β-Trace Protein and Cystatin C as Predictors of Long-Term Outcomes in Patients With Acute Heart Failure
Do β-trace protein (BTP) and cystatin C, biomarkers of renal dysfunction, prognosticate long-term outcomes in patients with acute decompensated heart failure (ADHF)?
This was a prospective cohort study of 220 consecutive patients hospitalized with ADHF (systolic and diastolic failure). BTP, cystatin C, blood urea nitrogen (BUN), and serum creatinine levels were measured and estimated glomerular filtration rate (eGFR) was calculated using the MDRD formula. The accuracy of the aforementioned measures to predict a composite outcome of death and/or HF hospitalization was assessed using receiver-operating characteristic curves with area under the curve (AUC) comparison. Cox multivariable analysis was used to determine the hazard ratio (HR [95% confidence interval]) for correlates of the primary outcome.
Of the 220 patients studied, 116 (53%) met the primary outcome (62 deaths, 76 readmissions) over a median (interquartile range) 500 (231-796) days of follow-up. Serum creatinine, BUN, BTP, cystatin C, N-terminal pro-B-type natriuretic peptide (NT-proBNP), and troponin T levels were significantly higher in subjects reaching the primary endpoint, whereas eGFR was lower. As BTP and cystatin C levels increased, adverse outcomes were more frequent. However, using ROC analysis, no serum marker had excellent discrimination of adverse HF outcomes (all AUCs <0.63), and none of the above markers was superior to the others (p > 0.05 for AUC comparison). On multivariable analysis, age (HR, 1.04 [1.01-1.06]), ejection fraction (HR, 0.98 [0.97-0.99]), New York Heart Association class >II (HR, 2.1 [1.5-3.0]), Log10 leukocyte count (HR, 8.6 [2.2-33]), inotrope use (HR, 2.2 [1.3-3.7]), Log10 troponin T (HR, 2.7 [1.5-4.7]), and Log10 BTP (HR, 3.2 [1.2-8.9]) were predictive of the primary outcome. In separate analyses (done to avoid colinearity with BTP and other renal markers), Log10 cystatin C (HR, 4.2 [1.3-13]) was also predictive of adverse events, but creatinine, eGFR, NT-proBNP, and BUN were not.
The authors concluded that BTP and cystatin C are superior to standard measures of renal function in predicting death and/or HF hospitalization in ADHF.
Prognostication in ADHF remains challenging. While renal dysfunction has consistently been shown to be a negative risk factor for adverse HF outcome, accurately assessing renal impairment in subjects with HF is fraught with difficulty due to the impact of age, poor nutrition and muscle mass, and other likely unmeasured factors. The authors make very strong conclusions about BTP and cystatin C with regard to risk factor assessment in ADHF. First, AUCs for all of the markers studied were very poor (all <0.63), as were positive and negative predictive values. Second, the AUCs for BTP or cystatin C were not statistically superior to any of the other studied markers on AUC comparison. Thus, neither BTP nor cystatin C can be considered more accurate than the other measures studied. Finally, the multivariable model included more variables than recommended for the number of events, and the models were non-nested (i.e., repeatedly run) due to colinearity. Thus, the ability of eGFR, BTP, cystatin C, and creatinine to predict poor outcome following ADHF in relation to each other was really not assessed.
Keywords: Follow-Up Studies, Kidney Function Tests, Biological Markers, Heart Failure, Glomerular Filtration Rate, New York, Cystatin C, Natriuretic Peptide, Brain
< Back to Listings