Does the right ventricle play a role in extended length of hospital stay of patients with heart failure with preserved ejection fraction?

Introduction

Extended length of hospital stay (LOS) of patients with heart failure with preserved ejection fraction (HFpEF) was correlated with higher in-hospital mortality, increased risk of rehospitalization, worse long-term quality of life and survival. Comorbidity burden and heart failure severity were the main drivers of extended LOS in HFpEF patients in previous studies, however scarce data is available on implications of the right ventricular (RV) performance in determining prolonged hospitalization of these patients.

Purpose

We aimed to assess the correlation of the basic RV echocardiographic parameters with extended LOS of HFpEF patients.

Methods

In this retrospective cohort study, we included HFpEF patients consecutively admitted to our cardiology department between January 2019-December 2019, after excluding acute coronary syndrome, pulmonary embolism, readmissions of the same patient, in-hospital death, and absence of echocardiographic data.

We determined extended LOS as a duration greater than the mean number of days of hospitalization of the entire cohort, respectively over 5 days.

For continuous variables associated with extended LOS, ROC analysis was used to determine cut-off values, sensitivity (Se), specificity (Sp), based on the Youden index.

Results

We analyzed 121 HFpEF patients with a mean age of 72.21±9.68 years, 38.22% male prevalence. 35% had an extended LOS.

Increased prevalence of extended LOS was proportional with increasing NYHA class from 30.2% in class II, 34.6% in class III and 100% in class IV (p for trend 0.037).

In ROC analysis, the following biomarkers were correlated with an extended LOS: NT-proBNP (AUC 0.67, 95%CI0.58–0.75, cut-off > 934pg/ml, Se 74%, Sp 56%, p=0.0014), RV diameter (AUC 0.62, 95%CI 0.53–0.71, cut-off >34mm, Se 48%, Sp 71%), tricuspid annular plane systolic excursion (TAPSE) (AUC 0.67, 95%CI 0.58–0.75, cut-off <=20mm, Se 79%, Sp 55%, p=0.001), pulmonary artery estimated systolic pressure (PASP) (AUC 0.65, 95%CI 0.56– 0.73, cut-off > 35mmHg, Se 67%, Sp 65%, p=0.007), TAPSE/PASP (AUC 0.68, 0.59–0.76, cut-off <=0.53, Se 70%, Sp 67%, p=0.0045), hemoglobin levels (AUC 0.60, 95%CI 0.51–0.69, p=0.07), serum potassium (AUC 0.60, 95%CI 0.51–0.69, p=0.07), leukocyte count (AUC 0.66, 95%CI 0.52–0.77, p=0.07).

Cardiovascular and non-cardiovascular conditions such as history of myocardial infarction, stroke, diabetes mellitus, hypertension, COPD, cirrhosis, chronic kidney disease, infections or malignancy were not associated with extended LOS.

In multivariable analysis after a stepwise approach, the final independent predictors of extended LOS were NTproBNP>934 pg/ml (HR 2.65, 95%CI 1.11–6.33, p=0.028) and TAPSE/PASP <0.46 mm/mmHg (HR 2.90, 95%CI 1.23–6.85, p=0.01).

Conclusion

In our cohort of HFpEF patients, the main determinants of extended LOS were the severity of heart failure expressed by the increased NT-proBNP levels and the RV-pulmonary artery uncoupling expressed by a reduced TAPSE/PASP.