Am J Cardiovasc Drugs https://doi.org/10.1007/s40256-018-0274-3

SYSTEMATIC REVIEW

Vitamins for Prevention of Contrast-induced Acute Kidney Injury: A Systematic Review and Trial Sequential Analysis Yongxing Xu1 Zhaoyan Gu3,4

•

Xinming Zheng2 • Boran Liang1 • Jianjun Gao1

•

Ó Springer International Publishing AG, part of Springer Nature 2018

Abstract Background To date, universally accepted preventive measures for contrast-induced acute kidney injury (CIAKI) do not exist, and they warrant further research. Objective The purpose of this study was to evaluate the efficacy of vitamins, including vitamin C and E, for prevention of CI-AKI. Methods We electronically searched the MEDLINE, EMBASE, and Cochrane databases. The outcome of interest was the incidence of CI-AKI. Results A total of 19 studies were included in this metaanalysis. Pooled analysis showed that vitamin C plus saline [relative risk (RR) = 0.63, 95% confidence interval (CI)

Electronic supplementary material The online version of this article (https://doi.org/10.1007/s40256-018-0274-3) contains supplementary material, which is available to authorized users.

0.49–0.82, p = 0.0005] and vitamin E plus saline (RR = 0.39, 95% CI 0.24–0.62, p \ 0.0001) significantly reduced the incidence of CI-AKI compared to saline alone. The effect of vitamin C plus saline was further confirmed by trial sequential analysis (TSA). However, TSA indicated that more trials are required to confirm the efficacy of vitamin E plus saline. There was no significant difference in preventing CI-AKI between vitamin C and N-acetylcysteine (NAC) (RR = 0.90, 95% CI 0.47–1.71, p = 0.75), between vitamin C plus NAC and saline (RR = 0.62, 95% CI 0.30–1.30, p = 0.20), as well as between vitamin C plus NAC and NAC (RR = 0.97, 95% CI 0.49–1.92, p = 0.93). Conclusions Vitamin C plus saline administration is effective at reducing the risk of CI-AKI. Evidence for the use of vitamin E plus saline in this context is encouraging, but more trials are required. Furthermore, this meta-analysis and TSA indicated insufficient power to draw a definitive conclusion on the effect of vitamin C plus NAC, versus saline or NAC alone, which needs to be explored further.

& Jianjun Gao [email protected] & Zhaoyan Gu [email protected] 1

Department of Nephrology, the 306th Hospital of Chinese PLA, 9 AnXiangBeiLi Road, Beijing 100101, China

2

Department of Nephrology, The Hospital of Shunyi District Beijing, No.3 Guangming South Street, Shunyi District, Beijing, China

3

4

Department of Endocrinology, Nanlou Division, Chinese PLA General Hospital, National Clinical Research Center for Geriatric Diseases, Beijing 100853, China Healthcare Department, Hainan Branch of Chinese of PLA General Hospital, Sanya 527400, China

Key Points Vitamin C plus saline administration is effective to reduce the risk of contrast-induced acute kidney injury (CI-AKI). Vitamin E may potentially protect against CI-AKI, but the result is inconclusive. Further large, well-designed trials are required to provide firmer evidence.

Y. Xu et al.

1 Introduction Contrast-induced acute kidney injury (CI-AKI) has become the third most common cause of acute kidney injury (AKI) in hospitalized patients [1], prolonging hospitalization time, increasing healthcare costs, and increasing mortality [2]. The complex pathophysiology of CI-AKI involves different mechanisms, such as vasoconstriction, oxidative stress, medullary ischemia, and the direct toxic effects of contrast medium [3]. Highly concentrated contrast increases fluid viscosity and decreases flow through medullary tubules and vessels [4]. The subsequent sustained reduction in renal blood flow contributes to renal injury in several ways: release of reactive oxygen species; ‘‘osmotic nephrosis’’ or vacuolization from direct toxic effects of contrast on tubular cells, leading to acute tubular necrosis; and ischemia of the outer regions of the medulla [3]. Of note, oxidative stress plays a critical role in CI-AKI [5]. The potent antioxidants vitamin C, vitamin E, and Nacetylcysteine (NAC) have been used in some studies in efforts to counter CI-AKI [6–8]. A number of clinical trials and systematic reviews explored this topic, but the question of whether vitamin plus saline exerted a more significant effect than saline alone did not receive a consistent answer [9–14]. To the best of our knowledge, two recent systematic reviews maintained that vitamin C afforded effective nephroprotection against CI-AKI [9, 10]. In contrast, two additional systematic reviews indicated that vitamin C had a statistically insignificant effect on CI-AKI compared with saline hydration alone [11, 12]. Most of these reviews were published earlier [15] or did not include studies published in abstract form [10–12, 16, 17]. Furthermore, some positive meta-analytic findings may become inconclusive when the statistical analyses take into account the risk of random error [18]. It is recommended that the interpretation of meta-analyses is done alongside a trial sequential analysis (TSA) [18, 19]. To date, no systematic review evaluating vitamin efficiency has employed TSA. Therefore, we performed a systematic review including randomized controlled trials (RCTs) published in any form, and used TSA to explore the associations between the use of antioxidative vitamins (including vitamin C and E) and the incidence of CI-AKI.

2 Methods The systematic review and meta-analysis were performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Statement [20]. The included trials met the following criteria: (1) each study was an RCT using the standard prophylaxis defined as saline hydration [6, 21] for CI-AKI in all patient

groups; (2) the participants underwent diagnostic and/or interventional procedures using contrast media; and (3) each study assessed the efficacy of a vitamin (including vitamin C and vitamin E) either alone or in combination with another strategy aimed at the prevention of CI-AKI. Of note, not all comparisons were included. We focused the review on comparisons for which two or more studies reported on the same comparison. The exclusion criteria were as follows: (1) the study did not report clinical outcomes of interest and the missing data were not available after contacting the corresponding author; or (2) the study was a quasi-randomized or non-randomized trial. The outcome was the development of CI-AKI, defined as an absolute increase in the baseline serum creatinine level of 44.2 lmol/L (0.5 mg/dL) or a relative increase of 25%, typically within 48–72 h after contrast injection. If data were not available for the first 48–72 h after treatment, we used data obtained within the first 5 days of treatment (the data point closest to 48–72 h was preferred) [17]. Electronic searches were performed using MEDLINE via Ovid (from inception to August 2016), EMBASE (from inception to August 2016), and the Cochrane Library database (Cochrane Central Register of Controlled Trials, no date restriction), with relevant text words and medical subject headings that covered vitamin C, ascorbic acid, L-ascorbic acid, vitamin E, tocopherol, controlled clinical trial, randomized controlled trial, randomized, and RCT [see the electronic supplementary material (ESM), Supplementary Table 1]. Conference proceedings, the ClinicalTrials.gov website, and the references of recent reviews and relevant studies were searched as well. We imposed no language restriction. Two independent reviewers (Z.Y. Gu and J.J. Gao) assessed the eligibility of the studies with a standardized approach. Where more than one publication of one study existed, only the publication with the most complete data was included. Where relevant outcomes were only published in earlier versions, these data were used. Discrepancies were resolved by discussion with a third individual (Y.X. Xu). Two authors (Z.Y. Gu and J.J. Gao) independently extracted information using a standardized data collection form (see the ESM, Supplementary Table 2). Any further information required from the original investigators was requested by written correspondence, and any relevant information obtained in this manner was included in the review. Two authors (Z.Y. Gu and J.J. Gao) independently assessed the risk of bias in all included studies using standard criteria. Seven different domains, including random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data

