Effect of Benign Prostatic Hypertrophy on Sleep Architecture in Patients With Obstructive Sleep Apnea

Abstract

PURPOSE: Obstructive sleep apnea (OSA) and Benign prostatic hypertrophy (BPH) affect sleep quality, Increase upper airway resistance and cyclic obstruction causing arousals defines OSA. Since BPH is also associated with increased urinary sensation and awakening causing interruption in sleep continuity. It is plausible that BPH with obstructive sleep apnea may worsen sleep quality more than OSA alone. We performed a single center retrospective case control study. METHODS: Patients who underwent polysomnography and found to have OSA during last 3 years were included. Patients who had clinical documentation of BPH constituted cases, while others were classified as controls. Demographics [age, sex, BMI], clinical factors affecting sleep [chronic pain, post-traumatic stress disorder (PTSD), benign prostatic hypertrophy (BPH), asthma], medications affecting sleep [rapid eye movement (REM) suppressants and stimulants] and polysomnographic parameters [total sleep time (TST), sleep efficiency (SE), sleep stages, REM onset, apnea-hypopnea index (AHI), arousal index, periodic leg movement disorder (PLMD) index] were recorded. Linear regression was performed to compare the primary sleep variables of TST and SE while adjusting for all significant confounding factors. All statistical analyses were performed using software R. RESULTS: There were 124 subjects; 62 cases (OSA and BPH) and 62 controls (OSA alone). The sample consisted of obese (32.5±3.9), older (63±11.1), and male (97.5%) subjects. There was no significant difference for the demographic variables age, BMI or sex between the BPH and non-BPH groups (p>0.05). There was no significant difference between the groups for the primary target variables; TST (268.68 ± 69.7 vs. 284.49 ± 66.4 minutes, p=0.19), or sleep efficiency (71.1±17.5 vs. 76.7±16.9 %, p=0.07). Only parameter found to be significant were; stage 1 % NREM sleep for BPH group (27.5±18.6 %) compared to the non-BPH (21.08±16.7, % p=0.02) and PLMI; BPH (30.27±40.1) compared to the non-BPH (14.48±24.5 p=0.009). In linear regression model with each parameter as dependable parameter while BPH as fixed factor and age, efficiency, arousal index as covariates both parameters were significant (p