Review of Clinically Relevant Factors to Long-Term Survival After Colic Surgery

Practitioners and owners use available information on long-term survival data after colic surgery in the decision-making process when discussing treatment options for horses requiring exploratory celiotomy for acute abdominal diseases. Veterinarians involved in all aspects of managing horses with surgical colic can use the same survival data to provide a more accurate prognosis before and after surgery.

1. Introduction

In the early 1970s, a published review of outcome after colic surgery¹  reported that only 10 of 38 horses were successfully treated by surgery, and most of those survivors had minor intestinal procedures. A later review of cases from the University of California at Davis from 1958 to 1971 reported that 48 of 75 horses with small intestinal strangulation had surgery and that 72.9% died.²  However, the mortality rate had improved to 46.6% in the last 14 years of the review period.²  A report on a substantial number of cases in 1984 described 80% short-term survival after surgery for a variety of small intestinal lesions, which was the most favorable result to that time. A study published in 1989 on 140 horses that had small intestinal resection between 1968 to 1986 reported that 51% of these horses died during hospitalization.³  However, this study differed from others by following up on the survivors over several years and demonstrated that 15 horses that had small intestinal resection and anastomosis, rightly regarded as the most challenging of colic surgeries, survived 3 to 9 years after surgery.³  It also demonstrated a trend toward improved survival rates in horses that had surgery toward the end of the study period compared with the beginning.³ 

During the 1980s, owner attitude toward colic surgery improved and referring veterinarians demonstrated that success in colic surgery was largely dependent on their decision to refer early. Failures in colic surgery were examined, and postoperative ileus and shock seemed to account for most postoperative deaths, followed by salmonellosis/colitis, long bone fracture, adhesions, hemorrhage, laminitis, wound infection, and ischemic muscle damage.⁴ This information formed the basis for current postoperative treatments and procedures to monitor postoperative progress.

Several studies reported excellent short-term results after surgery for different types of colic in the 1990s up to the early part of this century. However, a growing body of information was also emerging on long-term results, probably the product of time required to generate a sufficient database for meaningful study. The University of Liverpool emerged as a leader in the field of abdominal surgery in horses, led by Professor Barrie Edwards, in collaboration with Dr. Chris Proudman and Dr. Debbie Archer. This group has contributed substantial data on long-term survival and complications that represent advances that can improve survival of horses that undergo colic surgery.

Many studies evaluating long-term survival after colic surgery in horses have provided useful information on effects of disease, segment affected, and surgical procedure on outcome. Despite some conflicting information and variations in survival times, there is considerable evidence that survival rates have improved steadily over the years. This information should provide some encouragement to owners and veterinarians in practice about the prospects of long-term survival and acceptable quality of life for horses that have had colic surgery. Also, some recent studies have examined effects of colic surgery on return to sporting activity, which is highly relevant to an owner confronted with the prospects of an expensive surgery on a horse expected to perform at the highest level in a competitive sport.

Long-term data for survival is considerably more relevant than short-term data for an animal that could have a life expectancy of many years after colic surgery, depending on its age at surgery. Unfortunately, long-term data are considerably harder to obtain, largely due to difficulty in locating owners or trainers who have moved or locating horses that have changed ownership over a period of many years. The most commonly used method to examine long-term survival data is the Kaplan-Meier estimator (Fig. 1). This estimates the survival function as the fraction of patients living for a certain amount of time after treatment. The plot of survival function is a series of steps with each downward deflection representing loss(es) through death. An important advantage of the Kaplan-Meier curve is that it accounts for censored data, as occurs if a patient is lost to follow-up from the sample before the final outcome (death) is recorded (eg, sale, owner move, patient still alive at last recording interval). On the plot, small, vertical tick-marks indicate when a patient’s survival time has been censored, and these patients are not counted in the denominator for the subsequent data points. Results can also be compared as median survival times, which is the time when 50% of horses are still alive.
 

2. Case Selection

Data from short-term and long-term studies must be interpreted with full awareness that they deal with horses in which surgery was completed and were allowed to or were able to recover from general anesthesia. Therefore, horses that died or were euthanatized before and during surgery or during recovery are missing from the denominator. One reason for this approach is that those cases that undergo full treatment (surgery and aftercare) are relevant to a study on the effects of that treatment on outcome. The remainder might have been deliberately excluded from treatment because of financial constraints, presence of a concurrent disease, advanced age, or perception of a poor prognosis. Possibly some of these exclusions could be the result of an error in judgment, misdiagnosis, owner misunderstanding, or overly pessimistic perception about prognosis (on the part of owner and/or veterinarian). Those horses that died before they could recover from anesthesia possibly succumbed to severe complications of the disease itself (endotoxemic shock, uncontrollable hemorrhage, ruptured viscus) or they incurred a catastrophic fracture in the recovery stall. Such cases arguably represent examples of failed treatment. Regardless, studies on effects of treatment on long-term survival typically focus on horses that recover fully from surgery and anesthesia, largely because the reasons for exclusion of other cases cannot be readily deduced during retrospective examination of medical records. Nonetheless, any possible errors that underlie such unwitting case selection must be addressed to improve the way we treat horses with colic.  [...]