Article Text


Brain natriuretic peptide is a potentially useful screening tool for the detection of cardiovascular disease in patients with rheumatoid arthritis
  1. S M J Harney1,
  2. J Timperley2,
  3. C Daly3,
  4. A Harin1,
  5. T James4,
  6. M A Brown1,
  7. A P Banning2,
  8. K Fox3,
  9. S Donnelly5,
  10. B P Wordsworth1
  1. 1University of Oxford Institute of Musculoskeletal Sciences, Botnar Research Centre, Oxford, UK
  2. 2Department of Cardiology, John Radcliffe Hospital, Oxford, UK
  3. 3Kim Fox Research Department, Royal Brompton Hospital, London, UK
  4. 4Department of Biochemistry, John Radcliffe Hospital, Oxford, UK
  5. 5Department of Rheumatology, St George’s Hospital, London UK
  1. Correspondence to:
    Dr S M J Harney
    Institute of Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Nuffield Orthopaedic Centre, Windmill Road, Headington, Oxford OX3 7LD, UK; sinead{at}

Statistics from

Patients with rheumatoid arthritis (RA) have a significantly higher risk of coronary heart disease, despite being less likely to report symptoms of angina, and are more likely to experience unrecognised myocardial infarction and sudden cardiac death than non-RA controls.1 Furthermore, left ventricular diastolic dysfunction has been described in up to 40% of patients with RA.2

Traditional risk factors partially account for this increased risk, and primary prevention is important in this high risk group. Electrocardiography (with or without stress testing) and echocardiography are commonly requested investigations to detect underlying cardiac disease, but access, particularly to echocardiography, is mostly limited to those who are symptomatic. Brain natriuretic peptide (BNP) is a new cardiac biomarker, which is increased in ventricular dysfunction, both systolic and diastolic, and also left ventricular hypertrophy. Measurement of serum BNP levels has become a powerful adjunct to diagnosis and prognostic stratification of patients with suspected ventricular dysfunction in recent years.

This study aimed at investigating the extent to which appropriate primary preventative measures were being used in a group of patients with RA, and investigating the potential use of measurements of serum BNP levels in screening patients with RA for occult cardiac disease.

One hundred and twenty consecutive outpatients fulfilling the 1987 American College of Rheumatology criteria for RA were recruited over a 12 week period. Demographic data, age of onset, and duration of RA, current/past drugs, disease activity, and cardiovascular risk factors were recorded using structured questionnaires (available on request). Blood was taken to determine C reactive protein, erythrocyte sedimentation rate, BNP, thyroid function, and random lipid and glucose profiles. A one way analysis of variance test was used for all normally distributed data and the Mann-Whitney test was used for non-parametric data. Results are presented as mean (SD) unless otherwise stated. Table 1 summarises the clinical characteristics of the 120 patients.

Table 1

 Summary of clinical characteristics of the 120 patients (74 female, 46 male) in the initial study

Twenty six patients with RA from the initial group, selected specifically for the absence of a previous physician diagnosis of hypertension or ischaemic heart disease or symptoms of cardiac disease were studied further by Doppler echocardiography and compared with 32 healthy controls. The mean (SD) age of this subgroup was 63 (9.4) years, and 50% were female. Systolic dysfunction (ejection fraction <50%) was evident in 7 (27%) and diastolic dysfunction in 11 (42%) patients. Left ventricular hypertrophy was present in 14 (54%; mean (SD) mass 212 (66) g). BNP levels were significantly higher in patients with RA (mean 9.2 pmol/l, range 0.6–52.6) than in controls (mean 2.5 pmol/l range 0.6–10.4) (p = 0.004). BNP levels correlated with end diastolic volume (r2 = 0.83, p = 3×10−7), end systolic volume (r2 = 0.62, p<0.0001), and left ventricular mass (r2 = 0.4, p = 0.0009). Although the patients were older (mean age 63) than the controls (mean age 50), the correlations with BNP remained highly significant after adjustment for age and other covariates (including sex, full blood count, C reactive protein, erythrocyte sedimentation rate, and renal function), by logistic regression. Using a cut off point of 5 pmol/l, the sensitivity and specificity of BNP for detection of systolic dysfunction was 70% and 64% and of diastolic dysfunction 60% and 69%.

In this study, occult cardiac dysfunction was present in a worrying proportion of asymptomatic patients with RA. BNP has the potential to be a useful marker of occult cardiac disease in this population, despite potential confounding by age, sex, and subclinical renal disease. The cost of this assay is about one-tenth the cost of an echocardiogram, and so it would be a cost effective initial screening test in patients with RA. Larger studies are needed to confirm this finding, with longitudinal follow up to ascertain its prognostic usefulness.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.