Change in periodic limb movement index during treatment of obstructive sleep apnea with continuous positive airway pressure

Study Objectives: The following hypotheses were investigated. 1) severe obstructive sleep apnea (OSA) can mask concurrent periodic limb movement (PLM) disorder (PLMD), which becomes evident or worsens after treatment with continuous positive airway pressure (CRAP); 2) in patients with mild OSA, PLMs are not masked but may be triggered by subclinical hypopneas or respiratory effort-related arousals and improve after CPAP. Design: Retrospective analysis was performed on 2 polysomnographic studies per patient-1 baseline, the second with CPAP titration. The apnea-hypopnea index (AHI) and PLM index (PLMI) under the 2 conditions were statistically analyzed. Setting: University hospital sleep disorders center. Patients: Patients were selected if they had a baseline AHI of 5 or greater and CPAP titration resulted in reduced AHI. Also, each needed to have either a PLMI of 5 or greater on baseline PSG or during CPAP titration. Patients who started or discontinued a medication that could affect PLMs after the baseline PSG were excluded. Interventions: As clinically indicated, CPAP for OSA. Measurements and Results: Eighty-six patients qualified and were divided into 3 groups based on OSA severity. Significant correlations (P < 0.05) were found between AHI and PLMI on the baseline PSG (-0.50), between AHI on baseline PSG and PLMI on CPAP titration (0.49), and between PLMI on baseline PSG and on CPAP titration (-0.21). The increase in PLMI during CPAP titration in patients with severe OSA was statistically significant (P<0.001). The PLMI decreased with CPAP in 20 of 86 patients, mostly in the mild OSA subgroup. Regression of post-CPAP reduction of AHI and change in PLMI yielded a significant logarithmic relationship (R-2=0.3042). Conclusions: Severity of OSA may determine the effect of CPAP on PLMs. The PLMs may increase in moderate to severe OSA due mainly to "unmasking" of underlying PLMD. The PLMs may decrease in mild OSA post-CPAP due to resolution of PLMs associated with respiratory effort-related arousals. This suggests that PLMs may have more than 1 etiology and may be categorized as spontaneous (as in PLMD) and induced (when secondary to respiratory effort-related arousals).