. Kim O, Choi C, Kim B and Chae C 2000. Detection and Infect Genet Evol. 21: 205-213. differentiation of porcine epidemic diarrhoea Theuns S, Conceicao-Neto N, Christiaens I, Zeller M, virus and transmissible gastroenteritis virus in Desmarets LM, Roukaerts ID, Acar DD, Heylen E, clinical samples by multiplex RT-PCR. The Matthijnssens J and Nauwynck HJ 2015. Veterinary record. 146: 637-640. Complete genome sequence of a porcine epidemic diarrhea virus from a novel outbreak in Lee S and Lee C 2014. Outbreak-related porcine Belgium, january 2015. Genome Announc. 3(3). epidemic diarrhea virus strains similar to US Thompson JD, Higgins DG and Gibson TJ 1994. strains, South Korea, 2013. Emerg Infect Dis. 20: CLUSTAL W: improving the sensitivity of 1223-1226. progressive multiple sequence alignment through sequence weighting, position-specific Li W, Li H, Liu Y, Pan Y, Deng F, Song Y, Tang X and gap penalties and weight matrix choice. Nucleic He Q 2012. New variants of porcine epidemic Acids Res. 22: 4673-4680. diarrhea virus, China, 2011. Emerg Infect Dis. 18: Vui DT, Thanh TL, Tung N, Srijangwad A, Tripipat T, 1350-1353. Chuanasa T and Nilubol D 2015. Complete genome characterization of porcine epidemic Lin CN, Chung WB, Chang SW, Wen CC, Liu H, Chien diarrhea virus in Vietnam. Arch Virol. 160(8): CH and Chiou MT 2014. US-like strain of porcine 1931-1938. epidemic diarrhea virus outbreaks in Taiwan, Wang L, Byrum B and Zhang Y 2014. Porcine 2013-2014. J Vet Med Sci. 76: 1297-1299. coronavirus HKU15 detected in 9 US states, 2014. Emerg Infect Dis. 20: 1594-1595. Lowe J, Gauger P, Harmon K, Zhang J, Connor J, Yeske P, Loula T, Levis I, Dufresne L and Main R 2014. Role of transportation in spread of porcine epidemic diarrhea virus infection, United States. Emerg Infect Dis. 20: 872-874. Masuda T, Murakami S, Takahashi O, Miyazaki A, Ohashi S, Yamasato H and Suzuki T 2015. New porcine epidemic diarrhoea virus variant with a large deletion in the spike gene identified in domestic pigs. Arch Virol. 160(10): 2565-2568. Ojkic D, Hazlett M, Fairles J, Marom A, Slavic D, Maxie G, Alexandersen S, Pasick J, Alsop J and Burlatschenko S 2015. The first case of porcine epidemic diarrhea in Canada. Can Vet J. 56: 149- 152. Ouyang K, Shyu DL, Dhakal S, Hiremath J, Binjawadagi B, Lakshmanappa YS, Guo R, Ransburgh R, Bondra KM, Gauger P, Zhang J, Specht T, Gilbertie A, Minton W, Fang Y and Renukaradhya GJ 2015. Evaluation of humoral

Temeeyasen G. et al. / Thai J Vet Med. 2018. 48(4): 663-669. 669 บทคดั ย่อ การอุบัตใิ หมข่ องเช้ือไวรสั porcine epidemic diarrhea สายพันธุ์ดง้ั เดิมในประเทศไทย ซึ่งกอ่ ใหเ้ กิดโรคชนดิ อาการไมร่ ุนแรงในฝูงสุกรที่เคยมีการตดิ เชอ้ื สายพนั ธ์ุใหม่ กญั จน์ เตมยี ะเสน1 อัญชลี ศรีจังหวัด1 ธิตมิ า ไตรพิพฒั น์1 องั คณา ตนั ตธิ วุ านนท์2 เดชฤทธิ์ นลิ อบุ ล1* ในปี ค.ศ. 2014 พบวา่ การระบาดของโรคทอ้ งร่วงติดต่อในสุกร (โรคพีอีดี) ในฟารม์ สกุ รของประเทศไทยมลี กั ษณะเปลี่ยนแปลงไป โดยมคี วามรนุ แรงของโรคและอัตราการตายของลูกสุกรก่อนหย่านมนอ้ ยลงเมือ่ เปรยี บเทียบกับการระบาดกอ่ นหน้านั้น แตพ่ บความถ่ใี นการ ระบาดสูงขนึ้ จึงไดท้ าการสบื สวนหาสาเหตกุ ารระบาดของโรคโดยการตรวจสอบลกั ษณะทางพันธกุ รรมของเชอื้ ไวรัสพีอีดใี นฟารม์ ท่ีแสดง อาการของโรคไม่รุนแรง โดยไดด้ าเนินการสืบสวนโดยการรบั ตัวอย่างลาไส้จานวน 25 ตัวอยา่ ง จากสกุ รจานวน 5 ฝงู ที่มกี ารระบาดของโรคพอี ี ดี เพ่อื ทาการแยกเช้อื ไวรัสพีอีดีและวิเคราะหล์ าดับสารพนั ธุกรรมของยนี สไปคช์ นดิ ทั้งสาย พบว่าเช้อื ไวรสั พอี ีดไี อโซเลต อีเอเอส 1 ทไ่ี ดจ้ าก การเพาะแยกเชอื้ จัดอยูใ่ นกลุ่มสายพันธด์ุ ัง้ เดิม และมีลาดบั นิวคลีโอไทดแ์ ละลาดบั ของกรดอะมิโนท่เี หมือนกบั ไอโซเลต แอลแซสซี และไอโซ เลต เอสเอ็ม 98 ซึ่งเป็นไวรสั สายพนั ธุ์ดัง้ เดมิ ทีพ่ บในประเทศจนี และประเทศเกาหลใี ต้ถึง 98.4-99% และ 96.6-97.9% ตามลาดบั จากผล การศกึ ษานแี้ สดงให้เหน็ วา่ การระบาดของโรคพอี ีดีชนิดอาการไมร่ นุ แรงในประเทศไทยนนั้ มาจากการระบาดของเชอ้ื ไวรสั พอี ดี สี ายพันธุ์ ดง้ั เดมิ ซึง่ สาเหตขุ องการระบาดเข้ามาของเชื้อดงั กลา่ ว อาจมาจากการลกั ลอบนาวคั ซีนเช้ือเปน็ จากต่างประเทศเขา้ มาใช้อย่างไมเ่ หมาะสม คาสาคญั : การอบุ ัติ ใหม่ ไวรสั porcine epidemic diarrhea ยีนสไปค์ ประเทศไทย 1ภาควิชาจุลชวี วิทยา คณะสัตวแพทยศาสตร์ จฬุ าลงกรณ์มหาวทิ ยาลัย กรุงเทพฯ 10330 ประเทศไทย 2ภาควิชาวิทยาการเภสัชกรรมและเภสชั อตุ สาหกรรม คณะเภสชั ศาสตร์ จุฬาลงกรณ์มหาวิทยาลยั กรงุ เทพฯ 10330 ประเทศไทย *ผู้รับผิดชอบบทความ E-mail: [email protected]



Original Article Effectiveness of sentinel rodents for surveillance of exposure to undocumented bacterial pathogens in animal research facility Shih-Keng Loong1 Haryanti-Azura Mohammad-Wali2 Nurul-Asma-Anati Che-Mat-Seri1 Nur-Hidayana Mahfodz1 Dzuzaini Mohd-Ghazali2 Pooi-Fong Wong2,3 Sazaly AbuBakar1,4* Abstract The use of sentinel animals in animal research facilities is the cornerstone of animal health monitoring for exposure to natural pathogens of laboratory animals. This is because infection with these pathogens produces no overt signs of disease yet the infection may affect the outcome of research utilising these animals. One of the important pathogens includes Klebsiella pneumoniae, whose prevalence can be high in laboratory animals. This study sought to document the prevalence of K. pneumoniae and other undocumented bacterial pathogens in laboratory rodents for a 3- year period. K. pneumoniae and Chryseobacterium gleum were isolated from sentinel ICR mice housed in three different satellite animal laboratories, suggesting the effectiveness of the sentinel program. A novel strain of K. pneumoniae ST3125 was recovered from the gastrointestinal tract of the ICR mouse. β-lactamase and virulence genes were detected among K. pneumoniae and C. gleum strains, suggesting the acquisition of these genes from the users of the animal research facilities. Examination of the animal housing environment, feed and water specimens however, returned negative for the presence of K. pneumoniae and C. gleum suggesting that current hygiene practices were adequate in controlling transmission from the environment. Nevertheless, stringent hygiene practices and infection control protocols have to be applied in animal facilities to prevent the colonization and spread of pathogens capable of distorting experimental results. Keywords: Chryseobacterium gleum, infectious disease, Klebsiella pneumoniae, laboratory animals, Malaysia 1Tropical Infectious Diseases Research & Education Centre, University of Malaya, 50603 Kuala Lumpur, Malaysia 2Animal Experimental Unit, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 3Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 4Department of Medical Microbiology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 671-679.

672 Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. Introduction surveillance of potential infection from exogenous sources. In this study, the bacteriological status of The increasing demand for laboratory rodents housed in the three satellite animal animals, particularly rodents in biomedical research laboratories between the period of January 2014 and and biotechnology, reflects the rapid development of June 2017 was determined. This study aimed to detect medicine for human and veterinary applications. The K. pneumoniae and undocumented bacterial pathogens welfare of laboratory rodents has to be safeguarded to and subsequently, examine these strains to determine minimise discomfort, since stress may lead to non- their genetic and phenotypic characteristics. specific physiological effects which can jeopardise experimental results (Baumans, 2005). Additionally, Materials and Methods researchers in Malaysia have moral and binding legal obligations under the Biosafety Act 2007 and the Animal Housing Condition: Outbred females of ICR Animal Welfare Act 2015 to ensure that mice and Sprague Dawley (SD) rats with health experimentation done using these animals is certificates, aged four to six weeks old were introduced performed as humanely as possible and in compliance as sentinels in three different satellite animal with all ethical standards (Biosafety Act 2007; Animal laboratories in the Faculty of Medicine, University of Welfare Bill 2015). The animal research facility sentinel Malaya and were kept for eight weeks throughout the program was introduced to actively monitor the entire sentinel program. The rodents were housed in microbiological status of laboratory rodent colonies so open or individually ventilated cages (IVC), at a room as to ensure that the animals in the facility are not temperature of 19-21 °C, a relative humidity of 55-65 % exposed to undocumented pathogens (Brielmeier et al., and a 12/12 h light/dark cycle. Mice were serologically 2006). tested to be free from these pathogens; epizootic diarrhea of infant mice, mouse hepatitis virus, Guidelines for the design of sentinel Mycoplasma pulmonis, mouse parvovirus, minute virus programs were published by the Federation of of mice, pneumonia virus of mice, respiratory enteric European Laboratory Animal Science Associations orphan virus, Sendai virus and Theiler’s murine (FELASA) Working Group on Health Monitoring of encephalomyelitis virus, while rats were serologically Rodent and Rabbit Colonies in June 2001 (Nicklas et al., tested free from Kilham rat virus, Mycoplasma pulmonis, 2002). It is of vital importance to use quality animals pneumonia virus of mice, rat coronavirus, respiratory with known biological characteristics as research using enteric orphan virus, rat parvovirus, Sendai virus and animals so as to ensure the reliability and the Theiler’s murine encephalomyelitis virus. The sentinel reproducibility of experimental results. Exposure to animals were housed in pairs per cage and each cage infectious diseases will adversely affect the physiology was clearly identified as sentinels. The sentinel animals of the animals, inducing changes in behavior, growth were exposed to soiled bedding from each animal cage rate, relative organ weight and immune response during the cage changing process. About 25 grams of (Nicklas et al., 2002). The negative influence of the soiled bedding were transferred to the sentinel animal mouse parvovirus on mice’ reproductive and immune cage. The introduction of soiled bedding to different functions (Carty, 2008), is a case in point. Exposing sentinels was done each week to ensure exposure to as sentinel animals to soiled bedding is a common many animals as possible in the room, and the method for detecting viral and bacterial infections in handling of these sentinel animals was performed after rodents (Manuel et al., 2008). Amongst the viruses and care has been provided for all other animals in the bacteria of interest which have been identified using room. Overall, thirty one rats (n=31) and thirty mice this method are the mouse parvovirus (Watson, 2013), (n=30) were included as sentinel animals in this 3-year the mouse hepatitis virus (Smith et al., 2007), the study and at least two to three sentinel cages were murine norovirus (Manuel et al., 2008) and the placed in each satellite animal laboratory. Helicobacter sp. (Whary et al., 2000). Tissue Sample Collection: After eight weeks of Besides the environment, laboratory animal exposure to soiled bedding, the sentinel animals were infections may also be introduced by humans, sacrificed for full necropsy. The necropsy examination particularly the researchers who utilize the animal was done in a biosafety cabinet or on a clean and facilities. Even after applying the proper personal sanitized bench top. The rodents were anesthetized protective equipments, the risk of commensal bacteria intraperitoneally prior to blood collection by terminal transmission to laboratory animals still exists (Loong et bleeding (Loong et al., 2016a,b). Rats were anesthetized al., 2016a). Klebsiella pneumoniae is a commensal with a mixture of 50 mg/kg ketamine and 5 mg/kg microorganism in mammals, which can also cause xylazine, while mice were anaesthetized with a opportunistic infection at ectopic sites or turn mixture of 80 mg/kg ketamine and 10 mg/kg xylazine. pathogenic upon the acquisition of virulence genes that Following this, rats were given an overdose of aid spread and cause morbidity and mortality anesthesia or injection of euthanasia solution (Pendleton et al., 2013). Thus, it is imperative to control (pentobarbitone), whilst cervical dislocation was the spread of these pathogens in the animal research performed on mice to confirm death. Aseptic sampling facility. While it was not a requirement, we included of selected organs such as the trachea and lung, liver testing for the presence of virulence genes in and the gastrointestinal tract was then performed for commensal Escherichia coli into our existing animal bacterial culture of K. pneumoniae and other health monitoring program (Loong et al., 2016a). We undocumented bacterial strains. intend to expand the examination, covering K. pneumoniae and other undocumented bacterial pathogens to ascertain their genetic background for the

Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. 673 Bacterial Culture: Protocols for bacterial culture have Results been described by Loong et al. (2016a). Pure cultures of the bacterial isolates grown on MacConkey agar were A total of three (n=3) K. pneumoniae and one obtained and identified using 16S rDNA sequencing (n=1) Chryseobacterium gleum confirmed via 16S rDNA (Loong et al., 2016b). The isolated bacteria strains from sequencing were isolated from ICR mice in the course this study were tested with the extended-spectrum β- of the sentinel surveillance (Table 1). None of these lactamase (ESBL) phenotypic confirmatory method bacterial species was detected among the laboratory using antibiotic discs containing cefotaxime and rodents prior to the commencement of the sentinel ceftazidime with clavulanic acid on Mueller-Hinton program, suggesting acquisition from the satellite agar (CLSI, 2012). This study and all animal handling animal laboratories. The K. pneumoniae strains were procedures were approved by the Institutional Animal isolated from the mice’ gastrointestinal tracts while the Care and Use Committee, University of Malaya uncommonly found, yellow-pigmented C. gleum was (Reference no. 2013-11-12/AEU/B/WPF). isolated from the liver (Table 1). K. pneumoniae UM- AEU320 was cultured from an ICR mouse kept in open Molecular Assays: K. pneumoniae strains isolated from cage, whereas the other bacterial strains were cultured the animals were subjected to string test (Fang et al., from mice kept in individually ventilated cages (IVC) 2004). Amplification of virulence genes; aerobactin, allS, (Table 1). kfu, magA and rmpA, was performed following protocols established by Yu et al. (2008). β-lactamase K. pneumoniae strains UM-AEU250 and UM- genes (blaACC, blaACT, blaCTX-M, blaDHA, blaFOX, blaGES, AEU320 had the identical sequence type 1440 (ST1440), blaIMP, blaKPC, blaLAT, blaMOX, blaNDM, blaOXA-1, blaOXA-48, carried one β-lactamase gene (blaSHV) and one kfu blaPER, blaSHV, blaTEM, blaVEB and blaVIM) were amplified virulence gene (Table 2). The K. pneumoniae strain UM- following published protocols (Loong et al., 2016a) and AEU437 was characterized as a novel ST3125 carrying multi locus sequence typing (MLST) was performed two β-lactamase genes (blaSHV and blaVIM) and one kfu using modified primers from Guo et al. (2015). All virulence gene (Table 2). Sequencing of the amplified amplified DNA fragments were purified and β-lactamase genes revealed sequences similar to blaSHV sequenced in both directions, and MLST data was (accession no. FJ668811) and blaVIM (accession no. deposited at http://bigsdb.pasteur.fr/klebsiella/ HQ858608). They were all non-ESBL producers and klebsiella.html. Minimum spanning trees employing were negative for string test (Table 2). The constructed the MLST data were constructed using PHYLOViZ minimum spanning trees based on K. pneumoniae Online (Ribeiro-Gonҫalves et al., 2016) to visualise the MLST profiles suggest that ST1440 and ST3125 (novel evolutionary relationships of the K. pneumoniae strains ST) emerged respectively from ST17 (Figure 1) and isolated in this study. ST347 (Figure 2) as they were double locus variants of their respective founders. C. gleum strain UM-AEU276 Minimum Inhibitory Concentration and Transmission was tested as a non-ESBL producer although the Electron Microscopy: The minimum inhibitory Ambler class A ESBL, blaCGA-1 (Bellais et al., 2002) was concentrations (MIC) against amikacin, cefotaxime, amplified from its genomic DNA (accession no. ceftriaxone, ceftazidime, ciprofloxacin, levofloxacin, LT009413). MIC assays revealed that the C. gleum strain gentamicin, imipenem, meropenem and tetracycline UM-AEU276 was resistant to tetracycline but were determined for other undocumented bacterial susceptible to amikacin, cefotaxime, ceftriaxone, strains, according to the breakpoints for non- ceftazidime, ciprofloxacin, levofloxacin, gentamicin, Enterobacteriaceae (CLSI, 2012) using M.I.C.Evaluator imipenem and meropenem. Imaging of the yellow- Strips (Thermo Fisher Scientific, Cheshire, UK). pigmented C. gleum strain UM-AEU276 using Transmission electron microscopy was performed to transmission electron microscopy (Loong et al., 2017b) observe the structural features of other undocumemted at 40,000x magnification displayed a rod-shaped bacterial strains (Loong et al., 2017b). Additionally, morphology (Figure 3) similar to another member of swabs were taken from tabletops, taps, trolleys, wash the genus Chryseobacterium, C. joostei (Hugo et al., basins, animal feed and drinking water (Loong et al., 2003). Cultures from the swabs taken from the animal 2016a), to be cultured for K. pneumoniae and the housing environment, animal feed and drinking water undocumented bacterial strains. were free from K. pneumoniae and C. gleum, suggesting that these organims could have originated from other sources. Table 1 Bacteria isolated from sentinel rodents. Strain Mouse Collection date Organa Housing Satellite Bacteria identity conditionb laboratory strain K. pneumoniae IVC A C. gleum UM-AEU250 ICR February 2015 GIT IVC A K. pneumoniae Open cage B K. pneumoniae UM-AEU276 ICR July 2015 Liver IVC C UM-AEU320 ICR September 2015 GIT UM-AEU437 ICR June 2017 GIT aGIT = Gastrointestinal tract; bIVC = Individually ventilated cage

674 Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. Table 2 Genetic and phenotypic features of K. pneumoniae isolated from sentinel rodents. Strain ST ESBL-phenotypea β-lactamase gene Virulence gene UM-AEU250 ST1440 - blaSHV UM-AEU320 ST1440 - blaSHV kfu UM-AEU437 ST3125 - blaSHV, blaVIM a- = Non-ESBL producer kfu kfu Discussion virulence gene are associated with human liver abscesses (Zhang et al., 2015). The isolation of C. gleum The rodent health monitoring sentinel program described in this study which had been strains from clinical patient specimens (Lo and Chang, effective in detecting rare bacterial infections, as shown 2014) have also increased recently, implying that the by the detection of the recently characterized Bordetella strains from this study could have originated from the pseudohinzii (Loong et al., 2017a), was expanded to the hospital environment. However, the clinical C. gleum surveillance of K. pneumoniae and other undocumented bacterial infections. Two different K. pneumoniae STs strain reported by Garg et al. (2015) was susceptible to (ST1440 and ST3125), one of which is a novel ST3125 tetracycline, in contrast to our strain, suggesting that C. and C. gleum, were recovered in three satellite animal gleum UM-AEU276 could be under selective laboratories. The amplification of β-lactamase and tetracycline pressure. Both K. pneumoniae and C. gleum virulence genes in the K. pneumoniae and C. gleum strains suggest that these strains were not part of the can survive on inanimate surfaces and have been mice’ normal flora nor from within the laboratories, found to persist in the hospital environment (Garg et but were most likely introduced externally. al., 2015; Jin et al., 2015). Considering that K. Commensal bacteria which form the normal flora of pneumoniae and C. gleum could not be detected from the the mouse do not usually carry antimicrobial resistance nor virulence genes (Loong et al., 2016a), as carrying examination of the animal housing environment, feed these genes would have a negative effect on the and water, this could imply that the hygiene practices biological fitness of the commensals (Löhr et al., 2015). in the satellite animal laboratories were adequate for controlling transmission of infectious bacteria from the K. pneumoniae blaSHV and blaVIM carrying environment. Accordingly, this would also explain the strains are prevalent in the hospital environment relatively rare instances where undocumented worldwide (Nordmann et al., 2011; Yan et al., 2015; bacterial strains were isolated. In addition, the bacteria Kanamori et al., 2017), and those carrying the kfu culture protocols in this study would favor pathogenic strains (Loong et al., 2016b) over commensal bacteria which would not likely be cultured (Stewart, 2012). Figure 1 Minimum spanning tree showing K. pneumoniae ST1440 arising from the ST17-lineage. Each node represents a distinct ST and each branch represents a single locus variant. ST1440 is a double locus variant of ST17. Branch lengths are not proportional to genetic distance.

Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. 675 Figure 2 Minimum spanning tree showing K. pneumoniae ST3125 arising from the ST347-lineage. Each node represents a distinct ST and each branch represents a single locus variant. ST3125 is a double locus variant of ST347. Branch lengths are not proportional to genetic distance. Figure 3 Transmission electron microscopy imaging of C. gleum UM-AEU276 at 40,000x magnification. Analysis of K. pneumoniae genetic hypervirulent ST17 strains have been reported in Africa (Henson et al., 2017), Asia (Jin et al., 2015), relationships using minimum spanning trees Europe (Löhr et al., 2015), Latin America (Pasteran et displayed that the strains were derived from two al., 2012) and North America (Cerqueira et al., 2017), central founders; ST1440 emerged from ST17 (Figure 1) epitomizing their competence for colonization and while ST3125 emerged from ST347 (Figure 2). ST1440 global spread. In China, ST20 has been reported to was a double locus variant of ST17 and a single locus cause epidemics in a neonatal unit (Jin et al., 2015) and variant of ST20 (Figure 1). Multidrug-resistant and

676 Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. ST1440 has been grouped among high-risk clones in Bellais S, Naas T and Nordmann P 2002. Molecular and Taiwan (Yan et al., 2015). A 3-year outbreak in a medical center involving ST2252, which evolved from biochemical characterization of ambler class A ST1440, occurred despite strict infection control measures and ST2252 was also shown to spread extended-spectrum β-lactamase CGA-1 from carbapenem-resistance to other bacteria species (Kanamori et al., 2017). The novel ST3125 was derived Chryseobacterium gleum. Antimicrob Agents from ST347 and ST1023, previously found in pneumonia patients from Japan (Ito et al., 2015) and Chemother. 46(4): 966-970. from bloodstream infections in Shanghai (Xiao et al., 2017). Interestingly, ST2992, ST2995 and ST2594 Biosafety Act 2007. [Online]. Available: human rectal strains which were derived from ST3125, were postulated as human commensal strains from gut bch.cbd.int/database/attachment/?id=17640. microbiota (Gorrie et al., 2018). Accessed January 2, 2018. Overall, genetic analyses of the K. pneumoniae strains from this study revealed that ST1440 (Figure 1) Brielmeier M, Mahabir E, Needham JR, Lengger C, can potentially establish itself as the dominant clone in satellite animal laboratories as it belongs to a lineage Wilhelm P and Schmidt J 2006. Microbiological which has the propensity for colonization (Pasteran et al., 2012; Jin et al., 2015; Cerqueira et al., 2017). This was monitoring of laboratory mice and evident in the discovery of ST1440 from two different laboratories at different periods. The use of biocontainment in individually ventilated cages: individually ventilated cages (IVC) provided a measure of biocontainment when infected mice were a field study. Lab Anim. 40(3): 247-260. housed with non-infected mice, preventing pathogen transmission to other mice within the same facility Carty AJ 2008. Opportunistic infections of mice and (Brielmeier et al., 2006). K. pneumoniae UM-AEU320 (ST1440) was cultured from a sentinel mouse housed in rats: Jacoby and Lindsey revisited. ILAR J. 49(3): an open cage, perhaps exposing it to K. pneumoniae contaminated fomite from the environment 272-276. (Lautenbach et al., 2001). Meanwhile, ST3125 (Figure 2) which has multidrug resistant ancestors (Ito et al., 2015; Cerqueira GC, Earl AM, Ernst CM, Grad YH, Dekker Xiao et al., 2017) and presumably harmless human commensal descendants (Gorrie et al., 2018), could JP, Feldgarden M, Chapman SB, Reis-Cunha JL, potentially express hypervirulent phenotypes under antimicrobial selection pressure (Holt et al., 2015). Shea TP, Young S, Zeng Q, Delaney ML, Kim D, On the basis of our findings, as well as the Peterson EM, O'Brien TF, Ferraro MJ, Hooper DC, effectiveness of the sentinel program, we stress that stringent hygiene practices and rigorous infection Huang SS, Kirby JE, Onderdonk AB, Birren BW, control protocols must be vigilantly practiced in any animal laboratory to prevent the colonization and Hung DT, Cosimi LA, Wortman JR, Murphy CI spread of undocumented infectious pathogens. The health and infection status of laboratory rodents have and Hanage WP 2017. Multi-institute analysis of to be monitored closely to ensure that the animals are free of unwanted infections that could jeopardize the carbapenem resistance reveals remarkable interpretation and reliability of animal experiments. diversity, unexplained mechanisms, and limited Acknowledgements clonal outbreaks. Proc Natl Acad U.S.A. 114(5): This study was funded in part by the University of Malaya (RU016-2015 and RU005-2017). 1135-1140. We thank the team of curators of the Institut Pasteur MLST and all the genome MLST databases for curating CLSI 2012. Performance standards for antimicrobial the data and making it available at http://bigsdb.pasteur.fr. susceptibility testing; Twenty-second References informational supplement M100-S22: Wayne, PA, Animal Welfare Bill 2015. [Online]. Available: USA. https://www.cljlaw.com/files/bills/pdf/2015/ MY_FS_BIL_2015_19.pdf. Accessed January 2, Fang CT, Chuang YP, Shun CT, Chang SC and Wang 2018. JT 2004. A novel virulence gene in Klebsiella Baumans V 2005. Science-based assessment of animal pneumoniae strains causing primary liver abscess welfare: laboratory animals. Rev Sci Tech. 24(2): 503-513. and septic metastatic complications. J Exp Med. 199(5): 697-705. Garg S, Appannanavar SB, Mohan B and Taneja N 2015. Pyonephrosis due to Chryseobacterium gleum: a first case report. Indian J Med Microbiol. 33(2): 311-313. Gorrie CL, Mirceta M, Wick RR, Judd LM, Wyres KL, Thomson NR, Strugnell RA, Pratt NF, Garlick JF, Watson KM, Hunter PC, McGloughlin SA, Spelman DW, Jenney AWJ and Holt KE 2018. Antimicrobial resistant Klebsiella pneumoniae carriage and infection in specialized geriatric care wards linked to acquisition in the referring hospital. Clin Infect Dis. 67(2): 161-170. Guo C, Yang X, Wu Y, Yang H, Han Y, Yang R, Hu L, Cui Y and Zhou D 2015. MLST-based inference of genetic diversity and population structure of clinical Klebsiella pneumoniae, China. Sci Rep. 5: 7612. Henson SP, Boinett CJ, Ellington MJ, Kagia N, Mwarumba S, Nyongesa S, Mturi N, Kariuki S, Scott JAG, Thomson NR and Morpeth SC 2017. Molecular epidemiology of Klebsiella pneumoniae invasive infections over a decade at Kilifi County Hospital in Kenya. Int J Med Microbiol. 307(7): 422-429. Holt KE, Wertheim H, Zadoks RN, Baker S, Whitehouse CA, Dance D, Jenney A, Connor TR, Hsu LY, Severin J, Brisse S, Cao H, Wilksch J, Gorrie C, Schultz MB, Edwards DJ, Nguyen KV, Nguyen TV, Dao TT, Mensink M, Minh

Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. 677 VL, Nhu NT, Schultsz C, Kuntaman K, Newton Manuel CA, Hsu CC, Riley LK and Livingston RS 2008. PN, Moore CE, Strugnell RA and Thomson NR Soiled-bedding sentinel detection of murine 2015. Genomic analysis of diversity, population norovirus 4. J Am Assoc Lab Anim Sci. 47(3): 31- structure, virulence, and antimicrobial resistance 36. in Klebsiella pneumoniae, an urgent threat to public Nicklas W, Baneux P, Boot R, Decelle T, Deeny AA, health. Proc Natl Acad Sci USA. 112(27): E3574- Fumanelli M and Illgen-Wilcke B 2002. 3581. Recommendations for the health monitoring of Hugo CJ, Segers P, Hoste B, Vancanneyt M and rodent and rabbit colonies in breeding and Kersters K 2003. Chryseobacterium joostei sp. nov., experimental units. Lab Anim. 36(1): 20-42. isolated from the dairy environment. Int J Syst Evol Microbiol. 53(Pt 3): 771-777. Nordmann P, Naas T and Poirel L 2011. Global spread Ito R, Shindo Y, Kobayashi D, Ando M, Jin W, Wachino of carbapenemase-producing Enterobacteriaceae. J, Yamada K, Kimura K, Yagi T, Hasegawa Y and Emerg Infect Dis. 17(10): 1791-1798. Arakawa Y 2015. Molecular epidemiological characteristics of Klebsiella pneumoniae associated Pasteran F, Albornoz E, Faccone D, Gomez S, Valenzuela C, Morales M, Estrada P, Valenzuela with bacteremia among patients with L, Matheu J, Guerriero L, Arbizú E, Calderón Y, pneumonia. J Clin Microbiol. 53(3): 879-886. Ramon-Pardo P and Corso A 2012. Emergence of Jin Y, Shao C, Li J, Fan H, Bai Y and Wang Y 2015. NDM-1-producing Klebsiella pneumoniae in Outbreak of multidrug resistant NDM-1- producing Klebsiella pneumoniae from a neonatal Guatemala. J Antimicrob Chemother. 67(7): 1795- 1797. unit in Shandong Province, China. PLoS One. Pendleton JN, Gorman SP and Gilmore BF 2013. 10(3): e0119571. Clinical relevance in the ESKAPE pathogens. Kanamori H, Parobek CM, Juliano JJ, van Duin D, Expert Rev Anti Infect Ther. 11(3): 297-308. Cairns BA, Weber DJ and Rutala WA 2017. A Ribeiro-Gonҫalves B, Francisco AP, Vaz C, Ramirez M prolonged outbreak of KPC-3-producing and Carriҫo JA 2016. PHYLOViZ Online: web- Enterobacter cloacae and Klebsiella pneumoniae based tool for visualization, phylogenetic inference, analysis and sharing of minimum driven by multiple mechanisms of resistance spanning trees. Nucleic Acids Res. 44(W1): W246- transmission at a large academic burn center. W251. Antimicrob Agents Chemother. 61(2): e01516-16. Smith PC, Nucifora M, Reuter JD and Compton SP Löhr IH, Hülter N, Bernhoff E, Johnsen PJ, Sundsfjord 2007. Reliability of soiled bedding transfer for A and Naseer U 2015. Persistence of a pKPN3-like detection of mouse parvovirus and mouse CTX-M-15-encoding IncFIIK plasmid in a hepatitis virus. Comp Med. 57(1): 90-96. Klebsiella pneumoniae ST17 host during two years Stewart EJ 2012. Growing unculturable bacteria. J Bacteriol. 194(16): 4151-4160. of intestinal colonization. PLoS One. 10(3): Watson J 2013. Unsterilized feed as the apparent cause e0116516. of a mouse parvovirus outbreak. J Am Assoc Lab Lautenbach E, Patel JB, Bilker WB, Edelstein PH and Anim Sci. 52(1): 83-88. Fishman NO 2001. Extended-spectrum beta- Whary MT, Cline JH, King AE, Hewes KM, Chojnacky lactamase-producing Escherichia coli and Klebsiella D, Salvarrey A and Fox JG 2000. Monitoring pneumoniae: risk factors for infection and impact sentinel mice for Helicobacter hepaticus, H. rodentium and H. bilis infection by use of of resistance on outcomes. Clin Infect Dis. 32(8): 1162-1171. polymerase chain reaction analysis and serologic Lo HH and Chang SM 2014. Identification, testing. Comp Med. 50(4): 436-443. characterization, and biofilm formation of clinical Xiao SZ, Wang S, Wu WM, Zhao SY, Gu FF, Ni YX, Guo Chryseobacterium gleum isolates. Diagn Microbiol XK, Qu JM and Han LZ 2017. The resistance Infect Dis. 79(3): 298-302. phenotype and molecular epidemiology of Loong SK, Mahfodz NH, Che Mat Seri NAA, Klebsiella pneumoniae in bloodstream infections in Mohamad Wali HA, Abd Gani SA, Wong PF and AbuBakar S 2016a. Genetic characterization of Shanghai, China, 2012-2015. Front Microbiol. 8: commensal Escherichia coli isolated from 250. laboratory rodents. Springerplus. 5(1): 1035. Yan JJ, Wang MC, Zheng PX, Tsai LH and Wu JJ 2015. Loong SK, Mahfodz NH, Wali HAM, A Talib SA, Associations of the major international high-risk Nasrah SNA, Wong PF and AbuBakar S 2016b. resistant clones and virulent clones with specific Molecular and antimicrobial analyses of non- ompK36 allele groups in Klebsiella pneumoniae in classical Bordetella isolated from a laboratory Taiwan. New Microbes New Infect. 5: 1-4. mouse. J Vet Med Sci. 78(4): 715-717. Yu WL, Ko WC, Cheng KC, Lee CC, Lai CC and Loong SK, Tan KK, Sulaiman S, Wong PF, AbuBakar S Chuang YC 2008. Comparison of prevalence of 2017a. Draft genome of Bordetella pseudohinzii virulence factors for Klebsiella pneumoniae liver BH370 isolated from trachea and lung tissues of a abscess between isolates with capsular K1/K2 laboratory mouse. Genom Data. 12: 69-70. and non-K1/K2 serotypes. Diagn Microbiol Loong SK, Tan KK, Zainal N, Phoon WH, Mohd Zain Infect Dis. 62(1): 1-6. SN and AbuBakar S 2017b. Draft genome of the Zhang Y, Sun J, Mi C, Li W, Zhao S, Wang Q, Shi D, Liu emerging pathogen, Kocuria marina, isolated from L, Ding B, Chang YF, Guo H, Guo X, Li Q and Zhu Y 2015. First report of two rapid-onset fatal a wild urban rat. Mem Inst Oswaldo Cruz. infections caused by a newly emerging 112(12): 857-859.

6768 Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. hypervirulent K. pneumoniae ST86 strain of serotype K2 in China. Front Microbiol. 6: 721.

Loong S. K. et al. / Thai J Vet Med. 2018. 48(4): 671-679. 6797 บทคดั ย่อ Effectiveness of sentinel rodents for surveillance of exposure to undocumented bacterial pathogens in animal research facility Shih-Keng Loong1 Haryanti-Azura Mohammad-Wali2 Nurul-Asma-Anati Che-Mat-Seri1 Nur-Hidayana Mahfodz1 Dzuzaini Mohd-Ghazali2 Pooi-Fong Wong2,3 Sazaly AbuBakar1,4* The use of sentinel animals in animal research facilities is the cornerstone of animal health monitoring for exposure to natural pathogens of laboratory animals. This is because infection with these pathogens produces no overt signs of disease yet the infection may affect the outcome of research utilising these animals. One of the important pathogens includes Klebsiella pneumoniae, whose prevalence can be high in laboratory animals. This study sought to document the prevalence of K. pneumoniae and other undocumented bacterial pathogens in laboratory rodents for a 3-year period. K. pneumoniae and Chryseobacterium gleum were isolated from sentinel ICR mice housed in three different satellite animal laboratories, suggesting the effectiveness of the sentinel program. A novel strain of K. pneumoniae ST3125 was recovered from the gastrointestinal tract of the ICR mouse. β-lactamase and virulence genes were detected among K. pneumoniae and C. gleum strains, suggesting the acquisition of these genes from the users of the animal research facilities. Examination of the animal housing environment, feed and water specimens however, returned negative for the presence of K. pneumoniae and C. gleum suggesting that current hygiene practices were adequate in controlling transmission from the environment. Nevertheless, stringent hygiene practices and infection control protocols have to be applied in animal facilities to prevent the colonization and spread of pathogens capable of distorting experimental results. คำสำคัญ: Chryseobacterium gleum infectious disease Klebsiella pneumoniae laboratory animals Malaysia 1Tropical Infectious Diseases Research & Education Centre, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 2Animal Experimental Unit, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 3Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 4Department of Medical Microbiology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia *ผูร้ บั ผดิ ชอบบทความ E-mail: [email protected]



Short Communication Effects of supplementing varying levels of 1,3-diacylglycerol (DAG) on growth performance, apparent nutrient digestibility and selected fecal microbial populations in growing pigs Wen Chao Liu1# Kwan Sik Yun2# In Ho Kim3* Abstract The objective of the present study was to investigate the effects of 1,3-diacylglycerol (DAG) supplementation on growth performance, apparent digestibility and selected fecal microbial populations in growing pigs. A total of 80 crossbred pigs [(Landrace × Yorkshire) × Duroc] with an average initial body weight (BW) of 27.52 ± 0.20 kg were used in a 6 week feeding trial. Pigs were allotted to 4 dietary treatments based on their initial BW and sex (5 replications; 4 pigs per pen with 2 barrows and 2 gilts). Dietary treatments included: 1) CON, basal diets; 2) CON+0.075% 1,3-DAG; 3) CON+0.10% 1,3-DAG; 4) CON+0.15% 1,3-DAG. The final BW was linearly increased (P=0.0056) by 1,3-DAG supplementation. Average daily feed intake (ADFI) was improved (cubic, P=0.0312) by feeding 1,3-DAG during week 1-3. Overall (week 1-6), dietary 1,3-DAG supplementation led to a linearly increase in ADG (P=0.0042) and G:F (P=0.0235). In addition, dietary supplementation of 1,3-DAG linearly improved (P=0.0004) the apparent digestibility of gross energy and had a tendency (linear, P=0.0725) to improve the apparent digestibility of crude fat. Pigs fed 1,3-DAG supplemented diet also linearly decreased (P=0.0065) fecal E. coil populations at week 3 and linearly increased (P=0.0348) fecal Lactobacillus populations at week 6. In conclusion, these results show that 1,3-DAG can be used as a potential functional feed additive and exerts positive effects on growth performance, apparent energy digestibility and fecal microbiota in growing pigs. Keywords: 1,3-diacylglycerol, growing pigs, growth promoter 1Department of Animal Science, College of Agriculture, Guangdong Ocean University, Zhanjiang, Guangdong 524088, P. R. China 2Synergen Inc., 1106 Jung -Dong, Wonmi-Gu, Bucheon-Si, Gyeonggi-Do, South Korea 3Department of Animal Resource & Science, Dankook University, Cheonan, Chungnam 31116, South Korea #These authors contributed equally to this work *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 681-687.

682 Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. Introduction [(Landrace × Yorkshire) × Duroc] were used. Pigs with an average initial BW of 27.52 ± 0.20 kg, and were There is general consensus that energy intake sorted by sex and balanced by BW assigned to and utilization are often the first limiting factors for treatment. There were five replicate pens per treatment protein deposition in young and growing pigs (Nieto with four pigs (2 gilts and 2 barrows) per pen. Dietary et al., 2002), thus greatly affecting productive treatment groups included: 1) CON, basal diets; 2) performance. Dietary energy is mainly produced from CON+0.075% 1,3-DAG; 3) CON+0.10% 1,3-DAG; 4) the oxidation of lipids, carbohydrates and proteins. CON+0.15% 1,3-DAG. The tested product, 1,3-DAG Energy generated from the oxidation of fatty acids (50%) mixed with a carrier was obtained from a derived from dietary lipids is more than twofold commercial company (Il Shin Wells, Korea). The diet higher than that generated from starch (van Milgen et was in mash form and formulated to meet or exceed al., 2001). Therefore, dietary lipids can be an important the NRC (2012) nutrient requirements (Table 1). All the source of energy in pig diets. However, there are some pigs were reared in an environmentally controlled problems with lipids digestion and utilization in house with a slatted plastic floor. Throughout the trial young and growing pigs due to the immature period, a self-feeder and nipple waterer were equipped physiological functions and a low level of natural in each pen, in order to allow the pigs ad libitum access lipase production (Doreau and Chilliard, 1997). to feed and water. Diacylglycerol (DAG), a glycerol derivative, Sampling and measurements: At the beginning of the possesses 2 hydroxyl groups substituted by fatty acids trial and at the end of weeks 3 and 6, the BW of each through ester bond formation and exists either as 1,2- pig was measured and the average daily gain (ADG) DAG and 1,3-DAG (Flickinger and Matsuo, 2003). The was calculated. Meanwhile, the feed intake was most popular function of DAG is related to their recorded based on pens throughout the trial period to amphiphilic nature and surface-active properties, calculate average daily feed intake (ADFI) and gain: being well-known as emulsifier ingredients in the food feed (G:F). During the trial, the dead pigs in each pen industry (Martin et al., 2014). In this regards, DAG has were also recorded to correct the feed intake. been used synergistically with monoacylglycerol as an emulsifier to improve the fat digestibility (Flickinger At the end of week 5, pigs were fed a diet and Matsuo, 2003; Shimada and Ohashi, 2003; mixed with chromium oxide (Cr2O3, 0.2%) as an Nakajima, 2004). On the other hand, among the two indigestible marker to calculate the apparent total tract kinds of DAG, 1,3-DAG are hydrolyzed to free fatty digestibility (ATTD) of dry matter (DM) and nitrogen acids, 1(3)-monoglycerides (MG) during digestion. (N), cross energy (GE) and crude fat (Ball and Aherne, Several bioactivities of 1(3)-MG have been reported, 1987). On the last two days of the experiment, fresh including antiviral and antimicrobial (Lieberman et al., fecal samples from each pig were directly collected by 2006), antioxidant (Cho et al., 2010), and the massaging, and then they were uniformly pooled stimulation of fat utilization (Shimotoyodome et al., based on pen. All fecal samples and feed samples were 2012). Furthermore, DAG can be absorbed directly into stored in a -20 ° C freezer until analysis. Fecal samples the blood and enter hepatocytes via the portal vein for were dried at 57 °C for 72 hours prior, after which they rapid oxidation (Murata et al., 1997). Previous studies were ground to pass through a 1-mm sieve. The feed have demonstrated that DAG, particularly 1,3-DAG samples were ground to pass through a 1 mm sieve. has the ability to suppress body fat accumulation, Then according to (Method 930.15, AOAC 1995) and reduce postprandial hyperlipidemia and serum (Method 990.03, AOAC 1995) the DM and N content triacylglycerides (Flickinger and Matsuo, 2003; Yanai were analyzed. The fat content was determined by et al., 2007), suggesting that 1,3-DAG is involved in standard procedures (Method 954.02; AOAC, 2005). lipid metabolism. More recently, our group has shown The chromium content was determined by UV that dietary supplementation of 1,3-DAG could be absorption spectrophotometry (Shimadzu UV-1201, used as an emulsifier and induce a greater fat Shimadzu, Kyoto, Japan). The energy was determined digestibility and growth performance in broilers using a Compensated Jacket Calorimeter 6100 (Parr (Upadhaya et al., 2017). However, to the best of our Instrument Co., Moline, IL USA). Then the following knowledge, there is no relevant research to evaluate formula was used to calculate the ATTD: the effects of 1,3-DAG on growth performance in pigs. We hypothesized that dietary 1,3-DAG can rapidly ATTD (%) = [1-{(Nf × Cd)/(Nd × Cf)}] × 100 supply energy for growing pigs, and improve lipids where Nf = nutrient concentration in feces (% digestion and promote gut healthy status, DM), Nd = nutrient concentration in diet (% DM), Cd consequently, benefiting the growth performance. The = chromium concentration in diet (% DM), and Cf = present study was conducted to test this hypothesis. chromium concentration in feces (% DM). Materials and Methods At the end of week 3 and week 6, for determining the fecal microbiota, fresh fecal samples The Animal Care and Use Committee of were collected into sterilized EP tubes from two pigs Dankook University (Cheonan, South Korea) (one gilt and one barrow) in each pen. Fecal samples approved all animal feeding, management and were placed on ice and immediately transferred to the experimental data collection procedures of the present laboratory for microbial colony analysis. Fecal samples study. (1 g) from each pig were diluted with 9 mL of 1% peptone broth (Becton, Dickinson and Co., Franklin Experiment design, animals and housing: In this 6 Lakes, NJ, USA) and homogenized. The fecal samples weeks feeding experiment, a total of 80 crossbred pigs were then diluted 10-fold (ranging from 10-1 to 10-8) and subsequently cultured on Rogosa SL agar (Rogosa;

Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. 683 Difco Laboratories, Detroit, MI, USA) for counting cultured at 37 °C for 48 hours after inoculation. After Lactobacillus counts and MacConkey agar (Difco culture, colony counts were calculated and the results Laboratories, Detroit, MI) was used to calculate E. coli expressed as log10 transformed data. counts. All the dishes were inverted and anaerobically Table 1 Compositions of the basal diets (as-fed basis) Items 600 Ingredient, g/kg 161.9 Corn Soybean Meal 33% CP 60 Wheat 65 Distiller’s dried grains with solubles 25 Rapeseed meal 30 Molasses 30 Tallow 11 Dicalcium phosphate 8 Limestone 2.7 L-Lys HCl 2 Vitamin premix1 1 Mineral premix2 3 Salt 0.4 Choline, 25% Calculated composition, g/kg 3,302 Metabolizable energy, kcal/kg 155 Crude protein 60.3 Crude fat 6.5 Calcium 5.5 Total phosphorus 3.1 Avilable phosphorus 7.2 SID Lys 1.8 SID Met 6.1 SID Met+Cys Analyzed composition, g/kg 3863 157 Gross energy, kcal/kg 61 Crude protein 7.4 Crude fat 5.2 Calcium Phosphorus 1 Provided per kg of complete diet: 11,025 IU vitamin A; 1,103 IU vitamin D3; 44 IU vitamin E; 4.4 mg vitamin K; 8.3 mg riboflavin; 50 mg niacin; 4 mg thiamine; 29 mg d-pantothenic; 166 mg choline; 33 μg vitamin B12. 2 Provided per kg of complete diet: 12 mg Cu (as CuSO4•5H2O); 85 mg Zn (as ZnSO4); 8 mg Mn (as MnO2); 0.28 mg I (as KI); 0.15 mg Se (as Na2SeO3•5H2O). Statistical analysis: All experimental data was Inc., Cary, NC). The linear, quadratic and cubic analyzed as a randomized complete design using polynomial contrasts were used to determine MIXED procedure of SAS 2003 (v. 9.1, SAS Institute responses to supplemental graded levels of 1,3-DAG at

684 Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. 0, 0.075%, 0.10% and 0.15%. Tukey’s range test was 1,3-DAG on the performance of pigs, thus no more used for detecting the differences among treatment comparisons can be made with other studies. The DAG groups. Variability of the data was expressed as the possesses surface-active properties and is well known standard error of means (SEM). Differences were as an emulsifier ingredient in the food, pharmaceutical deemed significant when P<0.05. Trends were noted and cosmetic industries (Martin et al., 2014). It has been when 0.05 <P<0.10. suggested that dietary emulsifier could improve growth performance in weaning pigs (Zhao et al., Results and Discussion 2015), finishing pigs (Zhao et al., 2016), lactating sows (Zhao et al., 2016), and broilers (San Tan et al., 2016). Growth performance: As described in Table 2, at the Therefore, the positive effects on growth performance end of the experiment, dietary supplementation of 1,3- of growing pigs may be attributed in part to the DAG linearly increased the BW (P=0.0056). During emulsifier properties of 1,3-DAG. On the other hand, week 1 to week 3, dietary 1,3-DAG supplementation 1,3-DAG could be absorbed directly into the blood and led to an increase in ADFI (cubic, P=0.0312). During rapidly supply energy for the body (Murata et al., week 1 to week 6, dietary supplementation with 1,3- 1997). Furthermore, the hydrolyzate from 1,3-DAG DAG had a linear effect on improving the ADG exerts bactericidal and virucidal effects (Lieberman et (P=0.0042) and G:F (P=0.0235). Similar to our results, al., 2006), consequently promoting gut health status Upadhaya et al. (2017) showed that supplementation and growth performance. The exact mechanism behind of graded levels of 1,3-DAG linearly improved the the effects of 1,3-DAG on pig performance, however, weight gain and feed conversion ratios. However, to remain to be elucidated. our knowledge, this is the first report on the effects of Table 2 Effects of dietary 1,3-diacylglycerol (DAG) supplementation on growth performance in growing pigs1 Items2 CON DAG0.075 DAG0.10 DAG0.15 SEM3 Linear p-value Quadratic Cubic BW (kg) Initial 27.5 27.5 27.5 27.6 0.2 0.3153 0.5628 0.6036 Week 3 41.3 41.5 41.7 41.8 0.4 0.3602 0.9668 0.9554 Week 6 57.1a 57.8ab 58.3b 58.4b Week 1-3 0.2 0.0056 0.2145 0.8122 ADG (g) 655.1 666.3 671.6 ADFI (g) 1527.4 1503.2 1544.7 678.0 16.3 0.3144 0.8626 0.9745 G:F 0.429 0.443 0.435 Week 4-6 1502.1 12.0 0.5590 0.4648 0.0312 ADG (g) 754.3 777.5 792.6 ADFI (g) 1785.1 1772.0 1784.4 0.452 0.01 0.2734 0.9383 0.3544 G:F 0.422 0.439 0.445 Overall 789.2 14.1 0.1054 0.3701 0.8611 1799.8 44.3 0.7824 0.7644 0.9187 0.439 0.01 0.3746 0.4150 0.9870 ADG (g) 704.2a 721.7ab 732.9b 734.4b 5.1 0.0042 0.1370 0.8740 ADFI (g) 1656.0 1637.4 1664.3 1651.0 19.3 0.9060 0.9211 0.3627 G:F 0.425a 0.441ab 0.440ab 0.445b 0.004 0.0235 0.2124 0.2831 a,b Means in the same row with different superscripts differ (P<0.05). 1 Dietary treatments were as follows: CON: basal diet (antibiotic-free); DAG0.075: CON+0.075% 1,3-DAG; DAG0.10: CON+0.10% 1,3-DAG; DAG0.15: CON+0.15% 1,3-DAG. 2 There were five replication pens of 4 pigs/pen per treatment. BW, body weight; ADG, average daily gain; ADFI, average daily feed intake; G:F, gain to feed ratio. 3 Standard error of the means. Apparent nutrient digestibility: ATTD of gross energy 2016) and ATTD of gross energy in weaning pigs (Zhao was significantly improved (P=0.0004) by dietary 1,3- et al., 2015). Therefore, the improved growth DAG supplementation. In addition, dietary 1,3-DAG performance observed in our study is associated at supplementation tended to improve (linear, P=0.0725) least in part with the improvement in energy the ATTD of crude fat (Table 3). Similarly, Upadhaya digestibility elicited by 1,3-DAG. Nevertheless, dietary et al. (2017) reported that the addition of 1,3-DAG to supplementation of 1,3-DAG showed only a slight low energy diet increased the ATTD of gross energy in effect in improving ATTD of crude fat (P=0.0725) in broilers. Other studies also indicated that exogenous this study. These findings are partially in agreement emulsifier could improve apparent metabolisable with an earlier report showing that a diet with energy (AME) digestibility in broilers (San Tan et al., exogenous emulsifier had no effect on fat digestibility

Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. 685 but improved AME digestibility in broilers (San Tan et However, the data of ATTD is only from the end of trial al., 2016). However, the results from the present study in our study, therefore, it is necessary to cover more are not consistent with previous findings in that 1,3- stages in future research to study whether 1,3-DAG as DAG or emulsifier significantly increased the ATTD of a fat emulsifier can improve the digestibility of crude fat in broilers and weaning pigs (Zhao et al., 2015; fat. Meanwhile, the data from Zhao and Kim (2017) Upadhaya et al., 2017; Zhao and Kim, 2017). This is also indicated that the improvement in crude fat probably due to the inclusion level or the age of digestibility by emulsifier is more profound at a high animals. For instance, Zhao and Kim (2017) suggested inclusion level. Again, it is assumed that the beneficial that dietary emulsifier improved crude fat digestibility effects of 1,3-DAG on ATTD could be variable in broilers at d 14 of the age, whereas had no significant depending on other factors, such as administration effects on crude fat digestibility at d 28 of the age. strategies and the physical form of the fat in diet. Table 3 Effects of dietary 1,3-diacylglycerol (DAG) supplementation on apparent nutrient digestibility in growing pigs1 Items2, % CON DAG0.075 DAG0.10 DAG0.15 SEM3 Linear p-value Cubic 0.1951 Quadratic 0.9607 Dry matter 76.4 77.0 77.2 77.2 0.45 0.8907 0.8916 Nitrogen 75.9 76.0 75.9 76.0 0.49 0.0004 0.5737 0.9837 Gross energy 76.4a 77.2ab 77.9ab 78.5b 0.38 0.9707 0.8397 Crude fat 59.8 61.2 62.3 62.9 1.24 0.0725 0.7572 0.9288 a,b Means in the same row with different superscripts differ (P<0.05). 1 Dietary treatments were as follows: CON: basal diet (antibiotic-free); DAG0.075: CON+0.075% 1,3-DAG; DAG0.10: CON+0.10% 1,3-DAG; DAG0.15: CON+0.15% 1,3-DAG. 2 There were five replication pens of 4 pigs/pen per treatment. 3 Standard error of the means. Selected fecal microbial populations: Results of fecal Zhang et al. (2016) and Zhang and Houtman (2016). microbial populations are shown in Table 4, dietary Our results showed a linear decline in fecal E coli 1,3-DAG supplementation linearly decreased counts at week 3 and a linear increase in Lactobacillus (P=0.0065) fecal E. coil populations at the end of week 3, and linearly increased (P=0.0348) fecal Lactobacillus counts at week 6 with increasing dietary 1,3-DAG level. This may reflect the antibacterial actions of the populations at the end of week 6. 1,3-DAG is hydrolysed products of DAG and certainly warrants hydrolyzed to 1(3)-monolaurin by lipase in vivo. further research given the emphasis on antibacterial resistance and reduced antimicrobial use in animal Monolaurin has profound antiviral and antibacterial agriculture now. activity (Lieberman et al., 2006) with a number of different mechanisms proposed by Projan et al. (1994), Table 4 Effects of dietary 1,3-diacylglycerol (DAG) supplementation on selected fecal microbial populations in growing pigs1 Items2, log10cfu/g CON DAG0.075 DAG0.10 DAG0.15 SEM3 Linear p-value Cubic Quadratic Week 3 Lactobacillus 7.23 7.29 7.20 7.30 0.11 0.8742 0.9565 0.8660 E. coli 6.40a 6.32ab 6.24ab 6.15b 0.06 0.0065 0.8973 0.9302 Week 6 Lactobacillus 7.03 7.04 7.16 7.14 0.04 0.0348 0.7768 0.2238 E. coli 6.28 6.34 6.36 6.39 0.05 0.7080 0.7532 0.8510 a,b Means in the same row with different superscripts differ (P<0.05). 1 Dietary treatments were as follows: CON: basal diet (antibiotic-free); DAG0.075: CON+0.075% 1,3-DAG; DAG0.10: CON+0.10% 1,3-DAG; DAG0.15: CON+0.15% 1,3-DAG. 2 There were five replication pens of 4 pigs/pen per treatment. 3 Standard error of the means. However, we noted that the selective necessary to confirm whether 1,3-DAG has an effect on antibacterial effect of 1,3-DAG on fecal E. coil only improving lactobacilli in the animal's gut. occurred in week 3 but not in week 6 of the trial, this In conclusion, dietary supplementation with phenomenon may be attributed to a better developed graded levels of 1,3-DAG exerted positive effects on digestive system, enhanced resistance to intestinal growth performance, the ATTD of gross energy and disorders and improved immunity as pigs become fecal microbiota. These findings may provide an older. Regarding the fecal Lactobacillus results, there insight into a new nutritional strategy for growing was only a slight linear effect on Lactobacillus at week 6 pigs. (7.03 vs 7.04, 7.16 and 7.14), so future research is

686 Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. Acknowledgements San Tan H, Zulkifli I, Farjam AS, Goh YM, Croes E, Partha SK and Tee AK 2016. Effect of exogenous This research was funded by the Department emulsifier on growth performance, fat of Animal Resource and Science, Dankook University; digestibility, apparent metabolisable energy in Talent Research Start-up Project of Guangdong Ocean broiler chickens. J Bioche Microbiol Biotechnol. 4: University (R18007); Innovative Strong School 7-10. Engineering Youth Talent Project (2017KQNCX090) and Key Platform Projects: Innovative Platform for SAS 2003. Sas User’s Guide. Release 9.1 Ed. SAS Efficient Use of Animal Genetic Resources in the Institute, Inc., Cary, NC, USA. Tropics and Healthy Farming by Guangdong Provincial Department of Education; and Natural Shimada A and Ohashi K 2003. Interfacial and Science Foundation of Guangdong Province emulsifying properties of diacylglycerol. Food Sci (2018A030307023). Technol Res. 9: 142-147. References Shimotoyodome A, Osaki N, Onizawa K, Mizuno T, Suzukamo C, Okahara F, Fukuoka D and Hase T OAC 1995. Official Method of Analysis, 16th Edition. 2012. Dietary 1-monoolein decreases Association of Office Analytical Chemists, postprandial GIP release by reducing jejunal Washington, DC, USA. transport of glucose and fatty acid in rodents. Am J Physiol-Gastr L. 303: G298-G310. AOAC 2005. Official Methods of Analysis. 18th ed. Association of Official Analytical Chemists, Upadhaya SD, Park JW, Park JH and Kim IH 2017. Washington DC, USA. Efficacy of 1, 3-diacylglycerol as a fat emulsifier in low-density diet for broilers. Poult Sci. 96: Ball RO and Aherne FX 1987. Influence of dietary 1672-1678. nutrient density, level of feed intake and weaning age on young pigs. II. Apparent nutrient van Milgen J, Noblet J and Dubois S 2001. Energetic digestibility and incidence and severity of efficiency of starch, protein and lipid utilization diarrhea. Can J Anim Sci. 67: 1105-1115. in growing pigs. J Nutr. 131: 1309-1318. Cho K, Hong J and Lee K 2010. Monoacylglycerol Yanai H, Tomono Y, Ito K, Furutani N, Yoshida H and (MAG)-oleic acid has stronger antioxidant, anti- Tada N 2007. Diacylglycerol oil for the metabolic atherosclerotic, and protein glycation inhibitory syndrome. Nutr J. 6: 43. activities than MAG-palmitic acid. J Med Food 13: 99-107. Zhang MS, Sandouk A and Houtman JC 2016. Glycerol Monolaurate (GML) inhibits human T cell Doreau M and Chilliard Y 1997. Digestion and signaling and function by disrupting lipid metabolism of dietary fat in farm animals. Brit J dynamics. Sci Rep. 6: 30225. Nutr. 78: S15-S35. Zhang MS and Houtman JC 2016. Human Serum Flickinger BD and Matsuo N 2003. Nutritional Albumin (HSA) Suppresses the Effects of characteristics of DAG oil. Lipids 38: 129-132. Glycerol Monolaurate (GML) on Human T Cell Activation and Function. PloS one 11: e165083. Lieberman S, Enig MG and Preuss HG 2006. A review of monolaurin and lauric acid: natural virucidal Zhao PY, Li HL, Hossain MM and Kim IH 2015. Effect and bactericidal agents. Alterna Complem Ther. of emulsifier (lysophospholipids) on growth 12: 310-314. performance, nutrient digestibility and blood profile in weanling pigs. Anim Feed Sci Technol. Martin D, Moran-Valero MI, Vázquez L, Reglero G and 207: 190-195. Torres CF 2014. Comparative in vitro intestinal digestion of 1, 3-diglyceride and 1-monoglyceride Zhao PY, Cho JH, Balasubramanian B, Kathannan S rich oils and their mixtures. Food Res Int. 64: 603- and Kim IH 2016. Essential oil and emulsifier in 609. low energy density diets increase growth performance and meat quality of finishing pigs. J Murata M, Ide T and Hara K 1997. Reciprocal Anim Sci. 94 (supplement2): 130. responses to dietary diacylglycerol of hepatic enzymes of fatty acid synthesis and oxidation in Zhao PY, Zhang ZF, Lan RX, Liu WC and Kim IH 2016. the rat. Brit J Nutr. 77: 107-121. Effect of lysophospholipids in diets differing in fat contents on growth performance, nutrient Nakajima Y 2004. Water-retaining ability of digestibility, milk composition and litter diacylglycerol. J Am Oil Chem Soc. 81: 907-912. performance of lactating sows. Animal 11: 984- 990. Nieto R, Miranda A, García MA and Aguilera JF 2002. The effect of dietary protein content and feeding Zhao PY and Kim IH 2017. Effect of diets with different level on the rate of protein deposition and energy energy and lysophospholipids levels on utilization in growing Iberian pigs from 15 to performance, nutrient metabolism, and body 50kg body weight. Brit J Nutr. 88: 39-49. composition in broilers. Poult Sci. 96: 1341-1347. NRC 2012. Nutrient Requirements of Swine, 11th Ed. National Academy Press, Washington, DC, USA. Projan SJ, Brown-Skrobot S, Schlievert PM, Vandenesch F and Novick RP 1994. Glycerol monolaurate inhibits the production of beta- lactamase, toxic shock toxin-1, and other staphylococcal exoproteins by interfering with signal transduction. J Bacteriol. 176: 4204-4209.

