Effect of uric acid-lowering therapy on blood pressure: systematic review and meta-analysis

Abstract

Background: The purpose of this meta-analysis was to determine if uric acid-lowering therapy is associated with a decrease in blood pressure (BP) and serum creatinine levels.

Materials and methods: Medline, Cochrane, EMBASE, and Google Scholar databases were searched until 29 June 2016, with keywords: uric-acid-lowering therapy, allopurinol, febuxostat, uricosuric, and BP. Only randomized controlled trials were included. The primary outcomes were reduction in systolic BP (SBP) and diastolic BP (DBP), and secondary was reduction in serum creatinine level.

Results: Patients treated with allopurinol had greater reduction in SBP (standardized difference in means [SDM] = 0.321, 95% confidence interval [CI]: 0.145–0.497, p < 0.001), DBP (SDM = 0.260, 95% CI: 0.102 to 0.417, p = 0.001), and creatinine level (SDM = 0.312, 95% CI: 0.008 to 0.615, p = 0.044) than control patients. Subgroup analysis showed that allopurinol significantly decreased SBP whether or not antihypertensive drugs were being administered; a decrease in DBP was only seen in patients receiving antihypertensive drugs. Low-dose allopurinol (≤300 mg/day) was more effective at reducing SBP than high-dose (>300 mg/day) in patients receiving antihypertensive drugs.

Conclusions: These results support that allopurinol decreases BP and creatinine levels in patients with hyperuricemia.

KEY MESSAGES

• Allopurinol decreases SBP and DPB, and creatinine levels in patients with hyperuricemia.

• Allopurinol resulted in a significant decrease in SBP in patients with or without treatment of antihypertensive drugs.

• A dose of allopurinol of ≤300 mg per day might be more effective than a higher dose.

Keywords:

• Allopurinol
• blood pressure
• gout
• hypertension
• meta-analysis
• renal disease
• uric acid

Introduction

Uric acid (UA) is the final product of purine metabolism, and is produced by the catalysis of xanthine by xanthine oxidoreductase. UA is commonly elevated in patients with gout, and elevated UA levels, and hence gout, are typically treated with allopurinol, a xanthine oxidase inhibitor. In addition to the clear association of hyperuricemia with gout, evidence is accumulating that hyperuricemia is associated with a number of diseases including hypertension (1–4), cardiovascular disease (CVD) (5,6), renal disease (7–9), and metabolic syndrome (10). Studies have also suggested an association between elevated UA levels and cardiovascular risk (11–14).

Despite evidence indicating an association of elevated UA levels with a number of diseases, it remains unclear if UA is a modifiable risk factor, i.e., can UA-lowering therapy reduce the risks associated with elevated UA levels. For example, studies examining the effect of allopurinol on blood pressure (BP) have provided conflicting results (15–19). Prior reviews and meta-analysis have also been inconsistent indicating that allopurinol is associated with a small but significant reduction in BP (20,21), there is insufficient evidence to recommend the use of allopurinol as initial or adjuvant treatment of hypertension (22), UA-lowering therapy with allopurinol may retard the progression of chronic kidney disease (CKD) (23), there is limited evidence that allopurinol retards CKD progression or reduces cardiovascular events (7), and that adequately powered randomized controlled trials (RCTs) are needed to determine if UA-lowering therapies have a beneficial effect on renal disease and cardiovascular events (8). It also remains unclear if a decrease in BP does occur, if it can be attributed to allopurinol itself or a decrease in UA level.

Thus, the purpose of this study was to perform a systematic review of the literature and meta-analysis of high-quality studies to determine if UA-lowering therapy is associated with a decrease in BP, as well as decreases in serum creatinine levels and UA levels.

Materials and methods

Search strategy and study selection

PRISMA guidelines were followed in conducting this systematic review and meta-analysis (24). Medline, Cochrane, EMBASE, and Google Scholar databases were searched until 29 June 2016 using the following keywords: uric-acid-lowering therapy, allopurinol, febuxostat, uricosuric, BP. References lists of relevant studies were also examined for additional studies which might meet the inclusion criteria. Searches were conducted by two independent reviewers and a third reviewer was consulted for resolutions of any disagreements.

