Statins use and the risk of acute kidney injury: a meta-analysis

Abstract

The association between statins use and the risk of acute kidney injury (AKI) remains elusive. We aimed to evaluate the association of statins use with AKI risk by performing a meta-analysis. Twenty-one studies were included in our meta-analysis by searching electronic databases according to predefined criteria. No significant association between statins use and AKI risk was observed in overall populations, Caucasians, Asians, and patients undergoing cardiac and elective surgery (p = 0.816, 0.981, 0.18, 0.709, and 0.122). Statins use decreased the risk of contrast-induced AKI (CIN) (p = 0.005) and increased AKI risk in patients with community acquired pneumonia (CAP) (p = 0.006). Meta-regression analyses showed almost no impact on the pooled ORs of age and study length for overall populations. Exclusion of any single study had little impact on the pooled ORs. In conclusion, statins use is not associated with the risk of AKI in overall populations, Caucasians, Asians, and patients undergoing cardiac and elective surgery. Statins use decreases the risk of CIN and may increase the risk of AKI in CAP patients.

• Acute kidney injury
• meta-analysis
• statins
• risk

Introduction

Acute kidney injury (AKI), a sudden loss of kidney function, usually occurs over the course of hours to days, and is likely to develop in ambulatory outpatients and hospitalized patients; particularly critically ill patients.1 AKI is associated with substantial morbidity and mortality. Although recovery of kidney function occurs in the majority of patients surviving an episode of AKI, many patients remain dialysis dependent or are left with severe renal impairment.2 Severe AKI is associated with a mortality rate of 40–70%.3 Due to the harms of AKI mentioned above, effective prophylactic therapy of AKI appears imperative. To date, strenuous efforts have been made to prevent AKI, such as by administration with diuretics, vasoactive drugs, growth factors, or antioxidants.4 However, the prevention has not always been obtained by the use of these therapies.

Recently, statins has been shown to exert anti-inflammatory effects by blocking the infiltration of inflammatory cells and down regulating the expression of inflammatory mediators, statins also exert anti-oxidant effects, and improve endothelial function through the increased expression of nitric oxide.5,6 In addition, studies using animal models have shown that giving statins before an ischemic event significantly reduced AKI,7 which induced the hypothesis that statins use might prevent AKI in humans. The past few years have witnessed an increasing interest in testing this postulation. However, among studies8–28 on the association between statins use and AKI risk, the results were inconsistent. A previous meta-analysis by Li et al.29 showed that statins use decreased the incidence of contrast-induced AKI (CIN). However, it included only seven randomized controlled trials (RCTs), statins were also used in the control groups of three studies included in this meta-analysis. Another meta-analysis by Singh et al.30 demonstrated that there was no effect of statins therapy on the incidence of AKI. However, the patients in the nine studies recruited in this meta-analysis were all patients undergoing cardiac surgery, which made it difficult to extrapolate its conclusion to other disorders.

To clarify the association between the statins use and the risk of AKI in different populations and disorders, and examine the impact of covariates, such as age and study length, we performed an updated meta-analysis to provide a much more reliable finding on the significance of the association.

Methods

Search strategy

We attempted to search the published papers related to the association between the statins use and the risk of AKI in humans using PubMed, Embase, and Cochrane databases through September 2013. No restriction was imposed on search language. The used search terms were as follows: (1) acute kidney injury, acute renal injury (AKI); and (2) statins, statin, risk. Bibliographies of extracted articles and reviews were also scrutinized. If the same data were included in more than one study, we chose the study with the most complete analysis.

Study selection

We first performed a screening of titles and abstracts. A second screening was based on full-text review. Studies were considered eligible if they met the following criteria: (1) the study design was a case–control, cross-sectional, cohort, or prospective study; (2) the exposure of interest was statins use; (3) the outcome of interest was AKI; and (4) odds ratio (OR) and the corresponding 95% confidence interval (CI) (or data to calculate them) were reported.

Data extraction and quality assessment

We extracted any reported ORs, we also extracted study characteristics for each study. Data were recorded as follows: first author’s last name; year of publication; country origin and duration of follow-up; characteristics of study population and age at baseline; and number of AKI. In addition, we evaluated the quality of each cohort study included using Newcastle–Ottawa Quality Assessment Scale, which included the assessment for participant’s selection, exposure, and comparability. A study can be awarded a maximum of one score for each numbered item within the selection and exposure categories. A maximum of two scores can be given for comparability. We assessed the quality of RCTs using Jadad scores, which included the assessment for allocation concealment, double-blinding, and completeness of follow-up. A maximum of two scores can be given for allocation concealment and double-blinding. A maximum of one score can be given for completeness of follow-up. Two authors independently performed the data extraction with any disagreements resolved by discussion.

