Appraisal of the Current Evidence for Diagnostic Tests That Differentiate Medical and Surgical Colic

Abstract

Reasons for performing study Early identification of surgical cases of colic is critical to improving outcome and welfare. There have been a number of studies on diagnostic tests for colic, evaluating a range of tests to differentiate between medical and surgical cases. Objectives To systematically review and appraise the evidence on diagnostic tests for identifying surgical colic. Study design Systematic review. Methods The primary literature search was conducted in CAB Abstracts (1910–2014), WEB of Science (1950–2014) and MEDLINE (1946–2014) using search terms relating to equine colic. Publications were assessed against inclusion and exclusion criteria, and then reviewed using the QUADAS quality assessment tool. Results The primary search identified 5508 publications relating to equine colic; 976 related to diagnostic tests, 29 met the inclusion criteria and were assessed using the QUADAS tool. Of these, 16 papers reported on peritoneal fluid parameters, 21 on blood parameters, 10 on blood and peritoneal parameters, 3 on physical and blood parameters and one study on urine parameters. A range of different parameters were evaluated, with the majority of tests only being evaluated in single studies. None of the studies met QUADAS criteria 1 (representative spectrum of patients), 7/29 studies met criteria 2 (description of selection criteria) and 10/29 studies met criteria 9 (description of reference standard). Conclusions There are currently no published studies which use appropriate methodology to assess the accuracy of a diagnostic test in differentiating medical and surgical colic. This made direct application of the QUADAS tool challenging; predominantly due to the lack of a pre mortem ‘gold standard’ reference diagnostic test for colic. Very few studies enrolled a randomised selection of patients and there was also a propensity to case–control study design, both increasing the risk of bias and under/overestimating diagnostic accuracy. Ethical animal research: Not applicable. Sources of funding: Laila Curtis' studentship is funded by the School of Veterinary Medicine and Science, University of Nottingham. Tom Cullen is a Junior Clinical Training Scholar funded by the School of Veterinary Medicine and Science, University of Nottingham and Oakham Veterinary Hospital. Competing interests: None declared.