Impaired Recovery of Left Ventricular Function in Patients With Cardiomyopathy and Left Bundle Branch Block

Abstract

Background: Patients with left bundle branch block (LBBB) often respond to cardiac resynchronization therapy (CRT) with left ventricular ejection fraction (LVEF) improvement. Guideline-directed medical therapy (GDMT), not CRT, is first-line therapy for patients with reduced LVEF with LBBB. However, there are little data on how patients with reduced LVEF and LBBB respond to GDMT.

Objectives: This study examined patients with cardiomyopathy and sought to assess rates of LVEF improvement for patients with LBBB compared to other QRS morphologies.

Methods: Using data from the Duke Echocardiography Laboratory Database, the study identified patients with baseline electrocardiography and LVEF ≤35% who had a follow-up LVEF 3 to 6 months later. The study excluded patients with severe valve disease, a cardiac device, left ventricular assist device, or heart transplant. QRS morphology was classified as LBBB, QRS duration <120 ms (narrow QRS duration), or a wide QRS duration ≥120 ms but not LBBB. Analysis of variance testing compared mean change in LVEF among the 3 groups with adjustment for significant comorbidities and GDMT.

Results: There were 659 patients that met the criteria: 111 LBBB (17%), 59 wide QRS duration ≥120 ms but not LBBB (9%), and 489 narrow QRS duration (74%). Adjusted mean increase in LVEF over 3 to 6 months in the 3 groups was 2.03%, 5.28%, and 8.00%, respectively (p < 0.0001). Results were similar when adjusted for interim revascularization and myocardial infarction. Comparison of mean LVEF improvement between patients with LBBB on GDMT and those not on GDMT showed virtually no difference (3.50% vs. 3.44%). The combined endpoint of heart failure hospitalization or mortality was highest for patients with LBBB.

Conclusions: LBBB is associated with a smaller degree of LVEF improvement compared with other QRS morphologies, even with GDMT. Some patients with LBBB may benefit from CRT earlier than guidelines currently recommend.

Keywords: guideline directed medical therapy; heart failure; left bundle branch block; left ventricular functional recovery.

Figure 1 –

Study population consort diagram. The Echo database contains 157,586 unique patients. Black arrows show the number of patients removed for exclusionary criteria. Only 8,129 qualified for a baseline Echo. More patients with LBBB were subsequently excluded for receiving a cardiac device, LVAD, or transplant. Less NQRS patients died prior to a potential follow-up Echo. Only 659 patients were ultimately included in the study.

Figure 2/

Central Illustration –Mean LVEF with standard deviation shown for baseline and follow-up Echo, graphed for the 3 QRS groups. Patients from the 3 QRS groups have similar mean LVEF at baseline. NQRS patients have the highest rates of LV functional recovery, while LBBB patients have the lowest.

Figure 3 –

Kaplan-Meier Curves. A. Time-to-Mortality, B. Time-to-Heart Failure Hospitalization, and C. Time-to-Heart Failure Hospitalization or Mortality. Mortality data was acquired through the National Death Index (NDI). Heart failure hospitalization data was acquired through the medical record at Duke-affiliated hospitals. Because all Echos were performed prior to 2015, events were censored at December 31, 2014, matching the end date of the NDI query. All curves were censored at 10 years post follow-up Echo.

Figure 3 –

Kaplan-Meier Curves. A. Time-to-Mortality, B. Time-to-Heart Failure Hospitalization, and C. Time-to-Heart Failure Hospitalization or Mortality. Mortality data was acquired through the National Death Index (NDI). Heart failure hospitalization data was acquired through the medical record at Duke-affiliated hospitals. Because all Echos were performed prior to 2015, events were censored at December 31, 2014, matching the end date of the NDI query. All curves were censored at 10 years post follow-up Echo.

Figure 3 –

Kaplan-Meier Curves. A. Time-to-Mortality, B. Time-to-Heart Failure Hospitalization, and C. Time-to-Heart Failure Hospitalization or Mortality. Mortality data was acquired through the National Death Index (NDI). Heart failure hospitalization data was acquired through the medical record at Duke-affiliated hospitals. Because all Echos were performed prior to 2015, events were censored at December 31, 2014, matching the end date of the NDI query. All curves were censored at 10 years post follow-up Echo.