Prevention of Postoperative Complications Following Surgical Treatment of Equine Colic

1 Prevention of postoperative complications following surgical treatment of equine colic:

2 current evidence

4 Keywords: Horse; colic; postoperative complications; surgical site infection; postoperative

5 colic; postoperative ileus

7 Abstract

8 Changes in management of the surgical colic patient over the last 30 years have resulted in

9 considerable improvement in postoperative survival rates. However, postoperative

10 complications remain common and these impact negatively on horse welfare, probability of

11 survival, return to previous use and the costs of treatment. Multiple studies have investigated

12 risk factors for postoperative complications following surgical management of colic and

13 interventions that might be effective in reducing the likelihood of these occurring. The

14 findings from these studies are frequently contradictory and the evidence for many

15 interventions is lacking or inconclusive. This review discusses the current available evidence,

16 identifies areas where further studies are necessary and factors that should be taken into

17 consideration in study design.

1 18 Introduction

20 Colic is one of the most common causes of mortality in managed equine populations [1; 2]

21 accounting for 28% of reported horse deaths annually [3]. Most colic episodes that occur in

22 the general equine population resolve spontaneously or following medical treatment but

23 approximately 8% of episodes will require surgical treatment or euthanasia [4; 5]. Surgical

24 management of colic started to become more commonplace in the 1970’s. Since that time,

25 our knowledge of the pathophysiology and epidemiology of colic has improved significantly.

26 Advancements in surgical techniques, anaesthetic management and postoperative care have

27 improved morbidity and mortality rates [6; 7]. Current survival to discharge from hospital in

28 horses that recover from anaesthesia is reported to be 74-85% [8-14] with 63-85% of those

29 cases returning to athletic performance [15-18]. Postoperative complications are common,

30 have important welfare and economic consequences, and may impact negatively on horses’

31 likelihood of survival or future athletic use [18; 19]. Therefore, development of strategies

32 and/or therapies to minimise development of these complications based on available evidence

33 is important.

35 This review summarises the complications that can occur following surgical treatment of

36 colic, the prevalence of these complications, and the predisposing factors leading to their

37 occurrence. Particular focus has been placed on appraisal of study design, areas that require

38 future research and factors that should be considered in the design of such studies. For the

39 purpose of critical appraisal of relevant studies, a systematic electronic literature search of

40 Medline Database, Web of Science and Cab Direct was performed and only papers with level

41 4 or more of evidence (http://www.cebm.net/oxford-centre-evidence-based-medicine-levels-

42 evidence-march-2009/) were considered for inclusion. Expert opinions, in-vitro studies and

2 43 studies where interventions were tested in clinically healthy horses or cadaver studies were

44 excluded. In order to focus on studies relevant to current surgical management of colic, the

45 search was also limited to those conducted from 1985 onwards.

47 Incisional complications

48 Prevalence

49 Incisional complications vary from mild, self-limiting conditions to those that increase

50 duration and costs of hospitalisation, which are life-threatening or may prevent return to

51 athletic function. These include oedema, dehiscence, drainage, infection and hernia formation

52 (Supplementary Item 1). Surgical site infection / drainage has been reported in 11-42% [20-

53 24] of horses following laparotomy which is greater than surgical site infection (SSI) rates

54 reported following abdominal surgery in humans (13.3%) [25], dogs and cats (5.5%) [26] and

55 cattle (12.8%) [27]. Given that horses that developed an incisional hernia were 7-14 times

56 less likely to return to athletic use [18; 28], strategies to minimise the risk of SSI are therefore

57 important.

59 The definition of, and variable use of, terms such as incisional infection / suppuration,

60 drainage and surgical site infection (SSI) differ greatly between studies making it difficult to

61 make accurate comparisons. Frequently incisional drainage has been defined by its physical

62 characteristics (haemorrhagic, serous / serosanguinous or purulent) [29-31]. Some studies

63 have defined infection as purulent discharge only from the incision [21; 32; 33], whereas

64 others have defined this as any type of incisional drainage [23; 34-37] and irrespective of

65 bacterial growth following swabbing of the surgical site. The definition of incisional oedema

66 also varies between studies with some authors considering this to be a normal, physiological

67 consequence of laparotomy rather than a surgical complication [38]. Incisional oedema is

3 68 common in horses following laparotomy [29] with one study reporting a prevalence of 74.1%

69 [31]. Incisional dehiscence has a relatively low prevalence of 1-4% [29; 31; 32] and may be

70 limited to the skin and subcutaneous tissues (superficial) or include the linea alba (deep)

71 resulting in potentially fatal visceral prolapse [29]. Incisional hernia formation has been

72 reported in 6-17% of horses following laparotomy [18; 29; 32; 39] and was most commonly

73 identified 2-12 weeks postoperatively [39; 40]. Failure of studies to extend the follow-up

74 period beyond the time of hospitalisation is likely to underestimate the prevalence of

75 incisional complications as a proportion of SSI (13–100%) can still develop following

76 hospital discharge [35; 37]. The duration and quality of follow-up (owner vs. veterinary

77 reports) is also highly variable between studies.

79 Risk factors and prevention

80 Multiple studies have reported on prevalence of incisional complications and risk factors for

81 each and are summarised in Supplementary Item 1. Different pre-, intra- and postoperative

82 factors have been identified to alter the likelihood of incisional complications, and these vary

83 between studies. Preoperative antimicrobial prophylaxis is standard practice prior to horses

84 undergoing laparotomy [41; 42]. However, some studies have demonstrated poor adherence

85 to best practice in antimicrobial usage in terms of dosing and optimal timing before surgery

86 [37; 42]. In human medicine, guidelines for preoperative antimicrobial prophylaxis comprise

87 drug administration within 60 minutes of the surgical incision being performed and re-dosing

88 when surgery time surpasses two half-lives of the drug [43]. However, evidence of similar

89 best antimicrobial practice in equine colic surgery and any impact of this on the incidence of

90 SSI is lacking and is an area that requires future research. Postoperatively, the type of

91 antimicrobial administered has not been shown to be significantly associated with incisional

92 SSI [44]. The duration of antimicrobial therapy following laparotomy is variable between

4 93 studies and the optimal duration of therapy is also unknown. A recent study reported that the

94 prevalence of SSI did not differ in horses that received antimicrobials for 72 hour versus 120

95 hours postoperatively [24]. These findings are consistent with human studies in which

96 extended duration of antimicrobial prophylaxis failed to decrease the incidence of SSI after

97 gastrectomy compared with single dose prophylaxis [45].

99 Clean contaminated surgeries have been associated with increased risk of SSI suggesting a

100 potential role of intestinal bacteria in the establishment of incisional infection [33; 46].

101 However, many studies have reported no relationship between SSI and whether enterotomy

102 was performed or not [22; 30; 32; 36; 39]. In addition, bacterial cultures taken from

103 peritoneal fluid, enterotomy or resection and anastomosis sites intraoperatively and from the

104 linea alba after its closure failed to predict the occurrence of SSI [34; 47]. One study reported

105 poor intraoperative drape adherence, high surgery room contamination and isolation of

106 bacteria after anaesthetic recovery as significant risk factors for SSI [35]. These findings

107 suggest that environmental contamination of the surgical site during and following

108 anaesthetic recovery may play a major role in development of SSI, and that protection of the

109 abdominal incision during recovery from anaesthesia and in the early postoperative period

110 may be beneficial [34]. A prospective, randomised controlled trial assessing the use of an

111 abdominal bandage to protect the surgical incision placed immediately after recovery from

112 general anaesthesia resulted in 45% absolute risk reduction of incisional complications [31].

113 Protection of the abdominal incision with a stent bandage during anaesthetic recovery

114 followed by placement of an abdominal bandage following stent removal was reported to

115 result in a significant reduction in the likelihood of SSI in a more recent study [23]. Further

116 prospective, randomised controlled trials using well-defined outcomes and variables are

5 117 required to evaluate the role of stent / abdominal bandages and other interventions in altering

118 the prevalence of SSI in different hospital populations.

120 There is also no clear evidence about the optimal method for closing the abdominal incision

121 and various suture materials, suture patterns and use of skin staples have been evaluated

122 (Supplementary Item 1). Three layer closure (linea alba, subcutaneous tissue and skin) of the

123 laparotomy incision was reported in one study to decrease the risk of incisional drainage

124 during hospitalisation [31]. In contrast, Coomer et al. reported no significant difference in

125 incisional suppuration between the two (linea alba and skin) versus three layer closure

126 techniques [30]. Recently, a modified two layer closure technique was reported to decrease

127 the risk of incisional drainage [20]. The timing and duration of follow-up may explain the

128 differences between these studies as demonstrated by different risk factors identified by

129 Smith et al. [31] at different time points postoperatively. Abdominal bandages have been

130 shown to achieve different sub-bandage pressures dependent on the type of bandage used

131 [48] and this may have an impact on incisional healing. It is therefore important that studies

132 appraising different methods of abdominal wall closure take into account the effect of

133 whether abdominal bandaging was performed or not, the type of abdominal bandage used,

134 frequency of bandage changes and duration of use.

136 Prolonged duration of general anaesthesia has been associated with greater risk of incisional

137 complications [29; 31; 37] and a recent study also identified increased risk of incisional SSI

138 with hypoxaemia (PaO2 <80mmHg) [49]. Repeat laparotomy has been consistently identified

139 to increase the risk of SSI with a reported prevalence of 44-84% [9; 21; 32] and to increase

140 the risk of incisional hernia formation 12 fold [39]. SSI / incisional drainage and severe

141 oedema are factors consistently identified to increase the risk of incisional herniation [31; 34;

6 142 39; 44; 50]. There is no current evidence that the duration of time between initial and repeat

143 laparotomy nor location of the incision for repeat laparotomy (original midline versus

144 paramedian incision) has an effect on likelihood of hernia development. The use of a

145 commercial hernia belt was reported to reduce the incidence of incisional hernia development

146 in horses that developed other incisional complications [51]. However, the study was

147 retrospective in nature and did not have a comparison group, making it difficult to fully

148 assess the efficacy of this intervention.

150 Postoperative pain / colic

152 Postoperative pain / colic is a common complication following surgical management of colic,

153 occurring in 11-35% of horses postoperatively [19; 21; 50; 52]. It is also the most common

154 cause of postoperative death or euthanasia [13; 19; 46; 53; 54] and postoperative pain during

155 hospitalisation has been significantly associated with reduced survival [55; 56]. The

156 prevalence of postoperative colic varies depending on the location of the primary

157 gastrointestinal lesion. Rates of up to 61% have been reported following small intestinal

158 resection and anastomosis [10; 57] and 41% in horses that had strangulating large colon

159 lesions [58]. It is difficult to make accurate comparisons between studies as different

160 definitions (e.g. ‘pain’, ‘abdominal pain’, ‘colic/pain’) are used and rarely are objective

161 measures of pain used. It is therefore important that where postoperative pain is investigated

162 as an outcome or as a risk factor for development of other postoperative complications,

163 objective and validated measures specifically designed to measure equine visceral or post

164 abdominal surgical pain are utilised [55; 59; 60].

