Allopurinol, benzbromarone and risk of coronary heart disease in gout ...

International Journal of Cardiology 233 (2017) 85–90

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Allopurinol, benzbromarone and risk of coronary heart disease in gout patients: A population-based study Hsiu-Chen Lin a,b,1, Masao Daimon c,1, Ching-Hung Wang d,1, Yi Ho d,1, Yow-Shieng Uang d,1, Shuo-Ju Chiang e,f,⁎,1,2, Li-Hsuan Wang d,g,⁎⁎,1,2 a

Department of Pediatrics, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan Department of Laboratory Medicine, Taipei Medical University Hospital, Taipei, Taiwan Department of Clinical Laboratory, Tokyo University Hospital, Tokyo, Japan d School of Pharmacy, College of Pharmacy, Taipei Medical University, Taipei, Taiwan e Department of Internal Medicine, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan f Division of Cardiology, Department of Internal Medicine, Taipei Medical University Hospital, Taipei, Taiwan g Department of Pharmacy, Taipei Medical University Hospital, Taipei, Taiwan b c

a r t i c l e

i n f o

Article history: Received 22 August 2016 Received in revised form 2 January 2017 Accepted 1 February 2017 Available online 13 January 2017 Keywords: Gout Anti-hyperuricemic drugs Coronary artery disease Risk Dose-effect

a b s t r a c t Background: The effect of gout on the risk of developing coronary artery disease (CAD) is uncertain. Some studies have found that gout is a risk factor for acute myocardial infarction. This study examined the changes in risk of CAD in gout patients taking allopurinol and/or benzbromarone, and analyzed the dose-response relationship of both drugs with CAD incidence. Methods: The medical records of one million subjects from 2000 to 2011 were provided by the Taiwan National Health Insurance Research Database. Cox proportional hazard ratio was used to compare the risk of CAD in gout patients taking allopurinol or/and benzbromarone with those taking neither drug. Hazard ratios (HR) were adjusted for possible confounding factors, including age, gender, hypertension, hyperlipidemia, diabetes mellitus, chronic kidney disease, and relevant medications. Results: Of 8047 gout patients, 1422 were treated with allopurinol (Group A), 4141 with benzbromarone (Group B), and 2484 with both drugs (Group A/B) during the follow-up period. Our results showed the incidence of CAD after adjusting for covariates for Group A, Group B, and Group A/B did not significantly differ from the comparison group. However, after adjustment for covariates in dose-response analyses, treatment with over 270 defined daily doses (DDDs) of allopurinol, and over 360 DDDs of benzbromarone, was associated with a significantly reduced risk of CAD. Conclusion: We found that the use of allopurinol and benzbromarone, whether alone or in combination, had a linear dose-response relationship between the numbers of defined daily doses and the risk of CAD, especially in higher DDDs. © 2017 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Coronary artery disease (CAD) results from narrowing of the lumen of one or more coronary arteries. Acute obstruction is usually caused by ⁎ Correspondence to: S.J. Chiang, Department of Internal Medicine, School of Medicine, College of Medicine, Taipei, Medical University, No. 250, Wu-Hsing St., Taipei 11031, Taiwan. ⁎⁎ Correspondence to: L.H. Wang, School of Pharmacy, College of Pharmacy, Taipei Medical University, No. 250, Wu-Hsing St., Taipei 11031, Taiwan. E-mail addresses: [email protected] (S.-J. Chiang), [email protected] (L.-H. Wang). 1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data and their discussed interpretation. 2 Shuo-Ju Chiang and Li-Hsuan Wang contributed equally to this study.

http://dx.doi.org/10.1016/j.ijcard.2017.02.013 0167-5273/© 2017 Elsevier Ireland Ltd. All rights reserved.

rupture or erosion of an atherosclerotic plaque with consequent thrombus formation, leading to obstruction of the vessel. According to a survey in 2001, the prevalence of angina in Taiwan is 15.0%, with 15.1% in males and 14.4% in females [1]. Gout manifesting in a variety of ways, such as recurrent acute arthritis with intense pain, is considered to be caused by hyperuricemia with monosodium urate (MSU) crystals becoming deposited in joints [2]. When investigated in the Nutrition and Health Survey in Taiwan, its prevalence during 2005–2008 was estimated to be 8.2% in males and 2.3% in females [3]. The role of gout or hyperuricemia in the risk of CAD is uncertain. Some studies concluded that the association of uric acid with CAD or other cardiovascular outcomes is not due to uric acid per se but to its association with other risk factors, and accordingly that it does not