Inclusion criteria for the meta-analysis were: (1) RCTs; (2) patients studied had hyperuricemia with or without hypertension; and (3) reported quantitative outcomes of interest. Non-RCTs, cohort studiers, letters, comments, editorials, case reports, proceedings, personal communications and retrospective studies were excluded.

Data extraction and outcome measures

The following information/data were extracted from studies that met the inclusion criteria: the name of the first author, year of publication, study design, number of participants in each group, participants’ age and sex, medications used, body mass index (BMI), comorbid conditions, and outcomes. The primary outcome studied was reduction of systolic BP (SBP) and diastolic BP (DBP). Secondary outcomes were reduction of serum UA and serum creatinine levels.

Quality assessment

The risk-of-bias assessment tool outlined in the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0) was used to assess the methodological quality of each study (25). Briefly, six domains are evaluated: random sequence generation, allocation concealment, blinding of patients and personnel, blinding of outcome assessment, incomplete outcome data, and selective reporting risk. The assessment was conducted by two independent reviewers, with a third consulted for resolution of any disagreements. Risks of bias figures were generated using Cochrane Review Manager software 5.3.

Data analysis

All outcomes (SBP, DBP, UA, creatinine) were measured pre- and post-treatment, and changes in all outcome measures in each treatment group were calculated. The measure of effect size was the standardized difference in means, defined as the mean difference between patients treated with allopurinol and control patients standardized by standard deviation of score change. A positive value of the standardized difference in means indicates a greater reduction of SBP and DBP and UA and greater change in serum creatinine in the allopurinol group as compared to the control group. For all analysis, 95% confidence intervals (CIs) were calculated.

Study heterogeneity was examined using the χ²-based Cochran Q statistic and I². For the Q statistic, p values <0.10 were considered to indicate statistically significant heterogeneity. For the I² statistic, I² <25 % indicates low heterogeneity, while I² > 50% indicates moderate to high heterogeneity. Random-effects models (DerSimonian–Laird method) of analysis were used to calculate pooled estimates of standardized differences in means across studies (26).

A sensitivity analysis was conducted using a leave-one-out approach. Subgroup analyses were performed for primary outcomes according to (i) patients receiving and not receiving BP-lowering medications and (ii) dosage of allopurinol in patients taking BP-lowering medication. Publication bias was assessed for the primary outcome; the asymmetry of the funnel plot was evaluated by Egger’s test with a one-tailed p-value. The trim-and-fill method was performed to provide a summary effect adjusted for publication bias. All statistical analyses were performed using the statistical software Comprehensive Meta-Analysis, version 2.0 (Biostat, Englewood, NJ).

Results

Literature search and study characteristics

A flow diagram of study selection is shown in Figure 1. A total of 370 records were screened by review of abstract at title after duplicates were removed. Of these, 346 were excluded for not meeting the inclusion criteria, or fulfilling the exclusion criteria. Thus, the full texts of 24 articles were examined and nine were excluded, the reasons for which are shown in Figure 1. Ultimately, 15 studies were included in the meta-analysis (1,15,16,18,27–37).

Figure 1. Flow diagram of study selection.

Characteristics of the included studies are summarized in Table 1. The number of patients in the studies ranged from 28 to 179, and the mean age ranged from 14 to 69.7 years with the proportion of males ranging from 27.5% to 100%. Hypertension and diabetes mellitus were the two most common comorbid conditions reported. Hypertension was treated with a number of different medication types (Table 1).

Table 1. Summary of basic characteristics and outcomes of included studies.

