The 'Echo Heart Failure Score': an echocardiographic risk prediction score of mortality in systolic heart failure.

Aims Although many transthoracic echocardiographic (TTE) measurements have been shown to predict outcome in heart failure (HF), whether incremental risk prediction is afforded by their combination is unknown. We developed a simple echocardiographic risk score of mortality in HF patients. Methods and results We performed TTE in 747 systolic HF patients followed-up for 34 +/- 23 months. The Cox hazard model was used to evaluate the association between 14 TTE parameters and death. The Echo Heart Failure Score (EHFS) was derived by assigning the value of 1 to each independent predictor when present, and 0 when it was absent, and then by summing the number. The 3-year risk prediction improvement was tested by adding the EHFS to a model containing clinical predictors, and by calculating the C index and net reclassification improvement (NRI). Five baseline TTE variables (end-systolic volume index, left atrial volume index, mitral E-wave deceleration time, tricuspid annular peak systolic excursion, and pulmonary artery systolic pressure) remained independent predictors of mortality. The mortality rate (per 100 patients/year) significantly increased with EHFS ranging from 0 to 5 (EHFS = 0, 2.7%; 1, 5.2%; 2, 10.1%; 3, 13.7%, 4, 29.7%; 5, 36.9%; P < 0.0001). Patients with EHFS >= 3 had a mortality hazard ratio of 3.58 (95% confidence interval 2.74-4.78) compared with EHFS <3. Adding EHFS to the base model improved the C index (from 0.74 to 0.81, P < 0.0001), yielding a continuous NRI of 0.63 (P < 0.0001).