Association of Metabolic Score for Insulin Resistance With Resistant Hypertension and Hypertension in Obstructive Sleep Apnea

INTRODUCTION: Obstructive sleep apnea (OSA) is common in the presence of hypertension and refractory hypertension (RH); however, the pathophysiological interrelationship of such conditions is yet to be well explained with insulin resistance (IR) as a key mediator. The objective of the present investigation was (1) to evaluate the association of the metabolic score of insulin resistance (METS-IR) with hypertension and RH in patients with OSA and (2) to understand whether this association varies according to the severity of OSA. METHODS: It is a retrospective cohort, which included 680 adults with OSA diagnosis in 2020-2022. The participants were stratified into quartiles in accordance with their METS-IR. Polysomnographic evaluation and metabolic profiling including body mass index (BMI), fasting blood glucose, triglycerides, and high-density lipoprotein cholesterol (HDL-C) were conducted. Multivariate logistic regression analyses were used to determine the relationship between METS-IR and hypertension/RH, adjusting for sex, lifestyle, and variables related to OSA. Subgroup analyses were also done to compare associations between severe and non-severe cases of OSA cohorts. Additionally, receiver operating characteristic (ROC) curve analysis was employed to compare the discriminative performance of METS-IR, body mass index (BMI), and the triglyceride-to-high-density lipoprotein cholesterol (TG/HDL) ratio for both hypertension and RH. Post hoc power analyses were conducted across all groups (the total cohort, the severe OSA subgroup, and the non-severe OSA subgroup) to evaluate whether the analyses were adequately powered (> 80%) to detect the observed effect sizes. RESULTS: Elevated METS-IR levels were significantly associated with a higher prevalence of hypertension and RH. In patients with severe OSA, multivariate analysis revealed a robust, linear dose-response relationship between METS-IR and the risks of both hypertension and RH (p-trend < 0.01); conversely, in the non-severe OSA group, these associations were attenuated and did not exhibit a significant linear trend after full adjustment. ROC analysis revealed that METS-IR achieved the highest discriminative accuracy for both outcomes. For hypertension, the AUC of METS-IR (0.745) was significantly higher than that of BMI (0.729) and TG/HDL (0.636) (all p < 0.05). Similarly, for RH, METS-IR demonstrated superior discriminative ability (0.754) compared to BMI (0.746) and TG/HDL (0.610). CONCLUSIONS: METS-IR is significantly associated with hypertension and RH in patients with severe OSA, independent of BMI, although this association is attenuated in patients with non-severe OSA. These adequately powered results support the potential utility of METS-IR as a simple metabolic marker to identify high-risk phenotypes in clinical practice to manage severe OSA. Furthermore, METS-IR is a more robust marker of hypertension and RH than BMI or TG/HDL alone, suggesting that the integration of adiposity and metabolic parameters provides superior risk stratification in OSA patients.