7 166 In many cases the definitive cause of postoperative pain / colic is unknown as episodes may

167 resolve spontaneously or following medical therapy. In addition if euthanasia is performedon

168 the basis of unrelenting pain / colic and a second surgery is not an option, it is impossible to

169 be certain about the cause of colic recurrence unless post-mortem examination is performed,

170 which may be infrequent outside hospital facilities [61]. Postoperative colic may be due to

171 recurrence of the initial gastrointestinal lesion, intestinal obstruction related to adhesions or

172 anastomotic complications, or the development of gastrointestinal lesions unrelated to the

173 initial lesion [62]. Postoperative pain may also be related to POI and non-gastrointestinal

174 causes including incisional infection and peritonitis [60]. Horses that undergo surgical

175 correction of right dorsal displacement of the large colon have been shown to be more likely

176 to develop postoperative colic (prevalence 41.9%) compared to other forms of large intestinal

177 displacement including left dorsal displacement (8.3%) and non-strangulating volvulus

178 (20.5%) [63]. Left dorsal displacement of the large colon was associated with a recurrence

179 rate of 8.1-20% following surgical or non-surgical correction in two other studies [64; 65].

180 Postoperative colic was also 3 times more likely to occur in horses that underwent correction

181 of strangulating large colon torsion (volvulus) compared to other surgical lesions [44]. The

182 type of surgical procedure performed has also been shown to alter the likelihood of

183 postoperative colic. Mechanical obstruction at the site of an intestinal anastomosis due to

184 reduction in normal luminal diameter and functional obstruction due to altered motility of

185 adjacent segments of gut are hypothesised as possible causes of postoperative colic in horses

186 that have undergone small intestinal resection and anastomosis [12; 66]. Horses in which a

187 side-to-side jejunocaecostomy was performed are reported to be at increased risk of

188 developing colic postoperatively compared to those horses where an end-to-end

189 jejunojejunostomy was performed [67]. In addition, stapled side-to-side jejunocaecostomy

190 has been associated with a greater prevalence of postoperative colic (60%) compared to hand-

8 191 sewn side-to-side jejunocaecostomy (9%) [66]. Jejunocaecostomy has also been associated

192 with significantly reduced survival following hospital discharge compared to end-to-end

193 jejunojejunostomy and jejunoileostomy [12]. In the latter study a greater proportion of horses

194 undergoing jejunoileostomy underwent repeat laparotomy during hospitalisation due to

195 persistent postoperative reflux compared with the other 2 anastomosis groups but this

196 anastomosis had no significant effect on short- or long–term outcome. Whilst the nature of

197 the surgical lesion may limit the choice of small intestinal anastomosis, this knowledge

198 enables postoperative complications to be predicted more accurately.

201 Prevention

202 Prevention and management of postoperative pain is usually based on administration of

203 NSAIDs to provide analgesia and to reduce the inflammatory response. Due to concerns

204 about the detrimental effect of flunixin meglumine (a non-selective COX-1 and COX-2

205 inhibitor) on mucosal recovery of intestine, the use of other NSAIDs that are selective COX-

206 2 inhibitors including firocoxib and meloxicam has been suggested based on the results of in-

207 vitro studies utilising equine jejunum [68; 69]. However, other in-vitro studies have shown

208 that flunixin meglumine had no effect on recovery of equine colonic mucosa following

209 ischaemic injury [70] and that administration of systemic lidocaine reduced the inhibitory

210 effects of flunixin meglumine on recovery of the mucosal barrier [71]. A recent survey of

211 European Equine Internal Medicine and Surgery Diplomates demonstrates that at present,

212 flunixin meglumine remains the most frequently used NSAID following surgical treatment of

213 colic in Europe [72]. Currently there is no evidence of improved outcome with alternative use

214 of selective COX-2 inhibitors. In a randomised controlled trial comparing flunixin

215 meglumine and meloxicam in horses following surgical management of strangulating small

9 216 intestinal lesions [14], flunixin meglumine was significantly more effective than meloxicam

217 in reducing pain scores but there was no difference in outcome between the two groups.

218 Further randomised controlled trials using large numbers of horses in multicentre studies,

219 providing sufficient study power, are warranted to determine the effect of COX-2 inhibitors

220 such as firocoxib on outcome compared to flunixin meglumine. Continuous rate infusion

221 (CRI) of lidocaine or butorphanol are also used by some clinicians. Rigorous appraisal of

222 lidocaine CRI as a postoperative analgesic has not been performed. In a study by Malone et

223 al. [73], there was no significant difference in postoperative pain between horses that did or

224 did not receive lidocaine. However, pain was assessed subjectively, and because the study

225 was relatively small in size, may have lacked power to detect a difference between the two

226 groups. Butorphanol CRI has been shown to result in improved behavioural pain scores in a

227 prospective randomised, controlled, blinded clinical trial [74]. However, the effects of

228 butorphanol on gastrointestinal motility, particularly in horses at high risk of POI have not

229 been fully investigated. Evidence suggests that the nature of the primary gastrointestinal

230 lesion and requirement for certain surgical procedures to be performed (e.g.

231 jejunocaecostomy) may place some horses at an inherent increased risk of postoperative colic

232 both in the short- and long-term, which therefore may be difficult to prevent. Laparoscopic

233 ablation of the nephrosplenic space has been described as a technique to prevent recurrence

234 of left dorsal displacement of large colon [64; 75]. However, whilst this may physically

235 prevent recurrence of nephrosplenic entrapment, there is no evidence that this reduces the

236 likelihood of colic recurrence long-term. Colopexy may be indicated in some horses that have

237 recurrent large colon displacement or torsion [76]. However, complications including

238 postoperative colic or colon rupture can occur making appropriate case selection important

239 [77; 78]. Closure of the epiploic foramen to prevent intestinal entrapment within the foramen

10 240 is under current investigation [79; 80]. The results of clinical trials will be needed before

241 these can be appraised as methods to prevent EFE or its recurrence.

243 Postoperative ileus (POI) and postoperative reflux (POR)

245 Prevalence and risk factors

246 POI has been demonstrated by multiple studies to have a negative effect on survival [10; 21;

247 36; 81], to increase the likelihood of repeat laparotomy [44], development of laminitis and

248 incisional drainage and to significantly increase treatment costs [82]. The reported prevalence

249 of POI varies from 6.3-53% and was a reason for euthanasia or death of between 9-43% of

250 horses that had undergone surgical management of colic in these populations [19; 36; 46; 81;

251 83-86]. The criteria used to define POI differ making it difficult to make accurate

252 comparisons between some studies (Supplementary Item 2). Whilst there is some variation in

253 findings between studies, elevated heart rate and packed cell volume (PCV) on admission,

254 small intestinal lesions, evidence of intestinal ischemia and intestinal resection and

255 anastomosis are factors that have consistently been identified to increase the likelihood of

256 POI developing postoperatively.

258 More recent studies have used an alternative term “postoperative reflux” (POR), defined as

259 >2L reflux after nasogastric tube passage at any time during the postoperative period [12;

260 87]. The latter studies argued that without undertaking repeat laparotomy, in the majority of

261 cases of horses with postoperative reflux (POR) it is impossible to definitively identify the

262 underlying pathophysiology resulting in nasogastric reflux. This highlights the need for

263 common terminology, definitions and outcomes to be utilised by researchers. It would appear

264 prudent to utilise the more generic term “POR” in future studies but to also measure duration

11 265 and quantity of reflux and to assign the diagnosis of POI as an additional variable / outcome

266 measure where other causes of POR can be excluded.

268 Prevention

269 Early referral of horses for surgical management of colic, prior to development of

270 cardiovascular derangements and progression of intestinal distention and compromise is

271 therefore important in reducing the likelihood of POI developing. Evacuation of colonic

272 contents via a pelvic flexure enterotomy (PFE) reduced the risk of POI in one study [86].

273 However, a further study demonstrated that whilst PFE was protective for large intestinal

274 lesions it increased the risk of POI in horses with small intestinal lesions [82]. Therefore, in

275 horses considered to be at high risk of POI that have small intestinal lesions, evacuation of

276 colonic contents might best be reserved only for horses with large quantities of ingesta within

277 the large colon.

279 A variety of medical therapies that are currently used to prevent and manage POI include

280 flunixin meglumine, lidocaine, butorphanol, metoclopramide, erythromycin, morphine and

281 neostigmine [72]. However, there have been relatively few clinical trials undertaken to

282 determine the efficacy of these in preventing and managing POI. In a prospective, blinded,

283 randomised clinical trial [88] significant differences in ultrasonographic measurements of

284 jejunum, an intestinal function index and ultrasonographic identification of peritoneal fluid

285 were demonstrated between lidocaine CRI (initiated intraoperatively) and placebo (saline)

286 treated groups. There were no significant differences between time to first defecation or the

287 proportion of horses that survived to hospital discharge. This study included 18 horses with

288 large colon lesions and 8 horses with small intestinal lesions (4 treatment and 4 placebo). A

289 separate prospective, double–blinded, placebo-controlled trial conducted by Malone et al.

12 290 [73] demonstrated that treatment with lidocaine for 24 hours significantly reduced the

291 proportion of horses refluxing at 30 hours, the rate of reflux after initiation of treatment and

292 days of hospitalisation compared to the placebo (saline) group. There was no difference in

293 proportion of horses that survived to discharge home between the two groups. It is very

294 important to note that the latter study has some key limitations including the fact that it was

295 performed on a relatively small number of horses, with consequent low study power, and

296 included horses with both proximal duodenitis-jejunitis treated medically and POI (8

297 proximal duodenitis-jejunitis and 24 POI horses), which are lesions with different

298 pathobiology and relevance when considering postoperative complications. In a retrospective

299 study of 126 horses that recovered following small intestinal surgery, lidocaine therapy

300 (initiated intraoperatively and continued postoperatively) resulted in a 3 fold increase in

301 likelihood of survival to hospital discharge in horses with POI and a 3 fold reduction in

302 likelihood of POI developing [36]. In the latter study, some horses received concurrent

303 metoclopramide but this had no significant effect on likelihood of POI developing.

304 Continuous IV infusion of metoclopramide in POI was reported in another retrospective

305 study to result in a significant decrease in the rate and volume of gastric reflux obtained and a

306 shortened duration of hospitalisation compared to horses with POI where metoclopramide

307 was administered as an intermittent infusion or was not administered [89]. They found no

308 significant differences in the survival rate to hospital discharge among the treatment groups.

