Abstract
Objective—To determine the effect of colic surgery on return to function in Thoroughbred racehorses, identify clinical variables associated with successful return to racing, and compare racing performance between horses undergoing colic surgery and an untreated cohort.
Design—Retrospective cohort study.
Animals—59 Thoroughbred racehorses 2 to 5 years of age that underwent colic surgery and survived to hospital discharge and 90 untreated Thoroughbred racehorses equivalent in class.
Procedures—Medical records of patients evaluated for colic between January 1996 and July 2009 were reviewed, and horses with a Jockey Club Information Systems record were included. Physical examination and laboratory findings on hospital admission, lesion location and type, duration of surgery, duration of hospitalization, and any postoperative complications were recorded. The untreated cohort comprised 2 untreated horses randomly selected from runners in each treated horse's last race immediately prior to the date of colic surgery. Records were obtained from the Jockey Club Information Systems in April 2011. Only horses that raced at least once before and after surgery were included in the performance analysis. Number of starts, earnings per start, and total earnings were determined from race records for all horses. Quarterly earnings and number of starts for 12 quarters following the date of surgery were compared between treated and untreated horses via a Wilcoxon rank sum test. Longevity of racing was assessed by means of survival analysis. Poisson regression was used to compare rates of return to racing and active quarters aggregated across the first 12 quarters after surgery and for the available follow-up period for treated and untreated horses.
Results—45 of 59 (76%) horses that raced prior to surgery returned to racing. Return to racing was significantly associated with admission heart rate and blood lactate concentration. From quarters 3 to 12, treated and untreated horses had slight differences in the number of starts but no difference in earnings per quarter. Treated and untreated horses had no difference in total number of quarters raced, number of starts, or earnings after surgery. Treated horses had higher earnings per start, compared with untreated horses.
Conclusions and Clinical Relevance—In the present study, racing Thoroughbreds that underwent colic surgery and successfully returned to racing had no differences in performance variables, compared with their untreated cohorts.
Colic is an important cause of loss to the United States horse industry, with the total economic impact estimated at > $115.3 million/y.1 Several studies2,3 indicate a possible increased risk of colic in young Thoroughbreds in training. Hillyer et al2 showed a significant increase in the incidence of colic in Thoroughbreds on a flat-racing premises, compared with national hunt horses, and Tinker et al3 reported a significantly increased risk of colic in horses 2 to 10 years of age, with Thoroughbreds significantly overrepresented. Young horses in training had a significant increase in the incidence of colic, compared with mature horses that were used for pleasure riding or were kept at pasture.3 Despite the importance of colic to the equine industry, few studies have evaluated the impact of abdominal surgery for treatment of colic on postoperative performance.
Long-term survival and the ability of the horse to meet owner and trainer performance expectations following colic surgery are becoming increasingly important. Although there is a perception that horses do not return to their previous level of performance after colic surgery, recent studies4,5 evaluated the rate of return to athletic function after colic surgery and found that 85% to 90% of horses resumed sports activities at the expected level of performance after surgery. Only 1 study6 has evaluated racing performance in Thoroughbreds following colic surgery. The investigators evaluated racing outcome after colic surgery in Thoroughbreds < 2 years of age (those that had not raced prior to surgery).6 Young Thoroughbreds that underwent colic surgery were less likely to race, compared with their cohorts; however, if they did race, their pre- and postoperative performance was similar.6
No studies have evaluated clinical findings associated with the ability of a Thoroughbred racehorse to return to racing following colic surgery or compared racing performance of Thoroughbreds undergoing abdominal surgery for colic with that of horses of the same class that did not undergo surgery for colic. The purpose of the study reported here was to compare hospital admission clinical data, surgical findings, and postoperative complications as well as preoperative racing statistics between Thoroughbred racehorses that did and did not return to racing following colic surgery and to compare the total postoperative and quarterly number of starts and earnings for at least 2 years and career longevity between horses undergoing colic surgery (treated) and horses from the race prior to colic surgery (untreated). Our hypotheses were that severity of disease and postoperative complications would impact the ability of Thoroughbred racehorses to return to racing following surgery, and racing performance of horses that had abdominal surgery for colic would not be different from untreated horses by the third quarter following surgery.
