Free
Perioperative Medicine  |   April 2013
Preoperative Estimates of Glomerular Filtration Rate as Predictors of Outcome after Surgery: A Systematic Review and Meta-analysis
Author Affiliations & Notes
  • John F. Mooney, M.B.B.S.
    PhD Student, The George Institute for Global Health, Sydney, Australia, and Department of Cardiology, Westmead Hospital, Sydney, Australia.
  • Isuru Ranasinghe, M.B.Ch.B., M.Med.(Clin. Epi.)
    PhD Student, The George Institute for Global Health, and Department of Cardiology, Concord Repatriation General Hospital, Sydney, Australia.
  • Clara K. Chow, M.B.B.S., Ph.D.
    Associate Professor, The George Institute for Global Health, and Department of Cardiology, Westmead Hospital.
  • Vlado Perkovic, M.B.B.S., Ph.D.
    Associate Professor, The George Institute for Global Health.
  • Federica Barzi, B.Sc., Ph.D.
    Senior Lecturer, The George Institute for Global Health.
  • Sophia Zoungas, M.B.B.S., Ph.D.
    Associate Professor, The George Institute for Global Health, and School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia.
  • Martin J. Holzmann, M.D., Ph.D.
    Associate, Department of Emergency Medicine, Karolinska University Hospital and Department of Internal Medicine, Karolinska Institute, Stockholm, Sweden.
  • Gijs M. Welten, M.D., Ph.D.
    Department of Vascular Surgery, Erasmus Medical Centre, Rotterdam, The Netherlands.
  • Fausto Biancari, M.D., Ph.D.
    Adjunct Professor, Division of Cardiothoracic and Vascular Surgery, Oulu University Hospital, Oulu, Finland.
  • Vin-Cent Wu, M.D., Ph.D.
    Assistant Professor, Division of Nephrology, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan.
  • Timothy C. Tan, M.B.B.S., Ph.D.
    Clinical Research Fellow, Department of Cardiology, Westmead Hospital.
  • Alan Cass, M.B.B.S., Ph.D.
    Professor, The George Institute for Global Health.
  • Graham S. Hillis, M.B.Ch.B., Ph.D.
    Associate Professor, The George Institute for Global Health, and Department of Cardiology, Concord Repatriation General Hospital.
  • This article has been selected for the ANESTHESIOLOGY CME Program. Learning objectives and disclosure and ordering information can be found in the CME section at the front of this issue.
    This article has been selected for the ANESTHESIOLOGY CME Program. Learning objectives and disclosure and ordering information can be found in the CME section at the front of this issue.×
  • This article is featured in “This Month in Anesthesiology.” Please see this issue of Anesthesiology, page 9A.
    This article is featured in “This Month in Anesthesiology.” Please see this issue of Anesthesiology, page 9A.×
  • This article is accompanied by an Editorial View. Please see: Augoustides JG, Neuman MD, Fleisher LA: Estimated glomerular filtration rate: More bang for the buck. Anesthesiology 2013; 118:775–6.
    This article is accompanied by an Editorial View. Please see: Augoustides JG, Neuman MD, Fleisher LA: Estimated glomerular filtration rate: More bang for the buck. Anesthesiology 2013; 118:775–6.×
  • Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are available in both the HTML and PDF versions of this article. Links to the digital files are provided in the HTML text of this article on the Journal′s Web site (www.anesthesiology.org).
    Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are available in both the HTML and PDF versions of this article. Links to the digital files are provided in the HTML text of this article on the Journal′s Web site (www.anesthesiology.org).×
  • Received from the Cardiovascular Division, The George Institute for Global Health, Sydney, Australia. Submitted for publication April 1, 2012. Accepted for publication November 20, 2012. John F. Mooney is supported by a McGaughey Research Entry Scholarship, Royal Australasian College of Physicians, Sydney, Australia. Isuru Ranasinghe is supported by a National Health and Medical Research Centre postgraduate scholarship, Canberra, Australia. Clara K. Chow is supported by a National Health and Medical Research Centre Career Development Fellowship, Canberra, Australia. Vlado Perkovic is supported by a New South Wales Cardiovascular Research Network/Heart Foundation Career Development Fellowship, Melbourne, Australia. Sophia Zoungas is supported by a Heart Foundation Career Development Award, Melbourne, Australia. Timothy C. Tan is supported by a BJ Amos Travelling Fellowship, Westmead Hospital Association, Sydney, Australia. Graham S. Hillis is supported by a New South Wales Office for Science and Medical Research, Life Sciences Research Award, Sydney, Australia.
