Periodic limb movements of sleep and hypertension in patients with obstructive sleep apnea: a hypothesis-generating analysis from a polysomnographic cohort.

BACKGROUND

Hypertension in obstructive sleep apnea (OSA) is incompletely explained by intermittent hypoxia and respiratory events alone. Periodic limb movements of sleep (PLMS) induce repetitive nocturnal sympathetic activation and blood pressure surges, yet their contribution to established daytime hypertension in OSA is poorly characterised.

METHODS

In a retrospective cohort of 60 PSG-confirmed OSA patients, participants were stratified by PLMS Index (PLMS-I) threshold of ≥ 15 events/hr. Hypertension prevalence, cardiovascular risk indices (Framingham Risk Score, FRS; Charlson Comorbidity Index, CCI), gender distribution, and polysomnographic parameters were compared between groups using chi-square, independent t-test, Mann-Whitney U test, and Spearman correlation. Binary logistic regression was performed with hypertension as the outcome variable and PLMS-I group, age, and gender as covariates. Partial Spearman correlations between PLMS-I and FRS/CCI were computed after controlling for age.

RESULTS

Patients with PLMS-I ≥ 15/hr (n = 41) had a significantly higher hypertension prevalence than those with PLMS-I < 15/hr (n = 19): 76.7% vs. 23.3% of all hypertensive patients (p = 0.026). Gender distribution did not differ between groups (males: 70.7% vs. 57.9%; p = 0.326). On binary logistic regression adjusting for age and gender, high PLMS-I was associated with 2.4-fold increased odds of hypertension (OR = 2.421, 95% CI 0.640-9.161, p = 0.193), which did not reach statistical significance; age was an independent predictor (OR = 1.068/year, 95% CI 1.020-1.119, p = 0.005). PLMS-I correlated positively with FRS (r = 0.302, p = 0.023) and CCI (r = 0.346, p = 0.007); these associations were attenuated after partial correlation controlling for age (partial r = 0.146, p = 0.268 and partial r = 0.114, p = 0.391, respectively). The high-PLMS-I group demonstrated a significantly elevated arousal index (p = 0.001).

CONCLUSION

While high PLMS-I was associated with 2.4-fold increased odds of hypertension in OSA, this did not reach significance after age adjustment (OR = 2.421, p = 0.193), with age emerging as the primary predictor. Unadjusted correlations with FRS and CCI were attenuated after controlling for age. These hypothesis-generating findings highlight the confounding role of age and underscore the need for larger, adequately powered prospective studies to determine whether PLMS-I is a truly independent cardiovascular risk correlate in OSA.