Vitamins and CI-AKI

(attrition bias), selective reporting (reporting bias), and any other potential bias, were judged as low risk of bias, high risk of bias, or unclear risk of bias [22]. For dichotomous outcomes, we calculated relative risks (RRs) with 95% confidence intervals (CIs). For trials with endpoints featuring zero events in the treatment arm, risk ratios were calculated using the 0.5 cell correction [23]. The random-effects model by Dersimonian and Laird was our primary choice, which considers between-study variations [24] and is generally considered to be more conservative [25]. Heterogeneity of treatment effects among studies was statistically explored by calculating the chi-squared and I2 statistics [22]. Subgroup analyses of studies focusing on vitamin C plus saline versus saline alone were based on publication type, publication year, baseline kidney function, sample size, mean age of participants, route of administration, dose of vitamin C, and the definition and incidence of CI-AKI. Sensitivity analyses were performed to investigate whether the meta-analysis model chosen critically affected the results. Meta-regression was used to assess the association between the vitamin dose and the RR of postoperative AKI. Egger’s regression asymmetry test and Begg’s test were used to explore publication bias [26]. A two-sided p value of \ 0.05 was regarded as significant in all analyses. All analyses were performed with the aid of STATA software (version 12.0, Stata Corp, College Station, TX, USA) and Review Manager (version 5.3). False positive and false negative meta-analyses are common, but infrequently recognized [27]. When the number of participants in a meta-analysis is less than the required, based on a realistic and minimally important intervention effect, the constant application of a traditional 95% CI or a 5% statistical significance threshold will lead to too many false positive and false negative conclusions [19]. Frequent updating of a meta-analysis as new trials become available also increases the chances of making a false positive conclusion [27]. To obtain a more comprehensive assessment of meta-analyses, conducting a TSA was recommended [18, 19]. TSA is a method that introduces sequential analysis into traditional meta-analysis, aiming to control the increased risk of random errors with trial sequential monitoring boundaries and quantifying the reliability of cumulative data in meta-analyses [18, 19]. For TSA, the cumulative z curves were constructed with each cumulative z value calculated after including a new trial. Firm evidence for benefit or harm may be established if the cumulative z curve crosses the trial sequential monitoring boundary before reaching the required information size. In this case, further trials may turn out to be superfluous. In contrast, if the cumulative z curve does not cross any of the trial sequential monitoring boundaries, there is likely insufficient evidence to reach a conclusion, and additional trials may be required. Firm evidence for lack of the

postulated intervention effect can also be assessed with TSA. This occurs when the cumulative z score curve crosses into the futility area boundary, and then future trials are unlikely to change the trend of evidence. We performed TSA of the incidence of CI-AKI, with an overall 5% risk of a type I error and a power of 80%. We anticipated an intervention effect of a 25% relative risk reduction (RRR), which would be clinically important [12, 28]. We used the Copenhagen Trial Unit TSA software (version 1.0, http://www.ctu.dk/tsa).

3 Results 3.1 Search Results and the Characteristics of Included Studies Initially, we found 253 publications, of which 75 were removed because they were duplicates. After screening the titles and abstracts and full-text browsing, 19 studies [6, 8, 13, 14, 29–43] were included in this meta-analysis. Of note, some comparisons (vitamin C plus saline vs bicarbonate alone [40], vitamin C plus bicarbonate vs bicarbonate alone [44], vitamin C plus NAC vs bicarbonate infusion plus NAC [42], vitamin C plus NAC vs vitamin C [8], and vitamin C plus pentoxifylline vs saline [45]) were tested in only single trials. Therefore, these comparisons were not included. The literature selection flow is illustrated in Figure 1. Of the included RCTs, 15 focused on vitamin C and four on vitamin E. Fourteen studies were published in full-text articles and five in abstract form. The baseline characteristics of all included studies are shown in Table 1. The quality of studies was variable (see the ESM, Supplementary Figs. 1 and 2). Only one study was deemed to exhibit a truly low risk of bias in all domains [38]. Blinding was lacking or unclear in some trials. However, the outcomes of interest were objective and clinician awareness of treatment assignment did not compromise study validity. Thus, all trials exhibited low risks of performance and detection bias [46]. 3.2 Vitamin C plus Saline Versus Saline Alone We identified ten studies comparing vitamin C plus saline with saline alone. The vitamin C and control groups contained 712 and 807 patients, respectively. A total of 73 events (10.3%) were observed in the vitamin C group and 150 (18.6%) in the control group. The overall RR associated with vitamin C use was 0.63 (95% CI 0.49–0.82, p = 0.0005; Fig. 2). No heterogeneity was evident (I2 = 0%, p = 0.57). Our first sensitivity analysis based on a fixed-effect model also revealed that vitamin C reduced the incidence of CI-AKI (RR = 0.62, 95% CI 0.48–0.79,