310 Therefore, there is a need for prospective, large (i.e. sufficient study power), and potentially

311 multicentre randomised controlled clinical trials to be conducted on therapies and strategies

312 to manage POI. It is important that there is a consensus on the criteria on which POI is

313 defined and which is consistently used between studies, including rigorous differentiation of

314 horses with true primary POI and ileus secondary to mechanical causes [90; 91] and that

13 315 study populations are well defined e.g. only horses undergoing small intestinal surgery. This

316 is essential if multicentre studies are to be conducted effectively and in enabling accurate

317 comparisons to be made between studies. Treatments for POI can be expensive, almost

318 doubling the costs of treatment in one study [82]. The benefits of medical therapies for

319 prevention and treatment of POI, including therapies being considered for future development

320 [92] on outcome versus economic costs also need to be considered.

323 Intra-abdominal adhesions

325 Prevalence and risk factors

326 Intra-abdominal adhesion formation that was considered to be pathological in nature has been

327 reported in 9-27% of horses undergoing repeat laparotomy or post-mortem examination

328 following gastrointestinal surgery [61; 93; 94]. Given that adhesions may not result in clinical

329 signs of pain or gastrointestinal obstruction and definitive diagnosis is only possible at repeat

330 surgical exploration of the abdomen or necropsy, the prevalence of postoperative adhesions is

331 likely to be underestimated following surgical management of colic [61]. Confirmed

332 formation of adhesions was reported in 6-22% of horses following small intestinal surgery [9;

333 22; 95] , in 8-17% in foals and young horses (< 2 years old) following laparotomy for any

334 reason [15; 96-98] and in 32% of horses undergoing repeat laparotomy [61]. Comparison of

335 prevalence reported between studies is difficult due to differences in study design. For

336 example in the study by Gorvy et al. [61] only horses that underwent repeat laparotomy were

337 included whilst in the other studies, the primary surgical colic patient cohorts were used.

14 339 Factors that have been identified to increase the risk of adhesion formation include small

340 intestinal surgery [22; 94; 99] and performing intestinal resection and anastomosis [22; 50].

341 One study of 99 horses that underwent repeat laparotomy demonstrated that adhesions

342 develop irrespective of the site of the primary lesion [61]. This and other studies concluded

343 that resection and type of anastomosis were not associated with increased likelihood of intra-

344 abdominal adhesion formation [61; 95; 99].

346 Preventive measures

348 Clinicopathological parameters measured on hospital admission have been shown to have no

349 effect on likelihood of development of intra-abdominal adhesion formation after surgery for

350 management of colic. This would suggest that adhesion formation is more likely to be related

351 to surgical trauma rather than SIRS / hypercoagulable systemic state [61; 95]. Assessment of

352 coagulation and coagulopathies is not commonly performed peri-operatively in horses but is

353 an area for further research [100] to determine whether this may assist more accurate

354 prediction of complications such as adhesion formation and likelihood of postoperative

355 mortality and morbidity. Therapies that have been proposed to prevent adhesion formation

356 based on laboratory animal models, some models utilising healthy horses and work

357 performed in humans [101-105] include peritoneal lavage, systemic administration of low

358 dose heparin and chemical agents to physically separate serosal and peritoneal surfaces such

359 as 1% high molecular sodium carboxymethylcellulose (SCMC / CBMC), hyaluronate /

360 SCMC spray and bioresorbable barriers such as hyaluronate / SCMC membrane.

362 Local therapies that have been investigated in colic cases include use of a Hyaluronate /

363 SCMC membrane covering a continuous Lembert suture pattern. This showed no superiority

15 364 over an uncoated interrupted Lembert pattern for performing a one-layer jejunojejunostomy

365 in terms of postoperative survival rate and prevalence of post-operative complications [106].

366 This might not be surprising if the findings of Gorvy et al [61] are to be accepted i.e.

367 adhesions are not specifically related to the site of the primary lesion. Omentectomy has been

368 proposed as a method to prevent adhesion formation but had no significant effect on colic

369 related adhesions found at repeat laparotomy [99].

371 Abdominal-wide therapies have also been investigated in equine clinical studies. SCMC

372 solution was shown in one retrospective study to have no effect on rates of short- and long–

373 term survival and the incidence of postoperative complications following exploratory

374 laparotomy [107]. In contrast, another retrospective study by Fogle et al. [56] demonstrated a

375 significant positive effect of SCMC solution on postoperative survival. Due to the non-

376 randomised and retrospective nature of the latter two studies, both had some limitations due

377 to bias. A prospective, randomised controlled trial to assess the use of SCMC is warranted.

378 One study found a protective effect of use of intra-peritoneal heparin therapy in reducing

379 intra-abdominal adhesion formation [50]. However, as acknowledged by the study authors,

380 the findings from this study should be viewed with caution as only univariable analysis was

381 performed, preventing assessment of the effect of potential confounders.

383 Therefore, much of the evidence on how adhesions may be prevented is inconclusive, due to

384 the relatively small number of studies and their largely retrospective nature, with resultant

385 inherent biases. This demonstrates the need for prospective, randomised controlled trials to

386 fully appraise the evidence for use of surgical strategies or therapies to prevent adhesion

387 formation. Given that therapies used to prevent adhesions in humans are designed to be used

388 in a much smaller abdominal cavity, cost-benefit analysis of use of these products and effect

16 389 on outcome would also be important. One of the greatest challenges is in correctly

390 determining whether adhesions are present or not and proxy endpoints e.g. postoperative

391 colic where surgical or post-mortem examination cannot be performed may be the only

392 current practical outcome measure. Biomarkers for adhesion formation would be a potentially

393 important area for future investigation.

396 Postoperative diarrhoea

398 Postoperative diarrhoea increases costs of treatment, reduces survival and postoperative

399 performance, and increases the risk of nosocomial infection developing in other horses within

400 the hospital environment [18; 108]. Its prevalence varies widely depending on the location

401 and pathology of the primary gastrointestinal lesion and on the geographical location of

402 horses. Diarrhoea is usually defined as passage of unformed faeces for more than 24 hours

403 [21; 109] or on two or more consecutive occasions [110] following surgical treatment of

404 colic. However, in many studies case definition is poorly described.

406 Horses diagnosed with large intestinal lesions were significantly more likely to develop

407 diarrhoea postoperatively compared with other intestinal lesions [111]. Diarrhoea has been

408 reported to be a frequent complication following surgical correction of strangulating large

409 colon volvulus (28-45%) [58; 109], treatment of sand impaction of the large colon (46%)

410 [112] and following removal of large colon enteroliths (20%) [110]. Horses treated surgically

411 for small colon diseases were shown in one study to be 17 times more likely to develop

412 diarrhoea postoperatively compared to other surgical lesions [113], with the prevalence of

413 diarrhoea ranging from 11-70 % [113-115]. This complication was not as frequent following

17 414 small intestinal resection and anastomosis with reported prevalences of 2.7-25% [9; 10]. The

415 prevalence of postoperative diarrhoea would also appear to be higher in some studies based in

416 the USA (varying from 1.3-53.2%) [18; 82; 84; 111; 116] compared to the UK (3.2-3.8%)

417 [19; 21]. This is interesting and could potentially reflect differences in gastrointestinal

418 microbial ecosystems and environmental conditions (e.g. temperature, humidity) that vary

419 geographically. This observation merits further investigations.

421 Despite the high prevalence of diarrhoea following surgical treatment of colic, it is usually

422 self-limiting [109; 114]. Infectious agents such as Salmonella spp. and Clostridia spp. are

423 seldom isolated from affected horses [114; 116]. Studies that have conducted active

424 surveillance for faecal shedding of Salmonella spp. in horses admitted with clinical signs of

425 gastrointestinal disorders demonstrated that Salmonella spp. were isolated from the faeces of

426 a large proportion of horses that did not develop clinical salmonellosis postoperatively (43 –

427 74%) [108; 116; 117]. Molecular studies of the horse’s hind gut microbiota provides evidence

428 to suggest that colitis arises as a result of imbalance among the different microbial species in

429 the gut (dysbiosis) rather than a disease caused by a single bacterial species [118].

430 Development of diarrhoea in colic patients postoperatively is likely to be due to a

431 combination of factors including the nature of the primary gastrointestinal lesion and

432 concurrent changes in gut motility, surgical manipulation of the gastrointestinal tract and

433 administration of antimicrobials that may disrupt the normal microbial ecosystem [108; 119;

434 120]. Characterisation of hindgut microbial communities using non-culture dependent

435 techniques in normal horses and those that have undergone surgical management of colic

436 would assist in determining whether there is evidence for the use of interventions such as

437 administration of probiotics or certain feeds to manipulate these ecosystems, enabling these

438 to return quickly to a potentially ‘normal, pre-surgical state’.

18 439

440 Prevention

441 Clinical trials evaluating the efficacy of probiotics administered following colic surgery to

442 minimise shedding of Salmonella spp. and development of diarrhoea have not shown any

443 significant effect on these outcomes [116; 121]. A prospective, randomised controlled trial

444 evaluated the efficacy of postoperative administration of di-tri-octahedral smectite (versus

445 water placebo) for three consecutive days following surgery in 67 horses. These were horses

446 that had undergone surgical treatment of a large colon disorder, including pelvic flexure

447 enterotomy, and a faecal scoring system was used to objectively assess outcome. There was a

448 significant reduction in the prevalence of postoperative diarrhoea in the treatment group

449 (10.8%) versus the placebo group (41.4%) [122], which would support the use of this agent to

450 reduce the likelihood of postoperative diarrhoea in horses with surgical diseases of the large

451 colon.

453 SIRS, endotoxaemia and related complications

455 The systemic inflammatory response syndrome (SIRS) and presence of bacterial endotoxins

456 in the circulation (endotoxaemia) are common in horses that present with colic. SIRS was

457 evident in 28% of horses in a study by Epstein et al. [100] of which 37% had lesions

458 requiring surgical intervention. Plasma endotoxin (LPS) was also detected in 29% of horses

459 presenting with colic in a study by Senior et al. [123]. Markers of endotoxaemia / SIRS

460 including tachycardia, elevated PCV and altered mucous membrane colour have consistently

461 been shown to be one of the most reliable determinants of postoperative morbidity (and

462 mortality) in many studies assessed in this review. Therefore, strategies to prevent or

463 attenuate the effects of endotoxaemia / SIRS could theoretically reduce the prevalence of

19 464 postoperative complications. This is an area of current, active work in research models and is

465 therefore beyond the scope of this review. An update on current findings has recently been

466 summarised by Moore & Vandenplas [124]. In the studies appraised, standard postoperative

467 care following surgical management of colic includes use of NSAIDs, intravenous fluids and

468 in some studies, use of Polymixin B, in horses with evidence of or that are at high risk of

469 endotoxaemia / SIRS. Of note, there is no current evidence to support the use of polymixin B

470 once signs of SIRS are evident.

