Predictors of end-organ damage in patients with Obstructive Sleep Apnea according to the Baveno classification

Abstract

Introduction: Obstructive sleep apnea (OSA) is a heterogeneous disease. OSA severity is based on the apnea-hypopnea index (AHI), but its relevance is being questioned. Baveno’s classification is a multicomponent grading system for OSA severity, based on symptoms, comorbidities, and end-organ damage (EOD). We aimed to identify EOD predictors in OSA patients.

Methods: We retrospectively analyzed patients with OSA diagnosis. Clinical and polysomnographic (PSG) data (AHI, mean peripheral oxygen saturation (SpO2), time with SpO2<90% (T90), and oxygen desaturation index (ODI)) were reviewed. Patients were classified into Baveno groups; those with EOD were allocated in groups C and D. A multivariate logistic regression model was used to determine independent EOD predictors.

Results: Of 399 patients, 70.4% were male, with a mean age of 57.0±11.3 years; 87% had ≥1 cardiovascular risk factor. The patient distribution across Baveno groups was: 17.5% in group A, 54.6% in group B, 7.0% in group C, and 20.8% in group D. Patients with EOD were significantly older (p<.001), had a higher frequency of hypertension (HT) (p<.001), diabetes (p<.001) and dyslipidemia (p<.001). EOD was greater in cases with higher AHI (p=0.002), T90 (p=0.008), ODI (p=0.003), and lower mean SpO2 (p=0.033). The multivariate regression model showed that older age (p<.001), HT (p=0.001), diabetes (p=0.004) and dyslipidemia (p=0.001) were EOD independent risk factors. PSG results didn’t have a significant impact in predicting EOD.

Conclusion: This study highlighted the relevance of HT, diabetes, and dyslipidemia as independent predictors of EOD in OSA patients. PSG results proved to be unreliable EOD predictors.