Materials and Methods
Patient selection—Medical records of Thoroughbreds 2 to 5 years of age that were evaluated for colic at the George D. Widener Hospital for Large Animals at New Bolton Center, University of Pennsylvania; underwent exploratory celiotomy; and survived to hospital discharge between January 1996 and July 2009 were searched. Horses with a Jockey Club Information Systemsa record were included in the study. Horses were classified as having raced at least once prior to and after colic surgery, raced prior to surgery but failed to return to racing after surgery, and raced after surgery but not prior to surgery. Horses that had raced prior to surgery but not after surgery were only included as having failed to return to racing if they had raced within the 12 months preceding surgery.
Clinical variables and outcome—Information obtained from the medical record included physical examination and laboratory findings on hospital admission, lesion location (small intestinal, large intestinal, or other) and type (nonstrangulating, strangulating, or other), duration of surgery, and duration of hospitalization. Postoperative complications were identified and classified as incisional infection (persistent drainage for > 36 hours of serous, serosanguineous, or purulent fluid from the incision that occurred after the initial 48-hour postoperative period and was treated either locally or systemically), postoperative nasogastric reflux (> 2 L at any time after surgery), diarrhea (> 1 unformed bowel movement), salmonellosis (identification of Salmonella organisms in the feces or reflux as determined by selenite enrichment culture on at least 1 occasion, with or without clinical signs), postoperative colic, pneumonia, catheter-site infection (heat, signs of pain, and swelling, or drainage associated with the catheter site), and laminitis (increased palpable digital pulses, warm feet on palpation, and hoof sensitivity).7 These data were used to determine clinical variables associated with failure to return to racing.
Performance evaluation—Only horses that raced at least once prior to and after colic surgery were included in the performance evaluation analysis. Racing records for treated horses and 2 untreated horses randomly selected from runners in each treated horse's last race immediately prior to the date of colic surgery were obtained from the Jockey Club Information Systemsa in April 2011. The model for evaluating racing performance following surgery was based on a previous study8 in which the performance of horses undergoing a modified laryngoplasty and a randomly selected (analogous to the present study) untreated cohort of 3 horses for each treated horse was compared. Dates of the first race after the date of colic surgery were recorded for both treated horses and their untreated cohort and were used to calculate time from surgery to return to racing for treated horses. The date of the last race entered in the Jockey Club Information Systems was also recorded for all treated and untreated horses and was used to determine the number of months from the first to the last postsurgery race. Summary statistics calculated for each horse included the total number of races, total race winnings, total number of quarters in which racing occurred, races per quarter, total number of quarters in which dollars were earned, dollar winnings per active quarter, dollar winnings per start, and number of quarters idle (defined as quarters in which a horse did not race but records showed that it resumed racing in subsequent quarters). Total racing career was the sum of quarters in which racing occurred and the number of quarters idle. A quarter was defined as a 3-month interval of time.
Statistical analysis—Admission data, lesion, procedures performed, postoperative complications, duration of hospitalization, and the proportion of horses returning to racing, time to first start, number of starts, and earnings were evaluated descriptively. Continuous data are reported as mean ± SD if normally distributed and median and IQR if not normally distributed. Categorical data are reported as the proportion of horses in each category.
Our first hypothesis was addressed by determining the association between clinical variables and pre-race statistics and return to racing. The distribution of continuous data was determined on the basis of the Shapiro-Wilk test for normality. A Brown-Forsyth test for homogeneity of variance was also performed. If the data were normally distributed and had equal variance, a 1-way ANOVA was performed. If the data were not normally distributed, a Wilcoxon rank sum test was used to determine whether there was a significant association between the variable and return to racing. Categorical data were analyzed with a χ2 test or, when there were < 5 values in an expected frequency cell, a Fisher exact test. Level of significance was P < 0.05.
