Prevalence and influence of hypouricemia on cardiovascular diseases in patients with rheumatoid arthritis

Demographic and clinical characteristics of RA patients

Among 1411 RA patients recruited in the cohort, 200 without complete laboratory measurement, 37 overlapped with other autoimmune diseases, 24 accompanied with malignancy, and 20 with gout were excluded. Totally, 1130 RA patients were eligible for analysis. Their mean age was 53.2 ± 12.6 years and 893 (79.0%) were female. The median disease duration was 62 months (range 20–121 months). There were 82.3% patients with active RA, and 22.9% treatment naïve patients who have not received glucocorticoid or disease-modifying anti-rheumatic drugs (DMARDs) therapy for 6 months before enrollment.

Distribution of SUA in RA patients with different stratification

The mean SUA level in all RA patients was 4.92 ± 1.72 mg/dL. The SUA level in males was significantly higher than that in females (5.79 ± 1.85 mg/dL vs. 4.68 ± 1.60 mg/dL, P < 0.001). The SUA level in postmenopausal female RA patients was higher than that in premenopausal patients (4.88 ± 1.74 mg/dL vs. 4.39 ± 1.34 mg/dL, P < 0.001, Fig. 1A).

Fig. 1figure 1

The levels of SUA and the prevalence of hypouricemia and hyperuricemia in RA patients with different stratification. The SUA levels in all, male, premenopausal and postmenopausal female RA patients A; The prevalence of hypouricemia and hyperuricemia in different gender B and treatment groups C; The prevalence of hyperuricemia in different age D, disease duration E, disease activity groups F; The prevalence of hypouricemia in different age G, disease duration H, disease activity groups I. *P < 0.05; **P < 0.01; ***P < 0.001. △Treatment naïve, without previous corticosteroids or disease-modifying anti-rheumatic drugs treatment for 6 months before recruited. RA rheumatoid arthritis; Remission (CDAI ≤ 2.8), LDA low disease activity (2.8 < CDAI ≤ 10), MDA moderate disease activity (10 < CDAI ≤ 22), HAD high disease activity (CDAI > 22)

The prevalence of hyperuricemia in RA patients was 12.0% (136/1130). The male RA patients had the highest prevalence of hyperuricemia than both females after and before menopause (27.4% vs. 11.2% vs. 3.1%, P < 0.05, Fig. 1B). The prevalence of hyperuricemia increased with age, disease duration, and disease activity in female RA patients, but not in male patients (Fig. 1D–F). The prevalence of hyperuricemia showed no difference between treatment naïve and treated RA patients (Fig. 1C).

On the other hand, the prevalence of hypouricemia in RA patients was 10.6% (120/1130). The premenopausal female RA patients had a higher prevalence of hypouricemia than males (13.1% vs. 6.3%, P < 0.05, Fig. 1B). The treated RA patients had a higher prevalence of hypouricemia than those treatment naïve patients (12.5% vs. 4.2%, P < 0.001, Fig. 1C). There was no significant difference in hypouricemia prevalence among different age, disease duration, or disease activity groups (Fig. 1G–I).

Pearson correlation analysis showed no significant association between SUA levels and RA disease activity in all RA or active RA patients (Fig. 2).

Fig. 2figure 2

The associations between SUA levels and RA disease activity. The associations between SUA levels and CRP A, ESR B, CDAI C in all RA patients. The associations between SUA levels and CRP D, ESR E, CDAI F in active RA patients. SUA serum uric acid, CRP C-reactive protein, ESR erythrocyte sedimentation rate, CDAI clinical disease activity index

Clinical characteristics of RA patients among different SUA groups

Compared with normouricemic group, RA patients with hyperuricemia were older (mean 58.3 years vs. 52.5 years), more male (47.8% vs. 18.0%), and having higher levels of ESR (median 49 mm/h vs. 35 mm/h) and CRP (median 12.85 mg/L vs. 5.41 mg/L), and higher HAQ-DI (median 0.75 vs. 0.38, all P < 0.05, Table 1). But, RA patients with hypouricemia showed no difference of demographic and RA disease characteristics, except for lower proportion of treatment naïve (9.2% vs. 24.1%) and higher proportion of previous treatment of glucocorticoid (60.8% vs. 46.9%), methotrexate (74.2% vs. 55.7%), and leflunomide (75.8% vs. 40.5%, all P < 0.05).

