Equine Colic: VI. Prognosis and Prevention

1. Introduction

Predicting survival in horses with colic is challenging. The prevalence of survival when considering all types of colic is very high, making prediction of non-survival difficult in individual cases. For the most part, the ability to predict the possibility of death for an episode of colic is needed when horses are unresponsive to initial treatment and there is a need to determine the chances that a treatment, such as surgery, will be successful when consideration of horse value and the costs of treatment are important.

One method of predicting survival is to apply previous knowledge of outcome for specific diseases. Case fatality for specific diseases varies from few deaths in cases of simple colic to 75% or more in some forms of strangulated intestine [1]. Case fatality of diseases requiring surgery is related to disease duration prior to surgical correction. Therefore, fatality rates vary between hospitals depending on the time inherent in recognition of the problem and travel time required to reach a surgical facility. Simple obstructions of the large colon such as displacements and impactions have a low case fatality rate (<10%) [1,2], whereas simple obstructions or impactions of the small intestine are associated with somewhat higher case fatality (≈30%) [3]. Impactions of the cecum have a higher case fatality rate than impactions of the large colon, but, overall, the fatality rate for impaction colic is relatively low (≈15%) [4].

Initial reports of survival of horses with strangulating lesions from 1970 - 1980 were bleak. Small intestinal volvulus, incarcerations in mesenteric rents, and incarcerations within the epiploic foramen had the highest fatality rates (75 - 95%), followed by volvulus of the large colon (77%) [1]. Strangulating lipomas still have a long-term fatality rate of 62% [5]. In recent years, reports of fatality rates have decreased with the change subjectively due to early referral, improved surgical experience, better anesthesia [6], and application of critical care during the postoperative period. A recent report states short-term case fatality for horses with small intestinal strangulation has decreased from 75% in 1983 to 25% in 2000 [7] and to 14% for 180 - 360° small intestinal volvulus [8]. Similarly, fatality from large colon volvulus reported as high as 72% is now reported as low as 10 - 20% in some hospitals [9].

Long-term follow-up has not been investigated for many intestinal diseases. The case fatality rate for horses after discharge from the hospital is highest during the first 10 days after discharge, with most deaths occurring within the first 2 months [10]. Only horses with small intestine incarceration in the epiploic foramen had decreased survival more than 2 months after surgery. After 600 days, the survival rate was 75%, an improvement compared to previous studies. Furthermore, most horses that survive are able to return to their original use (>90%) [11].

Though it has been reported that clinicians can reliably predict the outcome after surgery for an acute abdomen, the accuracy suffers from recent experience of the clinician and the complications that arise in individual cases [12]. To try and improve on prognosticating individual cases of colic, several investigations have attempted to use single values, formulas or scoring systems to predict the odds of survival [11,13-30]. Parry et al [16]. proved that the cardiovascular values were of the greatest value in predicting survival. Other studies also identified lactate [13], anion gap [15], PCV [21], heart rate [16], mucous membrane color [26], peritoneal protein concentrations [26], and intraluminal presssure [18] as values which were significantly different between survivors and non-survivors. Unfortunately, none of these formulas provide a clear-cut separation between survivors and non-survivors, so these values cannot predict absolutely which horses will live and which will die. Nonetheless, some formulas and scores have increased the accuracy of predicting survival [17].

Figure 1. Table 2A. Model to predict death based on the age, sex, breed, pulse strength, heart rate, treatment medical or surgical, PCV, evidence of trauma, CRT, and prevalence of death from colic in the population. This example shows low odds for death based on a horse needing medical treatment. Eight of 100 horses with these values are predicted to die.

Figure 2. Table 2B. Model to predict death based on the age, sex, breed, pulse strength, heart rate, treatment medical or surgical, PCV, evidence of trauma, CRT, and prevalence of death from colic in the population. This example shows the odds for death based on a disease predicted to require surgery. Seventy-two out of 100 horses with these values are predicted to die.

Figure 3. Table 3. Analysis of bran from a farm, which routinely fed a bran mash to all horses daily. The total digestible nutrients (TDN) of the bran was as high as an equivalent amount of corn and in sufficient quantities could significantly increase the carbohydrate content of the diet. This increase lowers consumption of forage and causes dehydration of ingesta in the colon (TDN-total digestible nutrients; ENE T-total energy; NE-net energy; MCAL/lb-megacalorie per pound).