Our second hypothesis was addressed by comparing performance data for horses that underwent colic surgery (treated) with data for untreated horses randomly selected from the race prior to surgery. Performance data were examined for normality by means of the Shapiro-Wilk test. Most parameters were not normally distributed, so the Wilcoxon rank sum test was used to evaluate our second hypothesis by assessing differences in quarterly earnings between the treated and untreated cohorts. In addition, Poisson regression was used to compare rates of return to racing and active quarters aggregated across the first 12 quarters after surgery and for the entire follow-up period available for treated and untreated horses. Earnings were compared in a similar manner by means of regression analysis with a fixed effects model. To account for the likelihood that performance variables for individual horses would be correlated from one quarter to the next, inclusion of horse as a fixed effect and robust variance estimation were used to control for clustering in all regressions. Kaplan-Meier product limit estimates were used to evaluate the total duration of racing career in terms of both quarters (ie, the sum of active and inactive quarters), number of months between the dates of the first and last postsurgery races, and total number of races between treated and untreated horses. Data were not censored. Areas under the survival curves were used to determine the mean number of races and 95% CIs. Equality of survivor functions across treated and untreated cohorts were assessed by the Wilcoxon (Breslow) test. A value of P < 0.05 was considered significant. Statistical analyses were performed with statistical software b
Results
Patient selection—Between January 1996 and July 2009, 5,765 horses were referred to New Bolton Center for colic. Of these, 728 were 2- to 5-year-old Thoroughbreds at the time of evaluation; 89 Thoroughbreds for which race records were available were discharged from the hospital following colic surgery. On the basis of the 12-month exclusion criterion, 3 horses were excluded because they had not raced within 14 months preceding surgery. Fifty-nine horses had raced prior to surgery. Forty-five (76%) horses that were discharged from the hospital raced both prior to and after surgery. One horse had 4 colic surgeries within 1 year and then returned to racing; the duration of this animal's racing career was evaluated on the basis of the date of its last surgery. Fourteen (24%) horses raced prior to surgery but did not race after surgery. Twenty-seven horses had not raced prior to surgery but raced after surgery. Only horses that had raced prior to surgery were included in the analyses for clinical variables associated with return to racing. Horses that had not raced before surgery but raced afterward were not included in any analyses because there was no comparable untreated cohort.
Clinical variables and return to racing—Of the 59 horses that raced prior to surgery, there were 25 sexually intact females, 20 castrated males, and 14 sexually intact males. There was no significant difference in the percentage of horses of each sex that returned to racing. The median age of horses that returned to racing following colic surgery was 4 (IQR, 3 to 4) years, and the median age of horses that did not return to racing following colic surgery was 4 (IQR, 3 to 5) years (P = 0.11). Admission blood lactate concentration (P < 0.001) and heart rate (P = 0.03) were significantly lower in horses that returned to racing, compared with those of horses that did not return to racing (Table 1). Total plasma protein concentration (P = 0.08), duration of surgery (P = 0.09), and duration of hospitalization (P = 0.71) were not associated with whether horses returned to racing.
Clinical variables and preoperative racing data for 59 Thoroughbred racehorses 2 to 5 years of age that underwent colic surgery and did or did not return to racing after surgery.
| Variable | Returned to racing | Did not return to racing |
|---|---|---|
| Heart rate (beats/min)* | 44 (36–52) n = 38 | 51 (44–74) n = 12 |
| PCV (%) | 40 (36–45) n = 38 | 44 (39–46) n = 12 |
| Total plasma protein (g/dL) | 6.3 (6.0–6.8) n = 38 | 6.7 (6.7–6.9) n = 12 |
| Lactate (mmol/L)* | 0.8 (0.5–1.1) n = 30 | 1.8 (1.1–4.3) n = 12 |
| Duration of surgery (min) | 100 (86–129) n = 38 | 138 (79–177) n = 12 |
| Duration of hospitalization (d) | 7 (5–11) n = 45 | 7 (4–12) n = 14 |
| Total races before surgery | 8 (4–14) n = 45 | 14 (3–18) n = 14 |
| Total months racing before surgery | 9.5 (4.8–20.5) n = 45 | 19.3 (2.6–27.8) n = 14 |
| Total earnings before surgery (× $1,000) | 28.0 (13.5–67.1) n = 45 | 52.5 (5.8–68.9) n = 14 |
Data are median (IQR). The number of data points is indicated.
Significantly different between groups (P < 0.05).
Lesion location and type were recorded in 52 of 59 (88%) cases. There were 26 (50%) large intestinal nonstrangulating lesions, 15 (29%) small intestinal strangulating lesions, 5 (10%) large intestinal strangulating lesions, 3 (6%) small intestinal nonstrangulating lesions, and 3 (6%) lesions classified as other. There were 17 of 26 horses with a large intestinal nonstrangulating lesion, 14 of 15 horses with a small intestinal strangulating lesion, 5 of 5 horses with a large intestinal strangulating lesion, 2 of 3 horses with a small intestinal nonstrangulating lesion, and 2 of 3 horses with a lesion categorized as other that returned to racing. Significantly (P = 0.02) fewer horses with a nonstrangulating obstruction (primarily a large intestinal nonstrangulating obstruction) returned to racing, compared with horses with a strangulating obstruction. The percentages of horses that returned to racing with specific diagnoses were summarized (Table 2). The development of complications in the postsurgical period was not significantly associated with return to racing after surgery (Table 3); however, numbers for types of complications were low (ie, catheter-associated infection and laminitis).
