Incremental risk stratification using the H2FPEF score in patients with chronic kidney disease and unexplained dyspnea
A Al-Rubai, J C Christensen, F S D Davidovski, N E L Landler, F J O Olsen, E L F B Ballegaard, I M H S Sorensen, S S B Bjergfelt, B F R Feldt-Rasmussen, S B Bro, D H Hansen, C C Christoffersen, T B S Biering-SorensenAbstract
Background
Heart failure with preserved ejection fraction (HFpEF) is a common comorbidity in patients with chronic kidney disease (CKD) but remains sparsely investigated. The H2FPEF-score constitutes an important clinical tool to estimate the risk of prevalent HFpEF, but its association with adverse cardiovascular events has not been assessed in patients with CKD.
Purpose
To assess whether the H2FPEF score provides independent prognostic information regarding adverse cardiovascular events in patients with CKD and unexplained dyspnea.
Methods
The study is based on the CPH-CKD ECHO cohort, a prospective two-center study enrolling 825 outpatients with non-dialysis dependent CKD. Patients were examined with echocardiography, blood sampling, and questionnaires at baseline and followed through national registries. Exclusion criteria for the present analysis included conditions that might lead to dyspnea including valve disease, HF, left ventricular ejection fraction <50%, pulmonary disease and pulmonary hypertension. Dyspnea was defined as a NYHA score of ≥2. Patients with unexplained dyspnea were grouped according to H2FPEF score. The primary outcome was a composite of all-cause mortality and incident HF. Uni- and multivariable cox regression was used to assess the association between the endpoint and H2FPEF score or group, both within subjects with dyspnea, and with subjects free of dyspnea serving as the reference. Multivariable Cox regression was adjusted for sex, EGFR-stage, and albuminuria level at baseline. Forest plots were created to illustrate results of the Cox regressions.
Results
In total, 461 patients were included (median age: 61.0 years [IQR: 48.7, 69.6]; male sex: 268 (58.1%)) of whom 188 patients experienced unexplained dyspnea while 273 did not. Of those with dyspnea, 65 subjects scored 0-1 points on the H2FPEF score, 101 subjects scored 2-4, and 22 subjects scored ≥5. Subjects with higher scores were older and had a higher prevalence of comorbidities and lower eGFR. In addition, they displayed higher E/e’ and left ventricular mass index (table 1). During a median follow-up period of 8.9 years [IQR: 7.7, 9.7], 40 (21.3%) subjects with dyspnea met the endpoint, while the same was true for 48 (17.6%) subjects without dyspnea. Among subjects with dyspnea, higher H2FPEF score was associated with an increased risk of the endpoint (HR = 1.51 per 1-point increase (95% CI 1.26-1.81), p<0.001) which persisted in adjusted analysis (HR = 1.49 (1.23-1.81), p<0.001). Compared with subjects without dyspnea, subjects with dyspnea and an H2FPEF score ≥5 had a significantly higher risk of the endpoint (figure 1).
Conclusion
In patients with non-dialysis dependent CKD and unexplained dyspnea, a higher H2FPEF score was independently associated with an increased risk of all-cause mortality and incident HF.Baseline and objective measuresForest plot over hazard ratios