Left atrial reverse remodelling predicts prognosis in patients with acute decompensated heart failure

Abstract

We have read a recent article titled ‘Prognostic value of left atrial reverse remodelling in patients hospitalized with acute decompensated heart failure’, with great interest.¹ This study is significant as it provides insights into the left atrial reverse remodelling (LARR) that predicts long-term clinical outcome in patients with acute decompensated heart failure (ADHF). While recognizing the value of this study, we would like to make the following comments.

Firstly, although the Cox regression model is widely acknowledged for its utility, it may inadvertently lead to risk overestimation with potential competing risks. Consequently, for outcome prediction explored in this paper, especially when different variables are potentially interrelated, employing a competing risks model appears more fitting. Traditional survival analysis techniques might not adequately account for the influence of secondary events on the primary study outcome, whereas the competing risks model affords a more holistic analytical viewpoint.²

Secondly, the authors' inclusion of important information such as demographic information and laboratory tests to adjust for potential covariates is commendable. However, we suggest a further expansion of the scope of covariates. ADHF patients are at risk of developing cardiogenic shock. Considering studies have shown the association of cardiogenic shock with poor prognosis, it can be inferred that history of cardiogenic shock might be an important covariate affecting the clinical outcome.³ Additionally, factors such as race and physical condition should also be considered for a more comprehensive assessment of the stability and reliability of the results.

Last but not the least, diabetes is a well-recognized prognostic factor for morbidity and mortality in patients with heart failure.⁴ Conducting subgroup analyses could offer more nuanced insights. Furthermore, incorporating the age of diabetes onset, a predictor in the SCORE2-diabetes model, would be a valuable addition to these findings.⁵

This article is a significant step forward in our understanding of the relationship between LARR and clinical outcome. These findings have important implications for estimating the prognosis of ADHF patients. A more comprehensive prediction could be an intriguing topic for further investigation.

The authors declare that no potential conflict of interest.