https://orcid.org/0000-0001-5345-9788
Figures
Abstract
Aims
To compare the effectiveness of allopurinol with no treatment or placebo for the prevention of cardiovascular events in hyperuricemic patients.
Methods and results
Pubmed, Web of Science and Cochrane library were searched from inception until July 2020. Randomized controlled trials (RCT) and observational studies in hyperuricemic patients without significant renal disease and treated with allopurinol, versus placebo or no treatment were included. Outcome measures were cardiovascular mortality, myocardial infarction, stroke, or a combined endpoint (CM/MI/S). For RCT’s a random effects meta-analysis was performed. For observational studies a narrative synthesis was performed. Of the original 1995 references we ultimately included 26 RCT’s and 21 observational studies. We found a significantly reduced risk of combined endpoint (Risk Ratio 0.65 [95% CI] [0.46 to 0.91]; p = 0.012) and myocardial infarction (RR 0.47 [0.27 to 0.80]; p = 0.01) in the allopurinol group compared to controls. We found no significant effect of allopurinol on stroke or cardiovascular mortality. Of the 15 observational studies with sufficient quality, allopurinol was associated with reduced cardiovascular mortality in 1 out of 3 studies that reported this outcome, myocardial infarction in 6 out of 8, stroke in 4 out of 7, and combined end-point in 2 out of 2. Cardiovascular benefit was only observed when allopurinol therapy was prolonged for more than 6 months and when an appropriate allopurinol dose was administered (300 mg or more/day) or sufficient reduction of serum urate concentration was achieved (<0.36 mmol/l).
Conclusions
Data from RCT’s and observational studies indicate that allopurinol treatment reduces cardiovascular risk in patients with hyperuricemia. However, the quality of evidence from RCTs is low to moderate. To establish whether allopurinol lowers the risk of cardiovascular events a well-designed and adequately powered randomized, placebo-controlled trial is needed in high-risk patients with hyperuricemia.
Citation: van der Pol KH, Wever KE, Verbakel M, Visseren FLJ, Cornel JH, Rongen GA (2021) Allopurinol to reduce cardiovascular morbidity and mortality: A systematic review and meta-analysis. PLoS ONE 16(12): e0260844. https://doi.org/10.1371/journal.pone.0260844
Editor: Carmine Pizzi, University of Bologna, ITALY
Received: May 24, 2021; Accepted: November 17, 2021; Published: December 2, 2021
Copyright: © 2021 van der Pol et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: Dr. Cornel reports grants from ZonMw (LoDoCo 2 trial), personal fees from Amgen (advisory board), personal fees from Servier (advisory board), personal fees from Astra Zeneca (advisory board), outside the submitted work; All other authors declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Allopurinol, a well-known xanthine oxidase inhibitor, has been used for over 50 years to lower serum uric acid. According to clinical guidelines, this drug should be used to treat complications that result from hyperuricemia: frequent gout attacks, urate arthropathy, urate depositions in the skin (tophi) and urate nephropathy. Although urate nephropathy is not a well-defined entity, prove of urate stones and prevention of renal decline that results from tumour lysis syndrome are accepted conditions to start allopurinol [1, 2]. Considerably less agreement exists in clinical practice on the use of allopurinol in patients with asymptomatic or minimally symptomatic hyperuricemia, e.g. in patients who only rarely suffer a gout attack that can easily be managed by colchicine, prednisone or a cyclooxygenase(COX)-inhibitor. Hyperuricemia is associated with cardiovascular morbidity and mortality [3, 4], and an increased risk of renal failure [5]. Clinical guidelines on the use of a xanthine oxidase inhibitor differ in particular for patients with elevated cardiovascular risk or renal insufficiency and otherwise asymptomatic hyperuricemia. For example, Dutch guidelines for primary physicians do not recommend allopurinol for these patients [6, 7]. In contrast, the guidelines for rheumatologists suggest to discuss the option of allopurinol treatment in patients with cardiovascular disease or multiple cardiovascular risk factors, but do not state the arguments pro- or contra- that could be used in making this shared decision with the patient [7]. Three recently published randomized controlled trials (RCTs) fuelled this discussion further: the Cardiovascular Safety of Febuxostat and Allopurinol in Patients with Gout and Cardiovascular Morbidities trial (CARES) [8], the Febuxostat for Cerebral and Cardiorenovascular Events Prevention Study (FREED) [9], and the Long-term Cardiovascular Safety of Febuxostat Compared with Allopurinol in Patients with Gout Trial (FAST) [10]. The CARES trial was a large (>4000 patients with hyperuricemia and gout) randomized double blind clinical trial that compared febuxostat with allopurinol with cardiovascular safety as primary outcome. The patients who had been allocated to febuxostat showed more reduction in serum urate. However cardiovascular mortality was significantly lower in the group allocated to allopurinol as compared with febuxostat [8]. This finding has subsequently been substantiated by real life observational data in large cohorts [11, 12]. The FAST trial was a large (>6000 patients with gout and at least one additional cardiovascular risk factor) randomized, open-label, blinded-endpoint, non-inferiority trial of febuxostat versus allopurinol again with cardiovascular safety as the primary outcome. In this trial febuxostat was non-inferior to allopurinol for cardiovascular mortality, myocardial infarction, stroke or combined outcome [10], contrasting with the findings of the previously published CARES trial. This might be explained by a different incidence of cardiovascular disease in the medical history of the included patients. All patients included in the CARES trial had a history of previous cardiovascular disease as compared with only thirty-three percent of patients in the FAST trial. It should also be noted that rate of treatment discontinuation (25.3% vs 56.6%) and loss to follow-up (5.8% vs 45%) was lower in the FAST trial than the CARES trial. The FREED trial was a randomized trial with PROBE design comparing febuxostat with non-febuxostat therapy in around 1200 patients (age > 65 years) with asymptomatic hyperuricemia. Cardiovascular events, mortality and renal function were the composite primary outcome. Although febuxostat did not affect cardiovascular events in this trial, the decline in renal function was inhibited in those allocated to febuxostat as compared with usual therapy. Unfortunately, sample size was relatively low and 27.2% of the patients allocated to non-febuxostat arm used 100 mg allopurinol/day complicating the interpretation of the results.
This paucity of data on efficacy of allopurinol to prevent cardiovascular events in patients with asymptomatic or mildly symptomatic hyperuricemia triggered us to systematically review the literature on the efficacy of allopurinol compared with no treatment or placebo to prevent cardiovascular events in patients with hyperuricemia and normal or moderately reduced renal function.
