Hyperuricemia increases the risk of acute kidney injury: a systematic review and meta-analysis

Background Mounting evidence indicated that the elevated serum uric acid level was associated with an increased risk of acute kidney injury (AKI). Our goal was to systematically evaluate the correlation of serum uric acid (SUA) level and incidence of AKI by longitudinal cohort studies. Methods We searched electronic databases and the reference lists of relevant articles. 18 cohort studies with 75,200 patients were analyzed in this random-effect meta-analysis. Hyperuricemia was defined as SUA levels greater than 360-420 μmol/L (6–7 mg/dl), which was various according to different studies. Data including serum uric acid, serum creatinine, and incidence of AKI and hospital mortality were summarized using random-effects meta-analysis. Results The hyperuricemia group significantly exerted a higher risk of AKI compared to the controls (odds ratio OR 2.24, 95% CI 1.76-2.86, p < 0.01). Furthermore, there is less difference of the pooled rate of AKI after cardiac surgery between hyperuricemia and control group (34.3% vs 29.7%, OR 1.24, 95% CI 0.96-1.60, p = 0.10), while the rates after PCI were much higher in hyperuricemia group than that in control group (16.0% vs 5.3%, OR 3.24, 95% CI 1.93-5.45, p < 0.01). In addition, there were significant differences in baseline renal function at admission between hyperuricemia and control groups in most of the included studies. The relationship between hyperuricemia and hospital mortality was not significant. The pooled pre-operative SUA levels were higher in AKI group than that in the non-AKI group. Conclusions Elevated SUA level showed an increased risk for AKI in patients and measurements of SUA may help identify risks for AKI in these patients.


Background
Acute kidney injury (AKI) occurs commonly after cardiovascular surgery, in patients with sepsis, and after the administration of various nephrotoxins including contrast agents. The incidence of AKI has a significant effect on the outcomes. Prevention before any procedure is essential because no measures have been proven to effectively treat AKI. Therefore, if high-risk patients could be screened earlier, the clinician still would have opportunities to prevent AKI and further improve outcomes [1,2].
Uric acid is an end-product of purine degradation and is excreted via kidney. Many epidemiologic studies have suggested that hyperuricemia is associated with hypertension, cardiovascular diseases, diabetes mellitus and the progression of chronic kidney disease [3][4][5]. In addition, it is found that hyperuricemia is associated with acute kidney injury (AKI) in various statuses [6][7][8][9]. This meta-analysis was conducted to estimate whether hyperuricemia is an independent risk factor for incidence and prognosis of AKI. This effort hoped to raise awareness of the importance of hyperuricemia in the developing AKI.

Search strategy and data sources
We performed a computerized search to identify relevant published original studies (1985 to May 2016). Pubmed, Web of Science, Cochrwane Library, OVID and EMBASE databases were searched using medical subject headings (MeSH) or keywords. These words were "acute kidney failure, acute kidney injury, acute kidney dysfunction, acute kidney insufficiency, acute tubular necrosis, acute renal failure, acute renal injury, acute renal dysfunction, or acute renal insufficiency" and "hyperuricemia, or uric acid". This search was not limited to English language or publication type. We followed a prespecified protocol but this was not registered.

Selection criteria
An initial eligibility screen of all retrieved titles and abstracts was performed, and only studies reporting the relationship between serum uric acid (SUA) and AKI were selected for further review. The following included criteria were used for final selection: (1) studies reporting the incidence of AKI and pre-operative SUA Levels, (2) studies using clear definition of AKI, and hyperuricemia, (3) studies providing detailed information about the incidence of AKI, and/or hospital mortality. We restricted our search to clinical studies performed in adult populations. Studies without clear grouping or animal experimental studies were excluded.

