Development of a Colic Scoring System to Predict Outcome in Horses

Abstract

Acute abdominal pain in the horse is a common emergency presenting to equine practices. The wide variety of etiologies makes prognosticating survival a challenge. A retrospective, multi-institutional clinical study was performed to determine clinical parameters associated with survival of horses with colic, and to use them to develop a colic survival scoring system. The scoring system was then validated using clinical data in the prospective portion of the study. Medical records from 67 horses presenting for acute abdominal pain were evaluated to develop the colic assessment score. Twenty eight variables were compared between survivors and non-survivors and entered into logistic regression models for survival. Of these, six variables were included in the colic assessment score. A total colic assessment score range was from 0 to 12, with the highest score representing the lowest probability of survival. The optimal cutoff value to predict survival was seven resulting in an 86% sensitivity and 64% specificity with a positive predictive value of 88% and a negative predictive value of 57%. Data from 95 horses presenting for abdominal pain to two equine hospitals was then collected prospectively to validate the colic assessment score. Horses from the prospective portion of the study that received a score >7 were classified as predicted to die and those with a score ≤7 were predicted to survive. The classification was compared to the actual outcome, of which the sensitivity, specificity, positive and negative predictive values of the colic assessment score were 84, 62, 88, and 52%, respectively.

Clinical predictive models in equine medicine: A systematic review

“…Of the 37 colic models (excluding 2 independent model validations, Table S1), 14 20 had survival to hospital discharge as the primary study outcome 6,7,15–27 . The top 10 most commonly included predictors of short‐term survival are shown in Figure 2, with the top three being heart rate, packed cell volume (PCV), and mucous membrane characteristics (colour and capillary refill time).…”

“…Of the 11 neonatal illness models (excluding 7 independent validations, Table S2), 35–38 5 had short‐term survival outcomes, namely survival to the time of hospital discharge or survival to 10 days post‐partum 27,39–42 . Foals included in these models ranged in age from less than 4 to less than 14 days at the time of hospital admission.…”

“…Of the 37 colic models (excluding 2 independent model validations, Table S1), 14 20 had survival to hospital discharge as the primary study outcome. 6,7,[15][16][17][18][19][20][21][22][23][24][25][26][27] The top 10 most commonly included predictors of short-term survival are shown in Figure 2, with the top three being heart rate, packed cell volume (PCV), and mucous membrane characteristics (colour and capillary refill time). Four models (4/37, 11%) predicted the need for exploratory laparotomy; rectal palpation findings and peritoneal fluid colour were included as predictors in 3 of 4 of these models.…”

“…Of the 11 neonatal illness models (excluding 7 independent validations, Table S2), [35][36][37][38] 5 had short-term survival outcomes, namely survival to the time of hospital discharge or survival to 10 days post-partum. 27,[39][40][41][42] Foals included in these models ranged in age from less than 4 to less than 14 days at the time of hospital admission. Predictors included in multiple models were absolute white blood cell or neutrophil counts (3/6, 50%), serum levels of IgG (2/6, 33%), and anion gap findings (2/6, 33%).…”

Letter to the editor: A tool for calculating VetCOT score

“…Increasingly, predictive models are finding their way into veterinary medicine. [2][3][4] In both human and veterinary medicine, however, the development of new clinical predictive models consistently outpaces the implementation of existing predictive models in the clinic. 5 The reasons for this are manyfold.…”