Methods
This review is reported according to the PRISMA guidelines. The full systematic review protocol was prospectively submitted at the PROSPERO international prospective register of systematic reviews in March 2018 (registration number CRD42018089744, available from https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42018089744). All amendments to the review protocol were also registered on PROSPERO.
Search and study selection process
The comprehensive search string is included as S1 File. On July 22 2020, we searched PubMed, Web of Science and Cochrane library for RCTs or observational (cohort) studies in human adults with hyperuricemia without severe renal disease (defined as Modification of Diet in Renal Disease (MDRD) formula < 30 ml/min/1.73 m2) who were treated with a xanthine oxidase inhibitor (allopurinol or febuxostat) in any dose regimen or treatment duration. No restrictions on publication date or language were applied. Automated duplicate removal was used to remove duplicates (i.e. studies occurring more than once in our database after the search), if a DOI occurred more than once in the database the duplicate entries were automatically removed.
Two reviewers (KvdP, KW or GR) independently performed screening for eligibility based on title and abstract and assessment for final inclusion based on full-text. In case of discrepancies, the reviewers reached consensus through discussion. For details on the exclusion criteria see the flow chart of the study selection process (Fig 1) and the PROSPERO protocol. Only studies that used appropriate controls (placebo, no treatment) were included.
*there were 4 publications of 2 RCTs. Only 2 of these publications were included in the meta-analysis the other 2 were excluded.
Data extraction and risk of bias or quality assessment
Extraction of study characteristics and outcome data was performed by KvdP and independently checked by two reviewers (GR and JHC). Outcome data were extracted for four predefined primary outcomes: incidence of cardiovascular mortality, incidence of stroke, incidence of myocardial infarction and the combined incidence of these three. The risk of bias in RCTs was assessed by two independent using the Cochrane risk of bias tool in RevMan 5.3 [13]. The New-Castle Ottowa scale was used to assess the quality of observational studies (available at http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp). For randomized cross-over studies, we assessed risk of bias using the systems provided in the Cochrane handbook. All assessments were performed by independent reviewers (KvdP, GR or KW). In case of discrepancies, the reviewers reached consensus through discussion.
Data synthesis strategy
Each of the four outcomes was reported in at least 3 of the retrieved RCTs. In accordance with our predefined data-analysis plan, a meta-analysis was performed on each of these outcomes. The meta-analysis was performed using the OpenMeta[analyst] software package [14]. We calculated the risk ratio and corresponding 95% confidence interval (RR [95%CI]) and individual study effects were pooled using the DerSimonian and Laird random effects model. To include studies in the meta-analysis in which zero events occurred the numerator 0 was substituted with 0.5 (i.e. correction factor). Heterogeneity was assessed using the I², where an I² of >50% was considered substantial. In the predefined protocol subgroup analysis was only to be used to explore causes of heterogeneity in the results. Planned subgroup analyses on sex, age, dose, and various comorbidities were not performed because heterogeneity was 0% in all four overall analyses. For the same reason, reporting R2 was redundant and therefore omitted. Publication bias was assessed if at least 10 studies reported a certain outcome. We used visual inspection, Egger’s regression and trim and fill analysis to test for funnel plot asymmetry. For observational studies we performed a narrative synthesis for the four primary outcomes, taking into account the type of intervention and treatment regime and characteristics of the study population. A meta-analysis of observational studies was not included in the original protocol and therefore not performed. It was expected that there would be significant differences in design of observational studies resulting in large methodological and clinical heterogeneity. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria were used to assess the overall quality of the body of evidence [15–17]. A summary of findings table was created using GRADE PRO [18].
Patient and public involvement
Patients were involved in writing the discussion section dealing with communication of results to patients, in order to support shared decision making on the initiation of allopurinol in patients with asymptomatic or mildly symptomatic hyperuricemia. No patients were involved in designing or executing the study.
Results
Search and study selection
A flow chart of the study selection process is shown in Fig 1. After automated duplicate removal, 1995 unique publications were retrieved from Pubmed and Web of Science. The Cochrane library search did not reveal additional unique publications. After title and abstract screening, 506 eligible publications remained, which were assessed for final inclusion based on full-text. Ultimately, 49 publications met our inclusion criteria and underwent data extraction and quality assessment (26 publications of RCTs, 2 cross-over studies and 21 observational studies). In almost all trials cardiovascular events were not the primary outcome. We therefore extracted data on cardiovascular events from the reported adverse events in those trials. Therefore, trials not reporting adverse events were excluded.
Study characteristics
A complete list of all included studies and their general characteristics (e.g. sex and age of the participants, duration and dose of treatment) is presented as S1 Table.
Twenty-six publications of 24 RCTs reporting cardiovascular outcomes were retrieved [19–44]. One trial was published twice: once after two years of follow-up [22], and once after an additional 5 years follow up after patients had returned to pre-trial medication [43]. The latter publication contained more details about cardiovascular events and was therefore used in the meta-analysis. Two other publications appear to be on the same study population, with one reporting on carotid intima-media thickness [28], and the other on renal function [27]. The two publications report identical data on cardiovascular events, which were included once in our meta-analysis. Two randomized crossover trials were included [45, 46]. In the included trials, a total of 3080 patients have been allocated to either allopurinol (n = 1638) or no-urate-lowering therapy (n = 1442). In total 21 observational studies were included: 16 cohort studies (4 prospective [47–50] and 12 retrospective [51–62]), and 5 case-control studies [63–67].
Risk of bias and study quality
The risk of bias did not differ between outcome measures, therefore the results presented apply to all four outcomes. The individual risk of bias assessments of the 26 publications of 24 RCTs is shown as S1 Fig and S2 Table (for two randomized cross-over studies). For RCTs, a summary of the risk of bias assessment is presented in Fig 2. Out of all 26 trials, one trial was at low risk of all types of bias. Eight trials were at unclear risk of at least one type of bias. Seventeen trials were at high risk for at least one type of bias, often due to lack of blinding of selection, performance, or outcome detection. Overall, the quality of evidence derived from the RCT was classified as low for all outcomes. The individual quality assessments of the 21 observational studies are presented as S2 Table. All 5 case control studies were assessed as being of good quality. Six cohort studies were assessed as being poor quality [47–50, 54, 58], and these are therefore not included in the main text of this review (a complete overview of all studies, including those of poor quality, is available as S1 Table). The 10 remaining cohort studies were all assessed as being of good quality.
