The association of sleep disordered breathing with left ventricular remodeling in CAD patients: a cross-sectional study

Abstract

Background: There is still insufficient knowledge on the potential effect of mild to moderate sleep-disordered breathing (SDB) that is widely prevalent, often asymptomatic, and largely undiagnosed in patients with stable coronary artery disease (CAD). SDB affects 34% of men and 17% of women aged between 30 and 70. The objective of this study was to evaluate the association between SDB and left ventricular (LV) hypertrophy as well as structural remodeling in stable CAD patients.

Methods: The study was based on a cross-sectional design. Echocardiography and polysomnography was performed in 772 patients with CAD and with untreated sleep apnea. All study participants underwent testing by Epworth Sleepiness Scale questionnaire. Their mean age, NYHA and left ventricular ejection fraction were, respectively: 57 ± 9 years, 2.1 ± 0.5 and 51 ± 8%, and 76% were men. Sleep apnea (SA) was defined as an apnea-hypopnea-index (AHI) ≥5 events/h, and, non-SA, as an AHI <5.

Results: Sleep apnea was present in 39% of patients, and a large fraction of those patients had no complaints on excessive daytime sleepiness. The patients with SA were older, with higher body mass and higher prevalence of hypertension. LV hypertrophy (LVH), defined by allometrically corrected (LV mass/height^2.7) gender-independent criteria, was more common among the patients with SA than those without (86% vs. 74%, p < 0.001). The frequency of LVH by wall thickness criteria (interventricular septal thickness or posterior wall thickness ≥ 12 mm: 49% vs. 33%, p < 0.001) and concentric LVH (61% vs. 47%, p = 0.001) was higher in CAD patients with SA. The patients with SA had significantly higher values of both interventricular septal thickness and posterior wall thickness. Multiple logistic regression analysis showed that even mild sleep apnea was an independent predictor for LVH by wall thickness criteria and concentric LVH (OR = 1.5; 95% CI 1.04-2.2 and OR = 1.9; 1.3-2.9 respectively).

Conclusions: We concluded that unrecognized sleep apnea was highly prevalent among patients with stable CAD, and the majority of those patients did not report daytime sleepiness. Mild to moderate sleep apnea was associated with increased LV wall thickness, LV mass, and with higher prevalence of concentric LV hypertrophy independently of coexisting obesity, hypertension, diabetes mellitus or advancing age.

Keywords: Cad; Echocardiography; Hypertrophy; Left ventricular geometry; Polysomnography; Sleep apnea.

Fig. 1

Prevalence of left ventricular hypertrophy on the basis of wall thickness criteria (interventricular septal thickness or posteriorwall thickness ≥ 12 mm). LVH = left ventricular hypertrophy; SA = sleep apnea

Fig. 2

Frequency distribution of the four different LV–geometrical patterns in patients with vs without sleep apnea. The spreadsheet showed the distribution of LV mass index versus relative wall thickness (RWT). Apnea-hypopnea index (AHI) ≥5 n/h = sleep apnea. Left ventricular geometry, including concentric remodeling as well as eccentric and concentric hypertrophy based on normal/high RWT and left ventricular hypertrophy (LVH) based on left ventricular mass index (LVMI g/height ^2.7) criteria. The reference lines indicate the gender-independent LVH cut-off, (51 g/m ^2.7, vertical line), and the upper limit of RWT (0.42, horizontal line). In boxes: geometric pattern, percentage: without sleep apnea/with sleep apnea; χ²-test df = 3 p < 0.001

Fig. 3

Cross sectional predictors of left ventricular hypertrophy on the basis of wall thickness criteria (interventricular septal thickness or posterior wall thickness ≥ 12 mm). The independent variables entered in the multiple regression model were: age, sex, ESS score, hypoxemia during sleep SaO₂, age over 60, AHI (4 categories), BMI (4 categories), NYHA class, hypertension, diabetes mellitus, current smoking, history of MI as well as LVEF lower than 45% and excessive daytime sleepiness (ESS ≥ 10). Model: Hosmer & Lemeshow test χ² = 8.9 df = 8 p = 0.349; Nagelkerke R² = 0.213. Abbreviations: ESS = Epworth sleepiness scale, LVEF = left ventricular ejection fraction; AHI = apnea-hypopnea index; BMI = body mass index; OR = odds ratio; CI = confidence interval

Fig. 4

Cross sectional predictors of concentric left ventricular hypertrophy. The independent variables entered in the multiple regression model were: age, sex, ESS score, hypoxemia during sleep SaO₂, age over 60, AHI (4 categories), BMI (ORs per 5 kg/m² higher BMI above 25 kg/m²), NYHA class, hypertension, diabetes mellitus, current smoking, history of MI as well as LVEF lower than 45% and excessive daytime sleepiness (ESS ≥ 10). Model: Hosmer & Lemeshow test χ² = 5.1 df = 8 p = 0.746; Nagelkerke R² = 0.132. Abbreviations: ESS = Epworth sleepiness scale; LVEF; left ventricular ejection fraction; AHI = apnea-hypopnea index; BMI = body mass index; OR = odds ratio; CI = confidence interval