86

H.-C. Lin et al. / International Journal of Cardiology 233 (2017) 85–90

play a causal role in CAD development [4,5]. On the other hand, other studies found that hyperuricemia is a risk factor for acute myocardial infarction, and an independent risk factor for subclinical atherosclerosis in young adults [6,7]. Additionally, Taiwanese gout patients have been reported to have an increased risk for CAD whether or not other cardiovascular risk factors were present [8,9]. Medications for gout serve as urate-lowering agents for hyperuricemia or as anti-inflammatory agents for acute arthritis [2]. Uratelowering therapy is of two mechanisms: inhibition of urate production, and promotion of urate excretion. The former includes xanthine oxidase (XO) inhibitors [10], and the latter relies on uricosurics [11]. Allopurinol and benzbromarone are two of the most commonly used urate-lowering drugs in Taiwan. In randomized controlled trials, allopurinol appeared to improve endothelial dysfunction and reduce oxidative stress, prolong exercise time, and delay the onset of angina and ST depression during stress testing [12–14]. An open-label study indicated that coronary blood flow may improve in stable CAD patients with endothelial dysfunction after intravenous infusion of 200 mg oxypurinol (an active metabolite of allopurinol) [15]. However, research on the association of reduced CAD with allopurinol is limited, and all studies so far were conducted on small numbers of patients. Furthermore, although benzbromarone has been withdrawn from the market in several countries for over a decade, it is still prescribed in Taiwan, and to date there has been no research on the relationship between CAD and benzbromarone. Therefore, we conducted a study to observe how CAD risk changes with allopurinol and/ or benzbromarone therapy in gout patients using the nationwide population-based database in Taiwan.

400 mg allopurinol by oral or parenteral routes, and one DDD of benzbromarone corresponds to 100 g benzbromarone taken orally. 2.4. Adjustments for covariates We adjusted for several confounding factors, which were decided by several cardiovascular risk evaluation score systems, [16–18] including age, gender, hypertension, hyperlipidemia, diabetes mellitus, chronic kidney disease, and related medications (including stratins, angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), β-blockers, and antiplatelet drugs. 2.5. Statistical analyses All data were analysed using the Statistics Analysis System® (SAS®) 9.1 statistical package (SAS Institute Inc., Cary, NC, USA). Student's t-test and Pearson's chi-squared (χ2) test were used to examine differences in baseline characteristics and potential covariates among different groups of patients during the follow-up period. Cox proportional hazards regression was used to estimate the hazards of CAD for both the case and comparison groups with age, gender, and other covariates adjusted. Statistical significance was defined as a two-tailed p-value of b0.05.

3. Results 3.1. Baseline characteristics

The primary data of this nationwide population-based retrospective cohort study were sourced from the Longitudinal Health Insurance Database 2000 (LHID 2000), which contains medical registration files and original claim data of one million beneficiaries randomly drawn from the National Health Insurance Research Database (NHIRD) in Taiwan from 2000 to 2011. Instituted in 1995, the National Health Insurance program is a compulsory social insurance that covers over 99% of the population of Taiwan, and diagnoses and medical interventions such as medication records are included. "The NHIRD contained all the claimed record for national health insurance payment. Therefore, it can provide complete information for medication prescription and dispensing for individual patient." It is not possible to identify an individual using the NHIRD, and hence the present study was exempted from full review by the Joint Institutional Review Board of Taipei Medical University.

As shown in Fig. 1, of the 52,693 gout patients who met the inclusion criteria, 36,499 were treated with either allopurinol or benzbromarone and 16,194 were treated with neither drug. A total of 8047 gout patients remained and were regarded as the case groups. We classified them into subgroups as only allopurinol-treated (Group A, N = 1422), only benzbromarone-treated (Group B, N = 4141), and both ever-treated (allopurinol and benzbromarone, i.e., Group A/B, N = 2484) during the follow-up period. Each of the 8047 patients was exclusively matched with one of the remaining 16,194 subjects (no allopurinol or benzbromarone) by age and gender in the ratio of 1:1. Table 1 shows the differences in gender, age, comorbidities, and use of related medications. The average ages of all groups were between 50 and 60 years. Males accounted for more than three-fourths of the study population. After matching for age and gender, there were significant differences in the distributions of hypertension, hyperlipidemia, diabetes mellitus, chronic kidney disease and related medications between case groups A, B, and A/B and their comparison subjects. Therefore, the differences between the case groups and their comparison groups were adjusted in order to compare the incidences of CAD in the primary and secondary outcomes.

