|
|
||||||||
a Department of Internal Medicine, Rochester General Hospital and University of Rochester School of Medicine, Rochester, New York
b Department of Internal Medicine, Southern Illinois University, Springfield, Illinois
Accepted for publication September 9, 2008.
* Address correspondence to Dr Nigwekar, Rochester General Hospital, 1425 Portland Ave, Rochester, NY 14621 (Email: sagar.nigwekar{at}rochestergeneral.org).
| Cardiothoracic anesthesiology:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
| Abstract |
|---|
|
|
|---|
Methods: We searched MEDLINE, EMBASE, Cochrane Renal Health Library, and Google Scholar for randomized controlled studies that evaluated NAC in adult patients undergoing cardiovascular surgery. Acute renal failure, acute renal failure requiring dialysis, and mortality were the primary outcomes. Additional outcomes studied were length of intensive care unit stay, postoperative serum creatinine, creatinine clearance, renal biomarkers, and adverse effects of NAC.
Results: Twelve studies comprising 1,324 patients were found to be eligible. Meta-analytic estimates showed that NAC was not associated with reduction in acute renal failure (odds ratio [OR]: 0.89, 95% confidence interval [CI]: 0.68 to 1.15), acute renal failure requiring dialysis (OR: 1.09, 95% CI: 0.57 to 2.09) or mortality (OR: 0.95, 95% CI: 0.53 to 1.71). N-acetylcysteine was well tolerated but was not associated with any reduction in the length of intensive care unit stay. It had inconsistent effects on postoperative serum creatinine, creatinine clearance, and renal biomarkers. Subgroup analysis restricted to studies using intravenous NAC preparation showed a nonsignificant trend toward reduction in acute renal failure (OR: 0.81, 95% CI: 0.61 to 1.08) without any significant change in other outcomes.
Conclusions: Overall analysis of the existent literature shows that NAC is not beneficial in the prevention of post–cardiovascular surgery renal dysfunction. Routine use of NAC for this indication should be avoided.
Acute renal failure is among the most serious complications in patients undergoing cardiovascular surgeries. The incidence of acute renal failure in cardiac bypass and thoracoabdominal aortic surgeries depends on its definition and has been reported to be between 1% and 30% [1–3]. Acute renal failure in these settings is associated with increased mortality [4–6], more complicated hospital course [7, 8], higher risk for infections [9], and a substantial increase in the risk for permanent dialysis [10].
Acute renal failure after cardiovascular surgery is mostly due to renal parenchymal damage and is thought to be related to several factors, including hypovolemia due to operative complications, inflammation, ischemia-reperfusion, neurohumoral activation, aortic cross-clamping, and increased oxygen free-radical generation [3, 11]. N-acetylcysteine (NAC), with its pleiotropic effects, such as oxygen radical neutralization and endothelium-mediated vasodilatation [12] seems to be a promising agent to counteract some of these factors. However, randomized controlled studies evaluating the role of NAC in this setting have been largely underpowered and have produced conflicting results. In addition, NAC is thought to increase the risk of postoperative bleeding through its anticoagulant and platelet inhibitor activities [13]. The purpose of this review was to undertake a systematic analysis of randomized controlled studies to ascertain the therapeutic potential of NAC in the prevention of renal failure that occurs after cardiovascular surgery.
| Material and Methods |
|---|
|
|
|---|
To be included, the studies had to report at least one of the following renal outcomes: incidence of acute renal failure, incidence of acute renal failure requiring dialysis, postoperative serum creatinine, creatinine clearance, glomerular filtration rate, or renal biomarkers. Studies evaluating role of NAC in noncardiovascular surgical settings (such as radiocontrast nephropathy, sepsis, and so forth) and those that did not report the specified renal outcomes were excluded.
Data Extraction and Quality Assessment
Two authors (S.U.N., P.K.) independently assessed the studies for eligibility and extracted relevant data regarding study design and setting, participant characteristics, and outcome measures using a standardized data extraction form. We accepted the outcome definitions used by the original researchers. Descriptions such as "no major complications were observed in the study" were not considered as zero events. Only explicit descriptions of outcome events were tabulated.
The results of the individual studies were reported in many different ways, including mean and standard deviation (SD), standard error of the mean (SEM), or interquartile range (IQR). We converted SEMs and IQRs to SD, using appropriate formulas. We considered IQR to be 1.35 times the SD. Standard deviation was calculated as square root of sample size times the SEM. All data were converted to uniform measurements; thus, serum creatinine is presented as mg/dL and creatinine clearance as mL/min.
The method of all included studies was rated by means of the validated scale by Jadad and colleagues [14]. This scale considers randomization, blinding, and withdrawal/dropouts. Studies were considered of low quality if the Jadad score was from 0 to 2, of moderate quality if the score was from 3 to 4, and of high quality if the score was 5.
