Colic is a common problem in horses. Establishing a plan for treatment and predicting the possible outcome of a case of colic is important to both the attending veterinarian and the anxious owner of the horse. Abdominal fluid analysis is commonly used in the decision making process, but it does not provide definitive guidance in determining appropriate management of horses with colic.

A recent study has been conducted to assess the clinical utility of abdominal fluid analysis in predicting outcome (survival vs nonsurvival), lesion type (strangulating vs nonstrangulating), and whether medical or surgical treatment is indicated for horses with colic. (GO Freden, PJ Provost and WM Rand Reliability of using results of abdominal fluid analysis to determine treatment and predict lesion type and outcome for horses with colic: 218 cases (1991-1994) JAVMA 1998, Vol. 213, No. 7, pg. 1012).

Medical records for 218 horses were studied, the age of the horses ranged from 1 to 35 years, there were 115 geldings, 83 mares and 20 stallions and twenty different breeds were represented. The researchers studied the records of these horses and they were classified on the basis of outcome, type of treatment, lesion type and site of lesion. Abdominal fluid samples had been collected at the time of hospital admission, parameters of interest were: color, turbidity, specific gravity, red blood cell count and nucleated cell counts.

Results indicate that abdominal fluid color and specific gravity had a high positive prediction value for lesion type, and that patient age and abdominal fluid color had a high positive prediction value for patient outcome. Horses with darker fluid were more likely to have a poorer long term outcome. An evaluation of abdominal fluid color and specific gravity could provide clinicians with useful information regarding patient outcome and lesion type. The study also found that abdominal fluid nucleated cell count was not useful in management of horses with colic. The researchers advise that abdominal fluid alone cannot be reliably used to predict treatment type, lesion type, or outcome but it can provide important clinically relevant information