2.2. Study population

3.2. Primary outcome

2. Methods 2.1. Data source

Diagnoses were defined by the International Classification of Diseases, Ninth Revision, and Clinical Modification (ICD-9-CM) code. Patients with gout (ICD-9-CM codes 274.0, 274.8, and 274.9) between January 1, 2001 and December 31, 2008 were enrolled, including both inpatients and outpatients. Patients were excluded if they had a history of CAD (ICD-9-CM codes 410, 411, 413, and 414), heart failure (HF) (ICD-9-CM code 428), or cancer (ICD-9-CM codes 140–239), or were taking either allopurinol or benzbromarone before the diagnosis of gout. To ascertain the effects of allopurinol and benzbromarone on the risk of CAD, we excluded subjects who had been identified as having CAD before or within 180 days of taking either allopurinol or benzbromarone. We also excluded patients who had been taking either drug for fewer than 90 days within 180 days of starting either medication. In order to improve the accuracy of diagnosis, we included only those patients in whom the diagnosis of gout had been made either as an inpatient or at least twice as an outpatient. We excluded patients under 20 years of age to meet the requirement of the Institutional Review Board (IRB) of Taipei Medical University Hospital. 2.3. Primary and secondary outcomes The primary purpose of the study was to ascertain whether allopurinol and benzbromarone are associated with a reduced risk of CAD in gout patients. The primary outcome was occurrence of CAD during the follow-up period. We also endeavored to analyze the dose-response relationship between risk of CAD and the use of allopurinol and/ or benzbromarone. In order to calculate and compare the cumulative numbers of doses of different drugs, the concept of defined daily dose (DDD) was applied. The secondary outcomes were the hazard ratios of CAD in each case subgroup with different DDDs. According to the World Health Organization, one DDD of allopurinol corresponds to

Table 2 shows the incidence of CAD during the follow-up period. The risk of CAD for each study group is presented as the HR. The crude HRs indicated that the risk of CAD was significantly greater in Group A (HR = 1.27; 95% CI 1.02–1.57), Group B (HR = 1.27; 95% CI 1.12– 1.45), and Group A/B (HR = 1.46; 95% CI 1.22–1.74) than in their respective comparison groups. However, our results show the incidence of CAD during the follow-up period, after adjusting for covariates, the HRs for Group A (allopurinol-treated, adjusted HR = 1.07, 95% CI 0.86–1.33), Group B (benzbromarone-treated, adjusted HR = 1.05, 95% CI 0.92–1.21), and Group A/B (allopurinol/ benzbromarone treated, adjusted HR = 0.86, 95% CI 0.71–1.03) did not differ significantly. 3.3. Secondary outcome Patients of each case subgroup were stratified into five layers by number of DDDs to explore the relationships between CAD risk and doses of allopurinol, benzbromarone, and allopurinol plus benzbromarone. In Group A, the case subgroup receiving more than 270 DDDs of allopurinol showed a significant reduction of CAD risk after adjusting for covariates, the adjusted HR being 0.25 for the 271–360 DDD subgroup (95% CI 0.10–0.61) and 0.28 for the more


87

Fig. 1. Selection and classification of study population.

Table 1 Baseline characteristics of the case groups (Group A, B, and A + B) and their comparisons. Characteristics

Age, mean ± SD, years Male, no. (%) Hypertension, no. (%) Hyperlipidemia, no. (%) Diabetes mellitus, no. (%) Chronic kidney diseases, no. (%) Medications Statins, no. (%) ACEIs/ARBs, no. (%) β-Blockers, no. (%) Antiplatelet drugs, no. (%)

Group A

Group B

Group A/B

Case (N = 1422)

Comparison (N = 1422)

p-Value

Case (N = 4141)


p-Value

Case (N = 2484)


p-Value

57.46 1119 869 637 415 111

±15.22 (78.69) (61.11) (44.80) (29.18) (7.81)

57.46 1119 587 547 364 25

±15.22 (78.69) (41.28) (38.47) (25.60) (1.76)

1 1 b0.001 b0.001 0.03 b0.001

53.71 3105 2248 2124 1046 68

±14.98 (74.98) (54.29) (51.29) (25.26) (1.64)