472 Endotoxaemia has been identified to increase the likelihood of thrombophlebitis [44; 125;

473 126] and laminitis [127]. The prevalence of thrombophlebitis in postoperative colic patients

474 varies from 1.3-18% [8; 10; 14; 19; 21; 37; 42; 84; 109; 125]. Risk factors for catheter-

475 associated thrombophlebitis identified in one study were endotoxaemia, salmonellosis, low

476 systemic total protein and location in which they were hospitalised [126]. Other studies

477 identified tachycardia and elevated PCV at admission [44], prolonged duration of

478 catheterisation, postoperative diarrhoea [125], colic and endotoxemic shock (SIRS) [21; 44;

479 126] postoperatively to increase the risk of this complication. These findings are consistent

480 with horses that are likely to be in a hypercoagulable state, explaining the greater propensity

481 for horses that have undergone surgical management of colic to develop catheter associated

482 complications compared to other groups of horses [128]. Heparin therapy has been suggested

483 as a way in which the risk of thrombophlebitis might be reduced postoperatively. However, at

484 present there is a lack of evidence to support its use [129; 130]. Frequent ultrasonographic

485 examination of the catheter site in high-risk horses to monitor for increased thickness of the

486 vessel wall and visible thrombus formation would be indications to remove the catheter prior

487 to development of overt clinical signs of thrombophlebitis [131].

20 489 The frequency of postoperative laminitis following surgical management of colic varies

490 between studies, with prevalences of 0.4-12% reported [10; 19; 21; 22]. Distal limb

491 cryotherapy may be used to attenuate laminar damage [132]. There are no reports of its

492 effectiveness specifically in horses following surgical management of colic but there is

493 evidence to suggest that its use should be considered in horses with clinical evidence of

494 SIRS / endotoxaemia who are at increased risk of developing this complication.

497 Other postoperative complications and future interventions

499 The prevalence of other complications that may occur following surgical management of

500 colic is generally low. These include septic peritonitis [10; 14; 21; 57; 84], haemoperitoneum

501 [12], intraluminal haemorrhage [133], myopathies, gastric rupture [12], cardiac arrhythmias

502 and derangements in electrolytes [134], clotting disorders and disseminated intravascular

503 coagulation, pre- and postoperative renal azotaemia and renal failure [10; 84]. Treatment and

504 preventive strategies for these are detailed elsewhere [135-138].

506 Nutrition of the postoperative colic patient is an area that has been investigated infrequently

507 [139-141] and evidence about nutritional strategies to optimise patient survival and minimise

508 postoperative complications are lacking [142]. Intestinal microbiota and changes that may

509 precede colic development or develop subsequent to surgical management of colic are areas

510 of current research. Such studies may further our understanding of equine gut health and

511 interventions to minimise postoperative complications such as colic.

21 513 A study that measures ‘disease burden’ of surgical management of colic that considers the

514 prevalence, economic costs and welfare issues arising from postoperative complications may

515 be useful in order to prioritise future research in this area. It is important that future studies

516 have suitable outcome measures to assess the effectiveness of new therapies or management

517 strategies. Use of survival rates as sole measures of success should be avoided as euthanasia

518 is a complex area and issues such as selection bias may occur [143].

521 Conclusions

Early surgical intervention, prior to the development of cardiovascular derangements and marked intestinal distention, remains key in preventing or reducing many postoperative complications that may develop following surgical treatment of colic. There is a lack of current evidence of the efficacy of some interventions proposed to reduce the incidence of, or prevent, specific postoperative complications. In addition, there is a need to better understand the pathobiology of postoperative complications in order to develop better preventive strategies. The results between studies are often highly variable and sometimes contradictory due to inconsistent definitions and outcome measures used. Therefore, there is a need for clinical veterinary researchers to make concerted efforts to adopt common terminology, definitions, and outcome measurements and a continued need for well-designed and co- ordinated, prospective, large, suitably powered (multicentre, international) studies with common goals of research. New therapies and ways in which postoperative colic patients are managed may improve mortality and morbidity rates in the future. However, these may add to the expense of treatment, which has important implications in keeping surgical treatment of colic affordable for horse owners [144]. Appropriate and aggressive primary treatment of intestinal obstructions can be expensive but may provide a better outcome without the expense of treating postoperative complications (both short- and long-term). It is essential that studies investigating new therapies and management strategies not only undertake rigorous prospective testing of their efficacy in a suitable number of relevant clinical cases and using appropriate outcome measures, but also that they report and consider the economic costs versus benefits of these interventions before these are promoted for widespread clinical use.

22 523

524 Supplementary Item 1: Summary of studies that have specifically investigated prevalence rates and risk factors for postoperative incisional 525 complications

Risk factors significantly associated with outcome Recruitme Prevalence of surgical site complications Outcome (case Study identified on multivariable analysis (adjusted odds ratio References Study design, nt reported (%) definition) population, [OR] or hazard ratio [HR]; 95% confidence interval) data analysis /publicatio follow-up period Drainage / n year infection / Hernia Oedema Dehiscence suppuration Retrospective Incisional hernias Incisional drainage (17.8, 5.2 – 60.8), Chromic gut for 1979 – cohort, 210 horses, 4 (diagnosis confirmed linea alba closure (OR 55.0, 2.2 -110.5); postoperative 1987 / - 16% - - [39] multivariable months - 8 years by veterinary leukopenia (OR 4.6,1.3 – 16.0), Relaparotomy (OR 12, 1989 analysis surgeons) 2 – 73.5) Retrospective 1983 – 78 horses, 6 cohort, Incisional 1985 / months follow up 26% 5.7% - 3% Multivariable analysis not performed [32] univariable complications rates 1989 on 66 horses analysis 275 horses of Incisional Retrospective 1980 – which 161 had complications, Incisional complications: Weight, duration of surgery cohort, 36% 1987 / surgery for GIT 17% - 4% incisional drainage & Incisional drainage: Age, weight [29] * multivariable 1995 lesions, 1-82 incisional hernia Incisional hernia: Weight analysis months Prospective Preoperative peritoneal fibrinogen concentration (OR 1990 – cohort, 210 horses, not 25% Incisional infection 1.93,1.17 - 3.18), enterotomy (OR 1.93, 1.11 - 3.37), 1992 / 8% - - [33] multivariable stated (purulent discharge) polyglactin 910 for linea alba closure (OR 2.0, 1.13 – 1997 analysis 3.61) Prospective randomized Incisional infection Poor intraoperative drape adherence, high surgery room No stated / 53 horses, not clinical trial, 26% - - - (any evidence of contamination, isolation of bacteria after anaesthetic [35] * 1999 stated multivariable incisional drainage) recovery analysis Prospective 1998 – Incisional suppuration: None of the variables retained cohort, 341 horses, 3-33 Incisional suppuration 2000 / 16% 8.4% - - Incisional hernia: Incisional suppuration (OR 4.3, 1.45 – [19; 44] multivariable months Incisional herniation 2002 12.9), Heart rate at admission (HR 1.04,1.01 -1.06) analysis, 300 horses, Retrospective owners were first 1994 – cohort, contacted 12 Incisional 2001 / 29% 8% Multivariable analysis not performed [21; 50] univariable months post complications rates 2005 analysis discharge for hernia formation

23 Incisional drainage on hospital discharge: GA >110min (OR 13.76, 3.45 – 54.21), SC suture layer (OR 0.19, 0.05-0.76) Incisional drainage 14 days post discharge: Duration of colic signs prior to presentation (8-24h; OR 10.5,1.75 – 62.85, >24h; 13.29, 1.39 – 126.67), heart rate at presentation > 60 bpm (OR 28.38; 3.09 – 260.19), Primary lesion (large bowel obstruction; OR 0.03, 0.00 Prospective Incisional drainage – 0.58, small intestinal strangulation; 0.00 – 0.43), randomized 2003 - 85 horses, 3 (drainage started 12 Abdominal bandage (OR 0.08, 0.02 – 0.37), Pyrexia controlled trial, 2005 / 25% 14% 74% 1% [31] months hours post-surgery) & during hospitalization (OR 15.85, 1.49 – 168.41) multivariable 2007 incisional hernia Incisional drainage 30 days post discharge: Severe pain analysis on admission (OR 13.52, 2.45 – 74.5), Duration of pain 8-24h (OR 8.92, 1.48 – 53.95), abdominal bandage (OR 0.13, 0.03 – 0.63), Drainage at 14 days (OR 12.20, 2.80 – 53.11) Incisional drainage 3 months post discharge: Dehiscence at 14 days (OR 20.67, 1.63-262.70) Incisional hernia (3 months): Drainage at 30 days (OR 13.12; 1.56 – 110.48) Prospective randomised 2004 - Incisional suppuration No significant differences found between treatment (3 controlled 309 horses, 12 2006 / 21% - - - (Gross drainage of layer incisional closure) and intervention (2 layer [30] multicentre trial, months 2007 pus) incisional closure) groups Multivariable analysis Retrospective Incisional infection 2004 – Incisional closure by year 1 or 2 resident (OR 2.2, 1.16 - cohort, 356 horses, to (serous discharge for 2007 / 15% - - - 4.18), linea alba lavage (OR 0.38, 0.20 – 0.74), skin [38] multivariable hospital discharge >24 hours or purulent 2010 closure with staples (OR 3.85, 2.04 – 7.29) analysis discharge) Prospective randomized No significant differences found between control and Not stated / Incisional controlled study, 100 horses, 9 days 26% - - - study group (antibacterial-coated suture material for [145] 2010 complications rates univariable subcutaneous tissue closure) analysis 159 horses that had a laparotomy If a horse developed via a paramedian any of the incisional Retrospective incision and complications Quarter breed (OR 3.9, 1.3-11.7), Abdominal bandage cohort, 2003- survived for ≥30 15.5% 0.8 - 3.7% (drainage / infection, (OR 9.5, 2.9-30.8), Fever (>38.3 ºC for ≥ 4 consecutive [146] multivariable 2007/2011 days post-surgery, dehiscence, hernia or postoperative days) (OR 12.9, 2.8-58.2) analysis 121 horses were marked cutaneous followed for ≥90 scarring) days

24 Incisional infection Retrospective (discharge started 48 2006 - cohort, 113 horses, >2 hours postoperatively 2007 / 20% - - - Multivariable analysis not performed [37] univariable years (79%) and continued for >36 2012 analysis hours and needed treatment) Incisional infection Retrospective 2005 - (persistent serous cohort, 130 horses, 10 2011 / 11% - - - discharge with Stent bandage (OR 0.1, 0.02 – 0.50) [23] multivariable days post-surgery 2013 extensive edema or analysis purulent discharge) Retrospective Incisional drainage 2008 - cohort, 199 horses, not (persistent (>2 days) Skin suture pattern (referent modified subcuticular 2010 / 22% - - - [20] multivariable detailed serous discharge or closure of skin vs. Other pattern OR 4.9, 1.4 – 16.6) 2013 analysis purulent discharge) Prospective, intervention 2008 - Incisional drainage 92 horses, not No significant differences found between the two study multicentre study, 2009 / 42% - - - (drainage started 12 [24] detailed groups (72 vs 120 hours of postoperative antibiotics) univariable 2013 hours post-surgery) analysis Incisional infection (discharge started 72 hours post-surgery or Prospective purulent discharge Suture (Polyglocolic acid vs polydioxanone in 84 horses, 2 cohort, Not stated / combined with fever, subcutaneous layer ; OR 3.88; 1.2-12.3), anaesthesia weeks post 34.5% [49] Multivariable 2014 pain, or time (>2h OR 4.6,1.2 - 17.6), PaO2 <80mmHg (4.7, 1.3 hospital discharge analysis ultrasonographic - 13.9) evidence of subcutaneous fluid pockets) 526 GA general anaesthesia; SC subcutaneous; bpm beats per minute; * OR / CI was not reported