The most common diagnoses for Thoroughbred racehorses 2 to 5 years of age that underwent colic surgery (n = 52) and did or did not return to racing after surgery.
| Diagnosis | Returned to racing | Did not return to racing |
|---|---|---|
| Large colon impaction (n = 5) | 3 (60) | 2 (40) |
| Right dorsal displacement (n = 10) | 7 (70) | 3 (30) |
| Nephrosplenic ligament entrapment (n = 4) | 1 (25) | 3 (75) |
| Large colon volvulus (n = 4) | 4 (100) | 0 (0) |
| Cecal impaction (n = 6) | 5 (83) | 1 (17) |
| Small intestinal volvulus (n = 10) | 9 (90) | 1 (10) |
| Other (n = 13) | 11 (85) | 2 (15) |
Insufficient information was available in the medical record to record the final diagnosis for some patients. Data are No. of horses (%).
Postoperative complications for Thoroughbred racehorses 2 to 5 years of age that underwent colic surgery and did or did not return to racing after surgery.
| Complication* | Returned to racing | Did not return to racing |
|---|---|---|
| Incisional infection (10%) | ||
| Yes | 5 (100) | 0 (0) |
| No | 33 (73) | 12 (27) |
| Postoperative reflux (10%) | ||
| Yes | 3 (60) | 2 (40) |
| No | 34 (77) | 10 (23) |
| Diarrhea (6%) | ||
| Yes | 3 (100) | 0 (0) |
| No | 35 (74) | 12 (26) |
| Salmonellosis (18%) | ||
| Yes | 8 (89) | 1 (11) |
| No | 30 (73) | 11 (27) |
| Colic (16%) | ||
| Yes | 5 (63) | 3 (37) |
| No | 33 (79) | 9 (21) |
| Pneumonia (2%) | ||
| Yes | 1 (100) | 0 (0) |
| No | 37 (76) | 12 (24) |
| Catheter-associated infection (6%) | ||
| Yes | 1 (33) | 2 (67) |
| No | 37 (79) | 10 (21) |
| Laminitis (4%) | ||
| Yes | 1 (50) | 1 (50) |
| No | 37 (77) | 11 (23) |
Data are shown as No. of horses (%).
Percentage of horses with each complication is indicated.
Among treated horses, there were also no associations between presurgical total number of races (P = 0.23), total months raced (P = 0.49), or total earnings (P = 0.98) and whether horses returned to racing (Table 1). For horses returning to racing after surgery, median duration between the date of surgery and return to racing was 7.0 months (IQR, 5.5 to 10.0 months).
Performance variables—Number of starts prior to the date of surgery was not significantly (P = 0.23) different between treated horses (11; IQR, 5 to 16) and untreated cohorts (8; IQR, 4 to 14). Total earnings prior to the date of surgery was not significantly (P = 0.87) different between treated horses ($31,465; IQR, $9,715 to $72,020) and untreated cohorts ($28,000; IQR, $13,696 to $64,315). Duration of racing career prior to the date of surgery was not significantly (P = 0.94) different between treated horses (11 months; IQR, 3 to 18.5 months) and untreated cohorts (9.5 months; IQR, 5 to 19 months).
