HOMA-IR in acromegaly: a systematic review and meta-analysis

Purpose This review is aimed at examining whether the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) is higher in Caucasian, adult, treatment-naïve patients with acromegaly (ACRO) than in the reference population independently of diabetes presence and to evaluate the impact of treatment [surgery and somatostatin analogues (SSAs)] on its assessment. Methods We systematically reviewed in PubMed and Web of Science from July 1985 to December 2019, registered with the code number CRD42020148737. The inclusion criteria comprised studies conducted in Caucasian adult treatment-naïve patients with active ACRO in whom HOMA-IR or basal insulin and glucose were reported. Three reviewers screened eligible publications, extracted the outcomes, and assessed the risk of biases. Results Of 118 originally selected studies, 15 met the inclusion criteria. HOMA-IR was higher in ACRO than the reference population, with mean difference and (95% confidence intervals) of 2.04 (0.65–3.44), even in ACRO patients without diabetes, 1.89 (1.06–2.73). HOMA-IR significantly decreased after treatment with either surgery or SSAs − 2.53 (− 3.24– − 1.81) and − 2.30 (− 3.05– − 1.56); respectively. However, the reduction of HOMA-IR due to SSAs did not improve basal glucose. Conclusion HOMA-IR in treatment-naïve ACRO patients is higher than in the reference population, even in patients without diabetes. This finding, confirms that insulin resistance is an early event in ACRO. Our results also suggest that HOMA-IR is not an adequate tool for assessing insulin resistance in those patients treated with SSAs.