Can the Seattle Heart Failure Model Be Used to Risk Stratify Patients for an LVAD: Application to REMATCH Patients

Article Outline

• Copyright

Purpose: In the REMATCH trial left ventricular assist device (LVAD) therapy in patients with severe heart failure resulted in a 48% reduction in mortality compared to medical management. The ACC/AHA guidelines recommend a <50% 1 year survival for candidacy for LVAD implantation and CMS requires a survival of ≤2 years. The purpose of this analysis was to determine if the Seattle Heart Failure Model (SHFM-SeattleHeartFailureModel.org), a previously validated multivariate risk model, can be used to estimate the expected survival if the patient remained on medical therapy. This knowledge may be useful in selecting appropriate patients for LVADs. Methods and Materials: The SHFM was applied prospectively to data from the REMATCH trial. Variables missing in REMATCH were imputed from the primary database used to derive the SHFM. The use of inotropic agents and intra aortic balloon pump/ventilator were prospectively added to the previously published model. A hazard ratio of 1.17 was used for each inotrope used (COCHRANE meta analysis) and 2.92 for IABP/Ventilator use (COCPIT Model). Results: The SHFM was highly predictive of outcomes (P=0.0004). The use of dobutamine and/or milrinone had a similar hazard ratio as the COCHRANE meta analysis (1.09 95% CI 0.78–1.53, p=0.60 vs anticipated hazard ratio 1.17), although not significant. The use of IABP/Ventilator was also associated with risk (hazard ratio 2.1, 95% CI 1.1–3.9, p=0.02), similar to the anticipated hazard ratio of 2.92 from the COCPIT model. The addition of these variables to the SHFM improved the accuracy (1 year ROC 0.69 to 0.71). The 1 year predicted vs actual survival for the medical group (30% vs 28%) and LVAD group (49% vs 52%) were similar. 81% of patients had SHFM≤50% 1 year survival. There was no evidence that the benefit of the LVAD varied in the lower vs. higher risk patients. The lowest risk tertile had a hazard ratio of 0.55 (95% CI 0.28–1.09, p=0.09). Conclusions: The Seattle Heart Failure Model can be used to risk stratify end-stage heart failure patients, provided known markers of increased risk such as inotrope use and IABP/Ventilator support are included. This model may facilitate identification of high risk patients to evaluate for potential LVAD implantation prior to inotrope or IABP dependency.