Liu W.C. et al. / Thai J Vet Med. 2018. 48(4): 681-687. 687 บทคัดย่อ Effects of supplementing varying levels of 1,3-diacylglycerol (DAG) on growth performance, apparent nutrient digestibility and selected fecal microbial populations in growing pigs Wen Chao Liu1# Kwan Sik Yun2# In Ho Kim3* The objective of the present study was to investigate the effects of 1,3-diacylglycerol (DAG) supplementation on growth performance, apparent digestibility and selected fecal microbial populations in growing pigs. A total of 80 crossbred pigs [(Landrace × Yorkshire) × Duroc] with an average initial body weight (BW) of 27.52 ± 0.20 kg were used in a 6 week feeding trial. Pigs were allotted to 4 dietary treatments based on their initial BW and sex (5 replications; 4 pigs per pen with 2 barrows and 2 gilts). Dietary treatments included: 1) CON, basal diets; 2) CON+0.075% 1,3-DAG; 3) CON+0.10% 1,3-DAG; 4) CON+0.15% 1,3-DAG. The final BW was linearly increased (P=0.0056) by 1,3-DAG supplementation. Average daily feed intake (ADFI) was improved (cubic, P=0.0312) by feeding 1,3-DAG during week 1-3. Overall (week 1-6), dietary 1,3-DAG supplementation led to a linearly increase in ADG (P=0.0042) and G:F (P=0.0235). In addition, dietary supplementation of 1,3-DAG linearly improved (P=0.0004) the apparent digestibility of gross energy and had a tendency (linear, P=0.0725) to improve the apparent digestibility of crude fat. Pigs fed 1,3-DAG supplemented diet also linearly decreased (P=0.0065) fecal E. coil populations at week 3 and linearly increased (P=0.0348) fecal Lactobacillus populations at week 6. In conclusion, these results show that 1,3-DAG can be used as a potential functional feed additive and exerts positive effects on growth performance, apparent energy digestibility and fecal microbiota in growing pigs. คำสำคัญ: 1,3-diacylglycerol growing pigs growth promoter 1Department of Animal Science, College of Agriculture, Guangdong Ocean University, Zhanjiang, Guangdong 524088, P. R. China 2Synergen Inc., 1106 Jung -Dong, Wonmi-Gu, Bucheon-Si, Gyeonggi-Do, South Korea 3Department of Animal Resource & Science, Dankook University, Cheonan, Chungnam 31116, South Korea #These authors contributed equally to this work *ผู้รบั ผดิ ชอบบทความ E-mail: [email protected]



Short Communication Comparative ultrasonographic and computed tomographic images of the adrenal glands of healthy cats Panrawee Phoomvuthisarn1* Kiatpichet Komin1 Nan Choisunirachon1 Abstract Feline adrenal disorders that affect the size of the adrenal glands can be detected by diagnostic imaging including ultrasonography and Computed Tomography (CT). The objectives of this study were to compare the appearance of the adrenal gland in healthy cats between these two techniques and establish the reference values of the CT dimensions of the adrenal gland in cats. We found that there was no significant difference in the length of both adrenal glands between the two techniques but the cranial and caudal heights of the left adrenal glands and cranial height of the right adrenal gland obtained from the CT images in post-contrast enhanced-phase were significantly thicker than those obtained from ultrasonography. Our results suggest the reference values (mean±SD) of the left adrenal gland length, cranial, and caudal heights as 11.12 ± 2.65, 3.69 ± 0.72, and 3.70 ± 0.77 mm, respectively, and those of the right adrenal gland as 11.00 ± 2.33, 3.80 ± 0.99, and 3.53 ± 0.73 mm, respectively. The dimensions of both adrenal glands obtained from ultrasonography have more positive correlation with the age and body weight of the cats when compared with those obtained from the CT. Different sensitivity between the two techniques might be the cause and the images obtained from ultrasonography might be affected by the positioning and bodyweight of the patient. Keywords: Adrenal glands, Cats, Computed Tomography, Normal, Ultrasonography 1Department of Surgery, Faculty of Veterinary Science, Chulalongkorn University, Bangkok, Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 689-698.

690 Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. Introduction sensitivity to detecting other adrenal disorders involving a high degree of local invasion or Adrenal glands are important organs in the mineralization compared to other advance imaging mammal body. They are paired, ovoid organs localized such as Computed Tomography (CT) (Combes and craniomedially to kidneys in the retroperitoneal space. Saunders, 2014). The normal location of the left adrenal gland is ventrolateral area to the aorta, while the right adrenal CT is an advance diagnostic imaging method gland lies more cranial and dorsolateral to the caudal that uses multiple electromagnetic radiation, x-rays, to vena cava. They consist of two layers of tissue which provide transverse anatomical images (Goldman, 2007). are the adrenal cortex and medulla. The adrenal cortex secretes important steroid hormones such as CT is used widely as a diagnostic tool for various feline mineralocorticoids, glucocorticoids, and sex diseases such as nasopharyngeal polyps and tympanic hormones. Apart from this, adrenal medulla secrete bulla disease (Oliveila et al., 2012). In the same way, CT catecholamine which regulates the autonomic nervous has been used to examine many organs in the system (Prentice and Wood, 1975). abdominal cavity of cats for instance, kidneys and intestines (Reichle et al., 2002; Suwa and Shimoda, Feline adrenal disorders are becoming more recognised nowadays. This may have resulted from an 2017). However, there has been no published report of increase popularity in having cats as companion animals and relevant diagnostic techniques are the normal appearance of feline adrenal glands routinely performed. Feline adrenal disorders regarding before and after contrast enhanced-CT including Feline Cushing‘s Syndrome (FCS) and Feline hyperaldosteronism (Conn’s syndrome) result in images yet. Therefore, the objective of this study was to unregulated diabetes mellitus and skin atrophy. (Chiaramonte and Greco, 2007). Differentiation tests describe the characteristics of feline adrenal glands in such as high dose dexamethasone suppression test, healthy cats using CT to establish the reference values plasma endogenous ACTH concentration and for the length and width of these glands. diagnostic imaging have been used to make a definitive diagnosis of these diseases. Materials and Methods Disease that affects the size of the adrenal Animals: This study was a retrospective comparison of glands can be detected by diagnostic imaging such as the CT and ultrasonography images of the adrenal hyperplasia and tumours (Pagani et al., 2016). glands of 30 healthy cats presented for a study of the Congenital adrenal hyperplasia can be found as an appearance of abdominal lymph nodes. They were autosomal recessive inherited disease in cats divided into three groups as young, mature, and senior (Knighton, 2004; Owens et al., 2012), while various cats according to their age (Table 1). All cats were types of primary adrenal tumour have been diagnosed physically normal from physical examination, with a so far (Boord and Griffin, 1999; Millard et al., 2009; complete blood count and blood chemistry evaluation, Calsyn et al., 2010; Guerios et al., 2015,). However, two and serologic examination for infectious diseases main types of primary adrenal tumour in cats are including Feline Leukemia Virus (FeLV) and Feline adrenocortical and medullary tumours Immunodeficiency Virus (FIV). Cats with abnormal (pheochromocytoma) (Henry et al., 1993), which can results were excluded from this study. cause an abnormal appearance in adrenal glands (Rossmeisl et al., 2000; Boag et al., 2004; Smith et al., Abdominal Ultrasonography: Dorsal recumbency 2012) . As a result, a change in features and the size of positioned cats were placed in a V-shaped pad. Cat feline adrenal glands is becoming more of a concern for skin was clipped over the abdominal region, cleaned the detection of adrenal gland hyperplasia and adrenal with 70% alcohol solution, and then covered with gland tumour in cats. contact gel at both sides of the abdominal wall. B mode ultrasonography was performed using a linear Ultrasonographic imaging of feline adrenal microconvex transducer operating at 7.0 MHz (LOGIQ glands is currently assessed by various studies to P6, GE Thailand) to view the shape, dimension and the emphasise the clinical and problematic nature of feline echogenicity of the adrenal glands. The transducer was adrenal disorders (Moore et al., 2000; Combes and placed in a sagittal plane to visualize the adrenal Saunders, 2014). Ultrasonography has been used glands. The left adrenal gland was imaged widely to detect the size and shape of adrenal glands. craniolateral to the aorta and cranial to the left renal To emphasize this, there have been several studies artery, while the right adrenal gland was visualized using ultrasonography to visualize adrenal glands in dorsolaterally to the caudal vena cava. A transverse cats. These studies have reported the normal ultrasonography was not the main procedure as the appearance of adrenal glands from healthy cats (Cartee short axis obtained from this procedure is considered et al., 1993; Combes et al., 2013) and also an abnormal unreliable in cats (Combes and Saunders, 2014). appearance from sick cats with adrenal disorders (Zatelli et al., 2007; Combes et al., 2013). Computed Tomography: Mature cats were forbidden Ultrasonography of adrenal glands is used to diagnose food and water 6-8 hours prior to anaesthesia, while feline related-adrenal diseases. It detects adrenal young cats had it withheld for 4 to 6 hours. asymmetry which is good for unilateral adrenal Acepromazine (0.03 mg/kg) and atropine (0.04 mg/kg) tumours, however, bilateral adrenal tumours may were administered intramuscularly (IM) as the cause misleading results (Zimmer et al., 2000; Zatelli et sedative agents. Then, the cats were anesthetized with al., 2007). In addition, ultrasonography has a lower propofol (4 mg/kg, intravenously (IV)) as an induction anesthetic. After intubation, anesthesia was maintained with the inhalation of isoflurane 1-2 % in

Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. 691 oxygen. A CT scan was then performed after placing filter, 1.25 mm sliced thickness before injecting non- the cats in sternal recumbency with the head pointing ionic, iodine contrast medium, iohexol (Omnipaque to the 64-slice CT gantry (64-slice helical CT unit; 300®, Cork, Ireland) (300 mg I/kg), into the cephalic Optima CT660®, GE Thailand). vein as a post-contrast enhanced-phase. CT images were recorded in DICOM files for analysis. Images After the scout phase, a Field of View (FOV) were then processed using the Osirix® program to was created to cover from the 13th vertebral column to obtain values of long and short axis of both adrenal the perineum to obtain pre- and post-contrast glands. enhanced-images. Then, CT scanning for the pre- contrast enhanced-phase was done using a low pass Table 1 Groups of cats involved in this study Clinical Features Overall Values No. of patient Young (G1; 0-7 months) Mature (G2; 8-83 months) 30 Sex Senior (G3; ≥84 months) 10 10 Overall 10 Young (G1; 0-7 months) Female (Intact/spayed) 18 (8/10) Male (Intact/spayed) 12 (6/6) Mature (G2; 8-83 months) Female (Intact/spayed) 5 (4/1) Senior (G3; ≥84 months) Male (Intact/spayed) 5 (5/0) Age (months) Overall Female (Intact/spayed) 6 (4/2) Male (Intact/spayed) 4 (1/3) Female (Intact/spayed) 7 (7/0) Male (Intact/spayed) 3 (3/0) Mean±SD 44.80±46.06 Range (4.00-132.00) Weight (kg) Young (G1; 0-7 months) Mean±SD 6.10±1.28 Breed Mature (G2; 8-83 months) Range (4-7) Senior (G3; ≥84 months) Overall Mean±SD 22.7±12.03 Young (G1; 0-7 months) Range (11-48) Mature (G2; 8-83 months) Senior (G3; ≥84 months) Mean±SD 105.6±19.43 Overall Range (84-132) Young (G1; 0-7 months) Mean±SD 3.45 ± 1.13 Mature (G2; 8-83 months) Range (1.5-5.7) Senior (G3; ≥84 months) Mean±SD 2.78±0.80 Range (1.5-4.1) Mean±SD 3.33±1.27 Range (1.5-4.7) Mean±SD 4.11±0.84 Range (2.9-5.7) DSH 23 Mixed breed 3 Scottish Fold 2 American Short Hair 1 Persian 1 DSH 10 DSH 6 Mixed breed 3 American Short Hair 1 Persian 1 DSH 8 Scottish Fold 2