Data synthesis—randomized clinical trials
In general, few of the included trials reported events of cardiovascular mortality, myocardial infarction, stroke, or the combined incidence of these three. For the combined outcome 6 out of 26 trials reported an event, with 39 events in 1550 patients in the allopurinol treated group and 64 events in 1354 patients in the control arm resulting in a relative risk of 0.65 (95% CI 0.46 to 0.91; p = 0.012, I2 = 0%) in favour of allopurinol (Fig 3, for weights assigned to individual trials included in the meta-analysis see S3 Table). Allopurinol did not significantly affect cardiovascular mortality or stroke but led to a significant reduction in myocardial infarction (S2 Fig). Twenty out of 26 trials (including 2 cross-over studies) reported zero events. In these 20 trials 1059 patients were treated with allopurinol and 907 patients received placebo. Thus, studies that reported events included only a minority of randomized patients (938 out of the 2904 randomized patients).
Overall effect of allopurinol on the combined outcome: p = 0.01.
Publication bias assessment
Funnel plots for all four outcomes are shown in Figs 4 and S3. In all funnel plots a cluster of datapoints around the risk ratio of 1 and a standard error of 2 is observed, due to the large number of trials reporting 0 events in both treatment groups leads. Egger’s regression test was not significant for cardiovascular mortality and stroke (respectively p = 0.49 and p = 0.14) but indicated the presence of small study effects for myocardial infarction and the combined outcome (respectively p<0.0001 and p = 0.039). Trim and fill analysis indicated respectively 0, 13, 12 ad 9 missing studies for cardiovascular mortality, myocardial infarction, stroke, and the combined outcome. However, the aforementioned lack of diversity in the effect size and standard error of the datapoints hampers funnel plot asymmetry testing, which reduces the reliability of these results.
Egger’s regression P = 0.039; trim and fill +9 studies. Filled circles represent studies included in the meta-analysis, open circles represent studies added during trim and fill analysis.
Data synthesis—observational studies
Table 1 summarizes the findings of retrospective cohort studies and case control studies on cardiovascular mortality, myocardial infarction, stroke, and the combined outcome of cardiovascular mortality, myocardial infarction, and stroke.
Cardiovascular mortality.
Three cohort studies of sufficient quality were retrieved that compared current users with an untreated or previously treated cohort. All three cohort studies had a retrospective design. Two small studies from the same group (one in patients without gout, and one in patients with gout) observed a reduced cardiovascular and total mortality that was only significant in the study with gout patients [51, 52]. The third study with a much larger sample size compared users with non-users and observed a non-significant reduction in cardiovascular mortality and a significant reduction in total mortality [55].
Myocardial infarction.
Four cohort studies of sufficient quality were retrieved that compared a cohort of current users with an untreated or previously treated cohort. All four studies have a retrospective study design. In two of these studies, allopurinol treatment was associated with a dose dependent reduction in myocardial infarctions [56, 57]. Another study showed benefit associated with allopurinol treatment, but did not analyse a potential relation with dose [59]. One study did not show any benefit on myocardial infarction, but did so on the combined endpoint of stroke, myocardial infarction and cardiovascular mortality [55]. Unfortunately, the allopurinol dose was not reported in this study.
Four case control studies were retrieved that investigated the relation between use of allopurinol and acute coronary events. All four case-control studies were of good quality. De Abajo et al. observed an adjusted odds ratio of 0.52 (95% CI 0.33 to 0.83) in favour of allopurinol (versus no allopurinol). This benefit was fully driven by men (0.44; 95% CI 0.25 to 0.76) with a lack of benefit in women (0.90; 0.36 to 2.23). The benefit was only observed at higher doses (300 mg or higher) and prolonged treatment duration (> 180 days) [63]. Similar observations have been reported by Rodriguez-Martin et al. regardless the presence of gout [67]. Grimaldi-Bensouda et al. also found a reduced risk in those who used allopurinol (adjusted OR 0.73 (95% CI 0.54 to 0.99) [64]. These investigators did not observe a dose response relationship but the majority (>70%) of the allopurinol users in this study used only a dose of 200 mg/day or less. These authors also investigated the association between colchicine and myocardial infarction and found no colchicine-related protection (adjusted OR 1.17 (0.70 to 1.93)). In contrast, Liao et al. observed an increased risk of myocardial infarction in patients with prescribed allopurinol: adjusted OR 2.2 (95% CI 1.7–2.7), which occurred at all dose levels and increased with dose [66]. Liao et al. studied only patients with an age of at least 65 years.
Stroke.
Six cohort studies of sufficient quality were retrieved that compared a cohort of current users with an untreated or previously treated cohort (Table 1). All studies have a retrospective study design. Four of these studies also reported myocardial infarction, with one study reporting only the combined outcome of myocardial infarction and stroke [59]. Three of these studies reported a significant beneficial association between use of allopurinol and incidence of strokes [57, 59, 60], and three did not [53, 55, 62]. As for myocardial infarction, MacIsaac et al. reported dose dependence: only a dose of 300 mg/day or higher was associated with a lower risk for strokes [57]. One study only reported on stroke outcome and observed benefit associated with allopurinol in the analysis that was restricted to those who had a duration of exposure of at least half a year [60]. Larsen et al. did not observe a significant relation between allopurinol use and stroke as for myocardial infarction, but a significant association with the combined cardiovascular outcome of stroke, myocardial infarction and mortality [55].
One single case-control study was retrieved that studied the relation between allopurinol therapy and ischemic stroke [65]. They studied 14937 cases of first-time ischemic stroke and compared them with an equal number of age and gender matched controls. They included the cumulative duration of allopurinol use in their analysis. They adjusted for age (which was slightly different between controls and cases despite matching) and found an exposure dependent reduction of adjusted odds ratio for ischemic stroke: 1.12 (1.04–1.21) for exposure < 1 year, 0.97 (0.81–1.16) for 1–3 years and 0.74 (0.57–0.96) for more than 3 years of exposure.
Combined outcome.
Only two retrospective cohort studies with sufficient quality reported this combined outcome [55, 61]. Larsen et al. observed a significant association between allopurinol use and reduced cardiovascular death and event rate. Unfortunately, they did not study the impact of allopurinol dose on this relationship. Wei et al. showed a similar trend in a much smaller group of patients. Interestingly, within the group of allopurinol users they showed a significant association between allopurinol dose and cardiovascular events (see Table 1).