On the basis of the date of surgery for each treated horse and the date when racing statistics were obtained from the Jockey Club Information Systems for all treated and untreated horses, the least amount of time for which follow-up data were potentially available on a treated horse and its untreated cohort was 29 months (9.7 quarters) and the maximum was 180 months (60 quarters); median follow-up time available for treated horses and accompanying untreated horses was 97 months (IQR, 56 to 142 months), or 32.2 quarters (IQR, 18.7 to 47.3 quarters). The minimum number of quarters for which racing data were actually recorded in the information system for individual animals was 4 quarters for a treated horse and 1 quarter for an untreated horse; maximums were 32 quarters and 33 quarters for treated and untreated horses, respectively. Evaluation of dates for each animal's last recorded postsurgery race revealed that no horses had raced < 5 months prior to data collection. Moreover, only 6 treated horses and 8 untreated horses had raced within 12 months (4 quarters) prior to the collection date, suggesting that at least 87% of treated horses and 91% of their untreated cohorts and possibly more had completed their racing careers by the time racing records were obtained, which was supported by the additional finding that during the period when racing data were collected, with the exception of the postsurgical period of inactivity in treated horses, only 8 horses (2 [4.4%] treated and 6 [6.7%] untreated) were idle for > 3 consecutive quarters and then resumed racing. The longest idle period was 10 quarters in an untreated horse. Because it seemed likely that all possible race records had been collected for most of the study population, no additional steps were taken in the analyses to account for potentially missing data due to late recruitment (eg, in the survival analyses, all values are expressed relative to the date of surgery [time 0]).
Earnings per quarter and number of starts, expressed as mean ± SD of the mean from 1 to 12 quarters after the date of surgery for case and control horses, were summarized (Figure 1). Treated horses had significantly (P < 0.001) fewer starts per quarter and lower earnings per quarter in quarters 1 and 2. Although number of starts was still marginally lower, albeit not significantly (P = 0.05), for treated horses in quarter 3 (median, 2 starts; IQR, 0 to 4 starts) versus untreated horses (median, 1 start; IQR, 0 to 2 starts), there was no significant (P = 0.1) difference in quarter 3 earnings between the treated (median, $0; IQR, $0 to $9,650) and untreated (median, $500; IQR, $0 to $7,647) cohorts. In quarters 4 to 12, there was no significant difference in earnings in any quarter, even though the treated horses had slightly fewer races in both quarters 6 (median, 1.5 starts; IQR, 0 to 3 starts; P = 0.03) and 7 (median, 1 start; IQR, 0 to 3 starts; P = 0.05) than did untreated horses in quarters 6 (median, 3 starts; IQR, 1 to 4 starts) and 7 (median, 3 starts; IQR, 1 to 4 starts). Further, the Wilcoxon rank sum and regression analyses revealed that there were no significant differences in the either the number of race starts recorded for quarters 1 to 12 between treated (median, 13 starts; IQR, 7 to 20 starts) and untreated (16 starts; IQR, 6 to 24 starts; P = 0.27) cohorts or the incidence rate ratio (0.83; 95% CI, 0.65 to 1.05; P = 0.12). The earnings for that period were not significantly different between treated (median, $30,350; IQR, $12,751 to $91,723) and untreated (median, $30,818; IQR, $9,725 to $79,844; P = 0.39) cohorts (regression coefficient = 31,141; 95% CI, −47,104 to 109,387; P = 0.43). Similarly, when all available race records after the date of surgery were examined, the rate of race starts between treated and untreated horses was not significantly different (incidence rate ratio, 1.25; 95% CI, 0.96 to 1.63; P = 0.10), nor were total earnings (regression coefficient, 2,014; 95% CI, −5,600 to 9,629; P = 0.63), even though the earnings per start was slightly higher for treated horses than for their untreated counterparts (P = 0.04; Table 4). Because treated horses had a significantly (P < 0.001) higher number of quarters inactive (accounted for by the postsurgical period of inactivity), compared with untreated horses, the total career of treated horses (active plus inactive quarters) spanned more quarters than the career of untreated horses (P = 0.01). Despite this, the time from the date of the first race after surgery to the date of the last recorded race for each individual horse was not significantly different between treated and untreated horses (P = 0.96), indicating that a treated horse's career was longer due to the postoperative period of inactivity; however, once they resumed racing, the duration of racing for treated and untreated horses was the same.
Mean ± SD for earnings (A) and starts (B) per quarter for the first 12 quarters after surgery for 59 Thoroughbred racehorses 2 to 5 years of age that underwent colic surgery and survived to hospital discharge and 90 untreated Thoroughbred racehorses equivalent in class. *Significantly (P < 0.05) different between treated and untreated cohorts. Numbers above error bars are the number of horses included in the analysis.
