Prevalence and predictors of nocturia in obstructive sleep apnea-hypopnea syndrome--a retrospective study.

Abstract

OBJECTIVE: To determine the prevalence and predictors of pathological nocturia (PN) in patients with OSAHS. METHODS: Retrospective review of clinical and polysomnographic data obtained from patients evaluated at the Pulmonary Sleep Evaluation Center of the University of Pittsburgh Medical Center between November 1998 and September 1999. PN was defined as two or more urination events per night. OSAHS was defined as Apnea-Hypopnea Index (AHI) > or = 5. Group t-tests and Chi-square test were used to examine differences in subjects with and without PN. A multivariate approach utilizing logistic regression was performed to examine the relationships between polysomnographic variables [Arousal Index (ARI), AHI, Apnea-Hypopnea Time/Total Sleep Time x 100(%) (AHT/TST), % Total Sleep Time (TST)<90% Oxyhemoglobin Saturation (SPO2), nadir SPO2, Desaturation Event Frequency (average number of desaturations > 4% per hour sleep) (DEF)] and the presence of PN, while controlling for medical comorbidities. Data are expressed as mean + SD unless otherwise specified. Statistical significance was assessed at p < 0.05. RESULTS: Study population: n = 138 (50 females, 88 males), age: 49.7 +/- 12.3 years, BMI [Body Mass Index (kg/m2)]: 37.7 +/- 11.3. The overall prevalence of PN was 47.8%. The prevalence of PN was greater in females (60% vs. 40.9% in males). OSAHS patients with PN had a higher prevalence of peripheral edema, higher BMI, a greater %TST < 90% SPO2, lower nadir SPO2 and greater DEF. The logistic regression model indicated that age, ARI, AHI, AHT/TST, %TST < 90%, DEF were significant predictors of PN independent of BMI, neck circumference and medical comorbidities. CONCLUSIONS: Our data confirm that PN is common in OSAHS patients. The strongest predictors are age and selected polysomnographic variables reflecting OSAHS severity.