Grading of the quality of evidence
Grading of quality of the body of evidence is summarized in Table 2. The quality of evidence derived from the randomized controlled trials was classified as low (stroke and cardiovascular mortality as outcomes) to moderate (myocardial infarction and the combined outcome of myocardial infarction, stroke and cardiovascular mortality. Down-grading of trials was performed because of serious risk of bias (all outcomes) and imprecision (mortality and stroke). None of the trials was specifically designed to detect these outcomes. Twenty out of 26 trials reported zero events. In these 20 trials 1059 patients were treated with allopurinol and 907 patients received placebo. Thus, studies that reported events included only a minority of randomized patients (938 out of the 2904 randomized patients). This is further supported by the funnel plot analysis indicating overrepresentation of studies with a relative risk around 1. The results on stroke and cardiovascular mortality were further downgraded because of the low number of events, resulting in lack of statistical power and therefore imprecision of the results. We did not find significant inconsistency in the results of the included trials. All studies reporting myocardial infarctions showed benefit for allopurinol and no heterogeneity was found. For the combined outcome all but one study reporting events showed benefit for allopurinol, again tests for heterogeneity were insignificant.
Discussion
This systematic review explored the scientific literature on the impact of allopurinol on cardiovascular outcomes (cardiovascular mortality, myocardial infarction, stroke) in patients with hyperuricemia (with or without gout) and preserved renal function (average MDRD > 30 ml/min/1.73 m2). Both randomized clinical trials (allopurinol versus placebo or no treatment) and observational studies (retrospective and prospective cohort studies and case-control studies) were included in this review.
Principal findings
The meta-analysis of available randomized clinical trials showed a significant reduction in the combined outcome (cardiovascular death, myocardial infarction and stroke) that was driven by a significant reduction in myocardial infarction. No significant reduction in cardiovascular mortality or stroke was found. Data retrieved from observational studies was generally consistent with the results from the meta-analysis: the majority of these studies showed that use of allopurinol is associated with reduced cardiovascular events. In these studies, both myocardial infarction and stroke were reduced in patients on allopurinol therapy. Cardiovascular benefit was only observed when allopurinol therapy was prolonged for more than 6 months and when an appropriate allopurinol dose was administered (300 mg or more/day) or sufficient reduction of serum urate concentration was achieved (<0.36 mmol/l).
Strengths and weaknesses of the study
The quality of the body of evidence retrieved RCTs was low to moderate. Major reasons for down grading were reporting bias and imprecision due to low event-rates. Most studies did not report cardiovascular events as their primary outcome and events had to be extracted from adverse event reporting. Quality of this reporting was often unclear with studies not specifically listing cardiovascular events in their adverse event reporting or not specifying the nature of the event. The observational study data provides some additional insights into limitations of the current meta-analysis. First, the average follow-up time of the trials included in this meta-analysis was around 30–40 weeks, and in many studies < 8 weeks. This may account for the low incidence of cardiovascular events in the retrieved RCTs. Furthermore, two of the six trials with cardiovascular events used a dose less than 300 mg/day [40, 43], while in a third relatively large trial only 69% of the treated patients used the highest applied dose of 300 mg. Interestingly, serum urate concentration was not allowed as a criterium to up-titrate the dose to its maximum allowed dose of 300 mg/day in this trial [44]. This suggests that the meta-analysis may have underestimated the maximum benefit of allopurinol by using suboptimal dose. Although the reviewed observational studies used sophisticated epidemiological techniques to prevent bias as much as possible (propensity score matching to reduce bias by indication and covariate analyses to adjust for residual imbalance in baseline cardiovascular risk between treated and untreated patients), residual bias cannot be excluded. In particular the so called ‘healthy adherence bias should be mentioned. Adherence to allopurinol (and therefore dose and therapy duration) could be associated with a healthy lifestyle in general (healthy diet, physical activity, non-smoking). This healthy lifestyle rather than allopurinol exposure could play a role in the observed cardiovascular benefit. Recently, allopurinol has been associated with a slightly increased mortality during the first 30 days after initiation of this therapy [68]. This phenomenon, if confirmed, could have resulted in survival bias, overestimating the benefit of allopurinol adherence.
Comparison with other studies
Our result differs slightly from a recent meta-analysis by Bredemeier et al. [69]. They did not show a significant impact of allopurinol on major adverse cardiovascular events (MACE) in their overall analysis. However, they observed reduced MACE in the subset of trials that included patients with previous ischemic events. Two important differences between the two studies probably explain this difference: Bredemeier et al. included trials in patients without hyperuricemia and their meta-analysis did not include the trial recently published by Badve et al. in patients with a high cardiovascular risk profile [44]. In addition, their results suggest that allopurinol doses >300 mg/day may increase cardiovascular events, conflicting with observational study data. As mentioned, they included studies with patients with normal uric acid, and the achieved urate concentration in these patients on allopurinol may very well be much lower than in patients with hyperuricemia that use allopurinol. Experimental studies in humans indicate that extremely low urate concentrations are associated with vascular injury (‘J-curve’), hypothetically explained by a direct anti-oxidant effect of uric acid, which could explain these conflicting findings [70].
In addition, 6 other meta-analysis on the comparison between allopurinol and placebo on cardiovascular outcome have been published. Three had a similar approach to ours [71–73]. One reported only all-cause mortality and cardiovascular mortality in 2 observational studies and did not find a consistent effect of allopurinol versus placebo [74]. Two studies had a comparable research question but different patient group: heart failure [75], or patients undergoing cardiac revascularization [76]. Guedes observed a similar effect of allopurinol [71]. They included only one singe RCT and only four observational studies. The RCT is a study with a broad definition of cardiovascular outcome, including hospitalization for heart failure and arrhythmia [22]. This study was replaced in our analysis by a later publication on the same study with a longer follow-up [43]. Ying et al. did not exclude studies in patients without hyperuricemia, and otherwise included less RCT in their database in comparison to our systematic review. These differences may have accounted for their non-significant effect on Major Adverse Cardiovascular Events (MACE). They did not include observational studies [72]. Zhang et al. performed a network meta-analysis including trials that compared allopurinol with febuxostat or placebo [73]. They only included 10 RCTs in their review and did not include some relevant RCTs comparing allopurinol with placebo, nor observational studies. They did not observe a significant effect of either febuxostat nor allopurinol on MACE or cardiovascular mortality.
