Evaluation of NT-proBNP and High Sensitivity C-Reactive Protein for Predicting Cardiovascular Risk in Patients with Arthritis Taking Longterm Nonsteroidal Antiinflammatory Drugs

Christian T. Ruff, David A. Morrow, Peter Jarolim, Fang Ren, Charles F. Contant, Amarjot Kaur, Sean Curtis, Loren Laine, Christopher P. Cannon, Kay BruneThe Journal of Rheumatology 2011 · DOI

Abstract

NT-proBNP is a simple and robust noninvasive indicator of CV risk in patients with arthritis. Risk stratification based on NT-proBNP may facilitate identification of patients with arthritis who are at low CV risk during chronic NSAID treatment.

Introduction

Patients suffering from rheumatoid arthritis (RA) and osteoarthritis (OA) often require long-term management with nonsteroidal anti-inflammatory drugs (NSAIDs). However, chronic use of these medications—particularly COX-2 selective inhibitors—has been linked to an elevated risk of major adverse cardiovascular events, including myocardial infarction (MI), stroke, and heart failure (HF).

Clinicians face a significant challenge in identifying which patients can safely tolerate long-term NSAID therapy. While high-sensitivity C-reactive protein (hsCRP) is a well-established cardiovascular (CV) risk marker in the general population, its utility in arthritis patients is questionable due to the systemic inflammation inherent to the disease. This study evaluates whether N-terminal pro-B-type natriuretic peptide (NT-proBNP), a marker of cardiac wall stress, provides a more reliable tool for risk stratification in this specific population.

Methodology

This research was conducted as a prospective, nested biomarker study using data from the Multinational Etoricoxib and Diclofenac Arthritis Long-term (MEDAL) trial. The study analyzed a sub-cohort of 6,273 patients with RA or OA who had available baseline blood samples.

Baseline levels of NT-proBNP and hsCRP were measured using automated immunoassays. Patients were stratified into quartiles based on NT-proBNP levels and into three clinical categories for hsCRP (0–1, 1–3, and >3 mg/L). The primary endpoint was a composite of CV death, MI, stroke, and HF over a 2-year follow-up period. Statistical analysis utilized Cox proportional hazards models to adjust for traditional risk factors such as age, smoking status, hypertension, and diabetes.

Results

The Predictive Power of NT-proBNP

NT-proBNP demonstrated a highly significant, graded relationship with all cardiovascular outcomes. Patients in the highest quartile (Q4) of NT-proBNP levels faced a 3.53-fold increased risk of the composite CV endpoint compared to those in the lowest quartile (Q1).

Incidence of cardiovascular outcomes by NT-proBNP quartiles, showing a strong graded relationship across all endpoints.

Specific findings for NT-proBNP included:

  • CV Death: Highly significant trend (p < 0.0001).
  • Heart Failure: Highly significant trend (p < 0.0001).
  • Myocardial Infarction: Significant trend (p = 0.02).

Furthermore, a simple clinical heuristic—checking if the NT-proBNP level (in pg/ml) exceeded the patient's age—proved to be an effective method for identifying high-risk individuals. Conversely, patients with NT-proBNP levels below 100 pg/ml had a very low event rate (0.94%) over two years.

The Failure of hsCRP

In stark contrast to its performance in the general population, baseline hsCRP levels failed to predict thrombotic events or the composite CV endpoint in this arthritis cohort.

Incidence of cardiovascular outcomes by hsCRP levels, demonstrating a lack of significant predictive value for major events.

As shown in the data, there was no significant association between hsCRP categories and the composite endpoint (p = 0.65). This suggests that in patients with active arthritis, hsCRP levels primarily reflect joint inflammation rather than underlying vascular risk, making it an unreliable "noisy" predictor for this group.

Discussion

The findings suggest that NT-proBNP acts as a functional "stress test" in a bottle. Because it is released by the heart in response to ventricular stretch and wall tension, it identifies subclinical cardiac dysfunction that traditional risk factors and inflammatory markers miss. Importantly, NT-proBNP remained a potent predictor even after adjusting for age, hypertension, and prior CV history, indicating it captures a unique dimension of risk.

While this study is limited by the lack of a non-NSAID placebo control and the use of single-point biomarker measurements, the strength of the NT-proBNP correlation is compelling. Future clinical guidelines for arthritis management should consider incorporating NT-proBNP screening to improve patient safety during chronic NSAID therapy.

Conclusion

NT-proBNP is a robust, independent predictor of cardiovascular risk in arthritis patients taking long-term NSAIDs. Unlike hsCRP, which is confounded by systemic inflammation, NT-proBNP provides a clear signal of cardiac stress, allowing clinicians to better identify patients at low risk for major cardiac events during treatment.

Summary updated

Abstract

NT-proBNP is a simple and robust noninvasive indicator of CV risk in patients with arthritis. Risk stratification based on NT-proBNP may facilitate identification of patients with arthritis who are at low CV risk during chronic NSAID treatment.
BibTeX
@article{ruff2011evaluation,
  author = {Christian T. Ruff and David A. Morrow and Peter Jarolim and Fang Ren and Charles F. Contant and Amarjot Kaur and Sean Curtis and Loren Laine and Christopher P. Cannon and Kay Brune},
  title = {{Evaluation of NT-proBNP and High Sensitivity C-Reactive Protein for Predicting Cardiovascular Risk in Patients with Arthritis Taking Longterm Nonsteroidal Antiinflammatory Drugs}},
  journal = {The Journal of Rheumatology},
  year = {2011},
  doi = {10.3899/jrheum.100880},
}
Figure