    Received from the Cardiovascular Division, The George Institute for Global Health, Sydney, Australia. Submitted for publication April 1, 2012. Accepted for publication November 20, 2012. John F. Mooney is supported by a McGaughey Research Entry Scholarship, Royal Australasian College of Physicians, Sydney, Australia. Isuru Ranasinghe is supported by a National Health and Medical Research Centre postgraduate scholarship, Canberra, Australia. Clara K. Chow is supported by a National Health and Medical Research Centre Career Development Fellowship, Canberra, Australia. Vlado Perkovic is supported by a New South Wales Cardiovascular Research Network/Heart Foundation Career Development Fellowship, Melbourne, Australia. Sophia Zoungas is supported by a Heart Foundation Career Development Award, Melbourne, Australia. Timothy C. Tan is supported by a BJ Amos Travelling Fellowship, Westmead Hospital Association, Sydney, Australia. Graham S. Hillis is supported by a New South Wales Office for Science and Medical Research, Life Sciences Research Award, Sydney, Australia.×
  • Address correspondence to Dr. Mooney: The George Institute for Global Health, King George V Building, Royal Prince Alfred Hospital, Missenden Road, Camperdown NSW 2050 Australia. jmooney@georgeinstitute.org.au. This article may be accessed for personal use at no charge through the Journal Web site, www.anesthesiology.org.
Article Information
Perioperative Medicine / Cardiovascular Anesthesia / Renal and Urinary Systems / Electrolyte Balance / Quality Improvement
Perioperative Medicine   |   April 2013
Preoperative Estimates of Glomerular Filtration Rate as Predictors of Outcome after Surgery: A Systematic Review and Meta-analysis
Anesthesiology 04 2013, Vol.118, 809-824. doi:10.1097/ALN.0b013e318287b72c
Anesthesiology 04 2013, Vol.118, 809-824. doi:10.1097/ALN.0b013e318287b72c
Abstract

Background:: Kidney dysfunction is a strong determinant of prognosis in many settings.

Methods:: A systematic review and meta-analysis was undertaken to explore the relationship between estimated glomerular filtration rate (eGFR) and adverse outcomes after surgery. Cohort studies reporting the relationship between eGFR and major outcomes, including all-cause mortality, major adverse cardiovascular events, and acute kidney injury after cardiac or noncardiac surgery, were included.

Results:: Forty-six studies were included, of which 44 focused exclusively on cardiac and vascular surgery. Within 30 days of surgery, eGFR less than 60 ml·min·1.73 m−2 was associated with a threefold increased risk of death (multivariable adjusted relative risk [RR] 2.98; 95% confidence interval [CI] 1.95–4.96) and acute kidney injury (adjusted RR 3.13; 95% CI 2.22–4.41). An eGFR less than 60 ml·min·1.73 m−2 was associated with an increased risk of all-cause mortality (adjusted RR 1.61; 95% CI 1.38–1.87) and major adverse cardiovascular events (adjusted RR 1.49; 95% CI 1.32–1.67) during long-term follow-up. There was a nonlinear association between eGFR and the risk of early mortality such that, compared with patients having an eGFR more than 90 ml·min·1.73 m−2 the pooled RR for death at 30 days in those with an eGFR between 30 and 60 ml·min·1.73 m−2 was 1.62 (95% CI 1.43–1.80), rising to 2.85 (95% CI 2.49–3.27) in patients with an eGFR less than 30 ml·min·1.73 m−2 and 3.75 (95% CI 3.44–4.08) in those with an eGFR less than 15 ml·min·1.73 m−2.

Conclusion:: There is a powerful relationship between eGFR, and both short- and long-term prognosis after, predominantly cardiac and vascular, surgery.

Estimated glomerular filtration rate is a powerful and consistent predictor of risk of mortality, cardiovascular events, and acute kidney injury after surgery.

Supplemental Digital Content is available in the text.