Citation: Journal of the American Veterinary Medical Association 244, 2; 10.2460/javma.244.2.205
Performance variables for the patients in Table 1 that returned to racing after surgery, compared with those of their untreated cohorts.
| Variable | Treated | Untreated |
|---|---|---|
| Quarters inactive* | 4 (2–5) | 1 (0–3) |
| Quarters racing | 6 (3–8) | 5 (3–9) |
| Total career quarters* | 9 (7–13) | 6.5 (4–12) |
| First to last race after surgery (mo) | 17.5 (12–26) | 17.5 (8–35) |
| Total races after surgery | 13 (7–23) | 16 (6–28) |
| Total earnings after surgery (× $1,000) | 35.39 (16.60–96.95) | 30.82 (9.73–83.41) |
| Earnings per active quarter (× $1,000) | 8.50 (2.60–16.02) | 4.12 (2.25–10.11) |
| Earnings per start* (× $1,000) | 2.71 (0.94–8.79) | 1.65 (0.80–3.09) |
Data are shown as median (IQR).
Value is significantly (P < 0.05) different between treated and untreated horses.
Survival curves for total number of races after surgery, time from the first postsurgery race to the last race recorded, and total career duration were plotted (Figure 2). In the case of races after surgery, the cutoff for the survival analysis was set at 70 races. Only 2 horses (one treated and the other untreated) raced > 70 times; the maximum number of races recorded was 105 (a treated horse). Median number of races was 13 races (95% CI, 10 to 18 races) for treated and 16 races (95% CI, 14 to 19 races) for untreated horses; the survivor functions were not significantly (P = 0.43) different. Cutoffs of 72 months from the date of the first to the last recorded race and 30 quarters for total career duration were used in the analysis. Only 3 horses (1 treated and 2 untreated) had > 72 months between the first and last races; the maximum was 95.5 months for 1 untreated horse, and only 2 horses (one treated and the other untreated) had career durations > 30 quarters. The median number of races and survivor function for months between the first and last postsurgery race dates were not significantly different between treated (17.5 races; 95% CI, 15 to 22.5 races) and untreated (17 races; 95% CI, 13 to 22 races; P = 0.96) cohorts. However, there was a significant (P = 0.02) difference in median career duration and the survivor functions for treated (9 quarters (95% CI, 8 to 11 quarters) and untreated (6 quarters; 95% CI, 5 to 8 quarters). The finding that the 25% quartile was 7 quarters (95% CI, 5 to 8 quarters) for treated horses and only 4 quarters (95% CI, 2 to 5 quarters) for untreated horses, whereas the 75% quartile was 13 quarters (95% CI, 10 to 16 quarters) and 12 quarters (95% CI, 9 to 15 quarters) for treated and untreated horses, further supports the notion that enforced early period of inactivity after colic surgery is responsible for difference in total career quarters observed between treated and untreated cohorts.
Total duration of racing career for the horses in Table 1 after the date of surgery, as determined by use of Kaplan-Meier product limit estimates: for total races (A), months between the dates of the first and last recorded races after surgery (B), and total career quarters (active and inactive; C) for treated (solid line) versus untreated (dashed line) horses.
Citation: Journal of the American Veterinary Medical Association 244, 2; 10.2460/javma.244.2.205
Discussion
Results of the present study of 2- to 5-year-old Thoroughbred racehorses indicated that horses that underwent colic surgery were able to successfully return to racing, with no differences in performance variables, compared with their untreated cohorts. The only prior studies4,5 to investigate the potential for return to athletic activity in adult horses following colic surgery evaluated a wide range of disciplines, including show-jumping, dressage, pleasure riding, Standardbred racing, and 3-day eventing. Overall, sports activities were resumed by 85% to 90% of surviving horses, and by 6 months after surgery, 70% of horses had resumed full activity. We were interested in primarily evaluating return to athletic function in Thoroughbred racehorses for several reasons, including use of standardized performance variables obtained via the race records, and it is our opinion that the stigma associated with poor outcome following colic surgery is primarily encountered in the racing industry. The results of this study show that racing Thoroughbreds that undergo abdominal surgery for colic are able to successfully return to racing, with no significant differences in performance variables, compared with their cohorts. Importantly, by the third quarter following the date of surgery, there is little difference in the number of starts and no significant difference in earnings per quarter between treated horses and their untreated cohorts. In addition to the actual number of races, regression analyses showed that the rate at which racing occurred per unit time (quarter or months) was not significantly different between groups. Although the duration of racing career, measured in terms of all inactive and active quarters, was prolonged due to enforced inactivity for those horses that underwent colic surgery, once treated horses resumed racing, the number of months from the first to last recorded races was essentially identical to that of controls.