Outcome Measures
The primary outcomes of interest for the current review were acute renal failure, acute renal failure requiring dialysis, and mortality. Secondary outcomes analyzed included duration of intensive care unit stay, postoperative serum creatinine, creatinine clearance/glomerular filtration rate, and indicators of renal injury, for example, urine N-acetyl glucosaminidase and urine albumin to creatinine ratio. We also abstracted data regarding any adverse effects of NAC reported in the included studies, such as bleeding complications (assessed by the amount of blood loss, postoperative blood transfusion requirement, and incidence of reoperation).
Data Analysis and Quantitative Data Synthesis
We analyzed data according to guidelines in the Cochrane Reviewers' Handbook [15]. All the analyses were performed using RevMan 4.2.10 (Cochrane Collaboration, Oxford, United Kingdom). Dichotomous data outcomes from individual studies were analyzed according to the Mantel-Haenszel model to compute individual odds ratio (OR) with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the effects of treatment, the weighted mean difference was used. Statistical significance was set at the two-tailed 0.05 level for hypothesis testing. Statistical heterogeneity was analyzed using heterogeneity
2 (Cochrane Q) statistic and I2 test [15, 16]. The I2 values of 25%, 50%, and 75% correspond to low, medium, and high levels of statistical heterogeneity. Treatment effects were analyzed with the random-effects model. If there was a significant heterogeneity (I2 >25%), then we explored possible sources of heterogeneity (for example, participants, study quality, and so forth).
We performed the following subgroup analyses: (1) to explore the effect of nondialysis-dependent preexisting renal insufficiency, we performed an analysis from studies that specifically included patients with preexisting renal insufficiency; (2) to explore the effect of underlying surgery, we performed an analysis for cardiac versus vascular surgical studies; (3) to explore the effect of different formulations of NAC, we performed a separate analysis for studies using only intravenous preparations; and (4) to explore the effect of dose of NAC preparations, we performed a separate analysis for studies using high-dose NAC preparations. Studies included in the high-dose category were those that administered a cumulative dose of >150 mg/kg of NAC.
Sensitivity Analysis
We assessed the robustness of findings from the primary analysis to the effects of study population and baseline risk of any of the primary outcomes through sensitivity analyses by switching from random-effect to fixed-effect models, computing relative risks and repeating the analyses to assess the influence of study quality.
| Results |
|---|
|
|
|---|
|
|
Primary Outcomes
Acute renal failure and acute renal failure requiring dialysis were reported in seven and 11 studies, respectively. Meta-analytic estimate showed that the use of NAC was not associated with reduction in acute renal failure (OR: 0.89, 95% CI: 0.68 to 1.15; Fig 2) and acute renal failure requiring dialysis (OR: 1.09, 95% CI: 0.57 to 2.09; Fig 3). Mortality was reported in all studies, and meta-analytic estimate showed that NAC was not associated with any reduction in mortality (OR: 0.95, 95% CI: 0.53 to 1.71; Fig 4). There was no statistical heterogeneity among the studies for any of the primary outcomes.
|
|
|
Ten studies reported postoperative serum creatinine values [17, 19–27], four studies reported estimated creatinine clearance or glomerular filtration rate [18, 21, 26, 27], and one study reported 24-hour creatinine clearance [28]. Fischer and colleagues [21], Barr and associates [18], and Winjen and colleagues [28] demonstrated that the postoperative decrease in creatinine clearance was significantly attenuated by NAC, without any other significant difference between the NAC and control groups. Postoperative urine albumin to creatinine ratio [23, 28], urine N-acetyl glucosaminidase/creatinine ratio [23, 25], and serum cystatin C values [22, 23, 25] were not significantly different between the NAC and control groups.
Adverse Effects
We analyzed the adverse effects profile of NAC as reported in individual studies. Incidence of adverse effects such as nausea, headache, anaphylaxis, and hypotension was not different in the NAC group compared with the control group (OR: 0.90, 95% CI: 0.55 to 1.48). Amount of blood loss during the first 24 hours after the surgery was reported in one study, and it was significantly more in the NAC group compared with the control group; however, that did not translate into increased requirement of blood transfusion [25]. Similarly, Hynninen and coworkers [23] and Sisillo and associates [26] reported no difference between the NAC and control groups for the perioperative blood transfusion requirement. Risk of surgical reoperation was reported in 6 studies, and it was not different between the NAC and control groups (OR: 1.16, 95% CI: 0.62 to 2.17).
