Reply to Comments on Relationship between Hypoalbuminemia on Admission and Long-term Mortality in Patients with Acute Decompensated Heart Failure

The Authors Reply We appreciate the interest of Dr. Imamura in our recent paper, as well as his valuable comments (1). In our study, hypoalbuminemia was defined as a serum albumin level of 3.4 g/dL because the median serum albumin level was 3.4 g/dL, which is generally accepted as the cutoff value (2-4). In response to Dr. Imamura’s initial comment, we performed a receiver-operating characteristic (ROC) curve analysis, and 3.8 g/dL was found to be the ideal cutoff value for predicting the outcome in our dataset. However, hypoalbuminemia at admission, defined as a serum albumin level of 3.8 g/dL, was not associated with the long-term outcomes [hazard ratio (HR), 1.21; 95% confidence interval (CI), 0.70-2.22; p=0.511 in the pre-match model, and HR, 0.97; 95% CI, 0.45-2.07; p=0.934 in the post-match model]. The duration of follow-up varied across individual cases on the basis of the relatively long enrollment period (i.e., 5 years) and thus the numbers at risk decreased with the passage of time. In the Kaplan-Meier survival analyses, the two curves crossed after several years, as Dr. Imamura pointed out. However, the proportional hazards assumption was confirmed in both the pre-match and post-match models using log-minus-log survival graphs. It is often difficult to show the relationship between parameters/conditions in the acute phase and the long-term outcomes. However, many reports have investigated the relationships between parameters/conditions in the acute phase and long-term outcomes (5, 6). Again, we thank Dr. Imamura for his interest and comments on our paper.