53.71 3105 1551 1603 1039 59

±14.98 (74.98) (37.45) (38.71) (25.09) (1.42)

1 1 b0.001 b0.001 0.86 0.42

53.77 2056 1447 1253 673 141

±15.32 (82.77) (58.25) (50.44) (27.09) (5.86)

53.77 2056 780 919 551 34

±15.32 (82.77) (31.40) (37.0) (22.18) (1.37)

1 1 b0.001 b0.001 b0.001 b0.001

370 741 758 651

(26.02) (52.11) (53.31) (45.78)

228 439 545 502

(16.03) (30.87) (38.33) (35.30)

b0.001 b0.001 b0.001 b0.001

1185 1178 1934 1544

(28.62) (42.94) (46.70) (37.29)

662 1169 1570 1300

(15.99) (28.23) (37.91) (31.39)

b0.001 b0.001 b0.001 b0.001

663 1246 1248 1066

(26.69) (50.161) (50.24) (42.91)

371 592 785 634

(14.94) (23.83) (31.60) (25.52)

b0.001 b0.001 b0.001 b0.001

Group A: patients receiving allopurinol; Group B: patients receiving benzbromarone; Group A/B: patients receiving allopurinol and benzbromarone. SD = standard deviation; no. = number; ACEIs = angiotensin-converting enzyme inhibitors; ARBs = angiotensin receptor blockers.

88


Table 2 Risk of coronary artery disease as opposed to each comparison group. Group A

Incidence of CAD, no. (%) Days to CAD, mean ± SD Crude HR (95% CI) Adjusted HR† (95% CI)

Group B

Group A/B

Case (N = 1422)


Case (N = 4141)


Case (N = 2484)


196 (13.78) 490 ± 289 1.27⁎ (1.02– 1.57) 1.07 (0.86– 1.33)

151 (10.62) 486 ± 317 1 1

507 (12.24) 525 ± 290 1.27⁎⁎⁎ (1.12– 1.45) 1.05 (0.92– 1.21)

387 (9.35) 455 ± 321 1 1

304 (12.24) 532 ± 324 1.46⁎⁎⁎ (1.22– 1.74) 0.94 (0.71– 1.03)

204 (8.21) 469 ± 303 1 1

Group A: patients receiving allopurinol; Group B: patients receiving benzbromarone; Group A/B: patients receiving allopurinol and benzbromarone. CAD = coronary artery disease; no. = number; SD = standard deviation; 95% CI = 95% confidence interval. ⁎ p-Value b 0.05. ⁎⁎⁎ p-Value b 0.001. † Adjusted for age, gender, hypertension, hyperlipidemia, diabetes mellitus, chronic kidney diseases, statins, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, βblockers and antiplatelet drugs.

than 360 DDD subgroup (95% CI 0.12–0.63) relative to the less than 90 DDD subgroup (Table 3). Thus, the adjusted HR of CAD decreased as the number of DDDs of allopurinol increased. Unlike Group A, the case subgroups B had significantly lower risks of CAD than their counterparts only when taking more than 360 DDDs of benzbromarone (adjusted HR = 0.46, 95% CI 0.34–0.60) (Table 3). A negative association between the number of DDDs of benzbromarone and CAD risk was statistically significant (p b 0.001). Group A/B suggested that patients given a total of more than 360 DDDs of both allopurinol and benzbromarone (Group A/B) had a significantly lowered risk of CAD as compared with those who had used either drug below 90 DDDs. The adjusted HR was 0.44 (95% CI 0.31–0.63) for the more than 360 DDD subgroup.

4. Discussion Our study provides preliminarily data on the effects of allopurinol and benzbromarone on the development of CAD in gout patients. Our primary outcomes suggested that, after the adjustment for covariates, the risk of CAD was similar in the experimental and comparison subgroups. However, the secondary outcomes revealed that the risk of CAD diminished as the number of doses increased in Group A, Group B, and Group A/B. Specifically, the risk of CAD decreased when patients were treated with more than 270 DDDs of allopurinol, more than 360 DDDs of benzbromarone, or more than 360 DDDs of both drugs either separately or concurrently.

4.1. The mechanisms of protective effect of allopurinol and benzbromarone 3.4. Impacts of co-medications Table 4 showed that patients received other medications together with allopurinol and/or benzbromarone. In gout patients receiving statins, allopurinol and/or benzbromarone were still associated with lowered CAD risk. However, this effect was seen only in the patients taking ACEs/ARBs or antiplatelet drugs along with allopurinol or benzbromarone. The HRs were 0.74 (95% CI 0.58–0.94) and 0.79 (95% CI 0.63–0.97), respectively.