Data extraction and quality assessment
Two reviewers (X.X.L and H.J.C) examined the studies independently, and disagreement was resolved by discussion. Data extraction included country of origin, year of publication, study period, study design, inclusion criteria, definition of hyperuricemia or grouping according to SUA, conclusions and patient characteristics (age and sex). Hyperuricemia was defined as SUA levels greater than 360-420 μmol/L (6-7 mg/dl), which was various according to different studies. The primary outcomes were odds ratio (OR) of SUA to predict incidence of AKI. The definition of AKI in all these included studied used the AKI network criteria [10] with minor modification and defined as an increase ≥0.3 mg/dL in the serum creatintine level within 48 h in the hospital or ICU ( Table 1). The second outcomes included SUA levels in AKI and No-AKI group and hospital mortality in hyperuricemia and control group. The study selection, data extraction and reporting of results were all based on the Preferred Reporting Items for Systematic reviews and Meta-Analyses checklist [11]. The quality of the cohort studies was assessed independently by pairs of two authors, using the Newcastle-Ottawa scale (NOS) [12], which allocates a maximum of 9 points for quality of the selection, comparability, and outcome of study populations. Study quality scores were defined as poor (0-3), fair (4-6), or good (7)(8)(9).

Data synthesis and statistical analysis
Review Manager (RevMan, Cochrane Collaboration, version 5.3) and Comprehensive Meta-Analysis (CMA version 2.0, Biostat) were used to perform the metaanalysis. Pooled estimates were obtained for incidence of AKI and hospital mortality, which were reported using random-effects meta-analysis based on the methods of DerSimonian and Laird [13]. Meta-analyses were performed using OR for dichotomous outcomes. All confidence intervals (CI) were reported at 95%. P-value statistical significance was measured at 0.05. Heterogeneity across trials was evaluated with using theI 2 index and the Q test p value. A p value of less than 0.05 and anI 2 index of more than 25% indicated the presence of interstudy heterogeneity [14]. Publication bias was assessed by constructing a funnel plot and Egger's regression test.

Study selection
The article selection process is outlined in Fig. 1. The electronic database searches identified 1272 citations. After removal of duplicates and preliminary screening, 84 articles were selected for full-text review for their relevance to this study and 18 studies were included in this systematic review. At the full-text review stage, 30 articles were not about AKI, 18 did not involve hyperuricemia and 15 were review. Seven studies were excluded from the primary meta-analysis as they did not report the detailed information, and the corresponding authors were unable to provide the requisite data. Agreement between investigators at the full-text review stage was excellent as indicated by a κ of 0.8.

Study description and quality assessment
A detailed description of the included studies is provided in Table 1. The included studies were published between 2006 and 2016, and were carried out in a wide range of countries. The total number of patients included in the primary meta-analysis was 75,200 with a median (interquartile range) of 559 (122-1774) patients per study. The detailed information of age and gender was also listed in Table 1. Overall study quality was good with a mean NOS score of 7.5 out of a possible 9 (range, 7-9) and with 11 studies (91.7%) receiving a NOS greater than or equal to 7 ( Table 2).

Effects of SUA on the incidence of AKI
Eleven observational studies with 70,264 patients reported the incidence of AKI. The pooled rates of AKI incidence in hyperuricemia group and control group were  (Fig. 4).

Subgroup analysis
Although the pooled rates of AKI incidence after cardiac surgery in hyperuricemia and control group were 34.3%  We also conducted subgroup analysis of prospective and retrospective cohort studies (Fig. 6). The pooled ORs of hyperuricemia on AKI were 2.87 (95% CI 1.43-5.76) and 2.11 (95% CI 1.63-2.75) respectively. In addition, to reduce the bias of included patients, we also analyzed studies with or without equal renal function, which was defined as serum creatintine or estimated glomerular filtration rate (eGFR) without significant different at admission between hyperuricemia and control groups. There were significant differences in renal function at admission between hyperuricemia and control groups in most of the included studies, while only two studies with equal renal function were included, and the pooled OR was 3.21 (95% CI 1. 22-8.44, p = 0.02) (Fig. 7).

