Atrioventricular coupling as prognostic marker in cardiac resynchronization therapy

Background: Left ventricular (LV) ejection fraction (EF) is a widely used functional measure but is limited as a prognostic marker in patients with heart failure (HF) (1). Myocardial strain assessment from feature-tracking (FT) cardiac magnetic resonance (CMR) provides a more detailed assessment of cardiac function and is suggested to be more suitable for prognosis (2). LV global longitudinal strain (GLS), right ventricular (RV) GLS, left atrial (LA) and right atrial (RA) strain have independently shown prognostic in HF (3–6). However, the combined assessment of all four cardiac chambers, and the functional interplay between atrial and ventricular mechanics (atrio‑ventricular coupling, AVC) remain unexplored.

The aim of this study was to investigate the prognostic value of LV, RV, LA and RA strain, as well as AVC in patients with HF using FT-CMR.

Methods: This prediction model study prospectively analyzed 65 patients (68±9 years, 15 females) with HF indicated for cardiac resynchronization therapy (CRT). The follow-up period was five years, and the primary event was the composite outcome of major adverse cardiac events (MACE). Patients underwent CMR with standard balanced steady-state free precession cine images in short- and long-axis planes. FT longitudinal strain was analyzed in all four cardiac chambers. AVC was defined as left and right atrioventricular coupling index (LACI and RACI), calculated as the ratio of atrial and ventricular volumes at ventricular end diastole. Additionally, AVC was estimated by the slope of the linear regression line derived from time-aligned ventricular strain to atrial strain diagrams (LA-LV and RA-RV strain slope). Patients were grouped into tertiles from strain and AVC values. P-value < 0.05 was considered statistically significant.

Results: Of the 65 patients, 53 (82%) received CRT and 23 (35%) experienced MACE during the 5-years follow-up time. There was no difference in outcome between patients receiving CRT and those who did not (Chi-square p=0.2). Patients with events had worse LV-GLS (-7±3% vs -9±4%, p< 0.05), LA reservoir strain (12±7% vs 18±8%, p< 0.01), and RA reservoir strain (15±8% vs 20±9%, p< 0.01) compared to patients without. Patients with the worst tertile in LV-GLS and LA reservoir strain had higher probability of event than the other tertiles (Figure 1). Patients with highest tertile of LACI, LV-LA strain slopes or RACI were more likely to experience events (Figure 2). Cox proportional-hazards univariable models revealed that LV, RV, LA, and RA strain as well as AVC were associated with event. LA conduit strain was, in Cox proportional-hazards multivariable model adjusted for known risk factors (age, sex, BMI, LV EF, and NYHA), the only prognostic marker (Table 1).

Conclusion: LV, RV, LA, and RA strain measures in isolation were in varying degrees associated with worse prognosis, as was AVC. When adjusted for known risk factors, only LA conduit strain was independently associated with worse prognosis.

Table 1. Cox Proportional-Hazards Multivariable Model of Primary Event Non-Adjusted and Adjusted for LV-EF, NYHA, Age, Sex, and BMI.

	Non-adjusted hazard ratio (95% CI)	p	Adjusted hazard ratio (95% CI)	p
LV-GLS (%)	1.01 (0.77 – 1.33)	0.94	1.15 (0.85 – 1.55)	0.37
RV-GLS (%)	1.03 (0.90 – 1.16)	0.70	1.10 (0.90 – 1.34)	0.36
LA reservoir strain (%)	0.98 (0.79 – 1.23)	0.87	0.90 (0.68 – 1.20)	0.49
LA conduit strain (%)	0.72 (0.56 – 0.92)	0.01	0.69 (0.49 – 0.96)	0.01
RA reservoir strain (%)	1.05 (0.93 – 1.19)	0.42	1.15 (0.96 – 1.36)	0.12
LACI	1.38 (0.016 – 118)	0.89	0.17 (0.0001 – 96.1)	0.58
LV-LA strain slope	1.02 (0.16 – 6.31)	0.99	0.57 (0.05 – 6.57)	0.65
RACI	4.02 (0.25 – 64.6)	0.33	15.1 (0.26 – 895)	0.19
RV-RA strain slope	1.74 (1.39 – 7.71)	0.47	8.21 (0.90 – 74.6)	0.06

Statistically significant (P< 0.25) variables from Cox proportional-hazards univariable models were included in the multivariable model. P < 0.05 in the multivariable model was considered statistically significant.BMI: body mass index; EDV: left ventricular end-diastolic volume; ESV: left ventricular end-systolic volume; EF: left ventricular ejection fraction; GLS: left ventricular global longitudinal strain; HC: healthy controls; HFrEF: heart failure with reduced ejection fraction; LA: left atrial; LACI: left atrioventricular coupling index; LV: left ventricular; NYHA: New York Heart Association; RA: right atrial; RACI: right atrioventricular coupling index; RV: right ventricular.Figure 1. Kaplan-Meier plots of left ventricular global longitudinal strain (LV-GLS), right ventricular (RV)-GLS, left atrial (LA) reservoir strain, and right atrial (RA) reservoir strain in study patients grouped by tertiles.
Figure 2. Kaplan-Meier plots of left atrioventricular coupling index (LACI), left atrial (LA)-left ventricular (LV) strain slope, right atrioventricular coupling index (RACI), and right atrial (RA)-right ventricular (RV) strain slope in study patients grouped by tertiles.