As suggested in previous studies, uric acid and gout are associated, either indirectly or causally, with subclinical atherosclerosis and CAD, and patients with hyperuricemia or gout have an increased risk of CAD [4,6–9]. Use of allopurinol has been found to improve endothelial functions in hyperuricemic patients by lowering uric acid levels and reducing generation of free radicals, and the beneficial effect was suggested to be correlated with the reduction of uric acid levels and hence be not significant in patients with normal uric acid levels.[12].

Table 3 Dose-response relationship in each experimental group as oppose to its comparison group.

Group A 0–90 DDDs 91–180 DDDs 181–270 DDDs 271–360 DDDs N 360 DDDs Group B 0–90 DDDs 91–180 DDDs 181–270 DDDs 271–360 DDDs N 360 DDDs Group A/B 0–90 DDDs 91–180 DDDs 181–270 DDDs 271–360 DDDs N 360 DDDs

No. of patients

Incidence of CAD, no. (%)

Days to CAD, mean ± SD

Crude HR (95% CI)

Adjusted HR† (95% CI)

596 302 227 120 177

104 50 31 5 6

(17.45) (16.56) (13.66) (4.17) (3.39)

374 551 676 672 885

± ± ± ± ±

269 241 242 254 176

1 1.21 (0.88, 1.64) 1.18 (0.66, 1.41) 0.28⁎ (0.12, 0.69) 0.23⁎⁎⁎ (0.10, 0.52)

1 1.18 (0.86, 1.61) 0.87 (0.59, 1.27) 0.25⁎ (0.10, 0.61) 0.28⁎ (0.12, 0.63)

1144 1178 606 378 835

165 180 64 44 54

(14.42) (15.28) (10.56) (11.64) (6.47)

413 513 565 641 794

± ± ± ± ±

297 271 238 227 204

1 1.27⁎ (1.07, 1.51) 1.85 (0.66, 1.10) 0.93 (0.69, 1.27) 0.50⁎⁎⁎ (0.38, 0.67)

1 1.25⁎ (1.06, 1.49) 0.87 (0.67, 1.13) 0.92 (0.68, 1.25) 0.46⁎⁎⁎ (0.34, 0.60)

451 700 454 320 559

103 93 46 28 34

(22.84) (13.29) (10.13) (8.75) (6.08)

332 514 668 734 837

± ± ± ± ±

266 258 232 258 189

1 1.10 (0.87, 1.38) 0.81 (0.59, 1.10) 0.69 (0.47, 1.01) 0.47⁎⁎⁎ (0.33, 0.67)

1 1.19 (0.95, 1.51) 0.78 (0.57, 1.06) 0.71 (0.48, 1.04) 0.44⁎⁎⁎ (0.31, 0.63)

Group A: patients receiving allopurinol; Group B: patients receiving benzbromarone; Group A/B: patients receiving allopurinol and benzbromarone. CAD = coronary artery disease; no. = number; SD = standard deviation; 95% CI = 95% confidence interval. ⁎ p-Value b 0.05. ⁎⁎⁎ p-Value b 0.001. † Adjusted for age, gender, hypertension, hyperlipidemia, diabetes mellitus, chronic kidney diseases, statins, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, βblockers and antiplatelet drugs.


89

Table 4 Impact of statins, ACEIs/ARBs, β-blockers and antiplatelets drugs use on the risk of CAD in each case groups as oppose to its comparison groups. Patients receiving statins Group A

Incidence of CAD, no. (%) Crude HR (95% CI) Adjusted HR† (95% CI)

Group B

Group A/B

Case (N = 370)


Case (N = 1185)


Case (N = 663)


44 (11.89) 0.61⁎ (0.40–0.93) 0.59⁎ (0.38–0.92)

42 (18.42) 1 1

159 (13.42) 0.79 (0.62–1.01) 0.76⁎ (0.59–0.97)

106 (16.01) 1 1

97 (14.63) 0.79 (0.58–1.09) 0.66⁎ (0.47–0.92)

64 (17.25) 1 1

Patients receiving ACEIs/ARBs Group A


Group B

Group A/B

Case (N = 741)


Case (N = 1778)


Case (N = 1246)


139 (18.76) 0.93 (0.71–1.23) 1.09 (0.8–1.44)