What We Already Know about This Topic
  • Estimated glomerular filtration rate is a better estimate of kidney function than increased creatinine concentration

What This Article Tells Us That Is New
  • This meta-analysis of 49 studies finds that estimated glomerular filtration rate less than 60 ml·min·1.73 m−2 is associated with a three-fold increase in 30-day mortality

  • There was a strong nonlinear increase in mortality at lower preoperative estimated glomerular filtration rates

MAJOR surgery results in significant physiological stress and may be associated with adverse early and long-term outcomes. With an aging population surgery is now performed on more than 200 million patients annually, many with comorbidities and advanced disease. The ability to stratify risk ensures that patients are well informed, clinicians can decide on the most appropriate management, and that suitable perioperative care is planned.
Renal dysfunction is an established marker of an adverse outcome, in particular cardiovascular complications, in many settings.1–3  Its role in patients undergoing surgery is not, however, as clearly defined. Although individual studies have identified kidney dysfunction as predictor of worse perioperative outcome these have varied in terms of the measure used, the threshold(s) chosen, the surgical population, the outcomes selected, and the duration of follow-up.
Traditionally, renal function has been estimated using serum creatinine levels, often dichotomized around a largely arbitrary level. These have been incorporated into some perioperative risk scoring systems4–6  but not in all,7,8  reflecting residual uncertainty about the independence and strength of the relationship. Indeed, although increased creatinine levels (>2.0 mg/dl [177 µm]) are included in the most commonly used risk prediction score for noncardiac surgery, the Revised Cardiac Risk Index,5  they were not found to be significant predictors of outcome in the validation cohort used in this study.5  Likewise, although the American Heart Association/American College of Cardiology 2007 guidelines on noncardiac surgery9  identify “renal insufficiency” as an intermediate clinical risk factor they acknowledge the lack of good quality evidence in this setting.
It is now generally accepted that creatinine is an inexact measure of kidney function and is limited in defining mild renal impairment.10,11  A variety of more precise measures have been described and used for risk stratification. In particular, the estimated glomerular filtration rate (eGFR) can be derived by combining creatinine with other demographic parameters and is a more accurate indicator of renal function,12–14  identifies milder degrees of renal dysfunction,15–17  and predicts cardiovascular and renal outcomes in diverse populations.1,18  For these reasons, the National Kidney Foundation recommends12  that eGFR, calculated using the Cockcroft-Gault19  or Modification of Diet in Renal Disease study equations,13  be used to determine renal function in adults, particularly those with, or at risk of, cardiovascular disease.20 
Against this background, the aim of the current systematic review and meta-analysis of the literature is to clarify the nature, strength, and consistency of the relationship between eGFR and adverse outcomes (all-cause mortality, cardiovascular events, and acute kidney injury) after any type of surgery.
Materials and Methods
Study Selection
We included prospective or retrospective cohort studies that reported data on preoperative eGFR (or creatinine clearance) as predictors of postoperative adverse events. When eGFR was calculated by different formulae in the same article, and data were presented for each method separately, we used the most recent, validated, method of calculation. We excluded articles that were not cohort studies, articles that measured only postoperative renal function as a predictor, articles that did not describe the outcomes of interest, and articles on transplant surgery.
Outcomes
The outcomes of interest were all-cause mortality, composite cardiovascular events, and AKI. Composite cardiovascular events were defined as acute myocardial infarction, heart failure, cardiac arrest, cardiovascular death, and stroke, or as defined by the study authors. AKI was collected according to the definitions used in the studies reporting this outcome, though these varied considerably. Both short-term (in hospital or up to 30 days postoperatively) and long-term (any follow-up beyond 30 days) outcomes were assessed.
Search Strategy
A prospective study protocol was devised and followed (see Supplemental Digital Content 1, http://links.lww.com/ALN/A910, which outlines the search protocol).The Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines were followed.21  Two reviewers (J.F.M. and I.R.) independently searched the literature from Medline, EMBASE, CINAHL, and Cochrane central register of controlled trials up to May 17, 2010. No language restrictions were applied. A quality assessment tool was used to ensure selected articles were consistent with the study design and to maintain uniformity between reviewers. Disagreement on article inclusion was resolved by discussion with a third party (G.S.H. and C.K.C.).
