Acurácia da termografia infravermelha na detecção da tendinopatia ou risco de lesão tendínea: revisão sistemática com metanálise

Abstract

Introduction: The sympathetic nervous system (SNS) plays a role as a component of tendinopathy, increasing its innervation and intervening in local blood flow. This alteration can be directly portrayed by the skin temperature in the pathological tendon area. Due to its ability to detect variations in skin temperature, infrared thermography (IRT) can be a tool used to evaluate these alterations caused by tendinopathy processes. Objective: To systematically review the evidence on the use of IRT in the diagnosis of tendinopathy and in

the detection of risk of tendon injury. Methods: This is a systematic review with meta- analysis based on the PRISMA 2020 methodological guidelines, carried out from January to

December 2021 in the PubMed, PEDro, SciELO, Embase, CENTRAL and CINAHL databases, using descriptors for tendon injury and IRT. The screening, selection and information collection steps in the primary studies were performed concurrently by two independent researchers. The included studies were assessed for risk of bias using the QUADAS-2 tool. Evidence synthesis was performed quantitatively, through the analysis of sensitivity, specificity, summary ROC curve (SROC) and diagnostic odds ratio (DOR), using the Meta-DiSc 1.4 software. Results: Eight studies were selected for the qualitative synthesis, while for the meta-analysis only seven studies were included, which showed that the IRT has a sensitivity of 72% (CI 95%: 67% - 77%), specificity of 95% (CI 95 %: 92% - 98%) DOR of 75.94 (CI 95%: 12.04 - 479.14), with an accuracy (SROC) of 97%. When performing a sensitivity analysis, this review suggests that for lateral epicondylitis, IRT has a detection capacity with a sensitivity of 93% (CI 95%: 86% - 97%), specificity of 97% (CI 95%: 93% - 99 %) and DOR of 221.38 (CI 95%: 21.09 – 2321.13). As for tendinopathies of the shoulder region, IRT also demonstrated accuracy with a sensitivity of 63% (CI 95%: 56% - 70%), 100% specificity (CI 95%: 91% - 100%) and DOR of 60.71 (CI 95%: 8.10 - 454.87). Conclusion: The IRT constitutes an accurate tool to detect tendon injuries, with a high power of specificity, both in the evaluation of lateral epicondylitis and for tendonitis in the shoulder region.