A systematic review of equine infrared thermography : diagnostic applications for limbs and hoofs

Abstract

Equine infrared thermography (IT) is a useful diagnostic tool for identifying changes in peripheral blood flow, and tracking the effectiveness of treatment or healing process. Despite its growing popularity, literature on this imaging modality remains fragmented with many studies lacking rigorous methodologies. This systematic review aimed to evaluate the existing evidence on the clinical efficacy and diagnostic accuracy of IT, with a specific focus on its application to the limbs and hoofs. Research papers published between 1 January, 2000, and 1 August, 2024 were sourced from Agris, CABI, Google Scholar, PubMed, ScienceDirect, and Web of Science. One author screened the results, extracted and analysed the data using EndNote v21 and Covidence. PRISMA reporting guidelines were followed. Twenty-eight peer-reviewed studies published over the past 24 years, comprising 25 randomised control trials and 3 non-randomised control trials with a total study population of adult horses and ponies (n = 717), met the inclusion criteria. These studies were grouped into four categories: non-exercise, exercised, booted, and induced pathology. Statistically significant results were found in 32 of the 66 studies. The basic meta-analysis presented no statistical significance in the overall effect size for the non-exercise group, while the exercised, booted, and induced pathology groups demonstrated moderate statistical significance. The forest plots presented a wide confidence interval (CI) in one study from both the non-exercise [2.46, 6.80] and booted [2.94, 7.74] groups, indicating greater uncertainty in the study’s effect (Cohen’s d). The Galbraith plots pointed to publication bias and between-study heterogeneity across three of the groups, with the induced pathology group showing no such issues. The funnel plots implied moderate to considerable publication bias in the non-exercise and exercised groups. The Contour-enhanced funnel plots revealed varying degrees of publication bias and between-study heterogeneity in all four groups. Although this systematic review highlights the potential of equine IT, variations in study methodologies, small sample sizes, susceptibility to artefacts affecting image accuracy, and a lack of standardization in its application underscore the need for more rigorous and uniform research.