The current literature suggests a relationship between obstructive sleep apnea (OSA) severity and hematocrit. However, the degree that OSA contributes to clinically significant erythrocytosis is uncertain. The aim of this study is to evaluate this association in a large study sample controlling for multiple confounders.

We evaluated consecutive subjects with suspected untreated OSA using multivariate analysis to test the associations between apnea-hypopnea index (AHI) and hematocrit. Subjects were evaluated with sleep studies, comprehensive sleep questionnaires, and detailed electronic medical record reviews to document their medical comorbidities, and demographic and laboratory information.

1604 consecutive veterans (age 57.6 ± 13.4 years, 92% male) were included in the analysis with 77.4% diagnosed with OSA. However, few included subjects (1.6%) had clinical erythrocytosis. OSA severity defined by AHI was not associated with hematocrit or clinically significant erythrocytosis. Rather, awake oxygen saturation (-0.17 points, p < 0.001) and mean nocturnal oxygen saturation (-0.08 points, p = 0.04) were inversely proportional to hematocrit (per standardized Z-score). Other factors including active tobacco, increased alcohol ingestion and exogenous testosterone therapy were associated with higher hematocrit. Although AHI was not predictive of erythrocytosis, having severe OSA was predictive of nocturnal hypoxemia (adjusted OR 7.4, p < 0.001).

Hematocrit levels and presence of erythrocytosis appear not associated with OSA severity, but rather with hypoxemia as measured by awake and to a lesser extent mean nocturnal oxygen saturation. Nocturnal oximetry may provide diagnostic utility in the evaluation of unexplained secondary polycythemia and polysomongraphy may be warranted in those with unexplained nocturnal hypoxemia and erythrocytosis.