25 527 Supplementary Item 2: Summary of studies reporting prevalence of and risk factors for postoperative ileus

Study design and Study Years of Case definition P Risk factors identified in a multivariable model significantly Reference statistical analysis Populatio recruitment / re (p<0.05) associated with postoperative ileus (odds ratio [OR]; n publication year v 95% confidence interval) al e n ce Retrospective cohort, 148 July 1990 – July Nasogastric reflux of > 2 litres (L) plus HR ≥60 beats 2 PCV > 48 at admission (OR 6.9; 1.3 - 35.8), small intestinal [84] multivariable analysis surgical 1992 / 1994 per minute and abdominal pain 1 obstruction (OR 10.5; CI 2.6 - 41.7), small intestinal ischemia (OR colic % 16.2; 3.8 - 69.6) patients Retrospective cohort, 376 January 1990 - Nasogastric reflux of > 20 L during 24 hour period 1 PCV ≥ 45 at admission (OR 4.3; 2.2 – 8.3), duration of surgery > 2 [86] multivariable analysis surgical December 1996 / after surgery or reflux volume of > 8 L at any single 8 hours (OR 1.9; >1.0 - 3.7), strangulating small intestinal obstruction colic 2001 sampling time after surgery % (OR 3.2; 1.2 – 7.9), strangulating large intestinal lesion (OR 2.9; 1.1 patients – 8), non-strangulating small intestinal lesion (OR 5; 2 – 12.6), pelvic flexure enterotomy (OR 0.5; 0.2 - 0.9) Prospective cohort, 341 March 1998 – Nasogastric reflux of > 2 L on at least 2 consecutive 9 Pedunculated lipoma obstruction (OR 3.19; 1.35 - 7.53), PCV at [19; 44] multivariable analysis surgical August 2000 / occasions within 24 hours, at any time postoperatively % admission (OR 1.07; 1.01 – 1.13) colic 2002 patients Prospective cohort, 251 Not stated / 2004 As in [86] 1 PCV > 40% at admission (OR 2.9; 1.4 -6.1), duration of anaesthesia [82] multivariable analysis surgical 9 > 3 hours (OR 3.0; 1.4 - 6.7), Small intestinal lesion (OR 5.2; 2.3 colic % -11.7), [Model A – parsimonious multivariable logistic regression patients model; see paper for details of Model B- inclusive model] Retrospective cohort, 300 1994 – 2001 / Gastric reflux of > 2 L and no evidence of mechanical 1 Multivariable modelling not performed [21] univariable analysis surgical 2005 obstruction 8 colic % patients Retrospective cohort, 126 horses March 2004 – As in [86] 3 Heart rate at admission (OR1.05; 1.02 - 1.08), reflux > 8 L at [36] multivariable analysis diagnosed December 2006 / 3 admission (OR 4.16; 1.13 - 15.4), small intestinal resection (OR 6.4, with small 2009 % 2.23 - 18.4), prophylactic lidocaine (OR 0.31; 0.12 - 0.78) intestinal lesions at surgery Retrospective cohort, 233 horses 1995 – 2005 / As in [86] 1 PCV at admission (O.R. 1.05; 1.0 -1.10 ), age (O.R. 1.10; 1.04 - [147] multivariable analysis, diagnosed 2009 7 1.16) with small % intestinal lesions at surgery 528

26 529 References

531 [1] Ireland, J.L., Clegg, P.D., McGowan, C.M., Platt, L. and Pinchbeck, G.L. (2011) Factors 532 associated with mortality of geriatric horses in the United Kingdom. Prev Vet Med 101, 204- 533 218.

534 535 [2] Kaneene, J.B., Ross, W.A. and Miller, R. (1997) The Michigan equine monitoring system. II. 536 Frequencies and impact of selected health problems. Prev Vet Med 29, 277-292.

537 538 [3] Tinker, M.K., White, N.A., Lessard, P., Thatcher, C.D., Pelzer, K.D., Davis, B. and Carmel, D.K. 539 (1997) Prospective study of equine colic incidence and mortality. Equine Vet J 29, 448-453.

540 541 [4] Hillyer, M.H., Taylor, F.G. and French, N.P. (2001) A cross-sectional study of colic in horses on 542 thoroughbred training premises in the British Isles in 1997. Equine Vet J 33, 380-385.

543 544 [5] Proudman, C.J. (1992) A two year, prospective survey of equine colic in general practice. 545 Equine Vet J 24, 90-93.

546 547 [6] Moore, J.N. (2005) Five decades of colic: a view from thirty-five years on. Equine Vet J 37, 548 285-286.

549 550 [7] Mair, T.S. and White, N.A. (2005) Improving quality of care in colic surgery: time for 551 international audit? Equine Vet J 37, 287-288.

552 553 [8] Garcia-Seco, E., Wilson, D.A., Kramer, J., Keegan, K.G., Branson, K.R., Johnson, P.J. and Tyler, 554 J.W. (2005) Prevalence and risk factors associated with outcome of surgical removal of 555 pedunculated lipomas in horses: 102 cases (1987-2002). J Am Vet Med Assoc 226, 1529- 556 1537.

557 558 [9] Freeman, D.E., Hammock, P., Baker, G.J., Goetz, T., Foreman, J.H., Schaeffer, D.J., Richter, 559 R.A., Inoue, O. and Magid, J.H. (2000) Short- and long-term survival and prevalence of 560 postoperative ileus after small intestinal surgery in the horse. Equine Vet J Suppl, 42-51.

561 562 [10] Morton, A.J. and Blikslager, A.T. (2002) Surgical and postoperative factors influencing short- 563 term survival of horses following small intestinal resection: 92 cases (1994-2001). Equine Vet 564 J 34, 450-454.

565 566 [11] Wormstrand, B.H., Ihler, C.F., Diesen, R. and Krontveit, R.I. (2014) Surgical treatment of 567 equine colic - a retrospective study of 297 surgeries in Norway 2005-2011. Acta Vet Scand 568 56, 38.

27 570 [12] Stewart, S., Southwood, L.L. and Aceto, H.W. (2014) Comparison of short- and long-term 571 complications and survival following jejunojejunostomy, jejunoileostomy and 572 jejunocaecostomy in 112 horses: 2005-2010. Equine Vet J 46, 333-338.

573 574 [13] Mair, T.S. and Smith, L.J. (2005) Survival and complication rates in 300 horses undergoing 575 surgical treatment of colic. Part 1: Short-term survival following a single laparotomy. Equine 576 Vet J 37, 296-302.

577 578 [14] Naylor, R.J., Taylor, A.H., Knowles, E.J., Wilford, S., Linnenkohl, W., Mair, T.S. and Johns, I.C. 579 (2014) Comparison of flunixin meglumine and meloxicam for post operative management of 580 horses with strangulating small intestinal lesions. Equine Vet J 46, 427-434.

581 582 [15] Santschi, E.M., Slone, D.E., Embertson, R.M., Clayton, M.K. and Markel, M.D. (2000) Colic 583 surgery in 206 juvenile thoroughbreds: survival and racing results. Equine Vet J Suppl, 32-36.

584 585 [16] Tomlinson, J.E., Boston, R.C. and Brauer, T. (2013) Evaluation of racing performance after 586 colic surgery in Thoroughbreds: 85 cases (1996-2010). J Am Vet Med Assoc 243, 532-537.

587 588 [17] Hart, S.K., Southwood, L.L. and Aceto, H.W. (2014) Impact of colic surgery on return to 589 function in racing Thoroughbreds: 59 cases (1996-2009). J Am Vet Med Assoc 244, 205-211.

590 591 [18] Davis, W., Fogle, C.A., Gerard, M.P., Levine, J.F. and Blikslager, A.T. (2013) Return to use and 592 performance following exploratory celiotomy for colic in horses: 195 cases (2003-2010). 593 Equine Vet J 45, 224-228.

594 595 [19] Proudman, C.J., Smith, J.E., Edwards, G.B. and French, N.P. (2002) Long-term survival of 596 equine surgical colic cases. Part 1: patterns of mortality and morbidity. Equine Vet J 34, 432- 597 437.

598 599 [20] Colbath, A.C., Patipa, L., Berghaus, R.D. and Parks, A.H. (2013) The influence of suture 600 pattern on the incidence of incisional drainage following exploratory laparotomy. Equine Vet 601 J.

602 603 [21] Mair, T.S. and Smith, L.J. (2005) Survival and complication rates in 300 horses undergoing 604 surgical treatment of colic. Part 2: Short-term complications. Equine Vet J 37, 303-309.

605 606 [22] Phillips, T.J. and Walmsley, J.P. (1993) Retrospective analysis of the results of 151 607 exploratory laparotomies in horses with gastrointestinal disease. Equine Vet J 25, 427-431.

608 609 [23] Tnibar, A., Lin, K.G., Nielsen, K.T., Christophersen, M.T., Lindegaard, C., Martinussen, T. and 610 Ekstrom, C.T. (2013) Effect of a stent bandage on the likelihood of incisional infection 611 following exploratory coeliotomy for colic in horses: A comparative retrospective study. 612 Equine Vet J 45, 564-569.

28 613 614 [24] Durward-Akhurst, S.A., Mair, T.S., Boston, R. and Dunkel, B. (2013) Comparison of two 615 antimicrobial regimens on the prevalence of incisional infections after colic surgery. Vet Rec 616 172, 287.

617 618 [25] Pessaux, P., Msika, S., Atalla, D., Hay, J.M. and Flamant, Y. (2003) Risk factors for 619 postoperative infectious complications in noncolorectal abdominal surgery: a multivariate 620 analysis based on a prospective multicenter study of 4718 patients. Archives of surgery 621 (Chicago, Ill. : 1960) 138, 314-324.

622 623 [26] Mayhew, P.D., Freeman, L., Kwan, T. and Brown, D.C. (2012) Comparison of surgical site 624 infection rates in clean and clean-contaminated wounds in dogs and cats after minimally 625 invasive versus open surgery: 179 cases (2007-2008). J Am Vet Med Assoc 240, 193-198.

626 627 [27] Buczinski, S., Bourel, C. and Belanger, A.M. (2012) Ultrasonographic assessment of standing 628 laparotomy wound healing in dairy cows. Res Vet Sci 93, 478-483.

