Multifaceted Strategy Based on Automated Text Messaging After a Recent Heart Failure Admission: The MESSAGE-HF Randomized Clinical Trial

Abstract

Importance: Readmissions after an index heart failure (HF) hospitalization are a major contemporary health care problem.

Objective: To evaluate the feasibility and efficacy of an intensive telemonitoring strategy in the vulnerable period after an HF hospitalization.

Design, setting, and participants: This randomized clinical trial was conducted in 30 HF clinics in Brazil. Patients with left ventricular ejection fraction less than 40% and access to mobile phones were enrolled up to 30 days after an HF admission. Data were collected from July 2019 to July 2022.

Intervention: Participants were randomly assigned to a telemonitoring strategy or standard care. The telemonitoring group received 4 daily short message service text messages to optimize self-care, active engagement, and early intervention. Red flags based on feedback messages triggered automatic diuretic adjustment and/or a telephone call from the health care team.

Main outcomes and measures: The primary end point was change in N-terminal pro-brain natriuretic peptide (NT-proBNP) from baseline to 180 days. A hierarchical win-ratio analysis incorporating blindly adjudicated clinical events (cardiovascular deaths and HF hospitalization) and variation in NT-proBNP was also performed.

Results: Of 699 included patients, 460 (65.8%) were male, and the mean (SD) age was 61.2 (14.5) years. A total of 352 patients were randomly assigned to the telemonitoring strategy and 347 to standard care. Satisfaction with the telemonitoring strategy was excellent (net promoting score at 180 days, 78.5). HF self-care increased significantly in the telemonitoring group compared with the standard care group (score difference at 30 days, -2.21; 95% CI, -3.67 to -0.74; P = .001; score difference at 180 days, -2.08; 95% CI, -3.59 to -0.57; P = .004). Variation of NT-proBNP was similar in the telemonitoring group compared with the standard care group (telemonitoring: baseline, 2593 pg/mL; 95% CI, 2314-2923; 180 days, 1313 pg/mL; 95% CI, 1117-1543; standard care: baseline, 2396 pg/mL; 95% CI, 2122-2721; 180 days, 1319 pg/mL; 95% CI, 1114-1564; ratio of change, 0.92; 95% CI, 0.77-1.11; P = .39). Hierarchical analysis of the composite outcome demonstrated a similar number of wins in both groups (telemonitoring, 49 883 of 122 144 comparisons [40.8%]; standard care, 48 034 of 122 144 comparisons [39.3%]; win ratio, 1.04; 95% CI, 0.86-1.26).

Conclusions and relevance: An intensive telemonitoring strategy applied in the vulnerable period after an HF admission was feasible, well-accepted, and increased scores of HF self-care but did not translate to reductions in NT-proBNP levels nor improvement in a composite hierarchical clinical outcome.

Trial registration: ClinicalTrials.gov Identifier: NCT04062461.

Figure 1.. CONSORT Diagram

NT-proBNP indicates N-terminal pro–brain natriuretic peptide.

Figure 2.. Trial Outcomes

A, N-terminal pro–brain natriuretic peptide (NT-proBNP) levels at baseline, 30 days, and 180 days after randomization in the telemonitoring group and standard care group. Data for NT-proBNP are expressed as geometric means and 95% CIs. For this analysis, multiple imputation was used for missing values. The ratio of change from baseline to 180 days, was not significantly different in the telemonitoring group compared with the standard care group (0.95; 95% CI, 0.79-1.16). B and C, Hierarchical compositive analysis depicting total wins and ties in both groups (B) and for each component of the analysis (C). The composite outcome was hierarchically analyzed in the following order: (1) time to cardiovascular death; (2) time to first heart failure (HF) hospitalization; and (3) relative change of greater than 20% in NT-proBNP level from study inclusion to 180 days of follow-up.

Figure 3.. Prespecified Subgroup Analysis

eGFR indicates estimated glomerular filtration rate; HF, heart failure; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro–brain natriuretic peptide; NYHA, New York Heart Association.