Y. Xu et al. Figure 1 Flow diagram of literature selection process. CCTR cochrane central register of controlled trials, RCT randomized controlled trial

p = 0.0002). In our second sensitivity analysis, from which we excluded the three abstracts, the overall RR for CI-AKI associated with the use of vitamin C was 0.66 (95% CI 0.49–0.90, p = 0.008). Subgroup analyses were performed in terms of the incidence of CI-AKI in studies comparing vitamin C with saline hydration alone (Table 2). There might be some indication that the reduction in risk of CI-AKI may be associated with the dose of vitamin C, but this finding was not statistically significant (p = 0.10). No clear evidence of heterogeneity was evident when comparing the summary results of subsets of studies grouped by publication type, publication year, baseline kidney function, sample size, mean age of participants, route of administration, and the definition or incidence of CI-AKI. Meta-regression was further used to assess the association between the dose of vitamin C and the RR in terms of the incidence of CI-AKI. However, we found no evidence that the dose of vitamin C affected CI-AKI development (see the ESM, Supplementary Fig. 3). Visual estimation of a funnel plot (see the ESM, Supplementary Fig. 4), the Egger regression asymmetry test (p = 0.606), and Begg’s test (p = 0.721) used to evaluate

studies on vitamin C supplementation did not disclose any publication bias. Furthermore, TSA revealed that the cumulative z curve for the incidence of CI-AKI crossed the sequential monitoring boundary and that the required optimal sample size was 1979 (Fig. 3a). Although the required optimal sample size was not attained, the result indicated that vitamin C exerted a protective effect on CI-AKI compared with saline alone, with a 25% RRR, and that this finding was unlikely to be compromised by future RCTs. 3.3 Vitamin E plus Saline Versus Saline Alone Four studies comparing vitamin E plus saline versus saline alone were identified. The vitamin E and control groups contained 414 and 312 patients, respectively. Pooled analysis using a random-effects model showed that vitamin E afforded a clinically important benefit in terms of reducing the CI-AKI risk (RR = 0.39, 95% CI 0.24–0.62, p \ 0.0001; Fig. 2). We found no evidence of heterogeneity (I2 = 0%, p = 0.67). Pooled analysis using a fixedeffect model showed that the RR was essentially unchanged at 0.38 (95% CI 0.24–0.62, p \ 0.0001).

Publication type

Full text

Full text

Full text

Full text

Full text

Abstract

Study

Albabtain [8]

Brueck [13]

Briguori [42]

Boscheri [29]

Dvorsak, 2013 [30]

El-Fishawy [31]

At least 100 mL/h from 2 h before to 6 h after

50–100 mL for 2 h before and C 6 h after

500 mL at 2 h before and 500 mL during and 6 h after

1 mL/kg/h before and continued for 12 h

1 mL/kg/h from12 h before to 12 h after

50–125 mL/h from 2 h before to 6 h after

Saline hydration protocol

N/A

Coronary angiography and/or angioplasty

Iodixanol

Coronary angiography

Iopamidol

Iopromide

Coronary angiography

Coronary angiography or angioplasty

Iopromide

Ioxaglate

Contrast type

Coronary angiography and/or PCI

Coronary angiography or PCI

Imaging procedure

2 days

3–4 days

48 h

[25% above baseline in SCr or cystatin C [25% above baseline in SCr

48 h

72 h

4–5 days

Time point

C 25% above baseline in SCr

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

Increase of 0.5 mg/ dL in SCr

Decrease of creatinine clearance C 25% from baseline or increase of C 0.5 mg/dL in SCr

CI-AKI definition

Table 1 .Main characteristics of the included randomized controlled trials

50/6

50/2

2

41/3

2 1

40/2

69/8

2

1

74/5

111/ 11

2

1

107/ 11

98/ 24

3 1

192/ 53

2

58/5 64/5

3 4 192/ 62

62/5

2

1

57/2

N/ event

1

Arm

N/A

N/A

70.7 (50–88)

70.7 (49–85)

71 ± 28

71 ± 10

71 ± 9

69 ± 8

74 (69–77)

75 (70–79)

75 (69–79)

61.1 ± 10.9

64.0 ± 11.0

62.0 ± 9.4

58.7 ± 1.9

Age*

N/A

N/A

41.5

45

55

64

55

59

50.0

42.2

43.1

79.9

86.2

85.2

84.2

Diabetes (%)

N/A

N/A

1.51 ± 0.35

1.58 ± 0.27

1.73 ± 0.4

1.75 ± 0.4

1.95 (1.69–2.26)

1.93 (1.82–2.16)

1.5 (1.3–1.7)

1.5 (1.3–1.8)

1.5 (1.3–1.7)

1.22 ± 0.40

1.26 ± 0.43

1.45 ± 0.56

1.24 ± 0.44

Baseline creatinine (mg/dL)*

Placebo

IV 1200 mg NAC and 1.5 g ascorbic acid at 2 h before, 1200 mg NAC and 1.5 g ascorbic acid at the night and the morning after

Placebo

Oral 3 g ascorbic acid before and 2 g bid 2 days after

Placebo

Oral 1 g ascorbic acid at 20 min before

Oral 1200 mg NAC 1 dose before and 3 doses after

IV 3 g ascorbic acid before and 2 g q12 h after ? Oral 600 mg NAC 1 dose before and 600 mg 3 dose after

Placebo

IV 600 mg NAC at 24 h and 1 h before

IV 500 mg ascorbic acid at 24 h and 1 h before

Placebo

Ascorbic acid ? NAC

Oral 600 mg NAC bid 2 days before

Oral 3 g ascorbic acid at 2 h before and 2 g at 1 h and 2 g at 24 h after

Medical protocols

N/A

N/A

130.6(40–287)

144.6(35–15)

112 ± 67

99 ± 46

179 ± 102

169 ± 102

110 (80–150)

110 (80–160)

115 (90–150)

97.4 ± 99.4

95.6 ± 88.8

70.1 ± 60.4

88.3 ± 64.8

Contrast dose (mL)

Vitamins and CI-AKI

Publication type

Abstract

Full text

Abstract

Full text

Study

Grygier [32]

Habib [33]

Hamdi [34]

Jo [35]

Table 1 continued

1 mL/kg/h from 12 h before to 12 h after

N/A

1.0 mL/kg/h for 12 h after

Hydration for 12 h before PCI and continued for 12 h after PCI

Saline hydration protocol

Coronary angiography

Coronary angiography

Coronary angiography

PCI

Imaging procedure

Iodixanol

N/A

N/A

N/A

Contrast type

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

C 25% above baseline in SCr

48 h

2–3 days

48–72 h

2–5 days

[25% above baseline or increase of [ 0.5 mg/dL in SCr

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

Time point

CI-AKI definition

91/4

83/1

1

2

95/ 20

2

45/8 107/ 11

3 1

30/5

2

50/2

3

30/2

50/7

1

52/9

2

N/ event

1

Arm

64.3 ± 8.7

65.6 ± 8.1

N/A

N/A

63 ± 8.26

62.03 ± 9.37

63 ± 8.26

N/A

N/A

N/A

Age*

40.6

35.8

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Diabetes (%)

1.38 ± 0.52

1.27 ± 0.35

N/A

N/A

0.92 ± 0.27

0.89 ± 0.36

1.09 ± 0.45

N/A

N/A

N/A

Baseline creatinine (mg/dL)*

Oral 1200 mg NAC q12 h for 2 days before

Oral 3 g and 2 g ascorbic acid with 12-h time interval before and 2 g at 12 h after