Data Extraction and Quality Assessment
Relevant data, expressed as relative risk (RR), hazard ratio, or odds ratio, were extracted from each article and entered into a standardized database. For studies that used differing categories of eGFR, we extracted data into the equivalent stages of chronic kidney disease as defined by the Kidney Disease Outcomes Quality Initiative.12  Data were expressed as either short or long term depending on the duration of follow-up. To facilitate the analysis and interpretation of data, estimates of GFR derived from different formulae (almost exclusively the Cockcroft-Gault or modified diet in renal disease study equations) were considered to be equivalent.
Methodological quality for each study was assessed based on guidelines for assessing quality in prognostic studies.22  Two reviewers (J.F.M. and C.K.C.) independently used these guidelines to assess bias from six potential sources. The degree to which the study had taken adequate measures to eliminate the potential for bias in each domain was graded as “yes” (when there was clear evidence that adequate measures had been taken), “no” (when the measures taken were inadequate to eliminate bias), “not reported” (when the article did not report the measures taken in this domain), and “unclear” (where the adequacy of the reported measures was difficult to assess). The control for potential confounding factors was defined as being of adequate quality (“yes”) if some degree of multivariable assessment had been performed and the variables included had been clearly reported.
Statistical Analysis
Meta-analysis was performed where sufficient studies contributed data measuring an association between preoperative eGFR and the risk of a relevant outcome. Results were pooled using random effects meta-analyses to calculate summary estimates of RRs with confidence intervals (CIs) using the function Metan in STATA software version 11 (College Station, TX). Multivariable adjusted RRs were determined using the maximally adjusted model presented in the article. Where data were presented categorically by range of eGFR, the nature of the association was determined by using meta-regression models (Metareg in STATA). When comparing the utility of serum creatinine and eGFR in predicting 30-day mortality only unadjusted binary data were available.
The trend of the association between increasing levels of GFR and risk of mortality was examined using methodology outlined in a previous community-based systematic review and meta-analysis.1  Eight studies provided RRs of 30-day mortality by subgroups of GFR. The level of GFR in each subgroup was categorized by substituting the midpoint of the range for interval data and adding or subtracting 15 ml·min·1.73 m−2 from bounded data. The association between index levels of GFR and risk of mortality was estimated using a logistic model with a random intercept for study to take account of autocorrelation of estimates generated from the same study. The estimated risk and 95% CIs were plotted against the corresponding index levels of GFR using a bubble plot graph, with the size of the circles proportional to the size of the study (SAS/STAT version 9.2, Cary, NC).
The percentage of variability across studies attributable to heterogeneity, rather than chance, was estimated using the I2 statistic.23  Subgroup analysis was conducted to explore potential causes of heterogeneity and account for inherent bias due to selection, classification, and confounders among the different studies. Subgroups assessed included: surgery type (cardiac or noncardiac), age, sex, hypertension, heart failure, diabetes, method of GFR estimation (Cockcroft-Gault or modified diet in renal disease study equation), prospective or retrospective study, length of follow-up, emergency surgery, and advanced kidney disease (defined as dialysis dependence and/or an eGFR <15 ml·min·1.73 m−2). Publication bias was tested by the use of funnel plots (Metafunnel in STATA) using both Egger and Begg tests for heterogeneity.24 
Results
Characteristics of Included Studies
The search incorporated 4,828 articles of which 150 were selected for abstract review. A further 60 potential articles were identified from bibliography review. A total of 46 articles met inclusion criteria for the review.16–18,25–67  A flowchart outlining the search results is shown in figure 1.
Fig. 1.
Flow diagram of study selection.
Flow diagram of study selection.
Fig. 1.
Flow diagram of study selection.
×
Of the included studies, 28 involved cardiac surgery16–18,29,30,32,34,36–40,42–44,47,51,53,56–59,61–64,66,67  and 18 noncardiac surgery,25–28,31,33,35,41,45,46,48–50,52,54,55,60,65  of which 16 exclusively related to vascular surgery.26,27,31,33,35,41,45,46,48–50,52,54,55,60,65  The average age of study participants was 65 yr and 59% were male. Twenty-five studies excluded patients with severe renal disease.17,25,30,32,35–41,43,44,48,49,51,53,55,59–64,66,67  Most commonly, this was defined as dialysis dependence though some studies excluded patients with an eGFR below 30 ml·min·1.73 m−2. Six studies specifically excluded emergency surgical cases from their analysis.18,29,30,50,51,65  Fifteen studies were prospective15,17,18,25,28,29,35,38,42,43,51,57,59,62,65  and 31 retrospective.16,26,27,30–34,36,37,39–41,44–50,52–56,58,60,61,63,64,66  Of the 18 studies17,18,28,29,35,37,39,42–45,50,52,56,60,62,64,65  that performed follow-up after hospitalization, the average length of follow-up was 5.9 yr. The characteristics of the studies are outlined in table 1.