Study	Studydesign	Intervention	Dosage/duration/frequency	Number ofpatients	Co-treatment (number of patients)	Mean age(years)	Male (%)	BMI (kg/m^2)	Length offollow-up	Currentsmoking (%)	Hypertension(%)	Diabetes(%)
Segal et al. (27)	RCT	Chlorthalidone + allopurinol	5-week run-in on chlorthalidone (25 mg/day) and KCl, then randomized to allopurinol 300 mg/dL or placebo	69	Thiazide: 29 ACEI: 21 Calcium channel blocker: 14 ARB: 5 β-Blocker: 5 Potassium sparing diuretic: 2	50.8	27.5%	34.5 ± 7.3	8 weeks	NA	100.0%	NA
		Chlorthalidone + Placebo		70	Thiazide: 34 ACEI: 14 Calcium channel blocker: 9 ARB: 11 β-Blocker: 6 Potassium sparing diuretic: 2 Minoxidil: 1	51.1	28.6%	34.5 ± 6.8
Kim et al. (28)	RCT	Allopurinol	300 mg/day	36	NA	48.3	100.0%	25.9 ± 3.2	NA	47.2%	NA	NA
		Febuxostat	40 mg/day	35		49.6		26.4 ± 3.5		25.7%
		Febuxostat	80 mg/day	35		49.1		25.4 ± 2.6		34.3%
		Febuxostat	120 mg/day	36		51.2		25.9 ± 2.1		33.3%
		Control		37		51.8		25.8 ± 3.0		37.8%
Higgins et al. (29)	RCT	Allopurinol	300 mg/day	40	Anti-platelet: 38 ACE inhibitor or ARB: 24 Calcium channel blocker: 8 β-Blocker: 5 Lipid lowering: 38	66.9	60.0%	26.1 ± 4.52	NA	20.0%	50.0%	7.5%
		Placebo	Matched	40	Anti-platelet: 36 ACE inhibitor or ARB: 15 Calcium channel blocker: 5 β-Blocker: 18 Lipid lowering: 36	68.8	55.0%	27.7 ± 6.24		17.5%	35.0%	20.0%
Assadi et al. (30)	RCT	Enalapril + allopurinol	Allopurinol: 5 mg/kg/day in two divided doses (maximum daily 300 mg)	24	NA	16	54.0%	92(95% CI:89–93)	NA	NA	NA	NA
		Enalapril	0.15 mg/kg/day once daily (maximum daily 20 mg)	20		14	60.0%	88(95% CI:85–91)
Soletsky et al. (31)	RCT	Allopurinol	100 mg twice daily for 1 week, then 200 mg twice daily for 7 weeks	20	NA	14.5	60.0%	36.3(95% CI:33.9–40.9)	NA	NA	NA	NA
		Control	A placebo capsule twice daily for 1 week, then two placebo capsules twice daily for 7 weeks	20		14.1	65.0%	35.8(95% CI:31.9–39.7)
Kanbay et al. (15)	RCT	Allopurinol	300 mg/day	30	NA	54.4	53.0%	28.4 ± 2.9	NA	NA	NA	NA
		Control		37		50.4	48.0%	29.7 ± 3.4
Dogan et al. (32)	RCT	Allopurinol	900 mg/day	50	NA	50.5	50.0%	25.5 ± 4.5	12 weeks	0.0%	0.0%	100.0%
		Placebo		50		50	52.0%	25.8 ± 5.0
Noman et al. (33)	CrossoverRCT	Allopurinol	600 mg/day	28	Aspirin: 28 β-Blocker: 25 Oral nitrate: 13 Calcium antagonists: 6 Nicorandil: 5 ACE inhibitor: 11 Angiotensin-receptor blocker: 4 Statin: 27	65.2	79.0%	NA	6 month	4.0%	50.0%	11.0%
		Placebo		32	Aspirin: 32 β-Blocker: 27 Oral nitrate: 16 Calcium antagonists: 7 Nicorandil: 8 ACE inhibitor: 17 Angiotensin-receptor blocker: 2 Statin: 31	64	88.0%			13.0%	41.0%	13.0%
Goicoechea et al. (34)	RCT	Allopurinol	100 mg/day	57	RAAS blockers: 85 Calcium-channel blockers: 23 Statins: 49 Antiplatelet: 27	72.1	NA	NA	23.4 ± 7.8months	NA	NA	22.0%
		Control		56	RAAS blockers: 75 Calcium-channel blockers: 36 Statins: 44 Antiplatelet: 33	71.4						20.0%
Momeni et al. (35)	RCT	Allopurinol	100 mg/day	20	ACE inhibitors, angiotensin receptor blockers, OHA and/or insulin	56.3	45.0%	NA	NA	NA	NA	100.0%
		Placebo		20		59.1	45.0%
Feig et al. (12)	CrossoverRCT	Allopurinol	Either allopurinol, 200 mg twice daily, or placebo capsule twicedaily for 4 weeks, followed by a 2-week washout period then a 4-week crossover phase	30	NA	15.1	60.0%	33(95% CI:28–36)	NA	NA	NA	NA
		Placebo
Khan et al. (16)	RCT	Allopurinol	300 mg once daily, 8 weeks	14	ACE inhibitors: 78 β-Blocker: 28 Statins: 87 Antiplatelet: 93	68.2	79.0%	27.8 ± 3.64	NA	1.0%	NA	7.1%
		Placebo	300 mg once daily, 8 weeks	14	ACE inhibitors: 50 β-Blocker: 21 Statins: 93 Antiplatelet: 93	68.7	79.0%	29.8 ± 8.2		0.0%		0.0%
Siu et al. (36)	RCT	Allopurinol	100–300 mg/day	25	Lipid-lowering statins: 6	47.7	30.8%	NA	NA	NA	84.0%	24.0%
		Control	Continue the usual therapy	26	Lipid-lowering statins: 11	48.8	53.6%				73.0%	27.0%
George et al. (18)	CrossoverRCT	AllopurinolAllopurinolPlacebo	300 mg/day600 mg/day	30	ACE inhibitors: 19 Angiotensin receptor blockers: 8 Aspirin: 27 β-blockers: 20 Amiodarone: 2 Digoxin: 3 Thiazide diuretics: 2 Loop diuretics: 19 Spironolactone: 10 Statin: 24 Fibrates: 1	69.7	83.3%	29.7 ± 4.7	NA	10.0%	60.0%	60.0%
Sezai et al. (37)	RCT	Febuxostat	A maximum of 60 mg/day	71	ARB: 41 ACEI: 4 Calcium antagonist: 31 β-blocker: 34 Statin: 58 Furosemide: 26	67.4	81.7%	NA	NA	40.8%	81.7%	39.4%
		Allopurinol	A maximum of 300 mg/day	69	ARB: 37 ACEI: 5 Calcium antagonist: 34 β-blocker: 36 Statin: 55 Furosemide: 22	66.4	82.6%			46.4%	79.7%	33.3%

