(7) Heart Failure Phenotype Transitions in Early Diagnosis

Publication Number: 681

Background: Phenotyping heart failure (HF) patients based on left ventricular ejection fraction (LVEF) measurement is the first step before establishing treatment strategies.

Objectives: To describe patients’ transition patterns across LVEF categories after the first assessment.

Methods: This is a cohort study using two different data sources, ANMCO, a prospective registry of patients with HF who visited participating cardiology centres in Italy, and BIG-PAC, a database combining healthcare data from various primary and hospital centres in Spain. Patients with HF in ANMCO (new and prevalent patients) between January 2010 and December 2018, and in BIG-PAC (new patients only) between January 2012 and September 2020 were classified into HFrEF ( < 40%), HFmrEF (40-49%) and HFpEF (≥50%) and HFuEF (no EF record) based on their LVEF at index (according to ESC 2016 guideline). Patients without a second LVEF assessment during follow-up were excluded from the analysis. The proportion of HF phenotypes was reported for each assessment together with the transition among LVEF categories.

Results: The study included 5,304 and 12,498 eligible patients in ANMCO and BIG-PAC. At baseline (first diagnosis in BIG-PAC and enrolment date in ANMCO), HFrEF represented about 58% of HF patients in ANMCO [3,096/5,304] and 62% in BIG-PAC [7,798/12,498]). At the next measurement, 69% (2,146/3,096), 60% (764/1,269) and 84% (792/939) of patients with HFrEF, HFmrEF and HFpEF stayed in the same class in ANMCO while in BIG-PAC were 59% (4,638/7,798), 41% (362/885) and 55% (2,089/3,815) first and second categories agreed. About 31% (950/3,096) and 41% (3160/7,798) of HFrEF patients in ANMCO and BIG-PAC had a higher EF at the second measurement (HFrEF with improved EF) and about 23% of HFmrEF patients improve EF to ≥50% in both databases. Overall, 24% and 27% of patients show an increase in EF at the second measurement.
In ANMCO, more patients with HFmrEF had an EF > 50% at second measurement than EF < 40 ( 23% vs 17%), while in BIG-PAC the trend was the opposite (23% vs 36% HFmrEF to HFrEF). Among all patients, 7% (358/5304) and 17% (2132/12,498) in ANMCO and BIG-PAC had a lower EF at second measurement sufficient to change category. The proportion of patients with EF >=50% increased in ANMCO: 18% to 26%, but decreased in BIG-PAC: from 31% to 26% , with 21% of initial HFpEF patients having an EF < 40% at follow-up.

Conclusions: The proportion of HF phenotypes determined at second EF measurement are very similar in Italy and Spain. While about one quarter of patients shows an improvement in EF in both databases, the Italian Registry registry data shows fewer phenotype transitions overall owing, perhaps, to a population with more mature disease.