Table 1.
Characteristics of Included Studies
Characteristics of Included Studies×
Characteristics of Included Studies
Table 1.
Characteristics of Included Studies
Characteristics of Included Studies×
×
With regard to preoperative measurement of renal function, 26 articles calculated eGFR with the Cockcroft-Gault equation25,26,28,31,33,36,37,40,41,43–47,51,54,56–63,65,66  and 18 used the Modified diet in renal disease study equation.16–18,27,29,30,32,34,35,38,39,48–50,52,53,55,64  One article used serum cystatin C42 and derived eGFR with a specific cystatin c-based formula.68  Where data were presented as eGFR, the studies varied in definition of what was considered an abnormal eGFR, by either binary representation or stage of eGFR decline.
The methodological quality of each study is shown in appendix 1, table 2. Ten studies did not adequately report attrition.25,27,31,35,41,45,46,54,58,59  Otherwise most studies satisfied the qualitative assessments in the other domains of potential bias. Publication bias was detected in analysis of all cause mortality, demonstrated in the final sentence of paragraph 3 on this page, and also acute kidney injury, which is discussed in the second paragraph on page 14.
Appendix 1, Table 2.
Methodological Quality of Articles in Systematic Review
Methodological Quality of Articles in Systematic Review×
Methodological Quality of Articles in Systematic Review
Appendix 1, Table 2.
Methodological Quality of Articles in Systematic Review
Methodological Quality of Articles in Systematic Review×
×
All-cause Mortality
Data regarding the relationship between eGFR and all-cause mortality were available for eGFR as a categorical variable (14 studies representing 19,993 patients),17,18,26,28,29,38,39,43,47,52,56,63–65  and for multiple categories of eGFR (14 studies representing 586,516 patients),16–18,34,36,37,39,44,48–50,53,55  as a continuous variable (16 studies representing 30,734 patients),17,18,28,29,31,33,34,37,39,41–43,46,50,59,60  or as both a categorical and continuous variable (9 studies representing 20,182 patients).17,18,28,29,34,37,39,43,50 
An eGFR less than 60 ml·min·1.73 m−2 was associated with an increased risk of all-cause mortality in the short term (unadjusted RR 3.68, 95% CI 2.71–4.99; multivariable adjusted RR 2.98, 95% CI 1.95–4.96, I2 53.8%; fig. 2), and long-term (unadjusted RR 2.15, 95% CI 1.87–2.46; multivariable adjusted RR 1.61, 95% CI 1.38–1.87, I2 36.5%; fig. 2). Subgroup analysis (figs. 3 and 4) showed significant heterogeneity according to length of follow-up, with a reduced though still significant effect (RR 1.47, 95% CI 1.31–1.64; P = 0.03) from studies that followed outcomes for more than 5 yr.39,56,65  There was also a nonsignificant trend (P = 0.09) toward a stronger relationship in retrospective studies compared with that in prospective studies. Comparison between binary data calculated from the Cockcroft-Gault and the modified diet in renal disease study equations showed that the method used to estimate GFR had no significant effect on the RR that was observed.
Fig. 2.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
Fig. 2.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
×
Fig. 3.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 3.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Fig. 4.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 4.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Where data were present for categories of eGFR, the pooled results demonstrate an increased risk of death with a lower eGFR. Using meta-regression models, a significant inverse exponential association was observed between the category of eGFR and the risk of death during short-term follow-up (fig. 5). In relative terms, the estimated relative odds for death within 30 days of surgery associated with an eGFR of 60, 30, and 15 were 2.04, 4.17, and 6.00, respectively (with an eGFR of 90 ml·min·1.73 m−2 used as the reference).
