Background/Purpose : There is a high prevalence of premature atherosclerosis among patients with SLE, with a risk 7-9  times that of the general population. The traditional Framingham risk factor underestimates the risk for coronary artery disease (CAD) in patients with SLE. It has been suggested that a modified Framingham risk score (FRS) where each item is multiplied by 2 more accurately identifies patient at Moderate/High risk of CAD, and more accurately predicted subsequent CAD. The aim of this study was to determine whether the modified FRS (mFRS) more accurately reflected the prevalence of CAD(MI, angina, pacemaker) among patients in an inception cohort.

Methods : An inception cohort of SLE patients from 31 centres from 12 countries has been assembled according to a standardized protocol between 2000 and 2012 to study risk factors for atherosclerosis. Only patients with all variables necessary to calculate the FRS at enrolment and who did not have diabetes mellitus were included in this analysis.  Patients are followed at yearly intervals according to a standard protocol which included demographics, disease characteristics and classic risk factors for CAD as well as CAD events. Diagnosis of an event was confirmed using standard clinical criteria, relevant laboratory data and imaging where appropriate Sensitivity and Specificity (95% CI) of FRS and MFRS were evaluated in their prediction of future CAD.

Results: At enrolment 853 patients had sufficient data to calculate FRS. Of these, 140 patients had 8 years of follow-up available. 85% female, 50% Caucasian, 11.4% Black, 20.7% Asian 16.4% Hispanic and 1.4% other. 13.6% were current smokers, 22.2% past smokers. Age at diagnosis was 34.2 yrs and disease duration at enrolment was 5.0 mos. BMI was 24.4, and 25.7% were obese. 35%, were hypertensive and 42% had hypercholesterolemia. The 140 patients did not differ from those not followed for 8 years in either demographic features, disease characteristic, atherosclerotic risk factors. Table shows the calculated classic FRS and mFRS for the 140 patients.

Of the 140 patients 14 subsequently developed CAD, 8 of which are attributed to AS. The sensitivity of the FRS for CAD due to AS was 25.0 (3.2, 65.1) and specificity 97.7 (93.5, 99.5), whereas for the mFRS the sensitivity rose to 50.0 (15.7, 84.3) while the specificity decrease slightly to 86.4 (79.3, 91.7)

Conclusion: The mFRS, where each item is multiplied by 2, more accurately identifies patients at Moderate/High Risk of CAD. It provides higher sensitivity with little loss in specificity. Therefore the mFRS could be used to identify SLE patients for more intensive risk factor modification.