Meaning of the study and future research
The present meta-analysis shows a significant reduction in the incidence of combined cardiovascular events in hyperuricemic patients with preserved renal function (eGFR>30 ml/min/1.73 m2) treated with allopurinol. However, there are significant limitations in the quality of the data mainly due to imprecision of results because of its dependence on (likely insufficient) adverse event reporting. Due to these limitations, the results of this meta-analysis do not support its implementation in routine cardiovascular risk management. On the other hand, the analysis provides an important signal for a potentially impressive efficacy of allopurinol to prevent cardiovascular events. Therefore, there is still an urgent need to provide high level trial evidence for the use of allopurinol in cardiovascular protection. Observational study data provide important information on how such a trial should be designed. Based on this systematic review we stress the need for a well-designed randomized double-blind placebo-controlled trial that investigates the benefit of allopurinol on cardiovascular outcome. In this trial, allopurinol should be administered at a dose of 200–400 mg and preferably for at least 3 years. Furthermore, the statistical power should be sufficient to detect a reduction of at least 25% in cardiovascular events.
Supporting information
S1 Table. Characteristics of included studies.
a: RCH = retrospective cohort study, CACO = case-control study, RCT = randomized controlled trial, CROSS = crossover study, NA = not applicable, NR = not reported, CM = incidence of cardiovascular mortality, MI = incidence of myocardial infarction, S = incidence of stroke, ALLO = allopurinol, BENZ = benzbromarone, COL = colchicine; b: ALAT = alanine transaminase, AP = angina pectoris, ASAT = aspartate transaminase, BMI = body mass index, BNP = brain natriuretic peptide, BP = blood pressure, BUN = blood urea nitrogen, CKD = chronic kidney disease, Cr = creatinine, CRP = C-reactive protein, eGFR = estimated glomerular filtration rate, HDL = high-density lipoprotein, HF = heart failure, LDL = low-density lipoprotein, LVEF = left ventricular ejection fraction, NSTEMI = non-ST-elevation myocardial infarction, PAD = peripheral artery disease, STEMI = ST-elevation myocardial infarction, UA = uric acid.
https://doi.org/10.1371/journal.pone.0260844.s002
(DOCX)
S2 Table. Risk of bias CC CH Cross.
a: CACO = case-control study, PCH = prospective cohort study, RCH = retrospective cohort study.
https://doi.org/10.1371/journal.pone.0260844.s003
(DOCX)
S3 Table. Weights of studies in meta-analysis.
https://doi.org/10.1371/journal.pone.0260844.s004
(DOCX)
S1 Fig. Risk of bias of RCT’s.
Green circle represents low risk of bias; yellow circle represents unclear risk of bias; red circle represents high risk of bias.
https://doi.org/10.1371/journal.pone.0260844.s005
(DOCX)
S2 Fig. Forest plots.
a: Overall effect of allopurinol on cardiovascular mortality: p = 0.738; b: Overall effect of allopurinol on myocardial infarction: p = 0.015; c: Overall effect of allopurinol on stroke: p = 0.993.
https://doi.org/10.1371/journal.pone.0260844.s006
(DOCX)
S3 Fig. Funnel plots.
a: Egger’s regression P = 0.49; trim and fill +0 studies. Filled circles represent studies included in the meta-analysis, open circles represent studies added during trim and fill analysis; b: Egger’s regression P<0.0001; trim and fill +13 studies. Filled circles represent studies included in the meta-analysis, open circles represent studies added during trim and fill analysis; c: Egger’s regression P = 0.14; trim and fill +12 studies. Filled circles represent studies included in the meta-analysis, open circles represent studies added during trim and fill analysis.
https://doi.org/10.1371/journal.pone.0260844.s007
(DOCX)
References
- 1. Cortes J, Moore JO, Maziarz RT, Wetzler M, Craig M, Matous J, et al. Control of plasma uric acid in adults at risk for tumor Lysis syndrome: efficacy and safety of rasburicase alone and rasburicase followed by allopurinol compared with allopurinol alone—results of a multicenter phase III study. J Clin Oncol. 2010;(27):4207–13. pmid:20713865
- 2. Abou Chakra M, Dellis AE, Papatsoris AG, Moussa M. Established and recent developments in the pharmacological management of urolithiasis: an overview of the current treatment armamentarium. Expert Opin Pharmacother. 2020;21(1):85–96. Epub 2019/11/13. pmid:31714803.
- 3. Disveld IJM, Zoakman S, Jansen T, Rongen GA, Kienhorst LBE, Janssens H, et al. Crystal-proven gout patients have an increased mortality due to cardiovascular diseases, cancer, and infectious diseases especially when having tophi and/or high serum uric acid levels: a prospective cohort study. Clin Rheumatol. 2019;38(5):1385–91. Epub 2019/04/01. pmid:30929152.
- 4. Disveld IJM, Fransen J, Rongen GA, Kienhorst LBE, Zoakman S, Janssens H, et al. Crystal-proven Gout and Characteristic Gout Severity Factors Are Associated with Cardiovascular Disease. J Rheumatol. 2018;45(6):858–63. Epub 2018/04/17. pmid:29657151.
- 5. Tsai CW, Lin SY, Kuo CC, Huang CC. Serum Uric Acid and Progression of Kidney Disease: A Longitudinal Analysis and Mini-Review. PLoS One. 2017;12(1):e0170393. Epub 2017/01/21. pmid:28107415; PubMed Central PMCID: PMC5249245.
- 6.
Nederlands Huisartsen Genootschap (NHG). Artritis Guideline. richtlijnen.nhg.org: Nederlands Huisartsen Genootschap; 2017 [cited 2020 10-10-2020]. 2.0:[Available from: https://richtlijnen.nhg.org/standaarden/artritis.
- 7.
Nederlandse Vereniging voor Reumatologie (NVR). Gout Guideline: Nederlandse vereniging voor reumatologie (Dutch Federation of Rheumatology); 2014 [2020/10/19]. Available from: https://richtlijnendatabase.nl/richtlijn/jicht/jicht_-_startpagina.html.
- 8. White WB, Saag KG, Becker MA, Borer JS, Gorelick PB, Whelton A, et al. Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. N Engl J Med. 2018;378(13):1200–10. PubMed PMID: WOS:000428580500006. pmid:29527974
- 9. Kojima S, Matsui K, Hiramitsu S, Hisatome I, Waki M, Uchiyama K, et al. Febuxostat for Cerebral and CaRdiorenovascular Events PrEvEntion StuDy. European Heart Journal. 2019;40(22):1778–86. pmid:30844048
- 10. Mackenzie IS, Ford I, Nuki G, Hallas J, Hawkey CJ, Webster J, et al. Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet. 2020;396(10264):1745–57. Epub 2020/11/13. (20)32234-0. pmid:33181081.
- 11. Su CY, Shen LJ, Hsieh SC, Lin LY, Lin FJ. Comparing Cardiovascular Safety of Febuxostat and Allopurinol in the Real World: A Population-Based Cohort Study. Mayo Clin Proc. 2019;94(7):1147–57. PubMed PMID: WOS:312725651100011.