629 630 [28] Christophersen, M.T., Tnibar, A., Pihl, T.H., Andersen, P.H. and Ekstrom, C.T. (2011) Sporting 631 activity following colic surgery in horses: a retrospective study. Equine Vet J 43 (Suppl. 40). 632 3-6.

633 634 [29] Wilson, D.A., Baker, G.J. and Boero, M.J. (1995) Complications of celiotomy incisions in 635 horses. Vet Surg 24, 506-514.

636 637 [30] Coomer, R.P.C., Mair, T.S., Edwards, G.B. and Proudman, C.J. (2007) Do subcutaneous 638 sutures increase risk of laparotomy wound suppuration? Equine Vet J 39, 396-399.

639 640 [31] Smith, L.J., Mellor, D.J., Marr, C.M., Reid, S.W.J. and Mair, T.S. (2007) Incisional 641 complications following exploratory celiotomy: does an abdominal bandage reduce the risk? 642 Equine Vet J 39, 277-283.

643 644 [32] Kobluk, C.N., Ducharme, N.G., Lumsden, J.H., Pascoe, P.J., Livesey, M.A., Hurtig, M., Horney, 645 F.D. and Arighi, M. (1989) Factors affecting incisional complication rates associated with colic 646 surgery in horses: 78 cases (1983-1985). J Am Vet Med Assoc 195, 639-642.

647 648 [33] Honnas, C.M. and Cohen, N.D. (1997) Risk factors for wound infection following celiotomy in 649 horses. J Am Vet Med Assoc 210, 78-81.

650 651 [34] Ingle-Fehr, J., Baxter, G.M., Howard, R.D., Trotter, G.W. and Stashak, T.S. (1997) Bacterial 652 culturing of ventral median celiotomies for prediction of postoperative incisional 653 complications in horses. Veterinary Surgery 26, 7-13.

29 655 [35] Galuppo, L.D., Pascoe, J.R., Jang, S.S., Willits, N.H. and Greenman, S.L. (1999) Evaluation of 656 iodophor skin preparation techniques and factors influencing drainage from ventral midline 657 incisions in horses. J Am Vet Med Assoc 215, 963-969.

658 659 [36] Torfs, S., Delesalle, C., Dewulf, J., Devisscher, L. and Deprez, P. (2009) Risk factors for equine 660 postoperative ileus and effectiveness of prophylactic lidocaine. J Vet Intern Med 23, 606-611.

661 662 [37] Freeman, K.D., Southwood, L.L., Lane, J., Lindborg, S. and Aceto, H.W. (2012) Post operative 663 infection, pyrexia and perioperative antimicrobial drug use in surgical colic patients. Equine 664 Vet J 44, 476-481.

665 666 [38] Torfs, S., Levet, T., Delesalle, C., Dewulf, J., Vlaminck, L., Pille, F., Lefere, L. and Martens, A. 667 (2010) Risk factors for incisional complications after exploratory celiotomy in horses: do skin 668 staples increase the risk? Vet Surg 39, 616-620.

669 670 [39] Gibson, K.T., Curtis, C.R., Turner, A.S., McIlwraith, C.W., Aanes, W.A. and Stashak, T.S. (1989) 671 Incisional hernias in the horse. Incidence and predisposing factors. Vet Surg 18, 360-366.

672 673 [40] Cook, G., Bowman, K.F., Bristol, D.G. and Tate, L.P. (1996) Ventral midline herniorrhaphy 674 following colic surgery in the horse. Equine Veterinary Education 8, 304-307.

675 676 [41] Traub-Dargatz, J.L., George, J.L., Dargatz, D.A., Morley, P.S., Southwood, L.L. and Tillotson, K. 677 (2002) Survey of complications and antimicrobial use in equine patients at veterinary 678 teaching hospitals that underwent surgery because of colic. J Am Vet Med Assoc 220, 1359- 679 1365.

680 681 [42] Schaer, B.L.D., Linton, J.K. and Aceto, H. (2012) Antimicrobial Use in Horses Undergoing Colic 682 Surgery. J Vet Intern Med 26, 1449-1456.

683 684 [43] Bratzler, D.W., Dellinger, E.P., Olsen, K.M., Perl, T.M., Auwaerter, P.G., Bolon, M.K., Fish, 685 D.N., Napolitano, L.M., Sawyer, R.G., Slain, D., Steinberg, J.P. and Weinstein, R.A. (2013) 686 Clinical practice guidelines for antimicrobial prophylaxis in surgery. Surgical infections 14, 73- 687 156.

688 689 [44] French, N.P., Smith, J., Edwards, G.B. and Proudman, C.J. (2002) Equine surgical colic: risk 690 factors for postoperative complications. Equine Vet J 34, 444-449.

691 692 [45] Zhang, C.D., Zeng, Y.J., Li, Z., Chen, J., Li, H.W., Zhang, J.K. and Dai, D.Q. (2013) Extended 693 antimicrobial prophylaxis after gastric cancer surgery: a systematic review and meta- 694 analysis. World journal of gastroenterology : WJG 19, 2104-2109.

695 696 [46] Macdonald, M.H., Pascoe, J.R., Stover, S.M. and Meagher, D.M. (1989) Survival after Small 697 Intestine Resection and Anastomosis in Horses. Vet Surg 18, 415-423.

30 698 699 [47] Rodriguez, F., Kramer, J., Fales, W., Wilson, D.A. and Keegan, K. (2009) Evaluation of 700 Intraoperative Culture Results as a Predictor for Short-Term Incisional Complications in 49 701 Horses Undergoing Abdominal Surgery. Veterinary Therapeutics 10.

702 703 [48] Canada, N.C., Beard, W.L., Guyan, M.E. and White, B.J. (2015) Comparison of sub-bandage 704 pressures achieved by 3 abdominal bandaging techniques in horses. Equine Vet J 47, 599- 705 602.

706 707 [49] Costa-Farre, C., Prades, M., Ribera, T., Valero, O. and Taura, P. (2014) Does intraoperative 708 low arterial partial pressure of oxygen increase the risk of surgical site infection following 709 emergency exploratory laparotomy in horses? Vet J 200, 175-180.

710 711 [50] Mair, T.S. and Smith, L.J. (2005) Survival and complication rates in 300 horses undergoing 712 surgical treatment of colic. Part 3: Long-term complications and survival. Equine Vet J 37, 713 310-314.

714 715 [51] Klohnen, A., Lores, M. and Fiscer, A.T. (2007) Management of postoperative abdominal 716 incisional complications with a hernia belt: 85 horses (2001-2005). In: Veterinary Surgery, Ed: 717 2007 American College of Veterinary Surgeons Symposium October 18–21, Chicago, IL, 718 Blackwell Publishing Inc. pp E1-E29.

719 720 [52] van der Linden, M.A., Laffont, C.M. and Sloet van Oldruitenborgh-Oosterbaan, M.M. (2003) 721 Prognosis in equine medical and surgical colic. J Vet Intern Med 17, 343-348.

722 723 [53] Munoz, E., Argulles, D., Areste, L., Miguel, L.S. and Prades, M. (2008) Retrospective analysis 724 of exploratory laparotomies in 192 Andalusian horses and 276 horses of other breeds. Vet 725 Rec 162, 303-306.

726 727 [54] Mezerova, J. and Zert, Z. (2008) Long-term survival and complications of colic surgery in 728 horses: analysis of 331 cases. Veterinarni Medicina 53, 43-52.

729 730 [55] van Loon, J.P., Jonckheer-Sheehy, V.S., Back, W., Rene van Weeren, P. and Hellebrekers, L.J. 731 (2014) Monitoring equine visceral pain with a composite pain scale score and correlation 732 with survival after emergency gastrointestinal surgery. Vet J 200, 109-115.

733 734 [56] Fogle, C.A., Gerard, M.P., Elce, Y.A., Little, D., Morton, A.J., Correa, M.T. and Blikslager, A.T. 735 (2008) Analysis of sodium carboxymethylcellulose administration and related factors 736 associated with postoperative colic and survival in horses with small intestinal disease. Vet 737 Surg 37, 558-563.

738 739 [57] Semevolos, S.A., Ducharme, N.G. and Hackett, R.P. (2002) Clinical assessment and outcome 740 of three techniques for jejunal resection and anastomosis in horses: 59 cases (1989-2000). J 741 Am Vet Med Assoc 220, 215-218.

31 742 743 [58] Driscoll, N., Baia, P., Fischer, A.T., Brauer, T. and Klohnen, A. (2008) Large colon resection 744 and anastomosis in horses: 52 cases (1996-2006). Equine Vet J 40, 342-347.

745 746 [59] Pritchett, L.C., Ulibarri, C., Roberts, M.C., Schneider, R.K. and Sellon, D.C. (2003) 747 Identification of potential physiological and behavioral indicators of postoperative pain in 748 horses after exploratory celiotomy for colic. Applied Animal Behaviour Science 80, 31-43.

749 750 [60] Graubner, C., Gerber, V., Doherr, M. and Spadavecchia, C. (2011) Clinical application and 751 reliability of a post abdominal surgery pain assessment scale (PASPAS) in horses. Veterinary 752 Journal 188, 178-183.

753 754 [61] Gorvy, D.A., Edwards, G.B. and Proudman, C.J. (2008) Intra-abdominal adhesions in horses: A 755 retrospective evaluation of repeat laparotomy in 99 horses with acute gastrointestinal 756 disease. Vet J 175, 194-201.

757 758 [62] Mair, T.S. and Smith, L.J. (2005) Survival and complication rates in 300 horses undergoing 759 surgical treatment of colic. Part 4: Early (acute) relaparotomy. Equine Vet J 37, 315-318.

760 761 [63] Smith, L.J. and Mair, T.S. (2010) Are horses that undergo an exploratory laparotomy for 762 correction of a right dorsal displacement of the large colon predisposed to post operative 763 colic, compared to other forms of large colon displacement? Equine Vet J 42, 44-46.

764 765 [64] Rocken, M., Schubert, C., Mosel, G. and Litzke, L.F. (2005) Indications, surgical technique, 766 and long-term experience with laparoscopic closure of the nephrosplenic space in standing 767 horses. Vet Surg 34, 637-641.

768 769 [65] Hardy, J., Minton, M., Robertson, J.T., Beard, W.L. and Beard, L.A. (2000) Nephrosplenic 770 entrapment in the horse: a retrospective study of 174 cases. Equine Vet J Suppl, 95-97.

771 772 [66] Freeman, D.E. and Schaeffer, D.J. (2010) Comparison of complications and long-term survival 773 rates following hand-sewn versus stapled side-to-side jejunocecostomy in horses with colic. 774 J Am Vet Med Assoc 237, 1060-1067.

775 776 [67] Proudman, C.J., Edwards, G.B. and Barnes, J. (2007) Differential survival in horses requiring 777 end-to-end jejunojejunal anastomosis compared to those requiring side-to-side jejunocaecal 778 anastomosis. Equine Vet J 39, 181-185.