692 Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. Statistical analysis: The Computed tomographic and respectively. There were no significant differences in ultrasonographic measurements were compared using the measured dimensions of the gland between pre- paired student’s t-test and the results were considered and post-contrast enhanced-phases. In contrast to significant at p < 0.05. Pearson correlation test was ultrasonography, no correlation between the size of used to determine the relation between the size of the both adrenal glands and age or body weight was found adrenal glands and the age, body weight, and sex of the by the CT images, except for the positive correlation cats. between the cranial height of the left adrenal glands and body weight (Pearson r = 0.47) (Figure 4). In the Results and Discussion same way, no significant difference in these values was found between groups of cats divided by their age and Of the thirty cats included in this study body weight. twelve were males (six were intact and six were castrated) and eighteen females (eight were intact and Compared to ultrasonography, the cranial ten were spayed). Most of them were Domestic Short and caudal heights of the left adrenal gland obtained Hair (DSH) and some were Scottish Fold, Persian, from the CT images in both phases were thicker than American short hair and mixed breeds. The average those obtained from the ultrasonography (p = 0.03 and Body Condition Score (BCS) and weight was 3.16 ± 0.83 0.0002, respectively, in the pre-contrast enhanced- and 3.45 ± 1.13 kg, respectively. phase and p = 0.0005 and 0.0002, respectively, in the post-contrast enhanced-phase). In addition, the cranial The shapes of both adrenal glands were and caudal heights of the right adrenal gland obtained bilobed and similar to each other. Most of them were a from the CT images in the post-contrast enhanced- hypoechoic structure surrounded by hyperechoic area phase also tended to be thicker than those obtained (Figure 1A-B). The left adrenal gland localized at the from the ultrasonography but significant difference cranial and medial to the left kidney and lateral to the was found only in the cranial height between the two aorta, while all the right adrenal gland was at the methods (p = 0.008). cranial and medial to the right kidney and lateral to the caudal vena cava. Microscopic mineralization was The mean±SD of pre-contrast enhanced- found as a faint small hyperechoic focus in the adrenal attenuation value of the left adrenal gland was 32.96 ± gland in one cat (Figure 2A). Of the left adrenal grand, 12.77 HU, while that of the right adrenal gland was the mean±SD of ultrasonologic measurement of length, 32.14 ± 11.04 HU. The mean ±SD of post contrast cranial, and caudal heights were 10.65 ± 2.59, 3.10 ± enhanced-attenuation values of the left adrenal gland 0.86, and 3.04 ± 0.66 mm, respectively, while those of was 148.13 ± 44.50 HU, while that of the right adrenal the right adrenal glands were 11.32 ± 2.26, 3.24 ± 0.80, gland was 151.82 ± 49.11 HU. The differences between and 3.24 ± 0.95 mm, respectively. The cranial height of the HU values were not statistically significant except the left adrenal gland and length of the right adrenal for the difference between the pre- and post-contrast gland have a positive correlation with age (Pearson r = enhanced-phases. The attenuation values of both left 0.36 and 0.47, respectively). In addition, there were and right adrenal glands obtained from the post- positive correlations between body weight and the contrast enhanced-phase were significantly higher cranial height and length of the left adrenal gland the than those of the pre-contrast enhanced-phase (p < length, cranial, and caudal heights of the right adrenal 0.0001). gland, (Pearson r = 0.36, 0.53, 0.67, 0.52, and 0.45, respectively) (Figure 3). However, when comparing In this study, both adrenal glands were between groups, the length of the adrenal glands tends imaged by ultrasonography and CT scan. According to to increase with age but no significant difference was ultrasonography, the location and appearance of the found between groups, except the length of right adrenal glands agree with those of other studies adrenal gland in senile cats (12.05 ± 2.12) was (Combes et al., 2013; Combes and Saunders, 2014). significantly longer than young cats (10.18 ± 1.96), (P Regarding the length, cranial and caudal heights, we value = 0.009). found that the average length of the healthy cat’s left adrenal gland (of 10.65 ± 2.59 mm which agreed with Sagittal views were selected to measure the other reports (8.90 - 10.70 mm) (Combes and Saunders, width of the adrenal gland in both pre- and post- 2014). However, the cranial and caudal heights of the contrast-enhanced phases. In general, both adrenal left adrenal gland in this study were slightly thinner glands on the sagittal and dorsal planes were round to (3.10 ± 0.86, and 3.04 ± 0.66 mm, respectively) than oval in shape (Figure 1C-E). In the pre-contrast those of other reports (3.50 - 4.30 mm, respectively) enhanced-phase, the mean±SD of the left adrenal gland (Combes and Saunders, 2014). This may result from the length, cranial, and caudal heights were 10.88 ± 1.73, biologic variation among the cats and different 3.47 ± 0.62, and 3.65 ± 0.62 mm, respectively, while sensitivity of the ultrasonography. The right adrenal those of the right adrenal glands were 10.26 ± 2.36, 3.49 glands in this study were slightly larger than the left ± 0.75, and 3.39 ± 0.67 mm, respectively. In addition, we glands which were similar to other’s finding (Combes found distinct adrenal mineralization in the pre- and Saunders, 2014). The average length, cranial and contrast enhanced-phase in the same cats that had the caudal heights of the right adrenal glands were 11.32 ± mineralization by ultrasonography. In the post- 2.26, 3.24 ± 0.80, and 3.24 ± 0.95 mm, respectively, contrast enhanced-phase, the mean±SD of the left compared to other’s finding (9.80 - 10.80 mm and 3.60 adrenal gland length, cranial, and caudal heights were - 4.30 mm, respectively) (Combes and Saunders, 2014). 11.12 ± 2.65, 3.69 ± 0.72, and 3.70 ± 0.77 mm, In addition, the size of the adrenal glands tends to respectively, while those of the right adrenal gland increase with age and bodyweight. However, no were 11.00 ± 2.33, 3.80 ± 0.99, and 3.53 ± 0.73 mm, significant difference was found between groups of the cats and this needs further investigation. This study

Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. 693 agrees with other studies in dogs and cats that age and al., 1996; Besso et al., 1997), but one study stated that it body weight has little effect on the size of the adrenal was not specific for the adrenal glands and could be gland (Douglass et al., 1997; Combes et al., 2013; found in normal cats as an age-related change (Howell Soulsby et al., 2015) . and Pickering, 1964). From this study, adrenal mineralization can be detected more clearly with CT From ultrasonography and CT, adrenal compared to ultrasonography, especially in the pre- mineralization was found in the same cat, of 11 months contrast enhanced-phase. However, it depends on the old, as an incidental finding. Adrenal mineralization is degree of mineralization and ultrasonic plane waves. one of the characteristics of adrenal tumor (Grooters et Figure 1 Adrenal glands in a healthy cat (arrows). Ultrasonographic images of the left (A) and right (B) adrenal glands. CT images of the left (C, sagittal view and D, dorsal view) and right (D, sagittal view and E, dorsal view) adrenal glands.

694 Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. Figure 2 Adrenal mineralization (arrows) imaged by ultrasonography (A) and CT (B, sagittal view and C, dorsal view). By CT scan, post-contrast enhanced-images In the present study, the cranial and caudal were more appropriate to identify adjacent structure heights of the left adrenal gland and the cranial height around both adrenal glands. However, there was no of the right adrenal gland measured from the CT significant difference in the length, cranial and caudal images in the post- contrast enhanced-phase were heights of the glands between the pre- and post- significantly thicker than those measured by the contrast enhanced-phases. However, the attenuation ultrasonography. This might result from the value in the post-contrast enhanced-image was differences in positioning of the cats and in sensitivity significantly higher than the pre-contrast enhanced- between the two techniques. As previously reported, images. We report the normal value of these phases, the ultrasonographic scanning in a sagittal plane of the which might be helpful for differentiating adrenal glands obtained the cranio-caudal length longer than tumor in cats. The attenuation value can be used for the the transverse plane of the body (Combes and detection of adrenal adenoma in humans (Korobkin et Saunders, 2014). al., 1996). Regarding the age and body weight of the cats, we found no significant correlation with the sizes

Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. 695 of adrenal glands obtained by CT. These findings were and accuracy when compared to ultrasonography. The similar to those reported by other studies in dogs images obtained from the ultrasonography might be (Bertolini et al., 2006). According to the study in affected from the positioning and body weight of the humans, CT scans have higher sensitivity, specificity patient (Abrams et al., 1982). A Ultrasound-LAG B Ultrasound-RAG Cranial height =0.1317*Age +0.28 Length =0.1168*Age +1.04 0.6 P < 0.05, r =0.363 2.0 P < 0.01, r =0.466 0.4 * 1.5 Cranial Height Length * 0.2 1.0 0.5 0.0 50 100 150 0.0 50 100 150 0 0 Age * Age * C D Ultrasound-LAG Ultrasound-LAG 6 0.6 Cranial height =0.2839*BW +0.21 2.0 Length =0.2015*BW +0.83 P < 0.01, r =0.533 P < 0.05, r =0.363 * 0.4 1.5 Cranial Height 0.2 Length 1.0 0.0 0.5 0 24 6 0.0 24 E 0 BW BW * 0.6 F Ultrasound-RAG Ultrasound-RAG Caudal height =0.2536*BW +0.23 0.4 Cranial height =0.2730*BW +0.22 0.6 P < 0.05, r =0.451 P < 0.01, r =0.522 Cranial Height Length 0.4 Caudal Height 0.2 0.2 0.0 24 0.0 24 6 0 BW 60 BW G 2.0 Ultrasound-RAG * Length =0.2529*BW +0.78 1.5 P < 0.0001, r =0.6730 1.0 0.5 0.0 24 6 0 BW Figure 3 Correlations of age with cranial height of the left adrenal gland (A), and with length of the right adrenal gland (B) from ultrasonography. Correlations of body weight with cranial height (C) and length (D) of the left adrenal glands, and with cranial (E), caudal (F), and length (G) of the right adrenal gland from ultrasonography (* = p<0.05, r<0).

696 Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. 0.6 0.4 CT-LAG Cranial height =0.2244*BW +0.31 P < 0.01, r =0.474 Cranial Height * 0.2 0.0 24 6 0 BW Figure 4 Correlation of body weight with cranial height of the left adrenal gland from CT (* = p<0.05, r<0). In conclusion, the normal CT image values of Boord M and Griffin C 1999. Progesterone secreting the normal adrenal glands of healthy cats are reported to be reference values. Also, CT is one of the alternative adrenal mass in a cat with clinical signs of techniques to visualize the adrenal gland in cats. hyperadrenocorticism. J Am Vet Med Assoc. 214(5): 666-9. Calsyn JD, Green RA, Davis GJ and Reilly CM 2010. Acknowledgements Adrenal pheochromocytoma with This study was funded by Grants for contralateral adrenocortical adenoma in a cat. Development of New Faculty Staff, J Am Anim Hosp Assoc. 46(1): 36-42. Ratchadaphiseksomphot Endowment Fund, Cartee RE, Hudson JA and Finn-Bodner S 1993. Chulalongkorn University. We would like to Ultrasonography. Vet Clin North Am Small acknowledge the staff of the Surgery and Imaging Anim Pract. 23(2): 345-77. Unit, Small Animal Teaching Hospital, Faculty of Chiaramonte D and Greco DS 2007. Feline adrenal Veterinary Science, Chulalongkorn University for their disorders. Clin Tech Small Anim Pract, 22(1): help and support. 26-31. Combes A, Pey P, Paepe D, Rosenberg D, Daminet S, References Putcuyps I, Bedu AS, Duchateau L, De Fornel- Abrams HL, Siegelman SS, Adams DF, Sanders R, Thinaud P, Benchekroun G and Saunders JH Finberg HJ, Hessel SJ and Mcneil BJ 1982. Computed tomography versus ultrasound of 2013. Ultrasonographic appearance of the adrenal gland: a prospective study. Radiology. 143(1): 121-8. adrenal glands in healthy and sick cats. J Abrams HL, Siegelman SS, Adams DF, Sanders R, Feline Med Surg. 15(6): 445-57. Finberg HJ, Hessel SJ and Mcneil BJ 1982. Computed tomography versus ultrasound of Combes A, and Saunders JH 2014. Ultrasonogic the adrenal gland: a prospective study. Radiology. 143(1): 121-8. examination of feline adrenal glands: A Bertolini G, Furanello T, De Lorenzi D and Caldin M Review. Revue Vétérinaire Clinique. 49(1):1- 2006. Computed tomographic quantification of canine adrenal gland volume and 12. attenuation. Vet Radiol Ultrasound. 47(5): 444-8. Douglass JP, Berry CR and James S 1997. Besso JG, Penninck DG and Gliatto JM 1997. Ultrasonographic adrenal gland Retrospective ultrasonographic evaluation of adrenal lesions in 26 dogs. Vet Radiol measurements in dogs without evidence of Ultrasound. 38(6): 448-55. adrenal disease. Vet Radiol Ultrasound. 38(2): Boag AK, Neiger R and Church DB 2004. Trilostane treatment of bilateral adrenal enlargement 124-30. and excessive sex steroid hormone production in a cat. J Small Anim Pract. 45(5): Goldman LW 2007. Principles of CT and CT 263-6. technology. J Nucl Med Technol. 35(3): 115- 28; quiz 129-30. Grooters AM, Biller DS, Theisen SK and Miyabayashi T 1996. Ultrasonographic characteristics of the adrenal glands in dogs with pituitary- dependent hyperadrenocorticism: comparison with normal dogs. J Vet Intern Med. 10(3): 110-5. Guerios SD, Souza CHM and Bacon NJ 2015. Adrenocortical tumor in a cat secreting more than one type of corticosteroid. JFMS Open Rep. 1, 2055116915617970.

Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. 697 Henry CJ, Brewer WG, Mongomery RD, Groth AH, Zatelli A, D'Ippolito P, Fiore I and Zini E 2007. Ultrasonographic evaluation of the size of the Cartee RE and Griffin KS 1993. Clinical adrenal glands of 24 diseased cats without endocrinopathies. Vet Rec. 160(19): 658-60. vignette. Adrenal pheochromocytoma. J Vet Zimmer C, Horauf A and Reusch C 2000. Intern Med. 7(3): 199-201. Ultrasonographic examination of the adrenal gland and evaluation of the hypophyseal- Howell JM and Pickering CM 1964. Calcium Deposits adrenal axis in 20 cats. J Small Anim Pract, 41(4): 156-60. in the Adrenal Glands of Dogs and Cats. J Comp Pathol. 74: 280-5. Knighton EL 2004. Congenital adrenal hyperplasia secondary to 11beta-hydroxylase deficiency in a domestic cat. J Am Vet Med Assoc. 225(2): 238-41, 231. Korobkin M, Brodeur FJ, Yutzy GG, Francis IR, Quint LE, Dunnick NR and Kazerooni EA 1996. Differentiation of adrenal adenomas from nonadenomas using CT attenuation values. AJR Am J Roentgenol. 166(3): 531-6. Millard RP, Pickens EH and Wells KL 2009. Excessive production of sex hormones in a cat with an adrenocortical tumor. J Am Vet Med Assoc. 234(4): 505-8. Moore LE, Billers DS and Smith TA 2000. Use of abdominal ultrasonography in the diagnosis of primary hyperaldosteronism in a cat. J Am Vet Med Assoc. 217(2): 213-5, 197. Oliveila CR, O'brien RT, Matheson JS and Carrera I 2012. Computed tomographic features of feline nasopharyngeal polyps. Vet Radiol Ultrasound. 53(4): 406-11. Owens SL, Downey ME, Pressler BM, Birkenheuer AJ, Chandler DW and Scott-Moncrieff JC 2012. Congenital adrenal hyperplasia associated with mutation in an 11beta-hydroxylase-like gene in a cat. J Vet Intern Med. 26(5): 1221-6. Pagani E, Tursi M, Lorenzi C, Tarducci A, Bruno B, Borgogno Mondino EC and Zanatta R 2016. Ultrasonographic features of adrenal gland lesions in dogs can aid in diagnosis. BMC Vet Res. 12(1): 267. Prentice FD, and Wood JG 1975. Adrenergic innervation of cat adrenal medulla. Anat Rec. 181(4): 689-703. Reichle JK, Dibartola SP and Leveille R 2002. Renal ultrasonographic and computed tomographic appearance, volume, and function of cats with autosomal dominant polycystic kidney disease. Vet Radiol Ultrasound. 43(4): 368-73. Rossmeisl JH, Scott-Moncrieff JC, Siems J, Snyder PW, Wells A, Anothayanontha L and Oliver JW 2000. Hyperadrenocorticism and hyperprogesteronemia in a cat with an adrenocortical adenocarcinoma. J Am Anim Hosp Assoc. 36(6): 512-7. Smith RR, Mayhew PD and Berent AC 2012. Laparoscopic adrenalectomy for management of a functional adrenal tumor in a cat. J Am Vet Med Assoc. 241(3): 368-72. Soulsby SN, Holland M, Hudson JA and Behrend EN 2015. Ultrasonographic evaluation of adrenal gland size compared to body weight in normal dogs. Vet Radiol Ultrasound. 56(3): 317-26. Suwa A and Shimoda T 2017. Intestinal gastrointestinal stromal tumor in a cat. J Vet Med Sci. 79(3): 562-566.