Saline

3 g ascorbic acid 2 h before and 2 g the day after and the next day

Placebo

Oral 600 mg NAC q12 h for 2 days, 1 dose before coronary angiography and 3 doses after coronary angiography. Additionally, oral 3000 mg ascorbic acid before and 2000 mg on the night and morning after procedure

Oral 1200 mg NAC every 12 h for 2 days, 1 dose before coronary angiography and 3 doses after coronary angiography

Oral 1200 mg NAC before and 1200 mg bid for 2 days after ? IV ascorbic acid 2000 mg before PCI followed by oral 2000 mg in the night and morning after

Oral 1200 mg NAC before and 1200 mg bid for 2 days after

Saline

Medical protocols

203.6 ± 141.9

216.4 ± 136.1

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Contrast dose (mL)

Y. Xu et al.

Publication type

Full text

Abstract

Full text

Full text

Full text

Full text

Full text

Study

Kitzler [43]

Komiyama [36]

Nough [37]

Rezaei [38]

Spargias [14]

Tasanarong [39]

Tasanarong [6]

Table 1 continued

1 mL/kg/h for 12 h before and 12 h after

1 mL/kg/h for 12 h before and 12 h after

50–125 mL/h from 2 h before to 6 h after

1 mL/kg from 12 h before to 12 h after

0.5–1 mL/kg from 6 h before to 12 h after

1500–2000 mL total

0.45% saline 1 mL/kg/h for 12 h before and after

Saline hydration protocol

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

Iopromide

Coronary angiography or PCI

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

48 h

48 h

2–5 days

72 h

3 days

Increase of 0.5 mg/ dL or [ 2.5% above baseline in SCr C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

48 h

48 h

Time point

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

C 25% above baseline in SCr

CI-AKI definition

Iopromide

N/A

Iodixanol

Omnipaque, Visipaque, Ultravist

N/A

Iopromide

Contrast type

Coronary angiography or PCI

Coronary angiography or PCI

Coronary angiography

Coronary angiography

Coronary angiography or PCI

Nonionic radiocontrast CT

Imaging procedure

102/ 6

102/ 5

101/ 15

2

3

52/ 12

2 1

51/3

113/ 23

2 1

118/ 11

149/ 21

2 1

149/ 10

45/7

2 1

45/3

1

40/ 10

2

10/0

3 40/1

10/0

2

1

10/0

N/ event

1

Arm

66 ± 10

69 ± 9

64.4

62.7

63.7

40

68 ± 9 67 ± 9

43

23

27

35.6

35.6

N/A

N/A

N/A

N/A

30

0

20

Diabetes (%)

65 ± 11

64 ± 11

67 ± 9

67 ± 10

66 ± 11

59.2 ± 11.7

60.22 ± 11.7

N/A

N/A

74 ± 8.5

76.6 ± 9.5

73.3 ± 11.9

Age*

1.63 ± 0.53

1.58 ± 0.48

1.48 ± 0.48

1.67 ± 0.53

1.62 ± 0.44

1.36 ± 0.50

1.46 ± 0.52

1.3 (1.2–1.5)

1.3 (1.2–1.5)

1.11 ± 0.2

1.16 ± 0.32

N/A

N/A

1.33 ± 0.12

1.37 ± 0.51

1.37 ± 0.2

Baseline creatinine (mg/dL)*

137 ± 75

Oral 350 mg/day atocopherol for 5 days before and 2 days after

134 ± 69

134 ± 73

Oral 300 mg/day ctocopherol for 5 days before and 2 days after

Placebo (350 mg/day)

132 ± 58

150 ± 83

261 ± 128

287 ± 148

50 (50–100)

50 (40–100)

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Contrast dose (mL)

Placebo

Oral 525 IU vitamin E at 48 h, 24 h, morning before

Placebo

Oral 3 g ascorbic acid at 2 h before and 2 g bid 2 day after

Placebo

Oral 600 mg VE at 12 h before and 400 mg at 2 h before

Placebo

Oral 2 g ascorbic acid at 2 h before

Placebo

IV 3 g ascorbic acid before, 2 g after and 2 g at 12 h after

Placebo

Oral 1200 mg NAC 12 h, 6 h before and 6 h, 12 h after

IV 540 mg vitamin E 12 h, 6 h before and 6 h, 12 h after

Medical protocols

Vitamins and CI-AKI

TSA showed that the required information size was not attained. The cumulative z curve crossed the traditional boundary, but failed to cross the sequential monitoring boundary, indicating a lack of firm evidence for a 25% reduction in the incidence of CI-AKI with vitamin E plus saline versus saline alone (Fig. 3b). More trials are needed to build firm evidence. When the effect sizes of the 14 trials evaluating vitamin C and vitamin E were combined in a random-effects model, a significant RRR of 44% was evident (p \ 0.0001; Fig. 2). We did not detect heterogeneity (I2 = 0%, p = 0.48). A sensitivity analysis was performed using a fixed-effect model, but did not change the conclusion (RR = 0.56, 95% CI 0.45–0.71, p \ 0.0001). 3.4 Vitamins Versus N-Acetylcysteine (NAC)

*Values are presented as mean ± standard deviation or median (interquartile range)

bid bis in die, CI-AKI contrast-induced acute kidney injury, CT computer tomography, IV intravenously, N/A not available, NAC N-acetylcysteine, PCI percutaneous coronary intervention, SCr serum creatinine, q12 h every 12 h

133.7 ± 74.4 Saline 82/6 2

71.4 ± 5.7

28

1.248 ± 0.364

136.4 ± 71.6 IV 3 g ascorbic acid before and oral 0.5 g q12h for 2 days after

N/A Saline

26

1.286 ± 0.418

N/A

71.8 ± 8.1 74/4

68/9 2

1 48 h Iodixanol, Iopromide, Iohexol Coronary angiography or PCI Full text Zhou [41]

1 mL/kg/h from 4 h before to at least 12 h after

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

N/A

N/A

N/A IV 3 g ascorbic acid 2–4 h before and oral 1 g on day 1, 2 after N/A 64/8 N/A PCI Abstract Wenhua [40]

1 mL/kg/h for 6 h before and 12 h after

C 25% above baseline and/or increase of C 0.5 mg/dL in SCr

72 h

1

N/A

N/A

Medical protocols Contrast type Imaging procedure Saline hydration protocol Publication type Study

Table 1 continued

CI-AKI definition

Time point

Arm

N/ event

Age*

Diabetes (%)

Baseline creatinine (mg/dL)*

Contrast dose (mL)

