Left ventricular remodeling and systolic function changes in patients with obstructive sleep apnea: a comprehensive contrast-enhanced cardiac magnetic resonance study

Abstract

Background: A comprehensive assessment of left ventricular (LV) remodeling and systolic function using contrast-enhanced cardiac magnetic resonance (CMR) imaging in patients with obstructive sleep apnea (OSA) has not yet been reported. This retrospective case-control study aimed to explore and assess the myocardial structure, function, and tissue characteristic changes of LV remodeling in patients with OSA using the CMR method.

Methods: Fifty-one selected participants 32 OSA and 19 non-OSA underwent overnight polysomnography and CMR examination using T1 mapping and feature tracking techniques. Twenty age- and sex-matched healthy controls were also enrolled for comparison between the groups.

Results: Patients were grouped by apnea-hypopnea index (AHI): AHI <5 events/h as non-OSA group (n=19, 40.7±8.0 years), 5-30 events/h as mild-moderate OSA (n=13, 47.8±9.4 years), and >30 events/h as severe OSA (n=19, 39.0±10.0 years). The OSA group had a higher LV mass index (LVMI) to height^2.7 than the non-OSA and healthy control groups (21.0±3.8 vs. 16.4±3.1 and 16.3±3.2 mL/m^2.7, P<0.001). Compared with healthy controls, OSA patients had lower global circumferential strain values, although the LV ejection fraction was preserved. Late gadolinium enhancement was not detected in all participants, whereas the extracellular volume fraction was lower in patients with OSA than in the non-OSA and healthy control groups (24.4%±1.9% vs. 26.2%±2.5%, P=0.006 and 24.4%±1.9% vs. 26.5%±2.3%, P=0.004, respectively). The indexed cellular volume (iCV) of the myocardium was significantly higher in subjects with mild-to-moderate and severe OSA than in those without OSA (14.2±2.3 and 15.8±3.1 vs. 11.6±2.4 mL/m^2.7, P<0.05). On multivariate linear regression analysis of patients with two different models, OSA severity remained significantly associated with increased LVMI (β=0.348, P=0.004 and β=0.233, P=0.048, respectively) and iCV (β=0.337, P=0.004 and β=0.231, P=0.047, respectively) after adjusting for clinical risk factors.

Conclusions: LVMI is elevated in OSA with a normal LV ejection fraction, mainly with cellular hypertrophy. Cellular hypertrophy without focal fibrosis in OSA may be our main finding.

Keywords: Cardiac magnetic resonance (CMR); T1 mapping; feature tracking (FT); obstructive sleep apnea (OSA); ventricular remodeling.

Figure 1

The T1 mapping method. T1 mapping of a severe OSA patient (A-C) and a non-OSA patient (D-F). Delineating the endocardial (red) and epicardial (green) borders on pre- (A,D) and post-contrast (B,E) T1 maps with additional ROIs drawn in left ventricular blood pool. ECV (C,F) maps were computed automatically according to the hematocrit value. ECV, extracellular volume; myoc, myocardium; OSA, obstructive sleep apnea; ROI, region of interest.

Figure 2

Study flowchart. PSG, polysomnography; CMR, cardiac magnetic resonance; LVEF, left ventricular ejection fraction; LGE, late gadolinium enhancement; OSA, obstructive sleep apnea; AHI, apnea-hypopnea index.

Figure 3

Differences in left ventricular remodeling parameters among the non-OSA, the mild-moderate OSA and the severe OSA groups. (A-C) Conventional left ventricular structure parameters, including left ventricular mass index, left ventricular maximal wall thickness and left ventricular mass/volume ratio. (D-F) T1 mapping parameters, including indexed cellular volume, indexed extracellular volume, and indexed cellular volume minus indexed extracellular volume. *, P<0.05 between the severe OSA and the non-OSA groups; ^#, P<0.05 between the mild-moderate OSA and the non-OSA groups. OSA, obstructive sleep apnea.

Figure 4

Relationship between apnea-hypopnea index and left ventricular remodeling indexes. The apnea-hypopnea index and left ventricular structure parameters correlations were assessed with Spearman correlation after normality test.