Statistical analysis

OR was used to measure the association between the statins use and the AKI risk across studies. Heterogeneity of ORs among studies was tested by using the Q statistic (significance level at p < 0.10). The I² statistic, a quantitative measure of inconsistency across studies, was also calculated. The pooled ORs were calculated using either a fixed-effects model or, in the presence of heterogeneity, a random-effects model. Furthermore, 95% confidence intervals (CIs) were also calculated. We performed sensitivity analyses to assess the influence of a single study on the pooled effect estimates by omitting one study in each turn. Also, we performed subgroup analyses according to the study type. Additionally, we conducted meta-regression to evaluate whether effect estimate was associated with age at baseline, or study length. Potential publication bias was assessed by Begg’s test and Egger’s test at the p < 0.05 level of significance. All analyses were performed using STATA version 12.0 (Stata Corp, College Station, TX). p < 0.05 was considered statistically significant, except where otherwise specified.

Results

Literature search

We first retrieved 126 citations from the electronic databases. Of these, 103 publications were excluded due to the fact that they were editorials/reviews/case reports or did not state statins or AKI. Two studies were excluded because they did not have control group. Finally, 21 studies8–28 were identified for the analysis of the association between the statins use and the AKI risk (Figure 1).

Figure 1. Flow chart of study selection.

Study characteristics

The characteristics of enrolled studies are presented in Tables 1 and 2. Eighteen studies8–21,24–27 were conducted in Caucasians, and three22,23,28 in Asians. These studies were published between 2004 and 2013. Thirteen studies were retrospective cohort, seven were RCTs, and one study was prospective cohort. The age at baseline varied from 52.5 to 63 years. The study length ranged from 48 h to 4 years. The sample sizes ranged from 85 to 6,722,776.

Table 1 Characteristics of studies evaluating the association between statins use and AKI risk.

Study	Study design	Ethnicity	Age (years)	Statins/ control (n)	Therapy	Outcome statins/control	Length of follow-up	Definition of AKI
Pan et al. (2004)18	RC	Caucasians	62.9	943/720	Cardiac surgery	59/39	30 d	NR
Subramaniam et al. (2008)16	RC	Caucasians	64.9	654/654	Cardiac surgery	8/8	NR	NR
Virani et al. (2008)17	RC	Caucasians	65	1675/1326	Cardiac surgery	42/46	NR	ATN, AIN
Fedoruk et al. (2008)25	RC	Caucasians	64.3	203/244	Cardiac surgery	8/14	30 d	Isc > twice
Jo et al. (2008)28	RCT	Asians	65.6	118/118	Contrast media	3/4	48 h	Isc ≥ 0.5 mg/dL
Huffmyer et al. (2009)26	RC	Caucasians	64	1557/1203	Cardiac surgery	436/331	NR	eGFR > 50%
Allou et al. (2010)24	RC	Caucasians	62.1	222/208	Cardiac surgery	22/29	30 d	Isc > 50%
Argalious et al. (2010)13	RC	Caucasians	67.9	6152/4487	Cardiac surgery	819/467	4 year	Isc > 0.3 mg/dL
Billings et al. (2010)12	RC	Caucasians	59.9	174/150	Cardiac surgery	39/29	NR	RIFLE
Toso et al. (2010)20	RCT	Caucasians	75.5	152/152	Contrast media	15/16	5 d	Isc > 0.5 mg/dL
Acikel et al. (2010)23	RCT	Asians	59.8	80/80	Contrast media	0/1	48 h	NR
Ozhan et al. (2010)22	RCT	Asians	54.5	60/70	Contrast media	2/7	48 h	Isc ≥ 0.5 mg/dL
Bolesta et al. (2011)21	RC	Caucasians	67.6	356/207	Cardiac surgery	125/54	48 h	Isc ≥ 0.3 mg/dL
Mithani et al. (2011)11	RC	Caucasians	66	1434/670	Cardiac surgery	322/173	48 h	Isc ≥ 0.3 mg/dL
Molnar et al. (2011)27	RC	Caucasians	73	67,941/145,406	Elective surgery	1665/2355	14 d	Isc > 0.5 mg/dL
Patti et al. (2011)19	RCT	Caucasians	65.5	120/121	Contrast media	6/16	48 h	Isc ≥ 0.5 mg/dL
Prowle et al. (2012)10	RCT	Caucasians	68	42/43	Cardiac surgery	6/7	5 d	Isc > 50%
Quintavalle et al. (2012)14	RCT	Caucasians	70	202/208	Contrast media	9/37	1 week	Isc > 10%
Murugan et al. (2012)9	PC	Caucasians	69.6	413/1423	CAP	176/455	1 year	RIFLE
Brunelli et al. (2012)15	RC	Caucasians	60.7	10,779/67,321	Elective surgery	2522/14,406	30 d	Isc ≥ 0.3 mg/dL
Layton et al. (2013)8	RC	Caucasians	52.5	3,905,155/281,7621	Cardiac surgery	36,783/9661	1 year	Isc ≥ 0.3 mg/dL