Parry et al [17]. demonstrated a 93% accurate prediction of survival when combining systolic pressure, blood lactate, blood urea and PCV. Furr et al [26]. used heart rate, lactate concentration, peritoneal protein concentration, and mucous membrane color to determine a colic survival score with 67% sensitivity and 100% specificity ([Table 1]). This score predicted all horses that died, but some horses that died were predicted to live. Orsini et al [21]. developed a predictive model using blood lactate concentration and PCV to achieve 96% accuracy. Though accurate and useful in a hospital setting, all of these formulas suffer from a lack of simple factors, which can be used "horse-side" at the farm or stable when the clinician is trying to give an owner an accurate prognosis.

Even if these formulas could be used in practice during the initial examination for colic, the high prevalence of survival in a normal population of horses prevents these formulas from improving the predictability compared to the clinicians’ experience or tossing a coin. However, once a decision is made to classify a horse as a referral candidate, these prognostic systems can be helpful. With the advent of laptop computers and handheld devices, these formulas can be rapidly utilized to determine the odds that a horse will survive or need surgery. Reeves et al [31]. developed a model to predict the need for surgery and death using readily available clinical values and included the ability to specify the prevalence of death in the population being examined (Table 2). Though no formula is 100% accurate, this system can provide confidence in one’s medical judgment while accounting for the prevalence of surgery and survival in a particular practice.

Unless one is proficient in estimating or calculating prevalence of death and using a validated formula of easily obtained values from the physical examination, the safest way to help owners decide about the chance of survival is to provide a prognosis based on the disease classification. If the disease process can be categorized (simple colic, obstruction, strangulation, enteritis, ulceration, peritonitis), it is usually possible to give a reasonable estimate of the chance for survival. One important factor is the disease duration, which can alter the outcome of any disease depending on the severity of shock. Furthermore, as treatments and critical care have improved, case fatality rates, including long-term outcome, have also improved.

2. Prevention of Colic

Two types of factors should be considered when trying to prevent colic: farm factors and horse factors [1,32]. Farm factors include management, use, feeding, and environment. The associated risks on farms with high rates of colic include poor parasite control, high concentrate levels in the diet, multiple sources of concentrates (including supplements which contain high amounts of soluble carbohydrates), chronic water deficiency, excessive use of NSAIDs, acute changes in hay or grain diet, and horses in training that are confined and fed large amounts of carbohydrate and lesser amounts of roughage. Decreasing these risks with appropriate horse management has decreased the incidence of colic on farms with a higher-than-average annual colic rate.

Based on these known factors, colic prevention should start by making sure that horses have a constant source of fresh water, ensuring that forage makes up at least 60% or more of the diet, and that concentrates (soluble carbohydrates) are fed at the minimal level required to maintain weight and performance. The turn-out and exercise routines should be regular and consistent. Changes in feed should be completed over a 7 - 10 day period. Parasite control must be optimal.

Parasite control starts with fecal egg counts several times during the year or more often when horses are frequently moved on and off farms or stables. Though interval anthelminthic treatment would seem to be sufficient to control common parasites, small strongyles and tapeworms have both been linked to increased colic risk and may not be totally controlled by some programs [33,34]. Control of small strongyle infection on farms was reported to decrease the incidence of colic [33]. Anthelmintic resistance is now a concern, making monitoring of parasite control programs essential.

On farms with a high incidence of colic (>10 colic cases per 100 horses per year), careful monitoring of the daily management and measurement of the energy, protein, and fiber in the diet should be the first steps in assessing the farm for colic risk. An example of the need for this type of feed analysis is the unrealized carbohydrate overload, which can occur from feeding bran. Unbeknownst to many owners or managers, bran can have a high soluble carbohydrate content depending on the milling of the grain. If fed as a laxative in addition to a regular grain diet, the soluble carbohydrate intake in a horse’s diet can be doubled (Table 3).

When considering individual horse colic, internal factors, as well as the diet and environment, are potential contributors. Known factors which affect individual horses include acute confinement due to injury, lack of turn-out, gastric ulceration due to performance, and aerophagia. These factors are often associated with simple colic with no other diagnosis. Horse factors may be impossible to control, although some horses respond to turn-out 24 hrs a day and a total forage diet. This arrangement may not be compatible with certain types of performance, but often this ad lib exercise and feeding system will decrease recurrent colic. Other horse factors may play a role in diseases such as lipomas or incarcerations in internal hernias. However, factors which affect these types of intestinal disease have not been widely investigated.

Identifying horses at high risk may help to determine if there is a generalized colic risk on the farm or if specific individuals are responsible for a majority of colic episodes. Often the problem may be a basic diet or management issue such as lack of fresh water, feeding large amounts of concentrates with insufficient forage intake in the diet, or excessive use of medications such as NSAIDs. These factors will not always be apparent in the history of the farm or the owner’s assessment of management. Therefore, careful independent observations and record keeping are often prerequisites to understanding the colic risk factors on a farm.