- 12. Zhang M, Solomon DH, Desai RJ, Kang EH, Liu J, Neogi T, et al. Assessment of Cardiovascular Risk in Older Patients With Gout Initiating Febuxostat Versus Allopurinol. Circulation. 2018;138(11):1116–26. pmid:29899013.
- 13.
Review Manager (RevMan) software. Version 5.3 ed. Copenhagen: The Nordic Cochrane Centre: The Cochrane Collaboration; 2014.
- 14. Wallace BCD I.J.; Trikalinos T.A.; Lau J.; Trow P.; Schmid C.H. Closing the gap between methodologists and end-users: R as a computational back-end. J Stat Soft. 2012;49(5):1–15.
- 15. Singh JA, Yu SH. Are allopurinol dose and duration of use nephroprotective in the elderly? A Medicare claims study of allopurinol use and incident renal failure. Annals of the Rheumatic Diseases. 2017;(1):133–9. pmid:27296322
- 16. Vargas-Santos AB, Peloquin CE, Zhang YQ, Neogi T. Association of Chronic Kidney Disease With Allopurinol Use in Gout Treatment. JAMA Intern Med. 2018;178(11):1526–33. PubMed PMID: WOS:000449215200020. pmid:30304329
- 17.
The GRADE Working Group. GRADE handbook for grading quality of evidence and strength of recommendations. Updated October 2013. Available from www.guidelinedevelopment.org/handbook.2013.
- 18.
developed by Evidence Prime I. GRADEpro GDT: GRADEpro Guideline Development Tool [Software] available from gradepro.org. McMaster University; 2020.
- 19. Borgi L, McMullan C, Wohlhueter A, Curhan GC, Fisher ND, Forman JP. Effect of Uric Acid-Lowering Agents on Endothelial Function A Randomized, Double-Blind, Placebo-Controlled Trial. Hypertension. 2017;(2):243–8. pmid:28028194
- 20. Bowden RG, Shelmadine BD, Moreillon JJ, Deike E, Griggs JO, Wilsond RL. Effects of Uric Acid on Lipid Levels in CKD Patients in a Randomized Controlled Trial. Cardiol Res. 2013;4(2):56–63. PubMed PMID: WOS:000215198500002. pmid:28352421
- 21. Givertz MM, Anstrom KJ, Redfield MM, Deswal A, Haddad H, Butler J, et al. Effects of Xanthine Oxidase Inhibition in Hyperuricemic Heart Failure Patients: The Xanthine Oxidase Inhibition for Hyperuricemic Heart Failure Patients (EXACT-HF) Study. Circulation. 2015;(20):1763–71. pmid:25986447
- 22. Goicoechea M, de Vinuesa SG, Verdalles U, Ruiz-Caro C, Ampuero J, Rincon A, et al. Effect of allopurinol in chronic kidney disease progression and cardiovascular risk. Clin J Am Soc Nephrol. 2010;(8):1388–93. pmid:20538833
- 23. Hosoya T, Sasaki T, Ohashi T. Clinical efficacy and safety of topiroxostat in Japanese hyperuricemic patients with or without gout: a randomized, double-blinded, controlled phase 2b study. Clin Rheumatol. 2017;(3):649–56. pmid:27832384
- 24. Huang Y, Zhang CY, Xu ZQ, Shen JH, Zhang XG, Du HH, et al. Clinical Study on efficacy of allopurinol in patients with acute coronary syndrome and its functional mechanism. Hellenic Journal of Cardiology. 2017;(5):360–5. pmid:28093243
- 25. Jalal DI, Decker E, Perrenoud L, Nowak KL, Bispham N, Mehta T, et al. Vascular Function and Uric Acid-Lowering in Stage 3 CKD. J Am Soc Nephrol. 2017;(3):943–52. pmid:27620990
- 26. Kanbay M, Huddam B, Azak A, Solak Y, Kadioglu GK, Kirbas I, et al. A randomized study of allopurinol on endothelial function and estimated glomular filtration rate in asymptomatic hyperuricemic subjects with normal renal function. Clin J Am Soc Nephrol. 2011;(8):1887–94. pmid:21784838
- 27. Liu P, Chen Y, Wang B, Zhang F, Wang D, Wang Y. Allopurinol treatment improves renal function in patients with type 2 diabetes and asymptomatic hyperuricemia: 3-year randomized parallel-controlled study. Clin Endocrinol (Oxf). 2015;83(4):475–82. Epub 2014/11/18. pmid:25400252.