ACE: angiotensin-converting enzyme inhibitor; ARB: angiotensin-receptor blocker; BMI: body mass index; OHA: oral hypoglycemic agent; NA: not available; RAAS: renin–angiotensin–aldosterone system.

Reduction of BP

A total of 13 studies reported BP data (Figure 2(A,B)). One study (18) assessed the efficacy of two dosages of allopurinol (i.e., 300 mg and 600 mg), and for this reason it was split into two separate data sets for the analysis. There was significant heterogeneity among the included studies with respect to SBP data (I² = 42.0%, Cochran Q p = 0.049). Patients treated with allopurinol had greater reduction in SBP than those in the control group (standardized difference in means = 0.321, 95% CI: 0.145 to 0.497, p < 0.001) (Figure 2(A)). There was low heterogeneity (I² = 28.8%, Cochran Q p = 0.148) among the included studies with respect to DBP data. The pooled results indicated there was a greater reduction in DBP in patients treated with allopurinol (standardized difference in means = 0.260, 95% CI: 0.102 to 0.417, p = 0.001) (Figure 2(B)).

Figure 2. Forest plots for treatment effect of allopurinol on (A) reduction of systolic blood pressure (SBP), (B) reduction of diastolic blood pressure (DBP), (C) reduction of serum uric acid concentration, and (D) reduction of serum creatinine concentration.

Reduction of UA and creatinine levels

Thirteen studies reported data with respect to the association of allopurinol and serum UA concentration (Figure 2(C)). Significant heterogeneity was noted (I² = 94.7%, Cochran Q p < 0.001). The pooled results showed that treatment with allopurinol significantly reduced the serum level of UA (standardized difference in means = 1.548, 95% CI: 0.840 to 2.256, p < 0.001) (Figure 2(C)). One study, however, reported a significantly smaller reduction in UA in patients treated with allopurinol (15). Five studies reported data of the association of allopurinol treatment and serum creatinine level. No significant heterogeneity was noted (I²=0%, Cochran Q p = 0.418). Treatment with allopurinol resulted in a greater reduction is serum creatinine level (standardized difference in means = 0.312, 95% CI: 0.008 to 0.615, p = 0.044) (Figure 2(D)).

