Symptom burden and health related quality of life in heart failure patients: a multi-centre European study (STRATIFYHF project)
N C Okwose, F Bano, D A Ibirogba, S J Charman, M Dobric, R J Stefanetti, I Olivotto, A Fornaro, M J Bravo, A Preveden, A Milovancev, L Velicki, P M Seferovic, P Banerjee, D G JakovljevicAbstract
Background/Introduction
Patients with heart failure (HF) present with multiple signs and symptoms which can interfere with activities of daily living and impact generally on quality of life (QoL). However, the extent to which symptom burden and HF phenotype (i.e., HF with preserved, mildly reduced and reduced ejection [HFpEF, HFmrEF, HFrEF] respectively) influence perceived QoL remains insufficiently explored.
Purpose
To: (i) compare Minnesota Living with Heart Failure Questionnaire (MLHFQ) total scores between HF phenotypes; (ii) examine the relationship between symptom burden (≤2 vs ≥3 symptoms) and MLHFQ within and between phenotypes; and (iii) explore phenotype-symptom burden interaction effects on QoL.
Methods
Prospective data were collected from HF patients across five European clinical centres participating in the STRATIFYHF study. Signs/symptoms (fatigue, chest pain, palpitations, ankle swelling, dyspnoea, exercise intolerance, orthopnoea, peripheral oedema) were extracted from clinical assessments and used to quantify symptom burden. MLHFQ total scores (items 1–21) were summed to generate QoL score, with higher values indicating poorer QoL. Patients were categorised by HF phenotype and stratified by symptom burden (≤2 vs ≥3 symptoms). Group comparisons were performed using one-way ANOVA with Kruskal–Wallis verification; a two-way ANOVA evaluated the effect of HF phenotypes, symptom burden, and their effect on QoL (with gender as a covariate). Within phenotype comparisons were performed using Mann–Whitney or Welch t-tests as appropriate.
Results
A total of 168 patients mean (SD) age; 67 (11) years) were included (HFpEF n=34; HFmrEF n=35; HFrEF n=99). Females comprised 26% (44/168) with similar sex distribution across phenotypes: HFpEF 29% females, HFmrEF 23% females, and HFrEF 26% females. Median (IQR) MLHFQ total scores were HFpEF 23 (10–47), HFmrEF 26 (14–44), and HFrEF 22 (9–38), with no significant difference by phenotype (p=0.46). Symptom-burden effects on QoL varied by phenotype (p=0.001). Within-phenotype analysis showed a significant burden effect in HFpEF: patients with ≥3 symptoms 45 (21–53) vs ≤2 symptoms 7.5 (2–11) (p<0.001). In contrast, no significant differences were observed in HFmrEF ≥3 symptoms 23 (15–37) vs ≤2 symptoms 32 (13–45); p=0.74 or HFrEF ≥3 symptoms 22 (8–35) vs ≤2 symptoms 23 (12–39); p=0.58.
Conclusion
In HF patients, QoL did not differ significantly by phenotype, but the impact of symptom burden on QoL is phenotype-specific. Patients with HFpEF showed a marked deterioration in QoL when ≥3 symptoms are present. In HFrEF and HFmrEF however, quality of life was not related to symptoms. These findings support early, symptom-targeted interventions particularly in HFpEF, and highlight the value of phenotype-tailored symptom assessment at the time of diagnosis.