698 Phoomvuthisarn P. et al. / Thai J Vet Med. 2018. 48(4): 689-698. บทคดั ยอ่ การเปรียบเทียบลกั ษณะของภาพตอ่ มหมวกไตในแมวปกติทไ่ี ด้จากการบนั ทกึ ภาพคล่ืนเสียง ความถส่ี งู และการถา่ ยภาพรงั สีส่วนตดั อาศัยคอมพวิ เตอร์ พรรณระวี ภูมวิ ฒุ สิ าร1* เกียรติพิเชษฐ์ โคมนิ 1 แนน ชอ้ ยสุนริ ชร1 ความผิดปกตขิ องต่อมหมวกไตในแมวทมี่ ีผลกระทบต่อขนาดและรปู ร่างของต่อมหมวกไตสามารถตรวจพบได้ดว้ ยวธิ ีทางภาพ วนิ จิ ฉัย อาทิ การบันทกึ ภาพคลน่ื เสียงความถีส่ งู และการถา่ ยภาพรังสีสว่ นตดั อาศยั คอมพิวเตอร์ การศกึ ษาน้มี จี ุดประสงคเ์ พื่อเปรียบเทยี บ ลกั ษณะของต่อมหมวกไตในแมวปกตทิ ี่มสี ขุ ภาพดีจากการบันทกึ ภาพคลืน่ เสียงความถสี่ ูงและการถา่ ยภาพรงั สีส่วนตัดอาศยั คอมพิวเตอรแ์ ละ รายงานคา่ ปกติของขนาดของต่อมหมวกไตในแมวจากการถ่ายภาพรังสสี ว่ นตดั อาศยั คอมพวิ เตอร์ โดยการศกึ ษาคร้ังนีพ้ บวา่ ความยาวของตอ่ ม หมวกไตท่วี ัดจากทง้ั สองวธิ ีไม่มีความแตกต่างอย่างมนี ยั สาคัญทางสถติ ิ แต่คา่ ความกวา้ งดา้ นหน้าและด้านหลังของต่อมหมวกไตด้านซ้ายและ ค่าความกวา้ งด้านหนา้ ของต่อมหมวกไตดา้ นขวาจากภาพรงั สสี ว่ นตัดอาศยั คอมพวิ เตอรม์ ีค่ามากกว่าค่าที่ไดจ้ ากการบันทกึ ด้วยคลืน่ เสยี ง ความถี่สูงอยา่ งมีนัยสาคญั โดยค่าปกติของความยาว ความกว้างด้านหนา้ และ ดา้ นหลังของตอ่ มหมวกไตดา้ นซ้ายจากการศกึ ษาครั้งน้ีมคี า่ เทา่ กบั 11.12 ± 2.65 3.69 ± 0.72 และ 3.70 ± 0.77 มิลลเิ มตรตามลาดบั และตอ่ มหมวกไตด้านขวาเทา่ กบั 11.00 ± 2.33 3.80 ± 0.99 และ 3.53 ± 0.73 มลิ ลิเมตรตามลาดับ นอกจากน้ีขนาดของตอ่ มหมวกไตจากการบันทกึ ภาพคลนื่ เสยี งความถ่สี งู ยังมีความสมั พันธเ์ ชิงบวกกับ อายแุ ละนา้ หนักตวั ของแมวมากกวา่ ค่าท่ไี ดจ้ ากการถา่ ยภาพรงั สสี ่วนตัดอาศัยคอมพวิ เตอร์ ลักษณะดังกลา่ วอาจมีผลมาจากความไวและความ แม่นยาที่แตกตา่ งกันระหว่างสองวิธีโดยค่าที่ได้จากการบนั ทกึ ภาพคล่ืนเสียงความถสี่ ูงอาจไดร้ ับผลกระทบจากการจัดทา่ ทางระหว่างการ บันทึกภาพและนา้ หนักตวั ของแมวมากกวา่ การถ่ายภาพรังสีสว่ นตดั อาศัยคอมพิวเตอร์ คาสาคัญ: ต่อมหมวกไต แมว ภาพรังสสี ่วนตัดอาศยั คอมพิวเตอร์ ปกติ การบนั ทึกภาพคล่นื เสยี งความถ่ีสูง 1ภาควชิ าศลั ยศาสตร์ คณะสัตวแพทยศาสตร์ จฬุ าลงกรณ์มหาวิทยาลยั กรุงเทพมหานคร ประเทศไทย *ผูร้ บั ผดิ ชอบบทความ E-mail: [email protected]

Case Report Surgical treatment for tubular colonic duplication communicating to retroperitoneal space in an adult dog Nithida Boonwittaya1* Piyathip Choochalermporn2 Sunee Kunakornsawat3 Teerapol Sirinarumitr4 Abstract Canine colonic duplication is a rare congenital anomaly, usually symptomatic in the first year of life, with the common signs of bowel obstruction. Most importantly, colonic duplication in an adult dog has not been reported. This case document reports a non-communicating tubular colonic duplication with concurrent retroperitoneal mass in a 3- year-old female crossbred dog, presented for further diagnosis and treatment of the abdominal mass. Without any remarkable clinical signs, physical examination noted only abdominal distension with mild discomfort. Ultrasonography demonstrated a large mass in the mid-caudal abdomen, a tubular mass appearance on the colonic wall and the presence of peritoneal effusion. An exploratory laparotomy revealed a colonic tubular mass without communication with the true lumen. An opening tract from the tubular mass to a massive distended retroperitoneal sac was identified with a ruptured area. Surgical removal of the retroperitoneal sac and partial colectomy of the conjoined part were accomplished with full recovery and normal defecation within 12 months of the follow-up processes. The diagnosis of a non-communicating tubular colonic duplication was confirmed by the histopathological result. In consideration of the abdominal mass, colonic duplication should be listed as a differential diagnosis in any age of dog. Other associated abnormalities should be included in the investigation, in order to prevent any complications. Surgery is recommended to achieve a final diagnosis and an applicable treatment for successful outcomes. Keywords: colonic duplication, dog, partial colectomy, retroperitoneal space 1Surgery Unit, Kasetsart University Veterinary Teaching Hospital, Faculty of Veterinary Medicine, Kasetsart University, Phahon Yothin road, Chatuchak, Bangkok 10900, Thailand 2Kasetsart Veterinary Imaging and Radiotherapy Center, Kasetsart University Veterinary Teaching Hospital, Faculty of Veterinary Medicine, Kasetsart University, Phahon Yothin road, Chatuchak, Bangkok 10900, Thailand 3Department of Companion Animals Clinical Sciences, Faculty of Veterinary Medicine, Kasetsart University, Phahon Yothin road, Chatuchak, Bangkok 10900, Thailand 4Department of Pathology, Faculty of Veterinary Medicine, Kasetsart University, Phahon Yothin road, Chatuchak, Bangkok 10900, Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 699-706.

700 Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. Introduction space filled with mix echogenic content located near the right kidney (Figure 1). One part of the colon Colonic duplication is a remarkably rare showed a thickened wall with the focal tubular mass congenital malformation in veterinary medicine, and effect on it. An amount of mix echogenic peritoneal there have been only eight publicly reported single effusion was detected in the caudal part of the cases in dogs (Jakowski, 1977; Longhofer et al., 1991; abdomen. Intestinal mass with severe obstruction and Lorinson and Lorin, 1995; Shinozaki et al., 2000; Arthur intestinal leakage were suspected. Exploratory surgery et al., 2003; Landon et al., 2007; de Battisti et al., 2013; was performed for diagnosis and treatment on the Fernandez et al., 2017), one in a horse (Bassage et al., same day of referral. 2000) and one in a cat (Kramer et al., 2007). Other concurrent congenital abnormalities, including cecal Prior to surgery, the dog was given a systemic malformations, urogenital duplications and spinal antibiotic of marbofloxacin (4 mg/kg, IV; column abnormalities were noted in the first three Marbocyl®, Vetoquinol, France) at the outpatient canine cases, which were untreated and euthanized department. Premedication of midazolam (0.15 (Jakowski, 1977; Longhofer et al., 1991; Lorinson and mg/kg, IV; Midazolam Sandoz®; EVER Pharma Jena Lorin, 1995). All of the previous canine reported cases GmbH, Germany) and morphine (0.5 mg/kg, IM; were young patients. Clinical signs of colonic Morphine Sulfate Injection; M & H Manufacturing, duplication are nonspecific, varying from Co., Ltd. For Food and Drug Administration, Thailand) asymptomatic to severe progressive signs, depending were administered. Anesthesia was induced with on location and the association of the duplicated propofol (2 mg/kg, IV; Troypofol®; Troikaa Pharm., segment with other abdominal organs. The varieties of Ltd., India) and maintained with isoflurane symptoms often include defecation problems such as (AttaneTM; Piramal Critical Care, Inc., USA) in 100% rectal prolapse, dyschezia, tenesmus and constipation; oxygen. A loading dose of ketamine (0.5 mg/kg, IV; and other signs of gastrointestinal occlusion, Ketamine-Hameln, Hameln Pharm. Gmbh, Germany) abdominal distention and pain. A number of imaging and lidocaine (1 mg/kg, slow IV; Locana, L.B.S. techniques e.g. barium enema, colonoscopy, Laboratory, Ltd., Thailand) was administered followed abdominal ultrasonography and negative contrast by intravenous constant rate infusions of a mixture of computed tomography have been used for diagnosis in veterinary medicine. Moreover, exploratory morphine (0.3 mg/kg/h), lidocaine (50 g/kg/min) laparotomy, gross examination and histopathological evaluation are considered procedures for final and ketamine (10 g/kg/min) (MLK) in isotonic diagnosis and type identification. Treatment of this crystalloid fluid at 10 ml/kg/h. condition depends on type, location, shape, and size of the duplication (Fernandez et al., 2017). Previously in On abdominal exploration, a single tubular dogs, five of eight cases had surgical correction (10 x 3 cm) cavitary mass was grossly found attached performed with various procedures (Shinozaki et al., to the right side of the descending colon and extended 2000; Arthur et al., 2003; Landon et al., 2007; de Battisti in parallel from the cranial part to approximately 4 cm. et al., 2013; Fernandez et al., 2017) and all of them were cranial to the pelvic inlet (Figure 2). The mass was single colonic duplication without other concurrent equal in size of a diameter to the colonic tube and problems. adhered with both mesenteric- and antimesenteric borders. A shallow depression was detected as a septal This case document is the first to describe the wall on both borders between them with multiple concomitant conditions of tubular non-communicated branches of vessels presented on the grooves. There colonic duplication with a connecting tract to form a was blood supply sharing on attachments along the retroperitoneal mass in an adult dog. Surgical excision length of the segment. Functionally, the true colon had of the retroperitoneal sac and partial colectomy were a normal outflow without obstructive evidence. At the performed as treatments with successful outcomes. distal end of the mass, there was an opening tract communicating with the retroperitoneal space, which Case description was distended massively to be a large sac was approximately 10 x 15 cm in size. The retroperitoneal A 3-year-old 12-kg intact female crossbred sac located in the right mid- to the caudal abdomen and dog was presented with a week history of abdominal contained a large volume of light, brown, mixed distension at the Kasetsart University Veterinary mucous fluid with a dark brown substance. The Teaching Hospital (KUVTH), Bangkaen campus. ruptured opening hole of the sac was found at the Referral radiography noted the presence of a large, soft distal end near the level of the bladder, where the fluid opacity tissue extending from the mid- to caudal was leaking into the caudal abdominal cavity. The abdomen, whereas ultrasonography revealed a dilated fluid was collected for cytological evaluation and colon with hyperechoic content. Abdominal mass with bacteriologic culture. To identified the communication megacolon, severe constipation, colonic obstruction on the adhered site of the colonic tube, the and intestinal mass were suspected primarily. retroperitoneal sac was incised adjacent to the opening distal end of tubular mass and a finger was inserted On physical examination, abdominal into the tubular lumen to investigate. There was a distention and firmness were noted. On palpation, complete septal wall along the length and the cranial mild discomfort was elicited without any remarkable end of the tubular mass terminated in a blind pouch abnormalities. A complete blood count and serum limited to the descending colon. Other abdominal biochemistry results were within normal limits. organs were examined and appeared to be normal. The Abdominal ultrasonography at KUVTH identified a retroperitoneal mass was excised by undermining it large pear-shaped structure in the retroperitoneal free from the other retroperitoneal structures. The right

Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. 701 kidney, ureter and other retroperitoneal structures submitted for histopathological evaluation. End-to- were examined and abnormalities were not detected. end anastomosis was performed by full thickness one- A partial colectomy was performed to remove the layer closure with 4-0 polydioxanone suture in a conjoined segment. The lumen of the descending colon simple interrupted pattern. The leakage was checked was occluded at 4 cm from both ends of the adhered thoroughly and the mesenteric defect was closed. The site after the fecal content had been milked away from abdomen was irrigated with warm isotonic saline the lumen. The vasa recta vessels were ligated at the solution. A closed active suction drain (MILA point of transection and the major colic vessels were International Inc., USA) was placed into the abdomen preserved intact. The descending colon was excised to establish peritoneal drainage and the abdomen was unreservedly approximately 3 cm cranially and closed routinely. caudally to the conjoined part. The tissues were Figure 1 Sagittal ultrasonographic images of mid-(A) and caudal abdomen (B) (C): Notice a large fluid-filled area with a mixture of echogenic cellular components located in the retroperitoneal space (X marker) near the right kidney (RK). The peritoneal free fluid (FF) is visible near the descending colon (asterisk). Focal fluid-filled tubular mass effect (cross) is evident on some part of the colonic wall (arrows).

702 Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. Figure 2 Intraoperative photographs of mid-to caudal abdomen: (A) An equal-sized duplicated colon (white arrow) is present on the right border of the descending colon (black arrow), the branches of vessel are visible along the groove on the antimesenteric conjoined area (white arrowheads) and the tract is shown at the distal end of the duplicated lumen (Tract). (B) The right lateral view of the descending colon shows the sharing blood supply on the mesenteric border (black arrowheads), the connecting tract (Tract) communicating to the retroperitoneal space and the accumulated fluid-filled retroperitoneal sac (asterisk). Recovery from surgery happened No growth occurred on the bacterial culture uneventfully. The dog was treated postoperatively of the peritoneal effusion that leaked from the after admission in the KUVTH critical care unit. retroperitoneal mass. The cytology result was Defecation was noted as normal initially. One week mucinous non-septic suppurative inflammation. after surgery diarrhea and tenesmus were presented Histopathologic examination of the tubular lesion but then resolved gradually over a week. The close found two lumens separated by a septum. The walls of active suction drain was removed after 12 days these tubular structures were composed of mucosa, postoperatively. The dog had normal defecation and submucosa and both circular and longitudinal improved in appetite leading to a gain in weight up to muscular layers which were similar to the normal 15 kg (6 months after surgery). During the 12 months colon. The mucosa was lined with mostly goblet cells. of the postoperative follow-up, the condition was good Lymphocytes and plasma cell infiltration were and there were no other abnormal findings on physical presented throughout this layer. There were multiple examination. foci of mucosal ulceration. Moreover, some part of the septum were composed of loose connective tissue as

Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. 703 the serosal layer with blood vessels (Figure 3). The duplication was made as a conclusion from the gross diagnosis of non-communicating tubular colonic lesion and histopathological pattern in this dog. Figure 3 Photomicrograph of a section of the conjoined septum of the duplicated colon: Loose connective tissue with blood vessels as the serosal layer of the colon is present in some part of the septum and discontinuous. Moderate hemorrhage and edema of the septum is observed. H&E stain; bar = 100 m Discussion retroperitoneal space is unseen in any record. In this dog, it is assumed that the fusion between the Because of the rarity of canine colonic duplicated colon and retroperitoneal space may had duplication, the classification system has to be referred been a result of the defect during embryogenesis like from human medicine. Briefly, this congenital other defects that usually present together with malformation can be categorized into two types as type alimentary duplication (Spaulding et al., 1990; Arthur I and II mean without and with other anomalies, et al., 2003). Conversely, this fusion also may happen respectively, while the duplication may or may not after a burst of the over-distended lumen of type Ib communicate with a true colon in both types. In type I, duplicated colon after birth. Nevertheless, the the shape of the duplicated form is defined as a. etiologies of this congenital defect in the canine case spherical, b. tubular, c. double-tubular with should be investigated more in future. communications, d. loop duplication with separate blood supply and e. multiple duplications (mixing of Colonic duplication generally is detected in a., b. and c.) (Kottra and Dodds, 1971) (Figure 4). In juveniles especially when the patient shows clinical human type II hindgut duplication mostly occurs with signs of gastrointestinal obstruction, defecation urogenital and musculoskeletal malformation (Landon problems or other abnormal organ systems; including et al., 2007; Banchini et al., 2013). In previously urogenital and skeletal abnormalities (Landon et al., reported canine cases, three of them were type II 2007). However, signs can be hidden continuously for (Longhofer et al., 1991; Lorinson and Lorin, 1995; years until clinical problems show up (Fotiadis et al., Jakowski, 1997) and four of them were type I with 2005; Banchini et al., 2013). In dogs, this is the first communications (Shinozaki et al., 2000; Landon et al., report of colonic duplication in an adult (3 years old). 2007; de Battisti et al., 2013; Fernandez et al., 2017). In consequence of the connection of the duplicated There was only one reported case of a type I non- colon with retroperitoneal space, duplicated lumen communicating form without other abnormalities was able to constantly drain self-produced content out (Arthur et al., 2003). Moreover, other alimentary and did not distend to compress the true lumen or duplications have been reported in canine cases other organs nearby. The fecal movement in the true (Spaulding et al., 1990). colon was not diminished and the dog could urinate normally. With this incidence, the common clinical The duplication in this case was different. A sign of duplication defect was unidentified and the dog type Ib colonic duplication was suspected by A gross could endure suffering from colonic duplication until appearances. However, the fistula from the non- its maturity. This case shows that the onset of clinical communicating duplicated colon to retroperitoneal signs depends on type, location and organ association. space make the classification in this case unclear. In a Additionally, colonic duplication cases may not show female child with long segment colonic duplication, a the signs of gastrointestinal or defecation problem and rectovaginal, rectovulvar, or rectoperineal fistula can be presented in the young adult case. notably presented (Kaur et al., 2004) but a tract to

