BUENOS AIRES – Traditional tools such as the Framingham risk score have long been known to underestimate cardiovascular risk in people with systemic lupus erythematosus.
At the international congress on systemic lupus erythematosus, Dr. Michelle Petri of Johns Hopkins University, Baltimore, presented data in support of a novel risk-assessment formula to calculate the 10-year risk of a cardiovascular event (CVE). The tool combines the traditional cardiovascular disease risk factors of age, sex, high blood pressure, high cholesterol, smoking, diabetes, and body mass index (BMI), with three lupus-specific factors: a SLEDAI (SLE Disease Activity Index) score of 2 or higher, lupus anticoagulant, and a low mean complement 3 (C3) level.
Under both this and the Framingham algorithms, a 50-year old woman with SLE with a BMI of 23 kg/m2, systolic blood pressure of 150 mm Hg, and high cholesterol has about an 8% 10-year risk of a CVE. However, if she also has lupus anticoagulant, high disease activity, or a low C3 level, her risk rises to 15%-18% under the SLE-specific system, whereas her Framingham score, which does not measure these, would remain at 8%.
Courtesy of Keith Weller/Johns Hopkins Medicine
Dr. Michelle Petri
Dr. Petri told the conference that with rheumatoid arthritis, the advice has been to multiply an existing risk score by 1.5 when the patient has certain evidence of disease. "I thought we could do better than that in lupus, because we could derive a formula using actual data," she said.
"We know that traditional CV risk factors do predict actual events and also predict the progression of subclinical atherosclerosis. Even after we adjust, though, for every traditional cardiovascular risk factor, lupus patients still have a twofold increase in coronary calcium over controls – so obviously lupus-specific factors must be in the formula as well."
The data used to derive the new scoring system came from a cohort of 1,342 lupus patients (93% female) treated at the Johns Hopkins Lupus Center. None had a history of CVEs during the first 2 years of follow-up. Over the study period, 109 incident CVEs occurred in the cohort, including 52 strokes and 26 myocardial infarctions.
Dr. Petri and her colleagues used Cox proportional hazards models to determine the baseline variables affecting the risk of a subsequent CVE. Using the results, they derived a formula to calculate the 10-year risk of a CVE. For each of the risk factors, they investigators converted the hazard ratio into a simple integer score to simplify calculation in clinical practice. By adding up the integers and using a slide rule that stands in for the exponential equation, clinicians can translate the integer score into a risk percentage.
Dr. Petri acknowledged as limitations of the study its single-site design; the fact that the Johns Hopkins Lupus Center does not routinely perform cholesterol screening; and that the study reflects care from 1988 onward, where more recent data could reflect a better standard of care. "This needs to be independently validated," she said.
In another multivariate analysis of cardiovascular disease (CVD) risk factors in lupus, a cohort of 306 consecutive SLE patients at a clinic in Istanbul, Turkey, revealed pericarditis, lymphopenia, thrombocytopenia, and psychosis to be significantly associated with nonfatal CVD.
Dr. Murat Inanc of Istanbul University, the lead author of the study, said that 15.2% of the cohort (mean age, 40.2 years; 89% female) had clinical evidence of CVD. Many of them also had traditional risk factors. "The usual suspects are there – hypertension, high cholesterol, metabolic syndrome – but we also found disease-specific features more present in patients with CV involvement," Dr. Inanc said.
Patients with evidence of CVD were older and had longer disease duration, significantly more organ damage, and a higher number of American College of Rheumatology criteria for SLE. "We think disease severity takes almost equal importance with traditional risk factors in this population," he said. Treatment with cyclophosphamide was also seen as a risk factor, although it is likely related to disease severity, he told the congress.
Dr. Inanc noted that the conclusions that can be drawn from the study are limited by its single-site design and lack of data on subclinical CVD. Cumulative corticosteroid use was not investigated, and investigators did not have information on immunosuppressive use in the cohort.
Dr. Murray Urowitz
Dr. Murray Urowitz, of the Toronto Western Research Institute, presented findings from an international, multicenter inception cohort of 1,401 SLE patients. The goal of the study was to determine factors at diagnosis that are predictive of the development of premature atherosclerosis. Dr. Urowitz and his colleagues looked at age, sex, diabetes, smoking, obesity, low-density lipoprotein cholesterol, and creatinine. A multivariate analysis showed that only age and male sex were significant risk factors for atherosclerotic vascular events after a mean 5 years’ follow-up. "The implication here is these risk factors take more time to have their maximum effect, and in the first decade it’s hard to have more than age and sex," Dr. Urowitz told the congress.