Y. Xu et al.

Four relevant RCTs were identified, including three studies [8, 13, 35] that compared vitamin C with NAC and one [43] that compared vitamin E with NAC. In the study on vitamin E, no cases of CI-AKI developed in either arm. Hence, this study was excluded when the effect sizes of the trials were combined. The overall incidence of CI-AKI was 12.2% in the vitamin C group compared with 17.5% in the NAC group. On pooled analysis using a random-effects model, we found that the vitamin C/NAC difference was statistically insignificant, with an RR of 0.90 (95% CI 0.47–1.71, p = 0.75; see the ESM, Supplementary Fig. 5a). No significant heterogeneity was evident among the RCTs (I2 = 17%, p = 0.30). When a fixed-effect model was used, the result was consistent with the principal analysis (RR = 0.91, 95% CI 0.61–1.34, p = 0.62). TSA on the incidence of CI-AKI in patients receiving vitamin C versus NAC showed that the cumulative z curve did not cross either the conventional boundary or the trial sequential monitoring boundary. The required information size was not reached. Therefore, the evidence was insufficient and inconclusive (see the ESM, Supplementary Fig. 5b). 3.5 Vitamins plus NAC Versus Saline Alone Four studies compared vitamin C plus NAC versus saline alone [8, 31–33]. No study investigated the effect of vitamin E plus NAC versus saline. The vitamin C plus NAC group contained 188 subjects and featured 14 events (7.4%); the saline group contained 213 subjects and featured 28 events (13.1%). Using the random-effects model, pooled analysis revealed that vitamin C plus NAC did not exert a statistically significant effect in terms of reducing the CI-AKI risk (RR = 0.62, 95% CI 0.30–1.30, p = 0.20; see the ESM, Supplementary Fig. 6a) without any evidence of heterogeneity (I2 = 25%, p = 0.26). Sensitivity analysis

Vitamins and CI-AKI

Figure 2 Meta-analysis of the effect of vitamin C (VC) and vitamin E (VE) on the incidence of contrast-induced acute kidney injury (CI-AKI) compared to saline. CI confidence interval, RR relative risk

was conducted using the fixed-effect model, and the result was consistent with the principal analysis (RR = 0.59, 95% CI 0.33–1.08, p = 0.09). Based on an a priori anticipated intervention effect of an RR reduction of 25%, TSA indicated that the optimal sample size allowing a reliable conclusion to be drawn was 4008. The cumulative z curve did not break through the traditional monitoring boundary or the trial sequential monitoring boundary (see the ESM, Supplementary Fig. 6b). Thus, the conclusion may be a false negative.

heterogeneity was evident among the RCTs (I2 = 24%, p = 0.26). We also performed TSA of vitamin C plus NAC versus NAC on the incidence of CI-AKI (see the ESM, Supplementary Fig. 7b). The required optimal sample size was not achieved. The cumulative z curve did not break through the traditional boundary or the trial sequential monitoring boundary. Also, the cumulative z curve did not break through the boundary for futility. Thus, the statistical power was inadequate to allow a definitive conclusion to be drawn.

3.6 Vitamins plus NAC Versus NAC Alone We identified four studies [8, 32, 33, 42] comparing vitamin C plus NAC versus NAC. No study employed vitamin E plus NAC. Overall analysis using the random-effects model showed the difference was statistically insignificant when vitamin C plus NAC and NAC alone were compared (RR = 0.97, 95% CI 0.49–1.92, p = 0.93; see the ESM, Supplementary Fig. 7a). The fixed-effect model did not alter our initial qualitative interpretation. No significant

4 Discussion CI-AKI is one of the most common causes of hospitalacquired AKI, emphasizing the need for more effective protective regimens [1]. This systematic review and metaanalysis focused on whether antioxidant vitamins (vitamin C and vitamin E) were of clinical utility in terms of CI-AKI prevention. The results showed that vitamin C plus saline administration was effective. For vitamin E plus saline, the

Y. Xu et al. Table 2 Subgroup analyses of studies focused on vitamin C based on publication types, publication year, baseline kidney function, sample size, mean age of participants, route of administration, dose of vitamin C, and definition and incidence of CI-AKI Subgroup

Study

Relative risk (95% CI)

P value for heterogeneity

Full text

7

0.67 (0.50, 0.91)

0.49

Abstract

3

0.52 (0.27, 1.00)

Publication year C 2013*

Yes No

5 5

0.65 (0.47, 0.90) 0.62 (0.36, 1.05)

0.87

All patients with decreased kidney function

Yes

5

0.70 (0.51, 0.96)

0.26

No

5

0.51 (0.33, 0.80)

Publication type

Sample size C 143* Mean age(year) C 70*

Route of administration

Dose of vitamin C* Absolute or relative increase in creatinine when defining CI-AKI

Time point for CI-AKI B 72 h Time point for CI-AKI B 48 h Incidence of CI-AKI C 12%*

Yes

5

0.66 (0.49, 0.88)

No

5

0.54 (0.31, 0.94)

Yes

4

0.79 (0.55, 1.14)

No

4

0.47 (0.30, 0.72)

Unclear

2

0.51 (0.14, 1.90)

Orally

5

0.55 (0.33, 0.90)

Intravenously

2

0.50 (0.17, 1.46)

Both

2

0.87 (0.42, 1.78)

High C 7 g

5

0.48 (0.32, 0.73)

Low\7 g

5

0.75 (0.54, 1.05)

Both

6

0.52 (0.35, 0.79)

Relative Absolute

3 1

0.64 (0.34, 1.20) 0.76 (0.51, 1.14)

Yes

7

0.67 (0.49, 0.91)

No

3

0.48 (0.27, 0.86)

Yes

3

0.63 (0.22, 1.77)

No

7

0.63 (0.48, 0.84)

Yes

5

0.59 (0.41, 0.84)

No

5

0.69 (0.37, 1.31)

0.56 0.17

0.54

0.10 0.43

0.33 0.98 0.66

CI-AKI contrast-induced acute kidney injury, CI confidence interval *Grouped by the median value of included studies

evidence is encouraging, although inconclusive. Furthermore, we found that the evidence on the effect of vitamin C plus NAC, versus saline or NAC alone, was insufficient and inconclusive. Although other systematic reviews have been performed, our meta-analysis offers some strengths. First, this review is the most up-to-date and comprehensive metaanalysis available. Most previous reviews were published some time ago or did not include studies published in abstract form [9, 15, 47]. We did not impose any restriction on publication type, thus including abstracts. We comprehensively evaluated the efficacies of both vitamin C and vitamin E in terms of preventing CI-AKI. However, not all studies using vitamins were analyzed. We excluded regimens tested in only single RCTs. Second, we are the first to use TSA to determine the effects of vitamins in terms of preventing CI-AKI. The aim of a meta-analysis is to identify the benefit or harm of an

