Background/Purpose: Recent data report significant decreases in low-density lipoprotein (LDL) and total cholesterol (TC) levels in predominantly female Caucasian rheumatoid arthritis (RA) patients treated with hydroxychloroquine (HCQ). We evaluated the association of HCQ use with lipid profiles in Veterans Affairs registry (VARA) patients; a predominantly male cohort with multiple comorbidities.

Methods: VARA patients that had post-enrollment lipid profiles available were evaluated. [LDL, TC, high-density lipoprotein (HDL), triglycerides (TG)] values and HCQ status were extracted through links to national VA administrative and pharmacy databases. HCQ user was defined as at least 3 consecutive months of prescription prior to the index lipid value. Patient data included socio-demographics, MDHAQ, RA disease activity measures (TJC, SJC, ESR, CRP, 3vDAS28), treatment [DMARD (excluding HCQ), anti-TNF], statin and prednisone use, and presence of diabetes mellitus (DM). HCQ users and HCQ non-users were compared using chi-square tests for categorical variables and t-tests for continuous variables.  Multivariate analysis was performed, controlling for age, gender, race, 3vDAS28, prednisone, DMARD, statin use and DM.  

Results:

A total of 1012 VARA patients had at least one lipid profile following enrollment; 208 were excluded (< 3 consecutive months of HCQ use). Of 804 patients, the mean age was 62.5 yrs, predominantly male (91.7%) and Caucasian (79.6%), with established disease (10.6 yrs, SD12.2). Seventy-five percent received DMARDS, 24.6% biologic therapy. One hundred and sixty-six patients (20.7%) were HCQ users and 638 (79.3%) were HCQ non-users. Significant differences in HCQ-users versus non-users were age (64.4 vs 62.1 yrs, p<0.009), disease duration (13.0 vs 10.0 yrs, p=0.003), TJC (3.9 vs. 5.8, p < 0.001), SJC (3.5 vs. 5.0, p =0.001), 3vDAS28 (3.4 vs 3.8, p=0.001), and DMARD use (94.6 vs 70.5%, p<0.001).

HCQ users had statistically significant lower levels of TC, TC/HDL and LDL compared to HCQ non-users that persisted after multivariate analyses (Table). Despite no ethnic differences in HCQ use vs non-user status, prednisone, DMARD, and statin use or 3vDAS28 scores, multivariate analyses found significantly better lipid profiles in Caucasian but not African-American HCQ users versus non-users.

Conclusion:

In a cohort of US Veterans, RA patients taking HCQ had more optimal lipid profiles than those not using the drug. However, African Americans on HCQ did not demonstrate similar benefits as Caucasian patients. For a relatively inexpensive, low-risk drug, there may be a potential lipid lowering role for HCQ in some RA patients.

Table.  Univariate and Multivariate Associations of Lipid values in HCQ Users vs Non-Users in RA patients
	HCQ Usage			Univariate Analysis (p value)	Multivariate Analysis Model Coefficients (p value)
	Total n = 804 mean (mg/dl) (SD)	Current n = 166 mean (mg/dl) (SD)	Never n = 638 mean (mg/dl)  (SD)		Total Cohort (n=643)	Mena (n=595)	Caucasianb (n=547)	African Americanc (n=96)
TC	180.2 (37.7)	170.3 (31.4)	182.7 (38.8)	<0.001	-9.3 (0.004)	-9.1 (0.007)	-11.0 (<0.002)	-1.0 (0.898)
LDL	106.4 (33.4)	97.9 (30.1)	108.6 (33.9)	<0.001	-8.1 (0.007)	-7.9 (0.011)	-9.8 (0.003)	-0.5 (0.947)
HDL	46.0 (15.6)	47.7 (16.6)	45.5 (15.4)	0.128	1.5 (0.301)	1.6 (0.279)	1.7 (0.291)	0.6 (0.874)
TC/HDL	4.3 (1.6)	3.9 (1.2)	4.4 (1.6)	<0.001	-0.40 (0.001)	-0.402 (<0.002)	-0.45 (<0.001)	-0.18 (0.488)
aControlling for Age, Race, Gender, Prednisone Use, Statin Use, 3vDAS28, and Diabetes bControlling for Age, Race, Prednisone Use, Statin Use, 3vDAS28, and Diabetes cControlling for Age, Gender, Prednisone Use, Statin Use, 3vDAS28, and Diabetes

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/hydroxychloroquine-has-lipid-lowering-effects-in-us-veterans-with-rheumatoid-arthritis/