Table 1 Comparisons of demographic and clinical characteristics among RA patients with different SUA levels

As expected, compared with normouricemic group, RA patients with hyperuricemia had more traditional cardiovascular risk factors, including higher proportion of active smoking (39.7% vs. 14.3%), hypertension (44.1% vs.30.2%), T2DM (27.9% vs. 11.7%), and CKD (14.0% vs. 1.6%), and higher level of LDL-C (mean 3.32 mmol/L vs. 3.13 mmol/L) and TG (mean 1.58 mmol/L vs. 1.16 mmol/L, all P < 0.05). However, RA patients with hypouricemia showed higher levels of HDL-C (mean 1.50 mmol/L vs. 1.39 mmol/L), lower levels of TG (mean 0.98 mmol/L vs. 1.16 mmol/L), BMI (mean 21.1 kg/m2 vs. 21.9 kg/m2), and serum albumin (mean 33.3 g/L vs. 34.8 g/L, all P < 0.05).

CVD in RA patients among different SUA groups

There were 125 (11.1%) patients concomitated with CVDs, including stroke (3.5%), myocardial infarction (3.0%), heart failure (2.7%), angina pectoris (1.5%), and peripheral arterial disease (0.4%). As expected, compared with normouricemic group, RA patients with hyperuricemia had a higher rate of total CVDs (27.9% vs. 7.1%), including stroke (11.1% vs. 1.8%), myocardial infarction (7.4% vs. 1.9%), and heart failure (8.1% vs. 1.9%, all P < 0.05, Fig. 3). Surprisingly, RA patients with hypouricemia also had a higher rate of total CVDs (20.7% vs. 7.1%), including stroke (6.7% vs. 1.8%), myocardial infarction (5.8% vs. 1.9%), and peripheral arterial disease (1.7% vs. 0.1%, all P < 0.05).

Fig. 3figure 3

The associations between SUA levels and CVDs in RA patients. The rates of total CVDs A, stroke B, myocardial infarction C, heart failure D, angina E, and peripheral arterial disease F in RA patients with different SUA levels; the restricted cubic spline regression analysis shows the relationship between SUA levels and CVDs in all F, female G, and male RA patients H; OR (95% CI) was estimated with univariate logistic regression analysis. The red line represents the OR and the pink shaded area is the 95% CI. *P < 0.05; **P < 0.01; ***P < 0.001. CVD cardiovascular disease, SUA serum uric acid, RA rheumatoid arthritis, OR odds ratio, CI confidence interval

In turn, compared with those without CVDs, RA patients with CVDs had higher levels of SUA (mean 5.57 mg/dL vs. 4.83 mg/dL), higher rate of hypouricemia (20.0% vs. 9.5%), and hyperuricemia (30.4% vs. 9.8%). In addition, RA patients with CVDs were older (mean 63.1 years vs. 52.0 years), more male (30.4% vs. 19.8%), and having more serious RA disease and more traditional cardiovascular risk factors (all P < 0.05, Additional file 1: Table S1).

Associations between SUA levels and CVD in RA patients

The results of RCS regression showed a U-shaped relationship between SUA levels and total CVDs in all RA patients and female RA patients (both non-linear P < 0.001, Fig. 3F, G), but not in male RA patients (non-linear P = 0.262, Fig. 3H). After fully adjusted for the potential covariates (model 4, Table 2), multivariate logistic regression analysis revealed that compared with normouricemia, both hypouricemia [adjusted OR (AOR) = 4.707, 95% CI 2.570–8.620] and hyperuricemia (AOR = 3.707, 95% CI 2.174–6.321) were associated with higher CVDs in RA patients (both P < 0.05). The result of sensitivity analyses was consistent among different cut-off values (Additional file 1: Table S2). Among subgroup analyses, the association of hypouricemia and hyperuricemia with CVDs prevalence remained almost similar to all RA patients (Fig. 4).

Table 2 Multivariate logistic regression analysis of the association between SUA levels and CVD in RA patientsFig. 4figure 4

Subgroup analyses of the associations of SUA levels with CVDs in RA patients. BMI body mass index, T2DM, type 2 diabetes mellitus, CKD chronic kidney disease, TC total cholesterol, TG triglyceride, HDL-C high-density lipoprotein cholesterol, LDL-C low-density lipoprotein cholesterol, TNF tumor necrosis factor, AOR adjusted OR, CI confidence interval. *Normouricemic group was used as the reference in each subgroup analysis. #Adjusted by age, gender (male or female), active smoking (yes or no), BMI, hypertension (yes or no), T2DM (yes or no), TC, TG, LDL-C, HDL-C, serum albumin and CKD (yes or no), disease duration, RF positivity (yes or no), ACPA positivity (yes or no), ESR, CRP, CDAI, HAQ-DI, mTSS and previous treatment, including treatment naïve (yes or no), glucocorticoid (yes or no), methotrexate (yes or no), leflunomide (yes or no), hydroxychloroquine (yes or no), sulfasalazine (yes or no), cyclosporine A (yes or no), TNF inhibitors (yes or no), tocilizumab (yes or no), Janus kinase inhibitors (yes or no), statin (yes or no), aspirin (yes or no). & There was no patient with hypouricemia in CKD group

留言 (0)

沒有登入
gif