Fig. 5.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
Fig. 5.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
×
Many studies reported the risk of short- and long-term mortality in terms of the excess hazard associated with a defined decline in eGFR, most commonly a reduction of 10 ml·min·1.73 m−2. These data assume a linear relationship between declining eGFR and outcome. This was evident in some cohorts,18,29,42  but not in all,17  and in several studies29,33,46,59  was not reported. Accepting this limitation, an analysis of continuous data showed that, on average, for every 10 ml·min·1.73 m−2 decline of eGFR, the RR of death within 30 days was 1.27 (95% CI 1.23–1.31, I2 0%; fig. 6).
Fig. 6.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
Fig. 6.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
×
Major Adverse Cardiovascular Events
No studies reported data on cardiovascular events within the first 30 days of surgery. Meta-analysis was conducted using data from three studies representing 8,388 patients reporting cardiovascular events during long-term follow-up.29,37,39  The pooled results demonstrate an unadjusted RR of 1.81 (95% CI 1.60–2.02) for an eGFR less than 60 ml·min·1.73 m−2. After multivariable adjustment the RR was 1.49 (95% CI 1.32–1.67, I2 0%; fig. 7).
Fig. 7.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
Fig. 7.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
×
Acute Kidney Injury
The risk of AKI was determined in eight studies (representing 17,679 patients) as a binary variable,25,30,40,47,51–53,61  and in five studies (representing 554,403 patients) by category of eGFR.16,32,49,53,66  Unadjusted data were not available for analysis. Definitions of AKI varied markedly among the studies that contributed data and included an increase in serum creatinine by 25% by the third postoperative day and/or commencement of dialysis30 ; postoperativeserum creatinine more than 150 µm,47  or dialysis requirement alone.61  The risk associated with an eGFR less than 60 ml·min·1.73 m−2 was reported by six studies, and the multivariable adjusted RR of AKI was 3.13 (95% CI 2.22–4.41, I2 44%; fig. 7). Subgroup analyses (fig. 8) showed the prognostic importance of a lower preoperative eGFR was lower in studies that excluded emergent surgery (adjusted RR 1.68, 95% CI 1.09–2.59; P value for heterogeneity = 0.003). There was also a trend for studies with more than 1,000 participants to report a weaker association than studies with a smaller sample size. The risk associated with a reduced eGFR was also attenuated after adjusting for heart failure, with every 10% increase in heart failure rate reducing the RR for AKI by 32% (95% CI 7–50%; P value for heterogeneity = 0.015). Postoperative renal outcomes according to eGFR category were available from four studies but meta-analysis was not performed due to limited data and significant heterogeneity among study results.
Fig. 8.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 8.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Comparison of the Ability of Creatinine and eGFR to Predict 30-Day Mortality
Nine studies,17,18,26,28,29,47,57,60,63  representing 11,571 patients, provided unadjusted binary data of either creatinine, eGFR, or both to predict 30-day mortality in the same cohort. Both measures of renal function were strong predictors of this outcome (RR for eGFR < 60 ml·min·1.73 m−2 3.68, 95% CI 2.71–4.99, I2 = 18.2%; RR for creatinine above threshold defined by study 2.80, 95% CI 1.57–5.00, I2 = 69.3%; fig. 9). However, comparison of the two markers showed no difference (χ2 = 0.40, P = 0.40).
Fig. 9.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
Fig. 9.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
×
The pooled area under the receiver operating characteristic curves (c-statistic) was derived from five studies (representing 20,368 patients).36,39,41,59,64  The c-statistic for creatinine was 0.65 (95% CI 0.59–0.71) and for eGFR 0.70 (95% CI 0.64–0.78); there was, however, considerable heterogeneity in the results for both measures. Comparison between the two markers showed no significant difference (χ2 = 1.13, P = 0.29).