84 (19.13) 1 1

316 (17.77) 0.87 (0.74–1.04) 0.87 (0.73–1.04)

223 (19.08) 1 1

205 (16.45) 0.83 (0.66–1.05) 0.74⁎ (0.58–0.94)

109 (18.41) 1 1

Case (N = 758)


Case (N = 1934)


Case (N = 1248)


134 (17.68) 0.91 (0.70–1.18) 0.93 (0.71–1.21)

101 (18.53) 1 1

345 (17.84) 1.02 (0.87–1.20) 0.96 (0.82–1.14)

261 (16.62) 1 1

210 (16.83) 0.99 (0.80–1.24) 0.81 (0.64–1.02)

127 (16.18) 1 1

Patients receiving β-blockers Group A


Group B

Group A/B

Patients receiving antiplatelets Group A


Group B

Group A/B

Case (N = 651)


Case (N = 1544)


Case (N = 1066)


145 (22.27) 0.90 (0.71–1.15) 0.92 (0.71–1.18)

117 (23.31) 1 1

388(25.13) 1.01 (0.87–1.18) 0.97 (0.83–1.13)

308 (23.69) 1 1

232 (21.76) 0.86 (0.70–1.05) 0.79⁎ (0.63–0.97)

150 (23.66) 1 1

Group A: patients receiving allopurinol; Group B: patients receiving benzbromarone; Group A/B: patients receiving allopurinol and benzbromarone. CAD = coronary artery disease; no. = number; SD = standard deviation; 95% CI = 95% confidence interval; ACEIs = angiotensin-converting enzyme inhibitors; ARBs = angiotensin receptor blockers. ⁎ p-Value b 0.05. † Adjusted for age, gender, hypertension, hyperlipidemia, diabetes mellitus, chronic kidney diseases, statins, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, βblockers and antiplatelet drugs.

In another trial, patients were shown to have improved epithelial dysfunctions and reduced oxidative stress after treated with allopurinol [14]. Mechanistically, both XO and uric acid are considered to be involved in inflammatory responses and vascular regulation. When XO oxidizes hypoxanthine and xanthine, superoxide (O2•−) is formed from molecular oxygen, and then superoxide is either reduced via the iron-catalyzed Haber-Weiss reaction to other reactive oxygen species which contribute to oxidative stress, or reacted with nitric oxide (NO) to form peroxynitrite (ONOO−) [19,20]. In monolayers of arterial endothelial cells, not only does O2•− consume NO, but ONOO– anions formed by O2•− and NO also reduce the bioactivity of 5,6,7,8tetrahydrobiopterin (BH4), which is required for endothelial NO synthase (eNOS) stabilization, thereby reducing the production of NO. The decrease in NO flowing into smooth muscle cells attenuates phosphorylation of Ca2+ channels, reduces vascular relaxation, and finally leads to endothelial dysfunction [21–23]. Through the above mechanisms, allopurinol mitigates the effects of free radicals by inhibiting XO and reducing the production of uric acid, while benzbromarone has no effect on XO. Accordingly, the secondary outcomes revealed that allopurinol had a favorable effect on CAD prevention, when more than 270 DDDs had been taken, and benzbromarone when more than 360 DDDs had been taken.

effect on CAD. However, in patients taking ACE-I/ARB, allopurinol and benzbromarone alone do not have preventative effects on CAD. ACE degrades bradykinin, which preserves endothelial function via increased cellular expression of eNOS [25]. Abundant amounts of ACE are present in endothelial cells, and ACE-I has been shown to improve coronary artery endothelial cell function [26]. But not all ACE-Is have consistently been shown to benefit function. Accordingly, the protective effect was seen on in A/B group, not solely in group A and group B. As for β-blockers, they competitively inhibit the binding of endogenous catecholamines to β1 receptors, leading to reductions of heart rate, blood pressure, and myocardial contractility. In this way they lessen myocardial oxygen demand both at rest and during exertion or stress [27]. This reduces myocardial oxygen requirements, thereby preventing ischemia and angina. Therefore, the use of A, B, or A/B for protection from CAD has a different mechanism of action compared with βblockers. The add-on effect of A or B on β-blockers is not known. Our data revealed no difference of A or B in association with β-blockers. The antiplatelet agents exert CAD protection by inhibiting platelet aggregation, and partly also by regulating inflammation in vessels [28]. With regard to this anti-inflammatory effect, A and B are somewhat similar to antiplatelet drugs. Our results showed that only group A/B protects against CAD. The add-on effect of A or B to antiplatelet drugs is also unclear.