704 Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. Radiography and ultrasonography have been was reinforced and repaired with porcine small used as prior diagnostic procedures in cases of these intestinal submucosa (Arthur et al., 2003). According to abnormalities but some imaging results can lead to our case’s detail, there was the complete common misdiagnosis (Banchini et al., 2013). Even computed septal wall and the equal-sized duplicated tubular tomography confirmedly reported more precisely in adhered to both mesenteric- and antimesenteric canine colonic duplication investigation (de Battisti et borders of the true lumen. Similar to the recent al., 2013; Fernandez et al., 2017), exploratory publicity (Fernandez et al., 2017), there was fully laparotomy still has to be performed for final shared blood supply on the adherent site, thus, the evaluation and treatment. Because of the limited detail partial colectomy was preferred for treatment, in this on the source of peritoneal effusion and the colonic case, to prevent vessel damage complication abnormalities ultrasonographically in the dog of this postoperatively. In addition, excision of the common report, the exploratory laparotomy was most septum in our case may obtain the double sized appropriate. diameter of the new colonic tube with the questionable function of half part of the lumen that was the The first three cases, which were type II duplicated lumen, in which the intestinal content had colonic duplication, were untreated and euthanized. never been passed through before. Moreover, there The other five cases had surgical treatments have been incidences of malignant transformation performed. Three of them were communicating types associated with the alimentary duplication in human and the surgical procedures included incision of the cases (Orr and Edwards, 1975; Hickey and Corson, common septum (Shinozaki et al., 2000) mucosal 1981; Horie et al., 1986) the incidence in the veterinary stripping (Landon et al., 2007) and excision of the literature, until now, have not been reported. For all of these common septum (de Battisti et al., 2013). The fourth reasons to prevent future complications, partial case was a non-communicating colonic duplication colectomy tended to be the recommended choice of with lack of shared blood supply, excision of the surgical procedure in this case. duplicated tube was performed and the colonic wall Figure 4 Type I colonic duplications: (a) spherical, (b) tubular, (c) double-tubular with communication(s) and (d) loop duplication with separate mesentery and blood supply. The previous case report relating to tomography may be used to provide specific retroperitoneal abscess in dogs demonstrated that clinical information (Hylands, 2006; Marvel and MacPhail, signs commonly presented of nonspecific, chronic, fever and 2013). In our report, only mild abdominal discomfort was presented on attempted abdominal palpation on abdominal and/or back pain with progressive deterioration the visiting day without the history of other (Marvel and MacPhail, 2013). The most common remarkable signs. The retroperitoneal mass effect was etiologies were a foreign body from ingestion, noted and the dog had evidence of peritoneal effusion migrating plant material, infection from a bite wound, on ultrasonographic views, which influenced our reaction from suture material and unknown causes. decision to perform emergency abdominal exploration Ultrasonography is the first basic line and preferable on the referring day. Surgical removal with abdominal diagnostic tool for retroperitoneal space disease. lavage and the closed active suction drain placement Nevertheless, advanced imaging such as computed

Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. 705 were elected to eliminate the retroperitoneal mass and complicating colonic duplication. Cancer. 47(3): provide postoperative drainage. Because of the 602-609. complete septum on duplication and the result of Horie H, Iwasuki I and Takahashi H 1986. Carcinoid in cytology with bacterial culture, the leakage content in a gastrointestinal duplication. J Pediatr Surg. the abdomen was not contaminated with feces. 21(10): 902-904. Moreover, it is believed that the leakage had happened Hylands R 2006. Veterinary diagnostic imaging. recently prior to the visiting day. As a consequence, the Retroperitoneal abscess and regional cellulitis dog did not show clinical signs of severe peritonitis on secondary to a pyelonephritis within the left presentation. kidney. Can Vet J. 47(10): 1033-1035. Jakowski RM 1997. Duplication of colon in a Labrador So far this is the first report of partial Retriever with abnormal spinal column. Vet colectomy as a treatment of type I tubular non- Pathol. 14(3): 256-260. communicating colonic duplication in an adult dog Kaur N, Nagpal K, Sodhi P and Minocha VR 2004. with a concurrent tract connecting to retroperitoneal Hindgut duplication–case report and literature space, in which the removal was performed with a review. Pediatr Surg Int. 20(8): 640-642. good outcome. Kottra JJ and Dodds WJ 1971. Duplication of the large bowel. Am J Roentgenol Radium Ther Nucl Med. Conclusively, colonic duplication should be 113(2): 310-315. investigated as a differential diagnosis in both Kramer A, Kyles AE and Labelle P 2007. Surgical symptomatic and asymptomatic cases with unknown correction of colonic duplication in a cat. J Am large intestinal mass in any age of dog. Laparotomy Anim Hosp Assoc. 43(2): 128-131. with histopathology is necessary for final diagnosis Landon BP, Abraham LA, Charles JA and Edwards GA and surgical treatment is required. Alignment and 2007. Recurrent rectal prolapse caused by colonic condition of all abdominal organs must be seriously duplication in a dog. Aust Vet J. 85(9): 381-385. investigated perioperatively to be aware of other Longhofer SL, Jackson RK and Cooley AJ 1991. complications. Hindgut and bladder duplication in a dog. J Am Anim Hosp Assoc. 27(1): 97-100. Acknowledgements Lorinson VD and Lorin R 1995. Intraluminale kolon- und zakumduplikatur bei einer 12 wochen alten The authors wish to acknowledge Saroch münsterländerhundin. Wien Tierarztl Mschr. 82: Kaewmanee, Ph.D. and PodjanaWattananit, Ph.D. for 355-357. their comments and assistance in improving this Marvel SJ and MacPhail CM 2013. Retroperitoneal manuscript. abscesses in seven dogs. J Am Anim Hosp Assoc. 49(6): 378-384. References Orr MM and Edwards AJ 1975. Neoplastic change in duplications of the alimentary tract. Br J Surg. Arthur EG, Fox DB, Essman SC, Turnquist S and Bondy 62(4): 269-274. Jr PJ 2003. Surgical treatment of Shinozaki JK, Sellon RK, Tobias KM and Chamberlain noncommunicating duplication of the colon in a K 2000. Tubular colonic duplication in a dog. J dog. J Am Vet Med Assoc. 223(2): 210-214. Am Anim Hosp Assoc. 36(3): 209-213. Spaulding KA, Cohn LA, Miller RT and Hardie EM Banchini F, Delfanti R, Begnini E, Tripodi MC and 1990. Enteric duplication in two dogs. Vet Radiol. Capelli P 2013. Duplication of the transverse 31(2): 83–88. colon in an adult: case report and review. World J Gastroenterol. 19(4): 586-589. Bassage LH, Habecker PL, Russell EA and Ennulat D 2000. Colic in a horse associated with a massive cystic duplication of the ascending colon. Equine Vet J. 32(6): 565-568. De Battisti A, Harran N, Chanoit G and Warren-Smith C 2013. Use of negative contrast computed tomography for diagnosis of a colonic duplication in a dog. J Small Anim Pract. 54(10): 547-550. Fernandez N, Morrison L, Liuti T, Frame M and Yool D 2017. Type Ia (spherical) communicating colonic duplication in a dog treated with colectomy. J Small Anim Pract. 58(5): 298-300. Fotiadis C, Genetzakis M, Papandreou I, Misiakos EP, Agapitos E and Zografos GC 2005. Colonic duplication in adults: report of two cases presenting with rectal bleeding. World J Gastroenterol. 11(32): 5072-5074. Hickey WF and Corson JM 1981. Squamous cell carcinoma arising in a duplication of the colon: case report and literature review of squamous cell carcinoma of the colon and of malignancy

706 Boonwittaya N. et al. / Thai J Vet Med. 2018. 48(4): 699-706. บทคัดย่อ การผา่ ตัดรักษาภาวะล้าไสใ้ หญ่ซา้ ซ้อนชนิดท่อทเี่ ช่อื มต่อกบั ช่องหลงั ชอ่ งท้องในสนุ ขั โต นธิ ดิ า บุญวทิ ยา1* ปยิ ทิพย์ ชูเฉลิมพร2 สณุ ี คณุ ากรสวัสด์ิ3 ธีระพล ศิรินฤมิตร4 ภาวะลา้ ไส้ใหญ่ซา้ ซ้อนในสุนัข เปน็ ความผดิ ปกติแต่กา้ เนิดทพี่ บไดย้ าก โดยมักแสดงอาการในช่วงขวบปีแรกด้วยภาวะลา้ ไสอ้ ดุ ตนั ที่ สา้ คัญยังไมเ่ คยมรี ายงานในสนุ ขั โตมาก่อน รายงานสตั วป์ ว่ ยนีอธบิ ายถึงภาวะลา้ ไส้ใหญ่ซ้าซอ้ นชนดิ ทอ่ ท่เี ช่อื มต่อกบั ช่องหลงั ชอ่ งท้องในสุนัข พันธ์ผสมอายุ 3 ปี ซ่งึ ถูกสง่ ตัวเพื่อเขา้ รบั การวนิ ิจฉัยเพิ่มเติมและทา้ การรกั ษาเนืองอกในชอ่ งท้อง การตรวจร่างกายพบเพยี งช่องทอ้ งโตและไม่ สบายตัว จากการตรวจคลน่ื เสียงสะท้อนความถสี่ ูงพบกอ้ นเนือขนาดใหญบ่ รเิ วณชว่ งกลางถึงทา้ ยของชอ่ งท้อง ก้อนเนือท่ีผนังล้าไสใ้ หญ่ และ ของเหลวในชอ่ งท้อง การผา่ ตดั เปิดชอ่ งทอ้ งพบก้อนเนือทลี่ า้ ไส้ใหญ่มลี ักษณะเปน็ ท่อยาวและไมม่ รี ูเชอ่ื มตอ่ กับช่องภายในล้าไส้ บริเวณปลาย ก้อนเนอื พบทางเช่ือมไปยงั ถงุ ขนาดใหญใ่ นชอ่ งหลงั ช่องทอ้ งซง่ึ มีรรู ่วั การรกั ษาโดยการตดั ลา้ ไส้ใหญบ่ างส่วนในตา้ แหน่งซา้ ซอ้ นและสว่ นท่ีอยู่ ติดกันออก ร่วมกับ การเลาะถงุ ในชอ่ งหลงั ชอ่ งทอ้ งประสบผลส้าเรจ็ สุนขั ฟ้นื ตัวดีและอจุ จาระไดเ้ ปน็ ปกติตลอดระยะเวลา 12 เดอื นของการ ตดิ ตามอาการ ผลพยาธิวิทยายนื ยันภาวะลา้ ไสใ้ หญซ่ ้าซ้อนชนิดท่อที่ไมเ่ ชือ่ มต่อกับล้าไสจ้ รงิ ดังนันภาวะลา้ ไสใ้ หญซ่ า้ ซอ้ นควรถกู น้ามา พิจารณาในการวนิ ิจฉยั แยกโรคกรณีก้อนเนือช่องทอ้ งในสุนัขทกุ ชว่ งวยั รวมทงั ควรวินิจฉยั ความผดิ ปกตทิ ่ีเก่ยี วขอ้ งอน่ื ๆ เพ่ือป้องกัน ภาวะแทรกซ้อนท่ีอาจเกดิ ขึน สา้ หรบั การผ่าตดั นนั สามารถช่วยในการวนิ จิ ฉัยโรคขันสดุ ทา้ ย ร่วมกบั ให้การรักษาท่เี หมาะสมและประสบ ผลสา้ เรจ็ คา้ ส้าคญั : ภาวะลา้ ไสใ้ หญ่ซา้ ซ้อน สุนัข การตดั ลา้ ไส้ใหญ่บางส่วน ชอ่ งหลังชอ่ งท้อง 1หนว่ ยศลั ยกรรม โรงพยาบาลสัตว์มหาวทิ ยาลัยเกษตรศาสตร์ คณะสตั วแพทยศาสตร์ มหาวทิ ยาลยั เกษตรศาสตร์ ถนนพหลโยธนิ จตจุ ักร กรงุ เทพฯ 10900 2ศูนย์ภาพวินจิ ฉยั และรงั สรี ักษาทางสตั วแพทย์ โรงพยาบาลสตั ว์มหาวิทยาลัยเกษตรศาสตร์ คณะสัตวแพทยศาสตร์ มหาวทิ ยาลยั เกษตรศาสตร์ ถนนพหลโยธนิ จตจุ กั ร กรงุ เทพฯ 10900 3ภาควิชาเวชศาสตรค์ ลนิ กิ สัตว์เลย้ี ง คณะสัตวแพทยศาสตร์ มหาวทิ ยาลัยเกษตรศาสตร์ ถนนพหลโยธนิ จตุจกั ร กรงุ เทพฯ 10900 4ภาควชิ าพยาธวิ ทิ ยา คณะสัตวแพทยศาสตร์ มหาวทิ ยาลัยเกษตรศาสตร์ ถนนพหลโยธนิ จตจุ กั ร กรุงเทพฯ 10900 *ผรู้ ับผดิ ชอบบทความ E-mail: [email protected]

Diagnostic Forum ECG Quiz Chollada Buranakarl1* Saikaew Sutayatram1 History minute without cardiac murmur from auscultation using stethoscope. The systolic blood pressure was 120 A 10-year-old female Pomeranian dog was mmHg. The blood collection was performed and the presented to the Small Animal Teaching Hospital, results indicated the dog had anemia (PCV = 17%), Chulalongkorn University, on December, 2018, with a thrombocytopenia (42,000 per µL) and mild elevation history of anorexia, frequent vomiting, depress and of ALT (348 Unit/L). The azotemia was found with oliguria. Physical examination revealed that the dog BUN and creatinine of 127 mg/dL and 5.5 mg/dL, was depressed but responsive to the stimuli. The dog respectively. The concentrations of sodium (Na+) and was dehydrated (7%) with pale-pink mucous potassium (K+) were 134 and 7.75 mM, respectively. membrane. The capillary refilling time of 1.5 seconds Antibody test kit was positive for Ehrlichia canis. with strong femoral pulse were detected. The respiratory rate was 65 breath per minute with normal Electrocardiogram (ECG) was recorded as shown in sound while heart rate was approximately 60 beats per Figure 1. Figure 1 The ECG with a paper speed of 25 mm per second (A) and 50 mm per second (B) was recorded in a Pomeranian suffering from acute kidney injury. Please answer before turning to the next page. 1Department of Veterinary Physiology, Faculty of Veterinary Science, Chulalongkorn University Thai J Vet Med. 2018. 48(4): 707-708.