779 780 [68] Cook, V.L., Meyer, C.T., Campbell, N.B. and Blikslager, A.T. (2009) Effect of firocoxib or 781 flunixin meglumine on recovery of ischemic-injured equine jejunum. Am J Vet Res 70, 992- 782 1000.

32 784 [69] Little, D., Brown, S.A., Campbell, N.B., Moeser, A.J., Davis, J.L. and Blikslager, A.T. (2007) 785 Effects of the cyclooxygenase inhibitor meloxicam on recovery of ischemia-injured equine 786 jejunum. Am J Vet Res 68, 614-624.

787 788 [70] Matyjaszek, S.A., Morton, A.J., Freeman, D.E., Grosche, A., Polyak, M.M. and Kuck, H. (2009) 789 Effects of flunixin meglumine on recovery of colonic mucosa from ischemia in horses. Am J 790 Vet Res 70, 236-246.

791 792 [71] Cook, V.L., Shults, J.J., McDowell, M., Campbell, N.B., Davis, J.L. and Blikslager, A.T. (2008) 793 Attenuation of ischaemic injury in the equine jejunum by administration of systemic 794 lidocaine. Equine Vet J 40, 353-357.

795 796 [72] Lefebvre, D., Pirie, R.S., Handel, I.G., Tremaine, W.H. and Hudson, N.P. (2014) Clinical 797 features and management of equine post operative ileus: Survey of diplomates of the 798 European Colleges of Equine Internal Medicine (ECEIM) and Veterinary Surgeons (ECVS). 799 Equine Vet J.

800 801 [73] Malone, E., Ensink, J., Turner, T., Wilson, J., Andrews, F., Keegan, K. and Lumsden, J. (2006) 802 Intravenous continuous infusion of lidocaine for treatment of equine ileus. Vet Surg 35, 60- 803 66.

804 805 [74] Sellon, D.C., Roberts, M.C., Blikslager, A.T., Ulibarri, C. and Papich, M.G. (2004) Effects of 806 continuous rate intravenous infusion of butorphanol on physiologic and outcome variables 807 in horses after celiotomy. Journal of Veterinary Internal Medicine 18, 555-563.

808 809 [75] Farstvedt, E. and Hendrickson, D. (2005) Laparoscopic closure of the nephrosplenic space for 810 prevention of recurrent nephrosplenic entrapment of the ascending colon. Vet Surg 34, 642- 811 645.

812 813 [76] Markel, M.D., Meagher, D.M. and Richardson, D.W. (1988) Colopexy of the large colon in 814 four horses. J Am Vet Med Assoc 192, 358-359.

815 816 [77] Hance, S.R. and Embertson, R.M. (1992) Colopexy in broodmares: 44 cases (1986-1990). J 817 Am Vet Med Assoc 201, 782-787.

818 819 [78] Embertson, R.M. and Hance, S.R. (1991) Effects of colopexy in the broodmare. Proceedings 820 of the Annual Convention of the American Association of Equine Practitioners, 531-532.

821 822 [79] Munsterman, A.S., Hanson, R.R., Cattley, R.C., Barrett, E.J. and Albanese, V. (2014) Surgical 823 technique and short-term outcome for experimental laparoscopic closure of the epiploic 824 foramen in 6 horses. Vet Surg 43, 105-113.

33 826 [80] van Bergen, T., Wiemer, P., Bosseler, L., Ugahary, F. and Martens, A. (2015) Development of 827 a new laparoscopic Foramen Epiploicum Mesh Closure (FEMC) technique in 6 horses. Equine 828 Vet J.

829 830 [81] Archer, D.C., Pinchbeck, G.L. and Proudman, C.J. (2011) Factors associated with survival of 831 epiploic foramen entrapment colic: A multicentre, international study. Equine Vet J 43, 56- 832 62.

833 834 [82] Cohen, N.D., Lester, G.D., Sanchez, L.C., Merritt, A.M. and Roussel, A.J.J. (2004) Evaluation of 835 risk factors associated with development of postoperative ileus in horses. J Am Vet Med 836 Assoc 225, 1070-1078.

837 838 [83] Hunt, J.M., Edwards, G.B. and Clarke, K.W. (1986) Incidence, diagnosis and treatment of 839 postoperative complications in colic cases. Equine Vet J 18, 264-270.

840 841 [84] Blikslager, A.T., Bowman, K., Levine, J.F., Bristol, D.G. and Roberts, M.C. (1994) Evaluation of 842 factors associated with postoperative ileus in horses: 31 cases (1990-1992). J Am Vet Med 843 Assoc 205, 1748-1752.

844 845 [85] Rendle, D.I., Wood, J.L.N., Summerhays, G.E.S., Walmsley, J.P., Boswell, J.C. and Phillips, T.J. 846 (2005) End-to-end jejuno-ileal anastomosis following resection of strangulated small 847 intestine in horses: a comparative study. Equine veterinary journal; Colic special issue. 37, 848 356-359.

849 850 [86] Roussel, A.J., Jr., Cohen, N.D., Hooper, R.N. and Rakestraw, P.C. (2001) Risk factors 851 associated with development of postoperative ileus in horses. J Am Vet Med Assoc 219, 72- 852 78.

853 854 [87] Gazzerro, D.M., Southwood, L.L. and Lindborg, S. (2015) Short-Term Complications After 855 Colic Surgery in Geriatric Versus Mature Non-Geriatric Horses. Vet Surg 44, 256-264.

856 857 [88] Brianceau, P., Chevalier, H., Karas, A., Court, M.H., Bassage, L., Kirker-Head, C., Provost, P. 858 and Paradis, M.R. (2002) Intravenous lidocaine and small-intestinal size, abdominal fluid, and 859 outcome after colic surgery in horses. J Vet Intern Med 16, 736-741.

860 861 [89] Dart, A.J., Peauroi, J.R., Hodgson, D.R. and Pascoe, J.R. (1996) Efficacy of metoclopramide for 862 treatment of ileus in horses following small intestinal surgery: 70 cases (1989-1992). Aust 863 Vet J 74, 280-284.

864 865 [90] Merritt, A.M. and Blikslager, A.T. (2008) Post operative ileus: to be or not to be? Equine Vet J 866 40, 295-296.

34 868 [91] Freeman, D.E. (2008) Post operative ileus (POI): another perspective. Equine Vet J 40, 297- 869 298.

870 871 [92] De Ceulaer, K., Delesalle, C., Van Elzen, R., Van Brantegem, L., Weyns, A. and Van Ginneken, 872 C. (2011) Morphological changes in the small intestinal smooth muscle layers of horses 873 suffering from small intestinal strangulation. Is there a basis for predisposition for reduced 874 contractility? Equine Vet J 43, 439-445.

875 876 [93] Gerhards, H. (1990) Incidence and significance of spontaneous and postoperative peritoneal 877 adhesions in horses. Pferdeheilkunde 6, 277-280,283-288.

878 879 [94] Parker, J.E., Fubini, S.L. and Todhunter, R.J. (1989) Retrospective evaluation of repeat 880 celiotomy in 53 horses with acute gastrointestinal disease. Vet Surg 18, 424-431.

881 882 [95] Baxter, G.M., Broome, T.E. and Moore, J.N. (1989) Abdominal adhesions after small 883 intestinal surgery in the horse. Vet Surg 18, 409-414.

884 885 [96] Cable, C.S., Fubini, S.L., Erb, H.N. and Hakes, J.E. (1997) Abdominal surgery in foals: a review 886 of 119 cases (1977-1994). Equine Vet J 29, 257-261.

887 888 [97] Singer, E.R. and Livesey, M.A. (1997) Evaluation of exploratory laparotomy in young horses: 889 102 cases (1987-1992). J Am Vet Med Assoc 211, 1158-1162.

890 891 [98] Vatistas, N.J., Snyder, J.R., Wilson, W.D., Drake, C. and Hildebrand, S. (1996) Surgical 892 treatment for colic in the foal (67 cases): 1980-1992. Equine Vet J 28, 139-145.

893 894 [99] Kuebelbeck, K.L., Slone, D.E. and May, K.A. (1998) Effect of omentectomy on adhesion 895 formation in horses. Vet Surg 27, 132-137.

896 897 [100] Epstein, K.L., Brainard, B.M., Gomez-Ibanez, S., Lopes, M.A.F., Barton, M.H. and Moore, J.N. 898 (2011) Thrombelastography in horses with acute gastrointestinal disease. J Vet Intern Med 899 25, 307-314.

900 901 [101] Hague, B.A., Honnas, C.M., Berridge, B.R. and Easter, J.L. (1998) Evaluation of postoperative 902 peritoneal lavage in standing horses for prevention of experimentally induced abdominal 903 adhesions.

904 905 [102] Moll, H.D., Schumacher, J., Wright, J.C. and Spano, J.S. (1991) Evaluation of Sodium 906 Carboxymethylcellulose for Prevention of Experimentally Induced Abdominal Adhesions in 907 Ponies. Am J Vet Res 52, 88-91.

35 909 [103] Murphy, D.J., Peck, L.S., Detrisac, C.J., Widenhouse, C.W. and Goldberg, E.P. (2002) Use of a 910 high-molecular-weight carboxymethylcellulose in a tissue protective solution for prevention 911 of postoperative abdominal adhesions in ponies. Am J Vet Res 63, 1448-1454.

912 913 [104] Mueller, P.O.E., Harmon, B.G., Hay, W.P. and Amoroso, L.M. (2000) Effect of 914 carboxymethylcellulose and a hyaluronate-carboxymethylcellulose membrane on healing of 915 intestinal anastomoses in horses. Am J Vet Res 61, 369-374.

916 917 [105] Sheldon, H.K., Gainsbury, M.L., Cassidy, M.R., Chu, D.I., Stucchi, A.F. and Becker, J.M. (2012) 918 A sprayable hyaluronate/carboxymethylcellulose adhesion barrier exhibits regional adhesion 919 reduction efficacy and does not impair intestinal healing. J Gastrointest Surg 16, 325-333.

920 921 [106] Freeman, D.E. and Schaeffer, D.J. (2011) Clinical comparison between a continuous Lembert 922 pattern wrapped in a carboxymethylcellulose and hyaluronate membrane with an 923 interrupted Lembert pattern for one-layer jejunojejunostomy in horses. Equine Vet J 43, 708- 924 713.

925 926 [107] Mueller, P.O., Hunt, R.J., Allen, D., Parks, A.H. and Hay, W.P. (1995) Intraperitoneal use of 927 sodium carboxymethylcellulose in horses undergoing exploratory celiotomy. Vet Surg 24, 928 112-117.

929 930 [108] Schaer, B.L.D., Aceto, H., Caruso, M.A. and Brace, M.A. (2012) Identification of Predictors of 931 Salmonella Shedding in Adult Horses Presented for Acute Colic. J Vet Intern Med 26, 1177- 932 1185.

