Distribution of the oxygen desaturation index and polygraphy-derived respiratory parameters in self-reported apparently healthy adults: A cross-sectional study

Background Nocturnal desaturation in self-reported apparently healthy adults is poorly described in Latin America. We describe the oxygen desaturation index (ODI) and other respiratory parameters obtained from type III home respiratory polygraphy in these adults in Lima, Peru. Methods In this cross-sectional study, 150 adults aged 20–50 years with BMI < 30 kg/m² and no self-reported respiratory, cardiovascular or endocrine disease were recruited through non-probability sampling from a single specialized center; 133 were analyzed. Recordings were scored manually (AASM criteria). Without electroencephalography, the ODI (≥3% desaturations) and respiratory event index (REI) were calculated per hour of valid recording time. Ordinary least squares (OLS) regression models with HC3 robust standard errors were fitted for ln(REI + 1) and ln(ODI + 1), adjusting for age and BMI. Results Among 133 participants (mean age 35.5 ± 9.3 years; 54.1% men), 52 (39.1%) had an REI of ≥5 events/h of valid recording time. Median ODI was 7.4 vs 1.3 events/h and median REI 9.6 vs 2.1 events/h in men vs women (both p < 0.001). In adjusted models, male sex was associated with higher REI + 1 (Exp(β) = 2.08, 95% CI 1.54–2.80) and ODI + 1 (Exp(β) = 1.82, 95% CI 1.34–2.48; both p < 0.001), multiplicative contrasts, not ratios of the original indices. Age was positively associated with both outcomes. The ODI and REI, derived from the same recordings, were strongly correlated (r = 0.937, p < 0.001). Conclusions In this selected sample, REI and ODI were higher in men and older participants; the male-sex association persisted after adjustment. Given non-probability, single-center sampling, self-reported health status and a recording-time denominator, the findings are descriptive and do not estimate population prevalence or diagnostic validity.