4.2. The effects of co-medications 4.3. Strength Table 4 showed that both allopurinol and benzbromarone may reduce the risk of CAD in gout patients taking a statin. As there is evidence that statins might directly reduce vascular inflammation [24], allopurinol and benzbromarone might synergistically increase their protective

To date there have been few studies of the relation of allopurinol use to CAD, and none on the effect of benzbromarone. The effects of using only one or both of the drugs were studied by applying the concept of

90


the DDD. Additionally, the large size and long follow-up period of our study should ensure that the results are representative of the Taiwanese population. 5. Limitations First, the study is limited to data provided by the NHIRD, in which race, medication adherence, diet, exercise, clinical laboratory data, and radiologic interpretations are not recorded. Second, clinical data are a prerequisite for diagnosing and grading the severity of gout and CAD. Unavoidably, the baseline severity of gout in all study groups was inaccessible, and therefore we were not able to adjust for this covariate. Third, as the study was of a specific population, our findings cannot be generalized to people without gout. ″Fourth, our study is similar to other pharmacoepidemiological studies, there is no certainty that drugs were actually taken by participants. However, poor compliance of patients could have led to underestimation of the reduced risk effect of allopurinol and/or benzbromarone.″ 6. Conclusion Though we failed to demonstrate that allopurinol and benzbromarone could decrease risk of coronary heart disease in gout patients from primary outcome. From dose-response analysis, we found that the use of allopurinol and benzbromarone, whether alone or in combination, had a linear dose-response relationship between the number of defined daily doses and the risk of CAD. The reduced risk effect was significant when a large enough number of doses of allopurinol and/or benzbromarone had been taken. Since a considerable number of doses of allopurinol and/or benzbromarone were required for CAD risk reduction was observed in our study, the two drugs might be of value as an adjuvant medication against the development of CAD in patients with gout. However, this result needed be clarified with further studies in the future. Sources of funding This work was supported in part by a Grant-in-Aid for Scientific Research C (59003575) from the Japan Society for the Promotion of Science (Masao Daimon). The funders had no role in the study design, data collection and analysis, and preparation of the manuscript. Disclosures No conflicts of interest are declared. Acknowledgments None. References [1] Y.C. Lin, F.Y. Chu, C.C. Fu, J.D. Chen, Prevalence and risk factors for angina in elderly Taiwanese, J. Gerontol. A Biol. Sci. Med. Sci. 59 (2) (2004) 161–165. [2] T. Neogi, Gout, N. Engl. J. Med. 364 (5) (2011) 443–452. [3] S.Y. Chuang, S.C. Lee, Y.T. Hsieh, W.H. Pan, Trends in hyperuricemia and gout prevalence: nutrition and health survey in Taiwan from 1993–1996 to 2005– 2008, Asia Pac. J. Clin. Nutr. 20 (2) (2011) 301–308.