Sensitivity analysis

The direction and magnitude of the associations between treatment with allopurinol and SBP and DBP did not change considerably with the removal of individual studies in turn, indicated that the pooled results were reliable (Figure 3).

Figure 3. Sensitivity analysis for the treatment effect of allopurinol on (A) reduction of systolic blood pressure (SBP) and (B) reduction of diastolic blood pressure (DBP).

Subgroup analysis

After stratifying the data according to the use of BP-lowering drugs, the analysis showed that the decrease in SBP in patients receiving allopurinol was significant as compared to the control group regardless of whether patients were receiving BP-lowering drugs or not (without medication: standardized difference in means = 0.377, 95% CI: 0.005 to 0.750, p = 0.047; with medication: standardized difference in means = 0.295, 95% CI: 0.097 to 0.492, p = 0.003) (Figure 4(A)). Similarly, significant reduction of DBP was found in patients receiving allopurinol whether or not they were taking BP-lowering drugs (without medication: standardized difference in means = 0.426, 95% CI: 0.122 to 0.729, p = 0.006; with medication: standardized difference in means = 0.167, 95% CI: 0.003 to 0.331, p = 0.046) (Figure 4(B)).

Figure 4. Subgroup analysis according to use of blood pressure lowering drugs. (A) Reduction of systolic blood pressure (SBP). (B) Reduction of diastolic blood pressure (DBP).

A further subgroup analysis was performed according to dosage of allopurinol in studies recruiting patients taking BP-lowering drugs. The results demonstrated that treatment with low-dose allopurinol (≤300 mg per day) reduced SBP (standardized difference in means = 0.329, 95% CI: 0.116 to 0.542, p = 0.002) as compared to no allopurinol treatment, but this effect was not seen with high-dose allopurinol (>300 mg per day) treatment (standardized difference in means = 0.030, 95% CI: −0.476 to 0.536, p = 0.908) (Figure 5(A)). There was no difference of the effect of high- or low-dose allopurinol on reduction of DBP (low dose: standardized difference in means = 0.176, 95% CI: −0.004 to 0.356, p = 0.055; high dose: standardized difference in means = 0.087, 95% CI: −0.419 to 0.594, p = 0.735) (Figure 5(B)). These results, however, need to be interpreted with caution as there was only one study in the subgroup of >300 mg/day.

Figure 5. Subgroup analysis according to dose of allopurinol in studies recruiting patients using blood pressure lowering drugs. (A) Reduction of systolic blood pressure (SBP). (B) Reduction of diastolic blood pressure (DBP).

Publication bias

There was significant asymmetry of the funnel plot for the treatment effect of allopurinol on reduction of SBP (t = 2.03, df = 12, one-tailed p = 0.032). Results of the trim-and-fill method showed that there were three studies needed to minimize publication bias. After imputation, the pooled results became statistical insignificance (standardized difference in means after adjustment = 0.211, 95% CI: 0.018 to 0.404), for which the imputed pooled results remained statistically significant (Figure 6(A)). On the other hand, no publication bias was found for the effect of allopurinol of DBP (t = 0.105, df = 12, one-tailed p = 0.459) (Figure 6(B)).

Figure 6. Funnel plots for (A) reduction of systolic blood pressure (SBP) and (B) reduction of diastolic blood pressure (DBP).

Quality assessment

Results of the quality assessment of the included studies are shown in Figure 7. Fourteen of the included articles had random sequence generation, and six had allocation concealment. Only four studies had low risk of bias in blinding of outcome assessment, and other studies had an unclear risk of bias. Eleven studies had low risk of attrition bias, and all included studies had low risk of reporting bias. Overall the included studies had low risk in attrition bias and reporting bias, but unclear risk in selection bias and detection bias.

Figure 7. Quality assessment. (A) Risk of bias summary. (B) Overall assessment of risk of bias.

Discussion

The purpose of this meta-analysis was to provide high-quality data examining the effect of allopurinol on BP in patients with hyperuricemia. The overall results showed that patients receiving allopurinol had a greater reduction in SBP and DPB than control patients and this effect was seen whether or not patients were receiving BP-lowering drugs. The results also suggested that a dose of allopurinol of ≤300 mg per day might be more effective than a higher dose at reducing SBP, while dose (high or low) did not result is a different effect on DBP. The results regarding dose must be interpreted with caution as only one study was available for inclusion in the high-dose group. Allopurinol also reduced serum creatinine level and as expected reduced serum UA level.

