Background/Purpose: Accelerated atherosclerosis remains the major late cause of death in SLE.  Yet, the “traditional” cardiovascular risk equations (Framingham, Reynolds, SCORE) consistently underestimate the risk.  This may lead to under-recognition and under-treatment.  We sought to construct a data-driven risk equation of cardiovascular risk in SLE, based on data collected in a longitudinal cohort. 

Methods: To derive the score, risk factors were calculated based on variables measured in the first two years of cohort participation (mean systolic blood pressure, mean SLEDAI, etc).  Cox Proportional Hazards models were constructed to determine the variables that affected the risk of a subsequent CVE. Using the results, a formula to calculate the risk of a CVE within the next 10 years was derived.  There were 1342 patients, 93% female, 56% Caucasian, 38% African-American, and 6% other ethnicities.  There were 109 cardiovascular events:  52 strokes, 26 MI, 18 angina/CABG, and 13 claudications.

Results:

Table 1 shows the estimates for the association between predictors and risk of a CVE.

Table 1	Beta	Hazard Ratio (95% CI)	P-value	Integer Score
Age (for each  5 years over 40)	0.252	1.29 (1.15, 1.44)	<0.0001	1
Male (vs. female)	0.552	1.74 (0.98, 3.09)	0.060	2
Systolic Blood Pressure 140 or more	0.791	2.21 (1.37, 3.56)	0.0012	3
Cholesterol over 160	0.743	2.10 (1.16, 3.83)	0.015	3
Smoking	0.677	1.97 (1.31, 2.96)	0.0012	3
Diabetes	0.407	1.50 (0.91, 2.48)	0.11	2
Mean SLEDAI of 2 or more	0.730	2.08 (1.22, 3.53)	0.0069	3
History of Lupus Anti-coagulant	0.744	2.10 (1.40, 3.16)	0.0003	3
Low Mean C3	0.583	1.79 (1.14, 2.82)	0.012	2

Using this model, the risk of a CVE within 10 years is 1-0.9875^{(Hazard Ratio)}.  For example, if someone is 50 years of age, male, with high systolic blood pressure, then the hazard ratio is (1.0510)x(1.74)x(2.21) = 6.26.  The risk of a CVE in 10 years is then 1-0.9875^(6.26) = 7.6%.  In the absence of SLE risk factors, the estimated 10-year risk from our formula is higher than would be projected based on the Framingham formula.  This is especially true if there are SLE-related risk factors.  Table 2 shows a few scenarios. 

Table 2:  Examples

	Estimated 10-year risk based on our formula	Estimated 10-year risk based on Framingham formula
Woman, age 50, SBP=150, Chol=150	4.5%	4.7%
Woman, age 50, SBP=150, Chol=220	9.2%	7.8%
Woman, age 50, SBP=150, Chol=220, Lupus Anticoagulant	18.3%	7.8%
Woman, age 50, SBP=150, Chol=220, High disease activity	18.1%	7.8%
Woman, age 50, SBP=150, Chol=220, Low complement	15.8%	7.8%

Using the approach employed by the Framingham study, the regression coefficients were rounded to integers (Table 1) to create a simpler score. The score is the sum of the number of points earned for each risk factor.  For example, a 50 year old man with high SBP would get a score of 2+2+3=7.  Then, to calculate the 10 year risk, use Table 3. 

Table 3:  Percent risk of a CVD within 10 years

Score	0	1	2	3	4	5	6	7	8	9	10
% Risk	1.3	1.6	2.1	2.6	3.4	4.3	5.5	7.0	8.9	11.3	14.2

Conclusion: A data-driven SLE Cardiovascular Risk Score can better estimate 10-year cardiovascular risk than the Framingham equation.  Its use can lead to appropriate use of imaging and intervention.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/systemic-lupus-erythematosus-cardiovascular-risk-equation/