Figure 3 Trial sequential analysis (TSA) for contrast-induced acute c kidney injury (CI-AKI) in patients taking vitamin C (VC) vs saline (a) and vitamin E (VE) vs saline (b). In (a), the anticipated relative risk reduction (RRR) of 25% yielded a required information size of 1979. Although the required information size was not reached, the cumulative z curve (blue line) crossed both the conventional boundary and the trial sequential monitoring boundary favoring VC plus saline, establishing sufficient and conclusive evidence and suggesting that further trials are not required. In (b) the anticipated RRR of 25% yielded a required information size of 2470. The cumulative z curve (blue line) crossed the conventional boundary and was near to the trial sequential monitoring boundary favoring VE plus saline, but did not cross it. Hence, the result did not establish conclusive evidence, suggesting more trials are needed. The x axis represents the number of patients randomized; the y axis represents cumulative z score; the cumulative z curves (blue line) were constructed with each cumulative z value calculated after including a new trial; a z value of 1.96 corresponds to the conventional boundary in a two-sided hypothesis test; the middle area represents the futility area. TSA was performed with an overall 5% risk of a type I error and a power of 80%

Vitamins and CI-AKI

intervention as early as possible. Thus, meta-analyses are commonly updated when new trials are published, i.e., repeated analyses are performed on accumulating data. In this case, repeated significance testing, which is performed

with the conventional p value criterion, is prone to exacerbate the risk of random error [18]. When adjusted for random error risk, apparently conclusive neonatal metaanalyses might become inconclusive [18]. Some

Y. Xu et al.

investigators found that 17% of the statistically significant meta-analyses were potential false positives and 64% of those meta-analyses which did not achieve statistical significance were potential false negatives [27]. Similarly, Brok and colleagues reported that 39% of the statistically significant Cochrane neonatal reviews were potentially false positive and that 80% of the neonatal reviews which were not statistically significant were potentially false negative [48]. To overcome these limitations of conventional meta-analysis, TSA was recommended [18, 19]. TSA is a statistical approach that may reduce random error risk in cumulative meta-analyses for guarding against false positive results and premature dissemination of marginal or useless interventions [18, 49]. Vitamin C, an antioxidant, attenuates the oxidative damage caused by radiocontrast media and may effectively prevent CI-AKI. It has been reported that vitamin C effectively countered oxidative damage by reducing and scavenging reactive oxygen species that damage macromolecules such as lipids, DNA, and proteins [50]. A number of RCTs explored this topic, but the question of whether vitamin C plus saline exerted a more significant effect than saline alone did not receive a consistent answer [37, 42]. By combining the effect sizes of these RCTs, we found that vitamin C was clinically beneficial in terms of reducing the CI-AKI risk, as found by earlier systematic reviews [15, 47]. However, no prior study used TSA to determine the effect of vitamin C. To avoid overestimating intervention effects and minimize random errors, we performed TSA. Although the required information size was not reached, the cumulative z curve crossed both the conventional boundary and the trial sequential monitoring boundary favoring vitamin C plus saline, establishing sufficient and conclusive evidence and suggesting that further trials are not required. Some studies showed that vitamin C seemed to play an antioxidative role in a dose-dependent manner [7]. According to our analysis, there might be some indication that the reduction in risk of CI-AKI may be associated with the dose of vitamin C, but this finding was not statistically significant. The small sample size and the inadequacies of the studies may mean that subgroup analyses are of limited power when sought to detect differences. We found that although vitamin E significantly reduced the incidence of CI-AKI, TSA indicated that more trials were needed to confirm this conclusion. The antioxidative mechanism of vitamin E differs from that of vitamin C and NAC. Vitamin E is a lipid-soluble, non-enzymatic antioxidant, protecting tissues and cells from free-radical attack by stabilizing the cell membrane, maintaining cellular bioactivity [51]. However, only one included study [43] showed that intravenous vitamin E exerted no preventative effect on CI-AKI in patients undergoing elective computer

tomography (CT). The dose of elective CT was low (100 mL), which may explain why the incidence (0%) of CI-AKI did not differ between the vitamin E and control group. Some trials compared the efficacies of vitamin C and NAC in terms of CI-AKI prevention. After combining the effect sizes of these trials, we found that the difference between vitamin C and NAC was statistically insignificant, similar to the conclusion of another meta-analysis [47]. However, unlike the cited meta-analysis, we further performed TSA and found that the cumulative z curve did not cross either the conventional boundary or the trial sequential monitoring boundary. Therefore, the conclusion is not definitive. Based on our analysis, the combination of vitamin C and NAC may be more effective than saline alone in preventing CI-AKI. But this finding was not statistically significant, and more trials are needed to prove this. Our analysis also suggested that the combination of vitamin C and NAC might be no better than NAC alone. However, few studies are available, and they had small sample sizes; it was impossible to draw definitive conclusions. Our meta-analysis has several limitations. First, the studies included in this meta-analysis were heterogeneous in terms of patient populations and the types of radiological contrast agents used. Patients with various extents of kidney function and different complications were given radiocontrast material of various volumes at different doses. The routes of vitamin administration also varied. Second, most included studies were of low quality; only one study was deemed to exhibit a truly low risk of bias in all domains. Most trials did not report optimal randomization concealment. Of note, no study focusing on vitamin C was deemed to be of high quality, and thus, we did not perform subgroup analysis based on study quality.

5 Conclusion Vitamin C plus saline administration is effective to reduce the risk of CI-AKI, as confirmed by TSA with an anticipated intervention effect of 25% RRR. Vitamin E may potentially protect against CI-AKI, but TSA indicated that the result is inconclusive. Vitamin C may be as effective as NAC in preventing CI-AKI. The combination of vitamin C and NAC might afford no better protection than NAC alone. However, TSA revealed that these findings were not definitive. Further large, well-designed trials are required to provide firmer evidence. Acknowledgements The English in this document has been checked by at least two professional editors, both native speakers of English, from Textcheck Scientific and Technical Editing Service.

Vitamins and CI-AKI Author Contributions Research idea and study design: YX, JG, XZ; data acquisition: ZG, JG; data analysis/interpretation: ZG, JG, YX; statistical analysis: YX; the writing of the paper: ZG, XZ, BL; supervision or mentorship: JG, YX.

11.

Compliance with Ethical Standards Funding This study was supported in part by Hainan Natural Science Foundation of China to Z.Y.G (No.20168350) and Sanya Medical and Health Science and Technology Innovation Project of China to Z.Y.G (No. 2015YW24). Conflict of Interest Yongxing Xu, Xinming Zheng, Boran Liang, Jianjun Gao and Zhaoyan Gu declare that they have no conflicts of interest that might be relevant to the contents of this manuscript. Ethical Standards All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors. Informed consent was not involved.