- 28. Liu P, Wang H, Zhang F, Chen Y, Wang D, Wang Y. The Effects of Allopurinol on the Carotid Intima-media Thickness in Patients with Type 2 Diabetes and Asymptomatic Hyperuricemia: A Three-year Randomized Parallel-controlled Study. Intern Med. 2015;(17):2129–37. pmid:26328636
- 29. Madero M, Castellanos FER, Jalal D, Villalobos-Martin M, Salazar J, Vazquez-Rangel A, et al. A pilot study on the impact of a low fructose diet and allopurinol on clinic blood pressure among overweight and prehypertensive subjects: a randomized placebo controlled trial. Journal of the American Society of Hypertension. 2015;(11):837–44. pmid:26329473
- 30. McMullan CJ, Borgi L, Fisher N, Curhan G, Forman J. Effect of Uric Acid Lowering on Renin-Angiotensin-System Activation and Ambulatory BP: A Randomized Controlled Trial. Clin J Am Soc Nephrol. 2017;(5):807–16. pmid:28320765
- 31. Momeni A, Shahidi S, Seirafian S, Taheri S, Kheiri S. Effect of Allopurinol in Decreasing Proteinuria in Type 2 Diabetic Patients. Iranian Journal of Kidney Diseases. 2010;(2):128–32. pmid:20404423
- 32. Pichholiya M, Yadav AK, Luhadia SK, Tahashildar J, Aseri ML. A comparative study of efficacy and safety of febuxostat and allopurinol in pyrazinamide-induced hyperuricemic tubercular patients. Indian J Pharmacol. 2016;(5):522–5. pmid:27721537
- 33. Poiley J, Steinberg AS, Choi YJ, Davis CS, Martin RL, McWherter CA, et al. A Randomized, Double-Blind, Active- and Placebo-Controlled Efficacy and Safety Study of Arhalofenate for Reducing Flare in Patients With Gout. Arthritis & Rheumatology. 2016;(8):2027–34. pmid:26989892
- 34. Rosenfeld JB. Effect of long-term allopurinol administration on serial GFR in normotensive and hypertensive hyperuricemic subjects. Adv Exp Med Biol. 1974:581–96. pmid:4832585
- 35. Schumacher HR Jr., Becker MA, Wortmann RL, Macdonald PA, Hunt B, Streit J, et al. Effects of febuxostat versus allopurinol and placebo in reducing serum urate in subjects with hyperuricemia and gout: a 28-week, phase III, randomized, double-blind, parallel-group trial. Arthritis Rheum. 2008;(11):1540–8. pmid:18975369
- 36. Segal MS, Srinivas TR, Mohandas R, Shuster JJ, Wen XR, Whidden E, et al. The effect of the addition of allopurinol on blood pressure control in African Americans treated with a thiazide-like diuretic. Journal of the American Society of Hypertension. 2015;(8):610–9. pmid:26140739
- 37. Sezer S, Karakan S, Atesagaoglu B, Acar FN. Allopurinol reduces cardiovascular risks and improves renal function in pre-dialysis chronic kidney disease patients with hyperuricemia. Saudi J Kidney Dis Transpl. 2014;(2):316–20. pmid:24625997
- 38. Shi Y, Chen W, Jalal D, Li Z, Chen W, Mao H, et al. Clinical outcome of hyperuricemia in IgA nephropathy: a retrospective cohort study and randomized controlled trial. Kidney Blood Press Res. 2012;(3):153–60. pmid:22116196
- 39. Siu YP, Leung KT, Tong MK, Kwan TH. Use of allopurinol in slowing the progression of renal disease through its ability to lower serum uric acid level. Am J Kidney Dis. 2006;(1):51–9. pmid:16377385
- 40. Taheraghdam AA, Sharifipour E, Pashapour A, Namdar S, Hatami A, Houshmandzad S, et al. Allopurinol as a preventive contrivance after acute ischemic stroke in patients with a high level of serum uric acid: a randomized, controlled trial. Med Princ Pract. 2014;(2):134–9. pmid:24296871
- 41. Takir M, Kostek O, Ozkok A, Elcioglu OC, Bakan A, Erek A, et al. Lowering Uric Acid With Allopurinol Improves Insulin Resistance and Systemic Inflammation in Asymptomatic Hyperuricemia. Journal of investigative medicine. 2015;(8):924–9. pmid:26571421
- 42. Taylor TH, Mecchella JN, Larson RJ, Kerin KD, MacKenzie TA. Initiation of Allopurinol at First Medical Contact for Acute Attacks of Gout: A Randomized Clinical Trial. American Journal of Medicine. 2012;(11):1126-+. pmid:23098865
- 43. Goicoechea M, de Vinuesa SG, Verdalles U, Verde E, Macias N, Santos A, et al. Allopurinol and Progression of CKD and Cardiovascular Events: Long-term Follow-up of a Randomized Clinical Trial. American Journal of Kidney Diseases. 2015;(4):543–9. pmid:25595565
- 44. Badve SV, Pascoe EM, Biostat M, Tiku A, Boudville N, Brown FG, et al. Effects of Allopurinol on the Progression of Chronic Kidney Disease. N Engl J Med. 2020;382(26):2504–13. PubMed PMID: WOS:000545636400021. pmid:32579811
- 45. Nicotero JA, Scheib ET, Martinez R, Rodnan GP, Shapiro AP. Prevention of hyperuricemia by allopurinol in hypertensive patients treated with chlorothiazide. N Engl J Med. 1970;(3):133–5. pmid:4902227
- 46. Doehner W, Schoene N, Rauchhaus M, Leyva-Leon F, Pavitt DV, Reaveley DA, et al. Effects of xanthine oxidase inhibition with allopurinol on endothelial function and peripheral blood flow in hyperuricemic patients with chronic heart failure: results from 2 placebo-controlled studies. Circulation. 2002;(22):2619–24. pmid:12045167
- 47. Bayram D, Sezer MT, Inal S, Altuntas A, Kidir V, Orhan H. The effects of allopurinol on metabolic acidosis and endothelial functions in chronic kidney disease patients. Clinical and Experimental Nephrology. 2015;(3):443–9. pmid:25082656
- 48. Eliseev MS, Denisov IS, Markelova EI, Glukhova SI, Nasonov EL. [Independent risk factors for severe cardiovascular events in male patients with gout: Results of a 7-year prospective study]. Ter Arkh. 2017;(5):10–9. pmid:28631693
- 49. Kanbay M, Ozkara A, Selcoki Y, Isik B, Turgut F, Bavbek N, et al. Effect of treatment of hyperuricemia with allopurinol on blood pressure, creatinine clearence, and proteinuria in patients with normal renal functions. Int Urol Nephrol. 2007;(4):1227–33. pmid:17701281
- 50. Tashchuk VK, Al Salama MVO, Amelina TM. Effectiveness of Allopurinol and Quercetin inclusion in a complex treatment of stable angina pectoris—peculiarities of biomarkers and homeostatic indices changes. Zaporozhye Medical Journal. 2017;(3):265–9.
- 51. Chen JH, Lan JL, Cheng CF, Liang WM, Lin HY, Tsay GJ, et al. Effect of Urate-Lowering Therapy on All-Cause and Cardiovascular Mortality in Hyperuricemic Patients without Gout: A Case-Matched Cohort Study. Plos One. 2015;(12). pmid:26683302
- 52. Chen JH, Lan JL, Cheng CF, Liang WM, Lin HY, Tsay GJ, et al. Effect of Urate-lowering Therapy on the Risk of Cardiovascular Disease and All-cause Mortality in Patients with Gout: A Case-matched Cohort Study. Journal of Rheumatology. 2015;(9):1694–701. pmid:26077411
- 53. Ju C, Lai RWC, Li KHC, Hung JKF, Lai JCL, Ho J, et al. Comparative cardiovascular risk in users versus non-users of xanthine oxidase inhibitors and febuxostat versus allopurinol users. Rheumatology (Oxford). 2020;59(9):2340–9. Epub 2019/12/25. pmid:31873735.