Horses that did not return to racing after surgery had a significantly higher admission heart rate and blood lactate concentration. Heart rate and blood lactate concentration are a reflection of the patient's cardiovascular status and severity of disease. A high heart rate has been consistently associated with both development of postoperative complications9 and lower rate of survival to hospital discharge.10 Additionally, low total plasma protein11 and high blood lactate concentrations12 have been shown to be poor prognostic indicators in horses with strangulating small intestinal and large intestinal lesions, respectively. Although it has not necessarily been documented per se and is likely to be highly dependent on the cause of colic, it is probable that the duration of colic can impact the cardiovascular status at hospital admission. A longer duration of colic prior to initial examination may result in greater cardiovascular instability in horses with colic because of a greater degree of dehydration and has been associated with increased incidence of postoperative shock9,13 and decreased short-term survival rate.10 Whereas the impact of duration of colic prior to surgery on return to racing was not evaluated specifically in this study, this finding suggest that early referral and surgical intervention prior to the horse developing signs of shock may improve return to racing in horses treated surgically for colic. As a guideline for the decision to refer a horse with colic to a surgical facility, it has been previously shown that horses with signs of constant pain or pain that returns after administration of ≥ 1 dose of analgesic are significantly more likely to require surgery, compared to those that require either 1 dose of analgesic or none at all.14 Only horses that survived to hospital discharge were included in this study; therefore, horses thought to have a poor prognosis during surgery or developing life-threatening postoperative complications were euthanized and not considered.
Significantly fewer horses with nonstrangulating lesions (primarily large intestinal nonstrangulating obstructions) than strangulating lesions returned to racing. The reason for and clinical relevance of this finding are unknown and warrant further investigation. Identifying reasons for failure to return to racing would have been helpful and should be considered in future studies. Colonic dysfunction is a possible reason for this finding. It has been shown previously that horses with large colon disorders have a significant reduction in ICC density within the pelvic flexure, compared with control horses, and that horses with strangulating small intestinal lesions do not have a similar decrease in these cells.15 The ICC are gastrointestinal pacemaker cells, which are responsible for initiating slow-wave activity in the gastrointestinal tract, which is in turn integral in coordination of gastrointestinal motility.15 Histologic and immunohistochemical evaluation of the myenteric plexuses and ICC in horses with colic has revealed a significant increase in neuronal chromatolysis and necrosis throughout the gastrointestinal tract.16 Therefore, horses with large intestinal lesions may be more prone to chronic or recurrent colic, which could potentially impact the horse's ability to be a functional athlete. Long-term follow-up in this study was obtained purely via race records, so it was impossible to determine whether horses had further episodes of colic after surgery.
Diet has been associated with colic, particularly simple colon obstruction and distention.17 Compared with horses fed grass, horses fed a concentrate diet and horses with a simple colon obstruction and distention have a higher colonic lactic acid concentration corresponding to a greater population abundance of Grampositive lactic acid–producing bacteria, which favor fermentation of hydrolysable carbohydrates.18 These horses also have a lower proportion of obligate fibrolytic acid–intolerant bacteria, which would be expected to result in a reduction of fiber digestion.18 Racehorses are fed a high-concentrate diet to meet the energy demands associated with athletic performance. It could be postulated that intolerance of some racehorses to the high concentrated diet with the potential for associated colonic dysfunction could be the reason for a lower return to racing for horses with large intestinal nonstrangulating obstructions. Further investigation into the impact of large intestinal disease on athletic performance and nutritional modifications that optimize colonic function while meeting the energy demands of athletic horses is needed.
The main limitations of the present study are the impact of owner-trainer decision with regard to whether to return the horse to racing and the lack of follow-up via telephone conversation to determine the reason that horses did not return to racing. Because of the long period over which the study was performed and the frequency with which racehorses change owners or trainers, it would have been challenging to obtain this information. Additionally, we have to acknowledge the potential for selection bias in that racehorses for which owners and trainers approve the decision for colic surgery may be those with the best records or pedigree and thus it may be more likely that the stakeholders will ensure that these horses return to training and racing. In the present study, we had no way to assess the probability of selection bias. The finding that the career of previously raced horses that did not return to racing after surgery was not systematically different from those that did resume racing may mitigate against a role for selection bias, but without knowing the reasons why horses failed to race following surgery, it is difficult to draw firm conclusions.