12.

13.

14.

15.

References 1. Bilasy ME, Oraby MA, Ismail HM, Maklady FA. Effectiveness of theophylline in preventing contrast-induced nephropathy after coronary angiographic procedures. J Interv Cardiol. 2012;25(4):404–10. 2. McCullough PA. Contrast-induced acute kidney injury. J Am Coll Cardiol. 2008;51(15):1419–28. 3. Azzalini L, Spagnoli V, Ly HQ. Contrast-induced nephropathy: from pathophysiology to preventive strategies. Can J Cardiol. 2016;32(2):247–55. 4. Seeliger E, Sendeski M, Rihal CS, Persson PB. Contrast-induced kidney injury: mechanisms, risk factors, and prevention. Eur Heart J. 2012;33(16):2007–15. 5. Jeong BY, Lee HY, Park CG, Kang J, Yu SL, Choi DR, et al. Oxidative stress caused by activation of NADPH oxidase 4 promotes contrast-induced acute kidney injury. PLoS One. 2018;13(1):e0191034. 6. Tasanarong A, Vohakiat A, Hutayanon P, Piyayotai D. New strategy of alpha- and gamma-tocopherol to prevent contrast-induced acute kidney injury in chronic kidney disease patients undergoing elective coronary procedures. Nephrol Dial Transplant. 2013;28(2):337–44. 7. Khaledifar A, Momeni A, Ebrahimi A, Kheiri S, Mokhtari A. Comparison of N-acetylcysteine, ascorbic acid, and normal saline effect in prevention of contrast-induced nephropathy. ARYA Atheroscler. 2015;11(4):228–32. 8. Albabtain MA, Almasood A, Alshurafah H, Alamri H, Tamim H. Efficacy of ascorbic acid, N-acetylcysteine, or combination of both on top of saline hydration versus saline hydration alone on prevention of contrast-Induced nephropathy: a prospective randomized study. J Interv Cardiol. 2013;26(1):90–6. 9. Navarese EP, Gurbel PA, Andreotti F, Kolodziejczak MM, Palmer SC, Dias S, et al. Prevention of contrast-induced acute kidney injury in patients undergoing cardiovascular procedures—a systematic review and network meta-analysis. PLoS ONE. 2017;12(2):e0168726. 10. Ali-Hasan-Al-Saegh S, Mirhosseini SJ, Ghodratipour Z, SarrafanChaharsoughi Z, Rahimizadeh E, Karimi-Bondarabadi AA, et al. Strategies preventing contrast-induced nephropathy after coronary angiography: a comprehensive meta-analysis and systematic

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

review of 125 randomized controlled trials. Angiology. 2017;68(5):389–413. Giacoppo D, Gargiulo G, Buccheri S, Aruta P, Byrne RA, Cassese S, et al. Preventive strategies for contrast-induced acute kidney injury in patients undergoing percutaneous coronary procedures: evidence from a hierarchical Bayesian network metaanalysis of 124 trials and 28 240 patients. Circ Cardiovasc Interv. 2017;10(5):e004383. Subramaniam RM, Suarez-Cuervo C, Wilson RF, Turban S, Zhang A, Sherrod C, et al. Effectiveness of prevention strategies for contrast-induced nephropathy: a systematic review and metaanalysis. Ann Intern Med. 2016;164(6):406–16. Brueck M, Cengiz H, Hoeltgen R, Wieczorek M, Boedeker RH, Scheibelhut C, et al. Usefulness of N-acetylcysteine or ascorbic acid versus placebo to prevent contrast-induced acute kidney injury in patients undergoing elective cardiac catheterization: a single-center, prospective, randomized, double-blind, placebocontrolled trial. J Invasive Cardiol. 2013;25(6):276–83. Spargias K, Alexopoulos E, Kyrzopoulos S, Iokovis P, Greenwood DC, Manginas A, et al. Ascorbic acid prevents contrastmediated nephropathy in patients with renal dysfunction undergoing coronary angiography or intervention. Circulation. 2004;110(18):2837–42. Sadat U, Usman A, Gillard JH, Boyle JR. Does ascorbic acid protect against contrast-induced acute kidney injury in patients undergoing coronary angiography: a systematic review with meta-analysis of randomized, controlled trials. J Am Coll Cardiol. 2013;62(23):2167–75. Ali-Hasan-Al-Saegh S, Mirhosseini SJ, Ghodratipour Z, SarafanChaharsoughi Z, Dehghan AM, Rahimizadeh E, et al. Protective effects of anti-oxidant supplementations on contrast-induced nephropathy after coronary angiography: an updated and comprehensive meta-analysis and systematic review. Kardiol Pol. 2016;74(7):610–26. Su X, Xie X, Liu L, Lv J, Song F, Perkovic V, et al. Comparative effectiveness of 12 treatment strategies for preventing contrastinduced acute kidney injury: a systematic review and Bayesian network meta-analysis. Am J Kidney Dis. 2017;69(1):69–77. Brok J, Thorlund K, Wetterslev J, Gluud C. Apparently conclusive meta-analyses may be inconclusive–trial sequential analysis adjustment of random error risk due to repetitive testing of accumulating data in apparently conclusive neonatal meta-analyses. Int J Epidemiol. 2009;38(1):287–98. Wetterslev J, Jakobsen JC, Gluud C. Trial sequential analysis in systematic reviews with meta-analysis. BMC Med Res Methodol. 2017;17(1):39. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264–9 (W64). Alonso A, Lau J, Jaber BL, Weintraub A, Sarnak MJ. Prevention of radiocontrast nephropathy with N-acetylcysteine in patients with chronic kidney disease: a meta-analysis of randomized, controlled trials. Am J Kidney Dis. 2004;43(1):1–9. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The cochrane collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;18(343):d5928. Friedrich JO, Adhikari NK, Beyene J. Inclusion of zero total event trials in meta-analyses maintains analytic consistency and incorporates all available data. BMC Med Res Methodol. 2007;7:5. Dupuis C, Sonneville R, Adrie C, Gros A, Darmon M, Bouadma L, et al. Impact of transfusion on patients with sepsis admitted in intensive care unit: a systematic review and meta-analysis. Ann Intensive Care. 2017;7(1):5. Thompson SG, Pocock SJ. Can meta-analyses be trusted? Lancet. 1991;338(8775):1127–30.