Effects of SUA on hospital mortality
Five studies with 3735 patients provided the hospital mortality. The pooled rates of hospital mortality in hyperuricemia group and control group were 8.9% (95% CI, 2.1-30.8%) and 5.0% (95% CI, 1.0-21.9%) respectively (OR  (Figs. 2c and 8). The relationship between hyperuricemia and hospital mortality was not significant.

Publication bias
The funnel plots showed no evidence of publication bias. Egger's test for a regression intercept gave a p-value of 0.696 for effects of hyperuricemia on incidence of AKI, indicating no publication bias.

Discussion
AKI is one of the most serious complications with a reported mortality rate of 15% in hospitalized patients [15]. Our meta-analysis showed that HUA is a critical and potential risk factor for the incidence of AKI, not only in preoperative patients as reported previously but also in all hospitalized patients.
In this meta-analysis, we found that the pooled rates of AKI incidence in hyperuricemia group were much higher than that in the control group. The underlying reasons were analyzed as follows. Firstly,majority of uric acid is excreted by the kidneys and accounts for 70%. It should be noted that approximately 90-95% of the filtered uric acid from glomerular is absorbed, mostly by proximal tubules [16,17]. Secreted uric acid by the renal tubules is very little. Consequently the SUA concentration depends on glomerular filtration and subsequent tubular reabsorption function. There is mounting evidence to consider SUA as a clear marker for chronic kidney disease or an independent risk factor for the development of chronic kidney disease [18,19]. A number of studies demonstrated that pre-existing chronic kidney disease increases the risk of AKI. Ishani et al. reported that the incidence of AKI was 8.8% in patients with chronic kidney disease versus 2.3% in patients without chronic kidney disease [20]. Pannu N et al. found that the risk of AKI was 18-fold higher in patients with an eGFR less than 30 ml/min/1.73 m 2 than in those with an Fig. 6 Effects of hyperuricemia on incidence of acute kidney injury in prospective and retrospective studies Fig. 7 Effects of hyperuricemia on incidence of acute kidney injury in patients with or without equal renal function at admission eGFR more than 60 ml/min/1.73 m 2 [21]. Therefore, patients with increased SUA may already have the subclinical chronic renal dysfunction, leading them to be more vulnerable to AKI. In addition, we did an adjustment for the important covariate baseline GFR or serum creatinine. Unfortunately, there were only two included studies with equal renal function at admission, the results from which was more convincing.
Seconding, an elevated SUA concentration has been found to be associated with damage of impartment organs and result to many diseases such as hypertension [17,22], metabolic syndrome [23], atherosclerosis [24], myocardial infarction [25], diabetes mellitus [4], stroke [26] and so on. All of the above diseases are most common risk factor of AKI, which make it sense that the incidence of AKI in the hyperuricemic patients is higher than those in the normouricemic patients.
A number of studies supported that uric acid is an independent risk factor of cardiovascular disease. The incidence rate of cardiovascular disease in patients with hyperuricemia is higher than that in the normal population [27]. A meta-analysis showed that incidence of coronary heart disease (CHD) in the hyperuricemic patients was 1.34 times (95% CI 1. 19-1.49) than that in the normouricemic patients [5]. Patients with CHD combined with hyperuricemia have higher incidence of myocardial infarction. The global number of cardiac surgeries or PCI each year is approximately 2 million [28,29] and one of the most common and serious post-operative complications is AKI. A current meta analysis found that the incidence of AKI after cardiac surgery was 22.3% around the world (95% CI 19.8-25.1) [2]. The incidence of PCI-induced AKI has been estimated between 2% and 30% depending mainly on baseline renal function, which is increasing along with the higher prevalence of CHD year by year [15,29]. Our results suggest that higher pre-PCI SUA increased risk of AKI. We speculated that the patients with increased SUA maybe undergo more PCI, consequently have more incidence of AKI. In addition, it was found contrast agents have a uricosuric effect through enhancing renal tubular secretion of uric acid [30], which may promote renal injury caused by possible nephrotoxic effect of uric acid. However, there are more complex risk factors and mechanisms of AKI incidence after cardiac surgery than PCI, which led to less difference of the pooled rate of AKI between hyperuricemia and control group. Moreover, there need more studies to confirm the prognostic role of SUA in AKI incidence after cardiac surgery.
