Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


Multivariate Analysis of Vitamin D Insufficiency and Deficiency in SLE: The SOLD (Systemic Lupus Associates of Low Vitamin D) Project.

D'Souza2,  Caroline, Bello2,  Kayode Jibril, Fang2,  Hong, Magder3,  Laurence, Petri1,  Michelle A.

Timonium, MD
Johns Hopkins University
University of Maryland

Purpose:

Vitamin D is an important immunomodulator in animal models of autoimmune diseases. In humans, in addition to its role in bone health, it may be important in the prevention of diabetes, cardiovascular disease, and certain cancers. We explored clinical and laboratory associates of Vitamin D insufficiency (<32 ng/mL) and deficiency (<15 ng/mL) in SLE.

Methods:

810 SLE patients (mean age 44.94, 49.54% Caucasian, 40.36% African-American, 92.07% female) were included. 25 OH Vitamin D was measured from July 20, 2009 to April 22, 2010.

Results:

Univariate analyses suggested that age, ethnicity, hypertension, diabetes, physician global assessment of SLE activity (PGA), urine protein/creatinine ratio (>0.5), systolic blood pressure, cholesterol, and obesity were associated with both Vitamin D insufficiency and Vitamin D deficiency.

Table 1. Association between various factors and deficiency/insufficiency of Vit D in patients with SLE based on a multivariable logistic regression model

Variable Odds Ratio (95% CI) with Vit D DeficiencyP-ValueOdds Ratio (95% CI) with Vit D InsufficiencyP-Value
Age18–301.00 (reference) 1.00 (reference) 
 31–490.83 (0.47; 1.47)0.5241.20 (0.77; 1.85)0.419
 50+0.40 (0.21; 0.78)0.0070.63 (0.39; 0.99)0.048
RaceWhite1.00 (reference) 1.00 (reference) 
 Black5.92 (3.76; 9.33)0.0012.18 (1.57; 3.04)0.001
 Other1.89 (0.80; 4.51)0.1491.22 (0.68; 2.18)0.501
Urine Pr/Cr Ratio<0.51.00 (reference) 1.00 (reference) 
 >=0.51.01 (0.54; 1.91)0.9671.71 (0.90; 3.24)0.102
BMI<=301.00 (reference) 1.00 (reference) 
 >301.85 (1.23; 2.78)0.0032.51 (1.79; 3.53)0.001
Cholesterol<=2001.00 (reference) 1.00 (reference) 
 >2001.43 (0.93; 2.21)0.1021.79 (1.26; 2.55)0.001
Systolic BP<1401.00 (reference) 1.00 (reference) 
 >=1401.60 (0.97; 2.63)0.0651.60 (1.02; 2.51)0.039
DiabetesNo1.00 (reference) 1.00 (reference) 
 Yes2.05 (1.13; 3.69)0.0171.49 (0.85; 2.61)0.165

However, in the best multivariate logistical regression model (see table 1), age, ethnicity, and obesity remained statistically significant, but urine protein/creatinine ratio did not. In the best multivariate linear regression model (see table 2), urine protein/creatinine ratio was significantly associated with Vitamin D levels.

Table 2. Association between various factors and Vit D levels in patients with SLE based on a multivariate linear regression model

VariableEffect on Mean Vit DP-value
Age (per year)0.11 ± 0.040.005
Race (Black vs White)-6.32 ± 1.040.001
Systolic BP (per mmHg)-0.09 ± 0.030.004
Diabetes-2.43 ± 1.710.156
Urine Pr/Cr Ratio-1.55 ± 0.780.047
BMI-0.29 ± 0.070.001
Cholesterol (per mg/dL)-0.04 ± 0.010.001

Conclusion:

Younger age, African-American ethnicity and obesity are associated with both Vitamin D insufficiency and deficiency, after adjustment for other variables, in multiple variable models. The apparent association with disease activity in univariate analyses did not remain in multiple variable models. The association with urine protein/creatinine ratio, however, did remain in the multivariate linear regression model. This suggests that Vitamin D deficiency may play a role in renal lupus.

To cite this abstract, please use the following information:
D'Souza, Caroline, Bello, Kayode Jibril, Fang, Hong, Magder, Laurence, Petri, Michelle A.; Multivariate Analysis of Vitamin D Insufficiency and Deficiency in SLE: The SOLD (Systemic Lupus Associates of Low Vitamin D) Project. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1163
DOI: 10.1002/art.28929

Abstract Supplement

Meeting Menu

2010 ACR/ARHP