Y. Xu et al. 26. Egger M, Davey Smith G, Schneider M, Minder C. Bias in metaanalysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–34. 27. AlBalawi Z, McAlister FA, Thorlund K, Wong M, Wetterslev J. Random error in cardiovascular meta-analyses: how common are false positive and false negative results? Int J Cardiol. 2013;168(2):1102–7. 28. Roshanov PS, Dennis BB, Pasic N, Garg AX, Walsh M. When is a meta-analysis conclusive? A guide to trial sequential analysis with an example of remote ischemic preconditioning for renoprotection in patients undergoing cardiac surgery. Nephrol Dial Transplant. 2017;32(suppl 2):ii23–30. 29. Boscheri A, Weinbrenner C, Botzek B, Reynen K, Kuhlisch E, Strasser RH. Failure of ascorbic acid to prevent contrast-media induced nephropathy in patients with renal dysfunction. Clin Nephrol. 2007;68(5):279–86. 30. Dvorsak B, Kanic V, Ekart R, Bevc S, Hojs R. Ascorbic acid for the prevention of contrast-induced nephropathy after coronary angiography in patients with chronic renal impairment: a randomized controlled trial. Ther Apher Dial. 2013;17(4):384–90. 31. El-Fishawy HSN, Soliman A. Ascorbic acid and acetylcysteine for prevention of acute deterioration of renal function following elective aorto-iliac and coronary angioplasty. NDT Plus. 2010;3:300. 32. Grygier MJM, Araszkiewicz A, Kowal J, Mularek-Kubzdela T, Olasinska-Wisniewska A, Seniuk W, Popiel M, Lesiak M, Grajek S. Combined treatment with ascorbic acid and N-acetylcysteine prevents contrast-induced nephropathy in high-risk patients with acute myocardial infarction undergoing percutaneous coronary intervention. Eur Heart J. 2011;32:954–5. 33. Habib M, Hillis A, Hammad A. N-acetylcysteine and/or ascorbic acid versus placebo to prevent contrast-induced nephropathy in patients undergoing elective cardiac catheterization: the NAPCIN trial; A single-center, prospective, randomized trial. Saudi J Kidney Dis Transpl. 2016;27(1):55–61. 34. Hamdi SSW, Hraiech A, Jomaa W, Hamda KB, Maatouk F. Prevention of contrast induced nephropathy in patients undergoing coronarography with ascorbic acid. JACC. 2013;1:S22. 35. Jo SH, Koo BK, Park JS, Kang HJ, Kim YJ, Kim HL, et al. N-acetylcysteine versus AScorbic acid for preventing contrastInduced nephropathy in patients with renal insufficiency undergoing coronary angiography NASPI study—a prospective randomized controlled trial. Am Heart J. 2009;157(3):576–83. 36. Komiyama KTT, Tanabe Y, Harumizu S. Impact of bolus administration of ascorbic acid to prevent contrast induced nephropathy undergoing emergent cardiovascular procedure. Am J Cardiol. 2011;1:29A. 37. Nough H, Daryachahei R, Hadiani L, Najarzadegan MR, Mirzaee M, Hemayati R, et al. Ascorbic acid effect on CIN incidence in diabetic patient after coronary angiography. Adv Biomed Res. 2016;5:69. 38. Rezaei Y, Khademvatani K, Rahimi B, Khoshfetrat M, Arjmand N, Seyyed-Mohammadzad MH. Short-term high-dose vitamin E to prevent contrast medium-induced acute kidney injury in patients with chronic kidney disease undergoing elective

39.

40.

41. 42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

coronary angiography: a randomized placebo-controlled trial. J Am Heart Assoc. 2016;5(3):e002919. Tasanarong A, Piyayotai D, Thitiarchakul S. Protection of radiocontrast induced nephropathy by vitamin E (alpha tocopherol): a randomized controlled pilot study. J Med Assoc Thai. 2009;92(10):1273–81. Wenhua LDL, Yanbin Z, Tongda X, Hong Z, Fei H. Prevention of contrast-induced acute kidney injury with the sodium bicarbonate in ACS patients undergoing PCI. Heart. 2011;97:A110. Zhou L, Chen H. Prevention of contrast-induced nephropathy with ascorbic acid. Intern Med. 2012;51(6):531–5. Briguori C, Airoldi F, D’Andrea D, Bonizzoni E, Morici N, Focaccio A, et al. Renal insufficiency following contrast media administration trial (REMEDIAL): a randomized comparison of 3 preventive strategies. Circulation. 2007;115(10):1211–7. Kitzler TM, Jaberi A, Sendlhofer G, Rehak P, Binder C, Petnehazy E, et al. Efficacy of vitamin E and N-acetylcysteine in the prevention of contrast induced kidney injury in patients with chronic kidney disease: a double blind, randomized controlled trial. Wien Klin Wochenschr. 2012;124(9–10):312–9. Addad F, Triki M, Ibn El Hadj Z, Laaroussi L, Marrakchi S, Ben Amara W, et al. Does oral ascorbic acid plus sodium bicarbonate protect against contrast-induced nephropathy in diabetic and/or renal insufficiency patients undergoing coronary angiography? A multicentre randomised controlled trial. EuroIntervention, EuroPCR Abstracts and Posters 2014. Shakeryan F, Sanati H, Fathi H, Firouzi A, Zahedmehr A, Valizadeh G, et al. Evaluation of combination therapy with vitamin C and pentoxifylline on preventing kidney failure secondary to intravenous contrast material in coronary angioplasty. Iran Heart J. 2013;14(3):17–21. Smith OM, Wald R, Adhikari NK, Pope K, Weir MA, Bagshaw SM, et al. Standard versus accelerated initiation of renal replacement therapy in acute kidney injury (STARRT-AKI): study protocol for a randomized controlled trial. Trials. 2013;05(14):320. Feng Y, Huang X, Li L, Chen Z. N-acetylcysteine versus ascorbic acid or N-acetylcysteine plus ascorbic acid in preventing contrast-induced nephropathy: a meta-analysis. Nephrology (Carlton). 2017. https://doi.org/10.1111/nep.13068. Brok J, Thorlund K, Gluud C, Wetterslev J. Trial sequential analysis reveals insufficient information size and potentially false positive results in many meta-analyses. J Clin Epidemiol. 2008;61(8):763–9. Thorlund K, Devereaux PJ, Wetterslev J, Guyatt G, Ioannidis JP, Thabane L, et al. Can trial sequential monitoring boundaries reduce spurious inferences from meta-analyses? Int J Epidemiol. 2009;38(1):276–86. Fukumura H, Sato M, Kezuka K, Sato I, Feng X, Okumura S, et al. Effect of ascorbic acid on reactive oxygen species production in chemotherapy and hyperthermia in prostate cancer cells. J Physiol Sci. 2012;62(3):251–7. Wang X, Quinn PJ. Vitamin E and its function in membranes. Prog Lipid Res. 1999;38(4):309–36.