Publications

Introduction and Aims: Patients with heart failure have a poor prognosis: approximately 40% of patients die within the first year and 10% per year thereafter. Medication can improve the prognosis but many patients remain sub-optimally treated because, after an initial improvement, physicians may be reluctant to further increase drugs because of potential side-effects.
The natriuretic peptide BNP and NT-proBNP are biomarkers for heart failure which can provide feedback about cardiac function that can be used in combination with the patient's clinical condition to facilitate the decision of whether a further increase in medication might be beneficial.
Objective: to assess whether treatment guided by serial BNP or NT-proBNP monitoring improves outcomes compared with treatment guided by clinical assessment alone, in both primary and secondary care settings.
Method: Randomised-controlled trials comparing BNP or NT-proBNP-guided treatment of heart failure were identified through the EMBASE, MEDLINE and Cochrane-controlled-trials-register databases, with no language restrictions.
Results: Eleven trials involving a total of 2430 patients were included. Nine trials compared structured clinical care with and without BNP monitoring. Two trials had a third arm described as ‘usual care’. Preliminary results show that monitoring with BNP vs structured clinical care appears effective in reducing all-cause mortality; RR 0.77 (95% CI 0.63-0.94). However monitoring with structured clinical care vs usual care appears very effective in reducing all-cause mortality; RR 0.53 (95% CI 0.36-0.78). At least some of the impact of BNP monitoring may be due to co-interventions within structured clinical care which occur differentially between the study arms.
Conclusion: BNP monitoring includes powerful co-interventions, which impact on trial design, interpretation of results and implementation in routine practice.