Discussion
Renal dysfunction is a well-established marker of cardiovascular risk in many settings. In patients undergoing surgery, chronic kidney disease is recognized as an important determinant of a worse outcome and is included in several,4–6  though not all,7,8  risk scores. However, until recently, studies have almost exclusively used creatinine, usually used as a dichotomous variable so that it identifies patients with moderate to severe renal dysfunction. Likewise, the single previous systematic review on this area focuses on the excess perioperative risk associated chronic kidney disease, primarily identified using preoperative creatinine measurements.69 
The current study supports and adds to this existing literature. In particular, it addresses the prognostic importance of the preoperative eGFR. This reflects the general acceptance that, although the eGFR has limitations,70  it is a more accurate measure of renal function than creatinine and enables milder degrees of renal dysfunction to be detected.12,71  For these reasons, eGFR is now recommended as the optimal measurement of renal function and has in recent years been increasingly used in the perioperative setting. Our review consolidates this research. In particular, we have confirmed that eGFR predicts both short- (<30 days) and long-term mortality after, predominantly cardiac and vascular, surgery. The prognostic importance of the eGFR is apparent regardless of the method of calculation. We also confirm that there is a nonlinear relationship between preoperative eGFR category and early postoperative mortality but that even mild kidney dysfunction imparts a higher risk. The study also demonstrates that the relationship between renal function and postoperative prognosis persists despite correction for potential confounding variables and is consistent for several crucial outcomes and in several important subgroups. We found no evidence, however, that, when considered as a binary variable, eGFR was superior to creatinine in predicting 30-day mortality. Finally, but importantly, the review has identified significant gaps in the literature, in particular the lack of data exploring the relationship between more accurate and discriminatory measures of renal function and outcome in noncardiac and nonvascular surgery.
The strength of observed relationship between eGFR and adverse outcomes, in particular death, after surgery is very similar to those reported in large community-based studies15  and meta-analyses in nonsurgical populations.1,2  These have generally found an exponential relationship between renal dysfunction and all-cause mortality during long-term follow-up.1,2,15  Despite a tendency for patients with end-stage kidney disease or dialysis dependence to be less likely to undergo surgery or undergo it at a younger age,16  we found evidence of a similar relationship between categories of eGFR and early mortality. Thus, although some studies18,29,42  have reported a linear relationship between eGFR and outcome after surgery, it seems likely that the prognostic implications of a decline of 10 ml·min·1.73 m−2 in eGFR are greater in patients with more severe renal dysfunction.
A consistent relationship between eGFR and composite cardiac events was observed, but was limited by the small number of studies reporting this endpoint and restricted to people undergoing cardiac surgery. Similarly, preoperative chronic kidney disease was also found to be predictive of postoperative AKI. There was, however, significant heterogeneity across the pooled studies. In particular, studies that used preoperative eGFR to predict AKI after emergency surgery tended to report a stronger relationship, which may be explained by the presence of extrarenal factors such as sepsis and dehydration. The presence of heart failure and the sample size also influenced the magnitude of the observed effect. The heterogeneity identified may also reflect differences in the definition of AKI.
Potential Mechanisms
The mechanisms whereby renal dysfunction portends a worse prognosis after surgery are likely to be similar to those seen in the nonsurgical setting. These are complex and multiple, with kidney disease serving both as a marker and a mediator of an adverse outcome.72  Renal function declines with age and as a consequence of several other conditions that are associated with a worse perioperative outcome such as diabetes mellitus, dyslipidemia, and hypertension.73  Renal dysfunction both amplifies the adverse effects of these traditional risk factors and is associated with a greater prevalence of other abnormalities that may be detrimental to surgical outcome, such as anemia, abnormal calcium/phosphate homeostasis, and inflammation. A lower eGFR is also both a consequence and cause of reduced left ventricular systolic dysfunction and heart failure. In addition, kidney function is a sensitive indicator of vascular health and hemodynamic stability, both of which are important determinants of perioperative complications. There is also strong evidence that patients with renal disease frequently do not receive treatments that protect against vascular events74  and, even if they do, those with severe chronic kidney disease may derive less benefit.75,76 
Renal Function and Existing Risk Prediction Scores
In cardiac surgery there are limited data suggesting that the incorporation of more discriminatory and continuous measures of renal function can improve the precision of risk prediction scores.57,58  The European System for Cardiac Operative Risk Evaluation investigators have recently incorporated creatinine clearance by category to better estimate the risk of operative mortality in patients undergoing cardiac surgery.†††1  None of the currently recommended risk scores used in noncardiac surgery use the eGFR. Indeed some do not include any measure of renal function whereas others, such as the revised Cardiac Risk Index and the more recently described American College of Surgeons’ National Surgical Quality Improvement Program model,56  use creatinine but dichotomize the attributable risk.