Subgroup Analyses
To explore the effect of nondialysis-dependent preexisting renal insufficiency on outcomes, we performed a subgroup analysis on studies that specifically included patients with preexisting renal insufficiency (nondialysis dependent). Restricting the analysis to these studies did not change the effects of NAC on acute renal failure (OR: 0.79, 95% CI: 0.57 to 1.10), acute renal failure requiring dialysis (OR: 1.26, 95% CI: 0.62 to 2.58), or mortality (OR: 0.57, 95% CI: 0.25 to 1.30). After restricting the analysis to cardiac surgical patients, the effects of NAC on acute renal failure (OR: 0.86, 95% CI: 0.66 to 1.13), acute renal failure requiring dialysis (OR: 1.03, 95% CI: 0.53 to 2.00), and mortality (OR: 0.77, 95% CI: 0.39 to 1.50) remained nonsignificant. Restricting the analysis to studies that used intravenous NAC only did not change the effects on acute renal failure (OR: 0.81, 95% CI: 0.61 to 1.08), acute renal failure requiring dialysis (OR: 1.04, 95% CI: 0.37 to 2.94), and mortality (OR: 0.58, 95% CI, 0.24 to 1.41). Restricting the analysis to studies that administered high-dose NAC did not change the effects on acute renal failure (OR: 0.89, 95% CI: 0.57 to 1.39), acute renal failure requiring dialysis (OR: 0.75, 95% CI: 0.17 to 3.23), and mortality (OR: 0.92, 95% CI, 0.28 to 3.08).
Sensitivity Analysis
Sensitivity analyses performed by switching from random-effect to fixed-effect models, computing relative risks did not change the overall results for the primary outcomes. Restricting the analysis to studies of high quality did not change the effects of NAC on primary outcomes.
| Comment |
|---|
|
|
|---|
Our meta-analysis assessed whether NAC, an inexpensive medicine with antioxidant properties, would be beneficial in improving renal outcomes in patients undergoing cardiovascular surgery. We found that the prophylactic use of NAC was overall well tolerated but was not beneficial in improving acute renal failure, dialysis requirement, and mortality in the cardiovascular surgical setting when radiocontrast dye is not used. In our review, NAC did not reduce the length of intensive care unit stay and did not have consistent beneficial effects on surgery serum creatinine, creatinine clearance, and renal injury markers (urine N-acetyl glucosaminidase and urine albumin to creatinine ratio) after surgery.
Several mechanisms are thought to be playing a role in the development of post–cardiovascular surgery renal dysfunction. Release of reactive free oxygen radicals and neutrophil activation leading to increased levels of tumor necrosis factor-
and interleukin-8 are known to contribute to this renal dysfunction along with other factors such as hypovolemia, ischemia-reperfusion, neurohumoral activation, and aortic cross-clamping [11, 30]. In experimental studies, NAC has attenuated renal dysfunction through different anti-inflammatory and antioxidant effects such as antagonism of tumor necrosis factor-
and inhibition of the vascular cell adhesion molecule expression [30]. In radiocontrast nephropathy, NAC has been demonstrated to provide renal protection by acting as an antioxidant and arteriolar vasodilator through the nitrous oxide pathway [31]. Furthermore, NAC has been shown to prevent cardiopulmonary bypass pump–induced inflammatory response [32]. Despite these positive experimental and clinical findings, it is intriguing to see NAC was not found to be beneficial in the prevention of renal dysfunction in cardiovascular surgical setting. There may be several possible explanations for this. Mechanisms other than those inhibited by NAC possibly contribute more toward the pathogenesis of post–cardiovascular surgery acute renal failure. Antioxidant and vasodilatory actions of NAC are not enough to negate these other mechanisms. Although our review included studies involving NAC doses well above those used for radiocontrast nephropathy prevention [33], the optimal dose of NAC needed to prevent post–cardiovascular surgery renal dysfunction is unknown and may well be much higher than used so far in any study. The route of administration of NAC might have played role as bioavailability of oral NAC may be unpredictable in the perioperative setting. In subgroup analysis, however, we did not observe any benefit from intravenous NAC preparations.
Our systematic review has limitations. Some of the studies included patients undergoing cardiac surgery using off-pump technique and that could have attenuated the potential beneficial effects of NAC as cardiopulmonary bypass pump is known to induce a significant inflammatory response, which has been previously shown to be reduced by NAC [32]. In fact, Sisillo and colleagues [26] reported in their study that the use of NAC was associated with statistically significant reduction in acute renal failure when analysis was restricted to patients who underwent cardiac surgery using on-pump technique. However, NAC was not consistently beneficial in studies that specifically excluded patients who underwent cardiac surgery using off-pump technique. Individual study definitions of acute renal failure were not reported in some studies and varied among those that reported them, and that reduces the validity of the meta-analytic estimate for this outcome. Future studies related to acute renal failure should use standard definitions such as the one proposed by the Acute Kidney Injury Network [34]. The outcomes considered in our review were not necessarily the primary outcomes of interest to the study authors, and hence, the included studies were underpowered to detect any significant difference, especially for outcomes such as acute renal failure requiring dialysis and mortality. For definitive evaluation of outcome such as acute renal failure requiring dialysis, the sample size required will be approximately 2,400 (two-sided alpha 0.05, 80% power) [35]. Because the required sample size is large, it is unlikely that a future single randomized controlled trial will be able to address the role of NAC in this setting, and the findings of our comprehensive review provide the best available evidence to address its role in the prevention of post–cardiovascular surgery renal dysfunction.
In conclusion, overall analysis of the existent literature shows that the perioperative use of NAC is not beneficial in the prevention of post–cardiovascular surgery acute renal failure, and the routine use of NAC for this indication should be avoided.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |