Predictors for carotid and femoral artery intima-media thickness in a non-diabetic sleep clinic cohort

Abstract

Introduction: The impact of sleep disordered breathing (SDB) on arterial intima-media thickness (IMT), a surrogate measure for cardiovascular disease, remains uncertain, in part because of the potential for non-SDB vascular risk factor interactions. In the present study, we determined predictors for common carotid (CCA) and femoral (CFA) artery IMT in an adult, sleep clinic cohort where non-SDB vascular risk factors (particularly diabetes) were eliminated or controlled.

Methods: We recruited 296 participants for polysomnography (standard SDB severity metrics) and CCA/CFA ultrasound examinations, followed by a 12 month vascular risk factor minimisation (RFM) and continuous positive pressure (CPAP) intervention for participants with a range of SDB severity (RFM Sub-Group, n = 157; apnea hyponea index [AHI]: 14.7 (7.2-33.2), median [IQR]). Univariable and multivariable linear regression models determined independent predictors for IMT. Linear mixed effects modelling determined independent predictors for IMT change across the intervention study. P<0.05 was considered significant.

Results: Age, systolic blood pressure and waist:hip ratio were identified as non-SDB predictive factors for CCA IMT and age, weight and total cholesterol:HDL ratio for CFA IMT. No SDB severity metric emerged as an independent predictor for either CCA or CFA IMT, except in the RFM Sub-Group, where a 2-fold increase in AHI predicted a 2.4% increase in CFA IMT. Across the intervention study, CCA IMT decreased in those who lost weight, but there was no CPAP use interaction. CFA IMT, however, decreased by 12.9% (95%CI 6.8, 18.7%, p = 0.001) in those participants who both lost weight and used CPAP > = 4hours/night.

Conclusion: We conclude that SDB severity has little impact on CCA IMT values when non-SDB vascular risk factors are minimised or not present. This is the first study, however, to suggest a potential linkage between SDB severity and CFA IMT values.

Trial registration: Australian New Zealand Clinical Trials Registry, ACTRN12611000250932 and ACTRN12620000694910.

Fig 1. Study flow diagram.

Study flow diagram showing movement of participants and data through the recruitment and screening phases, the “Main Group” and “Risk Factor Minimisatation [RFM] Sub-Group” cross-sectional analyses, and the intervention study. SDB = sleep disordered breathing; BSL = blood sugar level; PSG = polysomnography; CPAP = continuous positive airway pressure; AHI = apnea hypopnea index; CCA IMT = common carotid artery intima-media thickness.

Fig 2. Main group IMT-age relationships.

Scatter plots of IMT versus age for CCA (A) and CFA (B). The back-transformed predicted values and 95% confidence bands from the linear regression of ln(IMT) on age are shown. R² for the relationship is higher for the CCA than for the CFA, but both have wide predictive bands for IMT indicating significant non-age-related variability.

Fig 3. RFM sub-group—Effect of CPAP use on ln(CCA IMT) and on ln(CFA IMT) by weight loss status.

Mean with 95%CIs showing the within participant change in ln(CCA IMT) and in ln(CFA IMT) in the RFM Sub-Group during the 12-month RFM intervention by weight loss status and CPAP use. The ln(CFA IMT) depended on both weight loss and CPAP use (p = 0.020) with those who lost weight and also used CPAP > = 4hrs/day demonstrating a 12.9% (95%CI 6.8, 18.7%, p = 0.001) reduction in CFA IMT from baseline compared to no significant change in those who gained/maintained weight or used CPAP <4hours/day.