Finally, it is well-known that AKI is resulted from multiple and interactive pathways. Uric acid itself can cause AKI due to several mechanisms ranging from direct tubular toxicity (crystal induced injury) [9] to indirect injury (secondary to vasoconstriction, oxidative stress, inflammatory and so on). In both animal and human models, uric acid is found to inhibit proliferation and migration of endothelial cell and cause dysfunction and apoptosis of endothelial cell [31,32]. Animal experimental studies suggest that uric acid may cause renal vasoconstriction via inhibiting of renal nitric oxide synthase to reduce product of nitric oxide in endothelial cell [31] and via stimulating of the renin-angiotensin system [32]. Renal vasoconstriction is a common pathogenic factor in the progression of AKI [33]. Inflammatory and oxidative stress are two of important mechanisms of AKI [34]. Experimentally, it has been found that uric acid activates inflammatory transcription factor nuclear factor-κB signaling pathway [35]. Increasing SUA also stimulates the expression of pro-inflammatory systemic cytokine i.e. tumor necrosis factor α [36], and the local chemokines, i.e. monocyte chemotactic protein 1 in the kidney [37]. High SUA levels induced oxidative damage of proximal tubule cell by activating nicotinamide adenine dinucleotide phosphate (NADPH) oxidase [38]. Therefore, SUA may be involved in the progress of AKI and contribute to higher incidence of AKI in the patients with hyperuricemia. Regardless of whether elevated SUA is solely a predictive factor of AKI or an independent risk factor of AKI, careful attention is warranted.
Thus, we wonder if uric acid lowering therapy could decrease the risk for developing AKI. At present, no trials showed that lowering SUA may provide benefit in preventing AKI. Allopurinol was once used in the hyperuricemic patients before cardiovascular surgery to reduce oxidative stress and then improve cardiovascular outcomes [39]. However, it was found that allopurinol couldn't prevent the incidence of AKI after cardiac surgery in these studies [40]. After that, researchers confirmed the protective role of allopurinol in the renal ischemia/reperfusion injury in rats [41,42]. In addition, in the cisplatin-induced AKI models, the uric acid lowering drugs rasburicase [43] and febuxostat [44] could attenuate renal injury by their antioxidant, anti-inflammatory, and cytoprotective effects. A prospective, randomized pilot trial with 26 cardiac surgery patients with hyperuricemia showed that there was no significant difference of postoperative serum creatinine between subjects receiving rasburicase and the control group. However, urine NGAL tended to be lower in the rasburicase group, which suggested that lowing uric acid before surgery might protect against renal tubular injury [45]. In Sezai A et al. study, febuxostat had a renoprotective effect with a significant earlier decrease of UA after cardiac surgery in hyperuricemic patients compared with allopurinol [46]. Therefore, we postulated that early intervention to decrease SUA levels may lower the risk of developing AKI.

Strengths and limitations
To the best of our knowledge, this study is the first to systematically evaluate the indicated effect of SUA on the incidence of AKI especially after cardiac surgery and PCI. It included data more than 75,000 patients from 18 studies. We analyzed these studies in detail considering the effect of renal function at admission and study design.
However, the present study may have limitations. Firstly, if there were more randomized controlled trials with high quality and large samples in this metaanalysis, these results would be more convincing. Secondly, Kanda et al. indicated that SUA level has a U-shaped association with loss of kidney function and low SUA (male <5 mg/dl; female <3.6 mg/dl) is also a candidate predictor of chronic kidney disease [47]. We are only focused on the role of hyperuricemia in AKI without referring hypouricemia which will need more studies in the future.

Conclusion
This meta-analysis demonstrated that elevated SUA levels could be associated with an increased risk of developing AKI especially in the patients after cardiac surgery and PCI.