Overall, the return to athletic function appears to be favorable, with 76% of horses that raced prior to colic surgery and surviving to hospital discharge racing at least once after surgery. Early referral and surgical intervention may improve the overall outcome and ability of a horse to return to athletic function. Importantly, horses with a large intestinal nonstrangulating lesion diagnosed at surgery may have a reduced rate of return to athletic function following surgical correction. Horses that returned to racing were able to pursue a successful racing career.
ABBREVIATIONS
| CI | Confidence interval |
| ICC | Interstitial cells of Cajal |
| IQR | Interquartile range |
Equineline, The Jockey Club Information Systems, Lexington, Ky.
Stata IC, version 11.2, StataCorp, College Station, Tex.
References
1. Traub-Dargatz JL, Kopral CA, Hillberg Seitzinger A, et al. Estimate of the national incidence of and operation-level risk factors for colic among horses in the United States, spring 1998 to spring 1999. J Am Vet Med Assoc 2001; 219:67–71.
2. Hillyer MH, Taylor FGR, French NP. A cross-sectional study of colic in horses on Thoroughbred training premises in the British Isles in 1997. Equine Vet J 2001; 33:380–385.
3. Tinker MK, White NA, Lessard P, et al. Prospective study of equine colic risk factors. Equine Vet J 1997; 29:454–458.
4. Launois T, Heiles PH, Desbrosse F, et al. Sports activity after colic surgery: post-operative outcome of 100 procedures, in Proceedings. 9th Int Cong World Equine Vet Assoc 2006;279–281.
5. Christophersen MT, Tnibar A, Pihl TH, et al. Sporting activity following colic surgery in horses: a retrospective study. Equine Vet J Suppl 2011;(40):3–6.
6. Santschi EM, Slone DE, Embertson RM, et al. Colic surgery in 206 juvenile Thoroughbreds: survival and racing results. Equine Vet J Suppl 2000; 32:32–36.
7. Freeman KD, Southwood LL, Lane J, et al. Post operative infection, pyrexia and perioperative antimicrobial drug use in surgical colic patients. Equine Vet J 2012; 44:476–481.
8. Aceto HW, Parente EJ. Using quarterly earnings to assess racing performance in 70 Thoroughbreds after modified laryngoplasty for treatment of recurrent laryngeal neuropathy. Vet Surg 2012; 41:689–695.
9. French NP, Smith JE, Edwards GB, et al. Equine surgical colic: risk factors for postoperative complications. Equine Vet J 2002; 34:444–449.
10. Mair TS, Smith LJ. Survival and complication rates in 300 horses undergoing surgical treatment of colic. Part 1: short-term survival following a single laparotomy. Equine Vet J 2005; 37:296–302.
11. Proudman CJ, Edwards GB, Barnes J, et al. Factors affecting long-term survival of horses recovering from surgery of the small intestine. Equine Vet J 2005; 37:360–365.
12. Johnston K, Holcombe SJ, Hauptman JG. Plasma lactate as a predictor of colonic viability and survival after 360° volvulus of the ascending colon in horses. Vet Surg 2007; 36:563–567.
13. Mair TS, Smith LJ. Survival and complications in 300 horses undergoing surgical treatment of colic. Part 2: short-term complications. Equine Vet J 2005; 37:303–309.
14. White NA, Elward A, Moga KS et al. Use of Web-based data collection to evaluate analgesic administration and the decision for surgery in horses with colic. Equine Vet J 2005; 37:347–350.
15. Fintl C, Hudson NPH, Mayhew IG, et al. Interstitial cells of Cajal (ICC) in equine colic: an immunohistochemical study of horses with obstructive disorders of the small and large intestines. Equine Vet J 2004; 36:474–479.
16. Pavone S, Gialletti R, Pepe M, et al. Histological and immunohistochemical studies of changes in myenteric plexuses and in interstitial cells of Cajal associated with equine colic. Res Vet Sci 2012; 93:350–359.
17. Archer DC, Proudman CJ. Epidemiological clues to preventing colic. Vet J 2006; 172:29–39.
18. Daly K, Proudman CJ, Duncan SH, et al. Alterations in microbiota and fermentation products in equine large intestine in response to dietary variation and intestinal disease. Br J Nutr 2012; 107:989–995.