Note: AKI, acute kidney injury; PCI, percutaneous coronary intervention; CAP, community acquired pneumonia; NR, not reported; CA, coronary angiograpghy; RC, retrospective cohort; PC, prospective cohort; RCT, randomized controlled trial; d, day; w, week; y, year; Isc, increased serum creatinine; ATN, acute tubular necrosis; AIN, acute interstitial necrosis.

Table 2. Statins protocol of included studies.

Study	Status	Duration	Dose
Pan et al. (2004)18	Past use	Short-term	Low
Subramaniam et al. (2008)16	Past use	Long-term	Low
Virani et al. (2008)17	Past use	Short-term	Low
Fedoruk et al. (2008)25	Past use	Long-term	Low
Jo et al. (2008)28	Current use	Short-term	Low
Huffmyer et al. (2009)26	Past use	Short-term	Low
Allou et al. (2010)24	Past use	Long-term	Low
Argalious et al. (2010)13	Past use	Long-term	Low
Billings et al. (2010)12	Past use	Long-term	Low
Toso et al. (2010)20	Current use	Short-term	High
Acikel et al. (2010)23	Current use	Short-term	Low
Ozhan et al. (2010)22	Current use	Short-term	High
Bolesta et al. (2011)21	Past use	Short-term	Low
Mithani et al. (2011)11	Past use	Short-term	Low
Molnar et al. (2011)27	Past use	Long-term	Low
Patti et al. (2011)19	Current use	Short-term	High
Prowle et al. (2012)10	Current use	Short-term	Low
Quintavalle et al. (2012)14	Current use	Short-term	Low
Murugan et al. (2012)9	Current use	Short-term	High
Brunelli et al. (2012)15	Past use	Long-term	High
Layton et al. (2013)8	Past use	Long-term	High

Quality scale

The number of awarded scores of enrolled cohort studies ranged from 5 to 8 (low quality: 1–3, median quality: 4–6, high quality: 7–9). Of all cohort studies included, two studies8,13 were awarded for five scores, two12,24 for six scores, nine11,15–18,21,25–27 for seven scores, and one9 for eight scores. The number of awarded scores of included RCTs ranged from 1 to 5 (high quality: 3–5), one study23 was awarded for one score, one22 for two scores, two10,14 for three scores, and three19,20,28 for five scores.

Association between statins use and AKI risk

Statins use was not associated with AKI risk among overall populations, Caucasians, and Asians (p = 0.816, 0.909, and 0.713, respectively, Figure 2). No obvious association between statins use and AKI risk was observed in patients undergoing cardiac and elective surgery (p = 0.632 and 0.689, respectively, Figure 3). Statins use decreased the risk of contrast-induced AKI (CIN) (p < 10⁻³, Figure 3), and increased the risk of AKI in patients with community acquired pneumonia (p = 0.827, CAP) (Figure 3).

Figure 2. Association between statins use and AKI risk in Caucasians, Asians, and overall populations.

Figure 3. Association between statins use and AKI risk based on different therapies.

Subgroup analyses

No significant association between statins use and AKI risk was observed in retrospective cohort studies (p = 0.467, Figure 4). Statins use decreased the risk of AKI in RCTs (p = 0.005, Figure 4), and increased AKI risk in the prospective cohort study (p = 0.006, Figure 4).

Figure 4. Association between statins use and AKI risk based on different study types.

Meta-regression analyses

Age and study length did not demonstrate significant influences on the ORs (as shown in Table 3).

Table 3. Characteristics of univariate meta-regression analysis.

Covariate	Co-efficient	95% CI	p
Age	0.022	−0.692 to 0.649	0.947
Length of follow-up	0.002	−0.043 to 0.047	0.924

Sensitivity analysis

Omission of any single study did not change the overall ORs significantly. The pooled ORs varied from 1.00 (95% CI: 0.48–1.53) to 1.07 (95% CI: 0.59–1.55).

Evaluation of publication bias

The Begg rank correlation test and Egger linear regression test indicated no evidence of publication bias across studies (Begg, p = 0.07 Egger, p = 0.06).

