ACCEL: STOP-HF: Can a Simple Screening and Management Program Prevent HF?

As cardiology legend Eugene Braunwald, MD, puts it, heart failure (HF) is "the last horizon of cardiovascular disease," and prevention of HF can be thought of as the "Holy Grail."

STOP-HF was a randomized, prospective, multicenter controlled trial aiming to, well, stop HF in its tracks. The study investigators noted that while disease management programs have been a major advance in HF care, the majority of such programs focus on the sicker segment of this population. The STOP-HF effort argues for a far wider application of disease management in HF so as to bring the benefits of specialist care, patient education, and regular follow-up to patients at an earlier stage in the natural history of the disease.1

Such expansion of this concentrated care approach would clearly provide access to a greater proportion of the HF and at-risk population to specialty-led strategies at critical time periods in the natural history of this syndrome. Do data justify this expansion? Prospective studies were clearly needed to clarify the incremental clinical benefit, if any, of screening for and aggressively managing risk factors to prevent HF development.

The St. Vincent's Screening TO Prevent Heart Failure (STOP-HF) trial is a unique prospective, longitudinal, randomized trial of about 2,000 Irish participants. The research program was started in 2004 with the main aim of seeing if people with HF risk factors can have their risk defined simply by the use of a blood test known as B-type natriuretic peptide (BNP), a protein released from the heart when it is under stress or strain. People with risk factors for developing HF, in particular those with high BP, high cholesterol, diabetes, or a prior MI were invited to partake in this study.

Preventing HF Before It Starts

STOP-HF asks whether an aggressive approach for people with high BNP can delay HF progression and CV outcomes. In the study, all patients received annual review and BNP assessment by their general practitioner, but the BNP results were not available to the doctors treating the control group. For those in the BNP-directed care arm, if measured BNP was >50 pg/ml at any time, patients would receive

  • shared-care with a cardiologist
  • cardiology review
  • echo-Doppler
  • other CV investigations
  • CV nurse coaching
  • regular cardiology follow-up

The primary endpoint was prevalence of HF (hospitalized) and asymptomatic LV dysfunction. For this primary endpoint, participants did significantly better overall in the intervention arm: 8.7% versus 5.3% in controls (p = 0.01).

Risk was higher overall with elevated BNP (any BNP >50 pg/ml), but again the intervention group showed significantly less risk of the primary endpoint: 9.5% versus 18.7% among controls (p = 0.003).The secondary endpoint was hospitalization for CV events (time to event and event rate).

For the secondary endpoint of time to first major acute cardiac event, patients in the intervention group did better overall (p = 0.04). However, across the 7 years of follow-up, the difference did not reach statistical difference when analyzed by any elevated BNP (p = 0.07), despite the fact that the Kaplan-Meier curves separated early and continued to separate from roughly year 2 to year 6.

BNP did increase over time, but this seemed to be attenuated in the intervention arm (26% vs. 13%; not significant).

Thus, the STOP-HF investigators concluded that BNP-based screening and collaborative care ended up targeting four in 10 at-risk patients and reduced the rates of LV dysfunction, HF, and emergency hospitalizations for major CV events. Or, as an early editorial put it that helped introduce the STOP-HF concept several years ago, disease management programs for HF may not be just for the "sick" HF population.1


1. McDonald K, Conlon C, Ledwidge M. Eur J Heart Fail. 2007;9:113-7.

Keywords: Cholesterol, Heart Failure, Disease Management, Risk Factors, General Practitioners, Diabetes Mellitus, Natriuretic Peptide, Brain

< Back to Listings