Re-appraisal of the obesity paradox in heart failure: a meta-analysis of individual data

Nick Marcks, Alberto Aimo, James L Januzzi, Giuseppe Vergaro, Aldo Clerico, Roberto Latini, Jennifer Meessen, Inder S Anand, Jay N Cohn, Jørgen Gravning, Thor Ueland, Antoni Bayes-Genis, Josep Lupón, Rudolf A de Boer, Akiomi Yoshihisa, Yasuchika Takeishi, Michael Egstrup, Ida Gustafsson, Hanna K Gaggin, Kai M EggersKurt Huber, Ioannis Tentzeris, Andrea Ripoli, Claudio Passino, Sandra Sanders-van Wijk, Michele Emdin, Hans-Peter Brunner-La Rocca*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


BACKGROUND: Higher body mass index (BMI) is associated with better outcome compared with normal weight in patients with HF and other chronic diseases. It remains uncertain whether the apparent protective role of obesity relates to the absence of comorbidities. Therefore, we investigated the effect of BMI on outcome in younger patients without co-morbidities as compared to older patients with co-morbidities in a large heart failure (HF) population.

METHODS: In an individual patient data analysis from pooled cohorts, 5,819 patients with chronic HF and data available on BMI, co-morbidities and outcome were analysed. Patients were divided into four groups based on BMI (i.e. ≤ 18.5 kg/m2, 18.5-25.0 kg/m2; 25.0-30.0 kg/m2; 30.0 kg/m2). Primary endpoints included all-cause mortality and HF hospitalization-free survival.

RESULTS: Mean age was 65 ± 12 years, with a majority of males (78%), ischaemic HF and HF with reduced ejection fraction. Frequency of all-cause mortality or HF hospitalization was significantly worse in the lowest two BMI groups as compared to the other two groups; however, this effect was only seen in patients older than 75 years or having at least one relevant co-morbidity, and not in younger patients with HF only. After including medications and N-terminal pro-B-type natriuretic peptide and high-sensitivity cardiac troponin concentrations into the model, the prognostic impact of BMI was largely absent even in the elderly group with co-morbidity.

CONCLUSIONS: The present study suggests that obesity is a marker of less advanced disease, but does not have an independent protective effect in patients with chronic HF. Categories of BMI are only predictive of poor outcome in patients aged > 75 years or with at least one co-morbidity (bottom), but not in those aged < 75 years without co-morbidities (top). The prognostic effect largely disappears in multivariable analyses even for the former group. These findings question the protective effect of obesity in chronic heart failure (HF).

Original languageEnglish
Pages (from-to)1280-1291
Number of pages12
JournalClinical research in cardiology
Issue number8
Early online date11 Mar 2021
Publication statusPublished - Aug 2021


  • Biomarkers
  • Body mass index
  • Co-morbidities
  • Disease severity
  • Heart failure
  • Obesity
  • RISK
  • FAT


Dive into the research topics of 'Re-appraisal of the obesity paradox in heart failure: a meta-analysis of individual data'. Together they form a unique fingerprint.

Cite this