The diagnostic accuracy and clinical utility of natriuretic peptide based screening for the detection of incident heart failure in the community: a systematic review and meta-analysis

Talk Code: 
Clare Goyder
Goyder CR, Jones NR, Roalfe AK, Plumptre C, Taylor KS, Hobbs FDR, Taylor CJ
Author institutions: 
Nuffield Department of Primary Care Health Sciences, University of Oxford, UK


Globally, approximately 26 million people are estimated to be living with heart failure (HF). The prevalence is also expected to rise further with population aging. HF is also associated with very high levels of morbidity and mortality, indeed patients with HF are now known to have a higher risk of death than people with most common types of cancer. There are multiple challenges to the accurate and timely diagnosis of HF in the community and most cases are only diagnosed in secondary care. Given these difficulties, and the importance of detection at an early stage, including increasing evidence to support preventative strategies, recent attention has focused on potential population screening. The role that natriuretic peptides (NP) might play in HF screening is controversial. Although both Canadian and American guidelines now advocate NP screening in high risk groups, the diagnostic accuracy of NPs in this context has not been established. This is the first systematic review to assess the accuracy of NP screening for incident HF and the potential impact that this may have.


Study design: Systematic review and meta-analysis. Data sources: Ovid Medline, Embase, Cochrane Database of Systematic Reviews, Cochrane CENTRAL, DARE, Science Citation Index. Included studies: Diagnostic accuracy studies, randomised controlled trials and observational studies including cohort studies. Outcomes: Diagnostic accuracy of NP screening in comparison with echocardiography, including sensitivity, specificity, PPV, NPV, diagnostic odds ratios, ROC curves. Clinical utility will be assessed by examining outcomes such as mortality, morbidity, hospital admissions and changes in quality of life measures.


2262 records were identified for inclusion. Following title and abstract screening, 211 full text studies were assessed for eligibility and 85 were included. There were 29 studies that examined diagnostic accuracy. In total this included 21213 participants, recruited from both high risk and low risk populations. 15 studies looked at NTproBNP, 11 studies at BNP and 8 studies analysed both biomarkers, further data analysis is currently ongoing. 56 studies assessed the clinical utility of NP screening, further data extraction and analysis is planned.


This is the first systematic review to assess how accurately NP levels detect incident heart failure and to determine what is actually changed by early diagnosis both for patients and health care systems. It may have important implications for developing future pathways in primary care HF prevention.

Submitted by: 
Clare Goyder
Funding acknowledgement: 
Clare Goyder and Nicholas Jones are Wellcome Trust doctoral fellows. Andrea Roalfe is supported by the NIHR Biomedical Research Centre Oxford (BRC) Oxford University Hospitals NHS Foundation Trust. Kathy Turner receives funding from the NIHR programme for applied research. Richard Hobbs acknowledges support from the NIHR School for Primary Care Research, NIHR CLAHRC Oxford, the NIHR Oxford BRC and Harris Manchester College, Oxford. Clare Taylor is an NIHR academic clinical lecturer.