[4] B.F. Culleton, M.G. Larson, W.B. Kannel, D. Levy, Serum uric acid and risk for cardiovascular disease and death: the Framingham Heart Study, Ann. Intern. Med. 131 (1) (1999) 7–13. [5] G. De Luca, G.G. Secco, M. Santagostino, et al., Uric acid does not affect the prevalence and extent of coronary artery disease. results from a prospective study, Nutr. Metab. Cardiovasc. Dis. 22 (5) (2012) 426–433. [6] E. Krishnan, J.F. Baker, D.E. Furst, H.R. Schumacher, Gout and the risk of acute myocardial infarction, Arthritis Rheum. 54 (8) (2006) 2688–2696. [7] E. Krishnan, B.J. Pandya, L. Chung, O. Dabbous, Hyperuricemia and the risk for subclinical coronary atherosclerosis-data from a prospective observational cohort study, Arthritis Res. Ther. 13 (2) (2011) R66. [8] S.Y. Chuang, J.H. Chen, W.T. Yeh, C.C. Wu, W.H. Pan, Hyperuricemia and increased risk of ischemic heart disease in a large Chinese cohort, Int. J. Cardiol. 154 (3) (2012) 316–321. [9] C.F. Kuo, K.H. Yu, L.C. See, I.J. Chou, et al., Risk of myocardial infarction among patients with gout: a nationwide population-based study, Rheumatology (Oxford) 52 (1) (2013) 111–117. [10] P. Pacher, A. Nivorozhkin, C. Szabo, Therapeutic effects of xanthine oxidase inhibitors: renaissance half a century after the discovery of allopurinol, Pharmacol. Rev. 58 (1) (2006) 87–114. [11] T. Iwanaga, D. Kobayashi, M. Hirayama, T. Maeda, I. Tamai, Involvement of uric acid transporter in increased renal clearance of the xanthine oxidase inhibitor oxypurinol induced by a uricosuric agent benzbromarone, Drug Metab. Dispos. 33 (12) (2005) 1791–1795. [12] G. Mercuro, C. Vitale, E. Cerquetani, S. Zoncu, M. Deidda, M. Fini, G.M. Rosano, Effect of hyperuricemia upon endothelial function in patients at increased cardiovascular risk, Am. J. Cardiol. 94 (7) (2004) 932–935. [13] A. Noman, D.S. Ang, S. Ogston, C.C. Lang, A.D. Struthers, Effect of high-dose allopurinol on exercise in patients with chronic stable angina: a randomised, placebo controlled crossover trial, Lancet 375 (9732) (2010) 2161–2167. [14] N.S. Rajendra, S. Ireland, J. George, J.J. Belch, C.C. Lang, A.D. Struthers, Mechanistic insights into the therapeutic use of high-dose allopurinol in angina pectoris, J. Am. Coll. Cardiol. 58 (8) (2011) 820–828. [15] S. Baldus, R. Köster, P. Chumley, et al., Oxypurinol improves coronary and peripheral endothelial function in patients with coronary artery disease, Free Radic. Biol. Med. 39 (9) (2005) 1184–1190. [16] W.B. Kannel, D. McGee, T. Gordon, A general cardiovascular risk profile: the Framingham Study, Am. J. Cardiol. 38 (1976) 46–51. [17] P.M. Ridker, J.E. Buring, N. Rifai, N.R. Cook, Development and validation of improved algorithms for the assessment of global cardiovascular risk in women: the Reynolds Risk Score, JAMA 297 (2007) 611–619. [18] R.B. D'Agostino Sr., R.S. Vasan, M.J. Pencina, et al., General cardiovascular risk profile for use in primary care: the Framingham Heart Study, Circulation 117 (2008) 743–753. [19] J.S. Beckman, T.W. Beckman, J. Chen, P.A. Marshall, B.A. Freeman, Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide, Proc. Natl. Acad. Sci. U. S. A. 87 (4) (1990) 1620–1624. [20] N.V. Blough, O.C. Zafiriou, Reaction of superoxide with nitric oxide to form peroxonitrite in alkaline aqueous solution, Inorg. Chem. 24 (22) (1985) 3502–3504. [21] R. Kietadisorn, R.P. Juni, A.L. Moens, Tackling endothelial dysfunction by modulating NOS uncoupling: new insights into its pathogenesis and therapeutic possibilities, Am. J. Physiol. Endocrinol. Metab. 302 (5) (2012) E481–E495. [22] L. Kuo, T. Hein, Vasomotor regulation of coronary microcirculation by oxidative stress: role of arginase, Front. Immunol. 4 (2013) 237. [23] P.M. Vanhoutte, Vascular biology. old-timer makes a comeback, Nature 396 (6708) (1998) 5–6 (213). [24] E. Ascer, M.C. Bertolami, M.L. Venturinelli, et al., Atorvastatin reduces proinflammatory markers in hypercholesterolemic patients, Atherosclerosis 177 (1) (2004) 161–166. [25] J. Zhuo, F. Mendelsohn, M. Ohishi, Perindopril alters vascular angiotensin converting enzyme, AT1 receptor and nitric oxide synthase expression in patients with coronary heart disease, Hypertension 39 (2 Pt 2) (2002) 634–638. [26] E. Braunwald, M.J. Domanski, S.E. Fowler, et al., Angiotensin converting enzyme inhibition in stable coronary artery disease, N. Engl. J. Med. 351 (20) (2004) 2058–2068. [27] S.T. O'Rourke, Antianginal actions of beta-adrenoceptor antagonists, Am. J. Pharm. Educ. 71 (5) (2007) 95. [28] I. Ikonomidis, F. Andreotti, E. Economou, C. Stefanadis, P. Toutouzas, P. Nihoyannopoulos, Increased proinflammatory cytokines in patients with chronic stable angina and their reduction by aspirin, Circulation 100 (8) (1999) 793–798.