Hazard ratios (HRs) and 95% confidence intervals (95% CIs) are based on the likelihood of developing a coronary artery disease event with each unit increase in mean high-sensitivity C-reactive protein (hsCRP) (log10 transformed; original units mg/liter), time-adjusted hsCRP (log10 transformed; original units mg/liter), or hsCRP quartile in relation to each unit increase in the Framingham Risk Score and adjusted mean Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K).
High-sensitivity C-reactive protein as a marker of cardiovascular risk in systemic lupus erythematosus
Version of Record online: 27 AUG 2012
Copyright © 2012 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 64, Issue 9, pages 3052–3053, September 2012
How to Cite
Nikpour, M., Harvey, P. J., Ibanez, D., Gladman, D. D. and Urowitz, M. B. (2012), High-sensitivity C-reactive protein as a marker of cardiovascular risk in systemic lupus erythematosus. Arthritis & Rheumatism, 64: 3052–3053. doi: 10.1002/art.34541
- Issue online: 27 AUG 2012
- Version of Record online: 27 AUG 2012
- Accepted manuscript online: 21 MAY 2012 10:55AM EST
In the general population, high-sensitivity C-reactive protein (hsCRP), a marker of inflammation, is a predictor of coronary artery disease (CAD) (1). Systemic lupus erythematosus (SLE) is an autoimmune inflammatory disease with a markedly increased risk of CAD (2). While in healthy subjects, levels of hsCRP are relatively stable over time, in patients with SLE, hsCRP takes a dynamic course, and levels fluctuate due to changes in disease activity, treatment, and infections (3, 4). This variability of hsCRP has cast doubt over its usefulness as a predictor of CAD in SLE.
In a prospective cohort study, we observed patients at the University of Toronto lupus clinic for the occurrence of incident CAD events (myocardial infarction [MI] and angina). Consent was obtained from all study participants. Levels of hsCRP (mg/liter) were measured in serum samples collected at baseline and at every visit during the study. The time-adjusted mean of all hsCRP measurements from each patient was calculated as the area under the curve of hsCRP against time, divided by the time from first to last measurement. Cumulative disease activity at each visit coinciding with hsCRP measurement was quantified using the adjusted mean Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) (5, 6).
Among 472 study participants, 89% were women, with a median SLEDAI-2K score and a median Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (7) score at first hsCRP measurement of 4.0 and 1.0, respectively. Mean ± SD disease duration at baseline was 13.8 ± 10.1 years. Analyses were based on a mean ± SD of 5.3 ± 2.5 serial hsCRP measurements per patient (2,523 hsCRP measurements in total). The time interval between hsCRP measurements was 1.3 ± 0.7 years.
Over a mean ± SD followup period of 3.8 ± 5.4 years (2,535.70 person-years in total), 20 new CAD events (8 MI, 12 angina, 0 sudden cardiac death) occurred. In univariate analyses, patients who experienced CAD events were more likely to be older (mean ± SD 47.6 ± 13.0 versus 41.4 ± 14.1 years; P = 0.047) and menopausal (62.5% versus 36.9%; P = 0.04) at study entry. Patients with CAD events had significantly higher 10-year Framingham Risk Scores at study entry (5.40 ± 5.17 versus 2.52 ± 3.67; P = 0.001) (8).
Results of multivariable regression analyses for CAD events are presented in Table 1. Mean and time-adjusted mean hsCRP were significantly predictive of CAD events. At each visit, higher hsCRP quartiles (as defined for the general population based on relative risk of CAD at 5 years) (9) were associated with a significantly increased likelihood of CAD events at subsequent visits (Table 1), with a level of ≥1.6 mg/liter denoting patients at substantially increased risk of CAD (hazard ratio 3.37 [95% confidence interval 1.09–10.41], P = 0.04). In each of these models, the Framingham Risk Score and adjusted mean SLEDAI-2K score were also significantly predictive of CAD events.
|Variable||Time-dependent covariate model|
|Mean hsCRP||Time-adjusted mean hsCRP||hsCRP quartiles†|
|HR (95% CI)||P||HR (95% CI)||P||HR (95% CI)||P|
|hsCRP||1.58 (1.05–2.39)||0.03||1.61 (1.06–2.43)||0.03||1.57 (1.02–2.43)||0.04|
|Framingham Risk Score||1.15 (1.06–1.26)||0.002||1.15 (1.05–1.26)||0.002||1.15 (1.05–1.27)||0.0025|
|Adjusted mean SLEDAI-2K‡||1.13 (1.04–1.24)||0.006||1.13 (1.04–1.24)||0.006||1.12 (1.02–1.22)||0.013|
A limitation of this study is that we were unable to evaluate the role of other cardiovascular risk factors such as obesity, physical inactivity, and metabolic syndrome. However, our findings have potential implications for clinical practice, making a case for measuring hsCRP in patients with SLE in order to identify those most likely to develop clinical CAD. Further studies are needed to confirm our findings and provide recommendations for management of elevated hsCRP levels in SLE.
Supported by the Centre for Prognosis Studies in the Rheumatic Diseases, the Smythe Foundation, the Ontario Lupus Association, the Lupus Flare Foundation, and the Lupus Society of Alberta. Dr. Nikpour's work was supported by the Arthritis Centre of Excellence and the Geoff Carr Lupus Fellowship.