Monitoring clinical changes in patients with heart failure: a comparison of methods

Although monitoring the clinical status of patients with heart failure rests at the core of clinical medicine, the ability of different techniques to reflect clinical change has not been evaluated. This study sought to describe changes in various measures of disease status associated with gradations of clinical change.

A prospective, 14-center cohort of 476 outpatients was assessed at baseline and 6 +/- 2 weeks to compare changes in 7 heart failure measures with clinically observed change. Measures included health status instruments (the Kansas City Cardiomyopathy Questionnaire [KCCQ], Short Form-12, and EQ-5D), physician-assessed functional class (New York Heart Association [NYHA]), an exercise test (6-minute walk), patient weight, and a biomarker (B-type natriuretic peptide). Cardiologists, blinded to all measures except weight and NYHA, categorized clinical change ranging from large deterioration to large improvement.

The KCCQ, NYHA, and 6-minute walk test were most sensitive to clinical change. For patients with large, moderate, and small deteriorations, the KCCQ decreased by 25 +/- 16, 17 +/- 14, and 5.3 +/- 11 points, respectively. For patients with small, moderate, and large improvements, the KCCQ increased by 5.7 +/- 16, 10.5 +/- 16, and 22.3 +/- 16 points, respectively (P < .01 for all compared with the no change group). New York Heart Association and 6-minute walk distance were significantly different for those with moderate and large changes (P < .05) but neither revealed a difference between those with small versus no clinical deterioration. The KCCQ had the highest c statistic for monitoring individual patients, followed by NYHA and 6-minute walk.

The KCCQ, followed by the NYHA and the 6-minute walk test, most accurately reflected clinical change in patients with heart failure.