933 934 [109] Ellis, C.M., Lynch, T.M., Slone, D.E., Hughes, F.E. and Clark, C.K. (2008) Survival and 935 complications after large colon resection and end-to-end anastomosis for strangulating large 936 colon volvulus in seventy-three horses. Vet Surg 37, 786-790.

937 938 [110] Pierce, R.L., Fischer, A.T., Rohrbach, B.W. and Klohnen, A. (2010) Postoperative 939 complications and survival after enterolith removal from the ascending or descending colon 940 in horses. Vet Surg 39, 609-615.

941 942 [111] Cohen, N.D. and Honnas, C.M. (1996) Risk factors associated with development of diarrhea 943 in horses after celiotomy for colic: 190 cases (1990-1994). J Am Vet Med Assoc 209, 810-813.

944 945 [112] Granot, N., Milgram, J., Bdolah-Abram, T., Shemesh, I. and Steinman, A. (2008) Surgical 946 management of sand colic impactions in horses: a retrospective study of 41 cases. Aust Vet J 947 86, 404-407.

948 949 [113] de Bont, M.P., Proudman, C.J. and Archer, D.C. (2013) Surgical lesions of the small colon and 950 post operative survival in a UK hospital population. Equine Vet J 45, 460-464.

36 952 [114] Prange, T., Holcombe, S.J., Brown, J.A., Dechant, J.E., Fubini, S.L., Embertson, R.M., Peroni, J., 953 Rakestraw, P.C. and Hauptman, J.G. (2010) Resection and anastomosis of the descending 954 colon in 43 horses. Vet Surg 39, 748-753.

955 956 [115] Frederico, L.M., Jones, S.L. and Blikslager, A.T. (2006) Predisposing factors for small colon 957 impaction in horses and outcome of medical and surgical treatment: 44 cases (1999-2004). J 958 Am Vet Med Assoc 229, 1612-1616.

959 960 [116] Parraga, M.E., Spier, S.J., Thurmond, M. and Hirsh, D. (1997) A clinical trial of probiotic 961 administration for prevention of Salmonella shedding in the postoperative period in horses 962 with colic. J Vet Intern Med 11, 36-41.

963 964 [117] Ekiri, A.B., MacKay, R.J., Gaskin, J.M., Freeman, D.E., House, A.M., Giguere, S., Troedsson, 965 M.R., Schuman, C.D., von Chamier, M.M., Henry, K.M. and Hernandez, J.A. (2009) 966 Epidemiologic analysis of nosocomial Salmonella infections in hospitalized horses. J Am Vet 967 Med Assoc 234, 108-119.

968 969 [118] Costa, M.C., Arroyo, L.G., Allen-Vercoe, E., Stampfli, H.R., Kim, P.T., Sturgeon, A. and Weese, 970 J.S. (2012) Comparison of the fecal microbiota of healthy horses and horses with colitis by 971 high throughput sequencing of the V3-V5 region of the 16S rRNA gene. PloS one 7, e41484.

972 973 [119] Ernst, N.S., Hernandez, J.A., MacKay, R.J., Brown, M.P., Gaskin, J.M., Nguyen, A.D., Giguere, 974 S., Colahan, P.T., Troedsson, M.R., Haines, G.R., Addison, I.R. and Miller, B.J. (2004) Risk 975 factors associated with fecal Salmonella shedding among hospitalized horses with signs of 976 gastrointestinal tract disease. Javma-Journal of the American Veterinary Medical Association 977 225, 275-281.

978 979 [120] Hird, D.W., Casebolt, D.B., Carter, J.D., Pappaioanou, M. and Hjerpe, C.A. (1986) Risk factors 980 for salmonellosis in hospitalized horses. J Am Vet Med Assoc 188, 173-177.

981 982 [121] Kim, L.M., Morley, P.S., Traub-Dargatz, J.L., Salman, M.D. and Gentry-Weeks, C. (2001) 983 Factors associated with Salmonella shedding among equine colic patients at a veterinary 984 teaching hospital. J Am Vet Med Assoc 218, 740-748.

985 986 [122] Hassel, D.M., Smith, P.A., Nieto, J.E., Beldomenico, P. and Spier, S.J. (2009) Di-tri-octahedral 987 smectite for the prevention of post-operative diarrhea in equids with surgical disease of the 988 large intestine: results of a randomized clinical trial. Veterinary Journal 182, 210-214.

989 990 [123] Senior, J.M., Proudman, C.J., Leuwer, M. and Carter, S.D. (2011) Plasma endotoxin in horses 991 presented to an equine referral hospital: correlation to selected clinical parameters and 992 outcomes. Equine Vet J 43, 585-591.

993 994 [124] Moore, J.N. and Vandenplas, M.L. (2014) Is it the systemic inflammatory response syndrome 995 or endotoxemia in horses with colic? Vet Clin North Am Equine Pract 30, 337-351, vii-viii.

37 996 997 [125] Lankveld, D.P., Ensink, J.M., van Dijk, P. and Klein, W.R. (2001) Factors influencing the 998 occurrence of thrombophlebitis after post-surgical long-term intravenous catheterization of 999 colic horses: a study of 38 cases. Journal of veterinary medicine. A, Physiology, pathology, 1000 clinical medicine 48, 545-552.

1001 1002 [126] Dolente, B.A., Beech, J., Lindborg, S. and Smith, G. (2005) Evaluation of risk factors for 1003 development of catheter-associated jugular thrombophlebitis in horses: 50 cases (1993- 1004 1998). J Am Vet Med Assoc 227, 1134-1141.

1005 1006 [127] Parsons, C.S., Orsini, J.A., Krafty, R., Capewell, L. and Boston, R. (2007) Risk factors for 1007 development of acute laminitis in horses during hospitalization: 73 cases (1997-2004). J Am 1008 Vet Med Assoc 230, 885-889.

1009 1010 [128] Monreal, L., Angles, A., Espada, Y., Monasterio, J. and Monreal, M. (2000) Hypercoagulation 1011 and hypofibrinolysis in horses with colic and DIC. Equine Vet J Suppl, 19-25.

1012 1013 [129] Young, D.R., Richardson, D.W. and Markel, M.D. (1989) The effect of low dose heparin 1014 therapy on complication and survival rates in horses following exploratory celiotomy. Equine 1015 Vet J Suppl, 91-93.

1016 1017 [130] Feige, K., Schwarzwald, C.C. and Bombeli, T. (2003) Comparison of unfractioned and low 1018 molecular weight heparin for prophylaxis of coagulopathies in 52 horses with colic: a 1019 randomised double-blind clinical trial. Equine Vet J 35, 506-513.

1020 1021 [131] Geraghty, T.E., Love, S., Taylor, D.J., Heller, J., Mellor, D.J. and Hughes, K.J. (2009) 1022 Assessment of subclinical venous catheter-related diseases in horses and associated risk 1023 factors. Vet Rec 164, 227-231.

1024 1025 [132] Van Eps, A.W. and Pollitt, C.C. (2009) Equine laminitis model: cryotherapy reduces the 1026 severity of lesions evaluated seven days after induction with oligofructose. Equine Vet J 41, 1027 741-746.

1028 1029 [133] Doyle, A.J., Freeman, D.E., Rapp, H., Murrell, J.A. and Wilkins, P.A. (2003) Life-threatening 1030 hemorrhage from enterotomies and anastomoses in 7 horses. Vet Surg 32, 553-558.

1031 1032 [134] Morgan, R.A., Raftery, A.G., Cripps, P., Senior, J.M. and McGowan, C.M. (2011) The 1033 prevalence and nature of cardiac arrhythmias in horses following general anaesthesia and 1034 surgery. Acta Vet Scand 53, 62.

1035 1036 [135] Dukti, S. and White, N. (2008) Surgical complications of colic surgery. Vet Clin North Am 1037 Equine Pract 24, 515-534, vii-viii.

38 1039 [136] Hackett, E.S. and Hassel, D.M. (2008) Colic: nonsurgical complications. Vet Clin North Am 1040 Equine Pract 24, 535-555, viii.

1041 1042 [137] Gray, S.N., Dechant, J.E., LeJeune, S.S. and Nieto, J.E. (2015) Identification, Management and 1043 Outcome of Postoperative Hemoperitoneum in 23 Horses After Emergency Exploratory 1044 Celiotomy for Gastrointestinal Disease. Vet Surg 44, 379-385.

1045 1046 [138] Epstein, K.L. (2014) Coagulopathies in horses. Vet Clin North Am Equine Pract 30, 437-452, ix.

1047 1048 [139] Durham, A.E., Phillips, T.J., Walmsley, J.P. and Newton, J.R. (2003) Study of the clinical 1049 effects of postoperative parenteral nutrition in 15 horses. Vet Rec 153, 493-498.

1050 1051 [140] Lopes, M.A. and White, N.A., 2nd (2002) Parenteral nutrition for horses with gastrointestinal 1052 disease: a retrospective study of 79 cases. Equine Vet J 34, 250-257.

1053 1054 [141] Durham, A.E., Phillips, T.J., Walmsley, J.P. and Newton, J.R. (2004) Nutritional and 1055 clinicopathological effects of post operative parenteral nutrition following small intestinal 1056 resection and anastomosis in the mature horse. Equine Vet J 36, 390-396.

1057 1058 [142] Valle, E., Bergero, D. and Gandini, M. (2014) Nutritional management of hospitalized horses 1059 for colic according to their risk category. In: The Eleventh International Equine colic Research 1060 Symposium, Dublin, Ireland.

1061 1062 [143] Christophersen, M.T., Dupont, N., Berg-Sorensen, K.S., Konnerup, C., Pihl, T.H. and Andersen, 1063 P.H. (2014) Short-term survival and mortality rates in a retrospective study of colic in 1588 1064 Danish horses. Acta Vet Scand 56, 20.

1065 1066 [144] Blikslager, A.T. and Mair, T.S. (2014) Trends in management of horses referred for evaluation 1067 of colic: 2004-2013. In: The Eleventh International Equine colic Research Symposium, Dublin, 1068 Ireland.

1069 1070 [145] Bischofberger, A.S., Brauer, T., Gugelchuk, G. and Klohnen, A. (2010) Difference in incisional 1071 complications following exploratory celiotomies using antibacterial-coated suture material 1072 for subcutaneous closure: Prospective randomised study in 100 horses. Equine Vet J 42, 304- 1073 309.

1074 1075 [146] Anderson, S.L., Vacek, J.R., MacHarg, M.A. and Holtkamp, D.J. (2011) Occurrence of 1076 Incisional Complications and Associated Risk Factors Using a Right Ventral Paramedian 1077 Celiotomy Incision in 159 Horses. Vet Surg 40, 82-89.

1078 1079 [147] Holcombe, S.J., Rodriguez, K.M., Haupt, J.L., Campbell, J.O., Chaney, K.P., Sparks, H.D. and 1080 Hauptman, J.G. (2009) Prevalence of and risk factors for postoperative ileus after small 1081 intestinal surgery in two hundred and thirty-three horses. Vet Surg 38, 368-372.

39 1082

1083 1084