- 54. Kim SC, Schneeweiss S, Choudhry N, Liu J, Glynn RJ, Solomon DH. Effects of xanthine oxidase inhibitors on cardiovascular disease in patients with gout: a cohort study. Am J Med. 2015;(6):653.e7-.e16. pmid:25660249
- 55. Larsen KS, Pottegard A, Lindegaard HM, Hallas J. Effect of Allopurinol on Cardiovascular Outcomes in Hyperuricemic Patients: A Cohort Study. Am J Med. 2016;(3):299-306.e2. pmid:26589484
- 56. Lin HC, Daimon M, Wang CH, Ho Y, Uang YS, Chiang SJ, et al. Allopurinol, benzbromarone and risk of coronary heart disease in gout patients: A population-based study. Int J Cardiol. 2017:85–90. pmid:28202260
- 57. MacIsaac RL, Salatzki J, Higgins P, Walters MR, Padmanabhan S, Dominiczak AF, et al. Allopurinol and Cardiovascular Outcomes in Adults With Hypertension. Hypertension. 2016;(3):535–40. pmid:26865199
- 58. Pérez Ruiz F, Richette P, Stack AG, Karra Gurunath R, García de Yébenes MJ, Carmona L. Failure to reach uric acid target of <0.36 mmol/L in hyperuricaemia of gout is associated with elevated total and cardiovascular mortality. RMD Open. 2019;5(2):e001015. Epub 2019/11/02. pmid:31673414; PubMed Central PMCID: PMC6803010.
- 59. Singh JA, Ramachandaran R, Yu SH, Curtis JR. Allopurinol use and the risk of acute cardiovascular events in patients with gout and diabetes. Bmc Cardiovascular Disorders. 2017. pmid:28288564
- 60. Singh JA, Yu S. Allopurinol and the risk of stroke in older adults receiving medicare. BMC Neurol. 2016;(1):164. pmid:27604082
- 61. Wei L, Mackenzie IS, Chen Y, Struthers AD, MacDonald TM. Impact of allopurinol use on urate concentration and cardiovascular outcome. Br J Clin Pharmacol. 2011;(4):600–7. pmid:21395653
- 62. Yen F-S, Hsu C-C, Li H-L, Wei JC-C, Hwu C-M. Urate-lowering therapy may mitigate the risks of hospitalized stroke and mortality in patients with gout. PLOS ONE. 2020;15(6):e0234909. pmid:32574194
- 63. de Abajo FJ, Gil MJ, Rodriguez A, Garcia-Poza P, Alvarez A, Bryant V, et al. Allopurinol use and risk of non-fatal acute myocardial infarction. Heart. 2015;(9):679–85. pmid:25561685
- 64. Grimaldi-Bensouda L, Alperovitch A, Aubrun E, Danchin N, Rossignol M, Abenhaim L, et al. Impact of allopurinol on risk of myocardial infarction. Annals of the Rheumatic Diseases. 2015;(5):836–42. pmid:24395556
- 65. Lai SW, Lin CL, Liao KF. Case-control study examining the association between allopurinol use and ischemic cerebrovascular disease. Journal of Investigative Medicine. 2019;67(1):48–51. PubMed PMID: WOS:000457712500007. pmid:30042112
- 66. Liao KF, Lin CL, Lai SW. Allopurinol use associated with increased risk of acute myocardial infarction in older people in a case-control study. Tzu Chi Med J. 2019;31(4):276–9. PubMed PMID: WOS:000486373600013. pmid:31867258
- 67. Rodríguez-Martín S, de Abajo FJ, Gil M, González-Bermejo D, Rodríguez-Miguel A, Barreira-Hernández D, et al. Risk of Acute Myocardial Infarction Among New Users of Allopurinol According to Serum Urate Level: A Nested Case-Control Study. J Clin Med. 2019;8(12):E2150. pmid:31817395.
- 68. Kim HJA, H.S.; Kim J.; Ghang B. OP0168 Cardiovascular event associated with initiating allopurinol and febuxostat—Acute gout attack and cardiovascular gout attack. Annals of the Rheumatic Diseases. 2020;79:106.
- 69. Bredemeier M, Lopes LM, Eisenreich MA, Hickmann S, Bongiorno GK, d’Avila R, et al. Xanthine oxidase inhibitors for prevention of cardiovascular events: a systematic review and meta-analysis of randomized controlled trials. BMC Cardiovasc Disord. 2018;18(1):24. pmid:29415653; PubMed Central PMCID: PMC5804046.
- 70. Sugihara S, Hisatome I, Kuwabara M, Niwa K, Maharani N, Kato M, et al. Depletion of Uric Acid Due to SLC22A12 (URAT1) Loss-of-Function Mutation Causes Endothelial Dysfunction in Hypouricemia. Circulation journal: official journal of the Japanese Circulation Society. 2015;79(5):1125–32. pmid:25739858.
- 71. Guedes M, Esperanca A, Pereira AC, Rego C. What is the effect on cardiovascular events of reducing hyperuricemia with allopurinol? An evidence-based review. Rev Port Cardiol. 2014;33(11):727–32. Epub 2014/12/03. pmid:25444231.
- 72. Ying H, Yuan H, Tang X, Guo W, Jiang R, Jiang C. Impact of Serum Uric Acid Lowering and Contemporary Uric Acid-Lowering Therapies on Cardiovascular Outcomes: A Systematic Review and Meta-Analysis. Front Cardiovasc Med. 2021;8:641062. Epub 2021/04/20. pmid:33869304; PubMed Central PMCID: PMC8044896.
- 73. Zhang S, Xu T, Shi Q, Li S, Wang L, An Z, et al. Cardiovascular Safety of Febuxostat and Allopurinol in Hyperuricemic Patients With or Without Gout: A Network Meta-Analysis. Front Med (Lausanne). 2021;8:698437. Epub 2021/07/03. pmid:34211992; PubMed Central PMCID: PMC8239361.
- 74. Hay CA, Prior JA, Belcher J, Mallen CD, Roddy E. Mortality in Patients With Gout Treated With Allopurinol: A Systematic Review and Meta-Analysis. Arthritis Care Res (Hoboken). 2021;73(7):1049–54. Epub 2020/04/15. pmid:32286732.
- 75. Kanbay M, Afsar B, Siriopol D, Dincer N, Erden N, Yilmaz O, et al. Effect of Uric Acid-Lowering Agents on Cardiovascular Outcome in Patients With Heart Failure: A Systematic Review and Meta-Analysis of Clinical Studies. Angiology. 2020;71(4):315–23. Epub 2020/02/01. pmid:32000517.
- 76. Ullah W, Khanal S, Khan R, Basyal B, Munir S, Minalyan A, et al. Efficacy of Allopurinol in Cardiovascular Diseases: A Systematic Review and Meta-Analysis. Cardiol Res. 2020;11(4):226–32. Epub 2020/07/01. pmid:32595807; PubMed Central PMCID: PMC7295562.