Discussion

Increasing attention has been paid to the association between statins use and AKI risk. The confirmation of renoprotective effect of statins would give an insight to early diagnosis or intervention of AKI. Several mechanisms accounted for the possible preventive effects of statins use on AKI. First, AKI is thought to be secondary to ischemia, inflammation, and increased radical generation.31 Statins, which are mainly used for their cholesterol-lowering effects, also have anti-inflammatory and anti-oxidant activity and improve endothelial function.5,6 Second, statins use restores renal vascular permeability and reduce renal tubular hypoxic injury.32 Third, animal study showed that statins use prevented ischemic renal failure in older rats after clamping both renal arteries.33 In terms of these facts, statins use may also lower the risk of AKI.

However, our meta-analysis of 21 studies did not yield expected results. Our investigation indicated that statins use was not associated with the risk of AKI in overall populations, Caucasians, Asians, and patients undergoing cardiac and elective surgery, and statins use decreased the risk of CIN and increased the risk of AKI in CAP patients. The inconsistent effect of statins preventing AKI observed in different populations might be due to the facts that all the studies were conducted in diverse regions with different origins, varied AKI definition and age at baseline, different times of statins use, and study length. For example, if we chose a lower increased serum creatinine for AKI definition, the incidence of AKI would be higher. Regrettably, we were unable to perform a subgroup analysis due to the lack of AKI definition in some included studies. These factors should be considered in interpreting the results although meta-regression showed no significant impact of age and length of follow-up on the final conclusions.

The renoprotective effects of statins use might depend on the healthy conditions. AKI is likely to occur in critically ill adult patients in whom its incidence can reach 65%.34 In our meta-analysis, the age ranged from 52.5 to 75.5 years, all the participants were comparatively older, and the less likely healthy conditions may confound the renoprotective effect of statins. Our investigation also found that statins use was not associated with AKI risk in overall populations, Caucasians, and Asians. The severity of diseases may also influence the effects of statins. For example, a number of factors unique to cardiac surgery including cardioplegia, cardiopulmonary bypass, and myocardial suppression may induce hemodynamic insults which may override the protective effects of statins in many instances.30 Our findings also indicated that there was no significant association between statins use and AKI risk in patients undergoing cardiac and elective surgery. In contrast, statins use decreased the risk of CIN, which may be due to the fact that the majority of patients receiving contrast media were not critically ill. This was verified by the results of a meta-analysis by McDonald et al.,35 in which the CIN group experienced a similar incidence of AKI to the control group. Also, the patients were less likely to receive contrast media for the reason of safety if they were critically ill. In the CAP patients, we observed an increased risk of AKI in statins users, which might be caused by the facts that there were significantly more patients with cardiovascular disease and diabetes in statins users than those in non-statins users, and statins users presented with more severe pneumonia than non-statins users. In addition, only one study regarding the association of statins use with AKI risk in CAP patients was included. More studies should be conducted in the future due to the limited evidence.

Although our investigation had a lot common with previous meta-analyses of the association between statins use and AKI risk, such as the use of statins and definition of AKI, the strength of this meta-analysis was obvious. Our study enrolled more studies than previous meta-analyses, which could generalize its results to the overall populations. Moreover, subgroup and meta-regression analyses provided the possible influencing factors on the risk between statins use and AKI risk.

Although no significant publication bias for overall populations was observed, several limitations should be considered in our meta-analysis. First, heterogeneities across the studies included might affect the results of our meta-analysis, although a random-effects model had been performed. Second, the characteristics including age, length of follow up, and study design across studies included were different, which might affect the results. Third, the protocol of statins might also affect the results. For example, the prospective long-term use of statins would produce more robust effects. Finally, the definition of AKI and time of statins use in some included studies was lacking, which might lead to inconsistent results. More studies with unified standards of various characteristics should be performed in the future.

Currently, the consensus was reached that identification and correction of volume depletion and avoidance of nephrotoxins are of most importance for AKI prevention.4

However, no permanent preventive effect of AKI was found in any drug, including emerging medicine and statins. These evidences indicate that AKI seems to be a multi-factorial disease. It should be cautious to select the drug for preventing AKI. Future studies should be focused on the incidence and influencing factors of AKI in different disorders and populations.

In conclusion, the results of our investigation suggest that statins use is not associated with the risk of AKI in overall populations, Caucasians, Asians, and patients undergoing cardiac and elective surgery; statins use reduces the risk of CIN, and may increase the risk of AKI in CAP patients. However, a large number of studies in different populations are needed in the future.

Declaration of interest

There is no conflict of interest for all authors. This study was supported by a grant of Research and innovation Project for College Graduates of Jiangsu Province, China (Grant number CXLX13_556).