Ihler CF, Larsen J and Skjerve E: Evaluation of clinical and laboratory variables as prognostic indicators in hospitalised gastrointestinal colic horses. Acta vet. scand. 2004, 45, 109-118. -The present prospective study included 106 horses referred to the Department of Large Animal Sciences, The Norwegian School of Veterinary Science, as non-responders to the initial colic treatment in general practise. In 14 of these cases a required surgical treatment was not performed due to economical or other reasons and were excluded from the study. Clinical and laboratory data were obtained at the arrival in the hospital. The outcome for all analyses was survival/ non-survival. A multivariable logistic regression was performed. The analyses were used in medically (46 horses) and surgically treated cases (46 horses) separately. The same analyses were also run for all 92 horses in a simulated "field" situation, where only clinical variables and D-dimer values were included. The fraction of survivors was 78% in the medical and 48% in the surgical cases. In total 63% of the horses survived. In the final multivariable logistic regression model packed cell volume (PCV) was the only important predictor for medically treated cases, and heart rate and presence of hyperaemic or cyanotic mucous membranes were the predictors in the surgically treated cases as well as in the simulated "field" situation. In conclusion, traditional variables as heart rate, mucous membranes and PCV were the important predictors for the outcome in hospitalised colic cases.