Although it has been known for decades that elevated UA levels are the cause of gout, and that drugs that inhibit the enzyme xanthine oxidoreductase reduce UA levels and hence treat gout, it has only recently been found that elevated UA levels may be associated with a number of clinical conditions including hypertension, CVD, and renal disease (1–9). Although, the mechanism by which UA may result in end organ damage has not been fully elucidated, evidence is suggesting that UA may adversely affect endothelial function (16,20,38–40). Studies of the effects of UA-lowering therapies on BP, CVD, and renal disease have shown inconsistent results, and some study has suggested that allopurinol itself may improve endothelial cell function and that the improvement is not related to lowering UA levels (14).

As a result of inconsistent results of various studies examining the effect of lowering UA levels on conditions other than gout, a number of systematic reviews have been conducted. Two systematic reviews have examined the effect of allopurinol on BP. A 2013 systematic review and meta-analysis by Agarwal et al. (20) included 10 longitudinal studies with 738 participants. The overall results showed that treatment with allopurinol was associated with small but significant decreases of BP (SBP decrease: 3.3 mm Hg, 95% CI: 1.4 to 5.3 mm Hg, p = 0.001; DBP decrease: 1.3 mm Hg, 95% CI: 0.1 to 2.5 mm Hg, p = 0.03). On the other hand, a Cochrane Database Systematic Review in 2013 (22) with strict inclusion criteria only found one double-blind, placebo-controlled cross-over trial that examine UA-lowering therapy in adolescents. Although the study did show that treatment with allopurinol was associated with a significant reduction of BP, the overall conclusions of the report were that there is insufficient evidence to recommend the use of UA-lowering therapy for the treatment of hypertension and that RCTs are needed.

Reviews have also provided similar conclusion with respect to the effect of UA-lowering therapies on CKD. A 2014 meta-analysis by Bose et al. (23) studying the effects of UA-lowering therapy on renal outcomes included eight trials with 476 participants. The results showed that allopurinol had no effects on proteinuria and BP, and a minimal effect on serum creatinine levels leading the authors to conclude that there is the possibility that allopurinol may retard the progression of CKD and that RCTs are necessary to address the issue. Another 2014 review that included four RCTs concluded that there is only limited evidence that allopurinol reduces CKD progression or cardiovascular events (7). Based on the four RCTs and 21 observational studies, the report did conclude that adverse events attributable to allopurinol were rate. The most recent meta-analysis examining the effect of UA-lowering therapy on CKD was published by Kanji et al. (8) in 2015. The literature search identified 19 RCTs with 992 participants, but because of marked heterogeneity between the studies, only trials that were greater than 3 months what compared allopurinol with an inactive control were included in the analysis. Allopurinol was associated with small but significant reductions in estimated glomerular filtration rate, serum creatinine, and SBP and DBP. Due to the limited number of longer-term comparator studies, the authors concluded that more adequately powered RCTs are needed to determine the effect of UA-lowering therapies on renal and cardiovascular outcomes.

There are limitations of the current analysis that should be taken into consideration. Overall, there was significant heterogeneity among included studies, which may result from the dosage and frequency of UA-lowering treatments and treatments for hypertension. We only examined the effect of UA-lowering treatments on BP and serum creatinine and UA levels, and did not study cardiovascular or renal outcome such as incidence of cardiovascular events or CKD progression. Other treatments to lower UA levels were not studied. As allopurinol is fully excreted by the kidneys, its use in patients with reduced renal function is limited and other drugs such as febuxostat may be more appropriate in this population (41).

In conclusion, the results of this meta-analysis provide further support that allopurinol decreases BP and creatinine levels in patients with hyperuricemia. Further study, however, is necessary to determine the mechanisms by which these effects occur, if there is a dose-related effect, and if allopurinol-associated decreases in BP and creatinine result in improved cardiovascular and renal outcomes.

Disclosure statement

The authors declare no conflict of interest.