For simplicity, it may remain preferable to use categories of eGFR to determine the postoperative risk. Our data suggest, however, that the thresholds used in traditional risk indices may need to be reviewed. For example, the revised Cardiac Risk Index attributes the excess risk associated with renal dysfunction when the creatinine is above 2 mg/dl (177 µm). For a 70-yr-old Caucasian male this equates to an eGFR of 33 ml·min·1.73 m−2 and for a black female of the same age an eGFR of 29 ml·min·1.73 m−2 In contrast, the current analyses suggest that, in keeping with other settings, the risk of early mortality after surgery begins to rise more steeply once the eGFR falls below 60 ml·min·1.73 m−2 (fig. 4). Thus, milder degrees of preoperative renal dysfunction, identified using the eGFR, may be associated with an important increase in postoperative risk, and reliance on a single cutoff level may limit the prognostic information available.
The Relative Prognostic Utility of Creatinine and eGFR
Several individual studies have assessed the relative ability of preoperative creatinine and eGFR levels to predict adverse postoperative outcomes.16,41,59,63  These have generally found the latter to be a better discriminator, though even in large data sets, the observed differences have sometimes been marginal.16  Likewise, in our meta-analysis of studies which provided unadjusted binary data we found that both measures are strong predictors. Although the risk ratio associated with an eGFR less than 60 ml·min·1.73 m−2 was higher than that observed for creatinine (dichotomized at various levels determined by the individual studies) no statistically significant difference was observed; though, as might be expected given the differences in the cutoff levels used, creatinine demonstrates greater statistical heterogeneity. Thus, although eGFR is regarded as a better indicator of renal function and is recommended for this purpose,71  the evidence that it will enhance postoperative risk prediction, when compared with creatinine, is limited.
Limitations of the Current Study
A key limitation of this review was the poor representation of noncardiac surgery in the literature. In addition, the remaining data relate almost exclusively to vascular surgery. Therefore, current evidence focuses heavily on selected populations with a high prevalence of overt atherosclerotic disease and a high cardiovascular risk profile, undergoing high-risk surgery. Further research is essential to better understand the importance of kidney function as a marker of risk in more diverse surgical populations77  and what role, if any, it plays in pathogenesis of postoperative events.
There were also limitations in terms of the outcome data that were available, with the majority of studies reporting only all-cause mortality. Few data were, therefore, available to explore the relationship between eGFR and composite and specific cardiac and renal outcomes. Likewise, we did not address the relationship between eGFR and other important, though inconsistently reported, outcomes such as postoperative atrial fibrillation, length of stay in intensive care, and duration of hospitalization. Another limitation is the retrospective nature of many studies, though the objective nature of the measurements and outcomes should reduce any tendency toward bias. In addition, significant heterogeneity was detected in the analyses measuring long-term all-cause mortality and AKI as outcomes. Potential causes have been sought and, although the predictive value remains significant even with adjustment, it remains possible that other confounders may not have been identified.
Clinical Implications
The current data support the use of eGFR as an indicator of the risk of postoperative complications, particularly after cardiac or vascular surgery. The strength of the relationship, its consistency among multiple subgroups, and the graded relationship support the importance of a declining eGFR as a risk factor for adverse postoperative outcome.
The results also suggest that reliance on dichotomous measures of renal function, as widely used currently, will considerably reduce the amount of prognostic information obtained. With better risk assessment there is the prospect of improved surgical outcomes. Potential interventions might be to limit the use of perioperative nephrotoxins and/or increase the intensity of postoperative observation in patients identified as being at high risk. In addition, because patients with kidney dysfunction are at increased risk for adverse cardiovascular outcomes after major surgery they may benefit from aggressive treatment of atherosclerotic risk factors and potentially the targeted use of therapeutic interventions such as statin therapy.78  Similarly, interventions such as N-acetyl cysteine79  might potentially be used to prevent renal injury.
In conclusion, we have found that kidney function, defined using eGFR, predicts cardiovascular events and AKI after, predominantly cardiac and vascular, surgery and, in particular, exhibits a powerful inverse nonlinear relationship with all-cause mortality in this setting.
The authors acknowledge Laurent Billot, M.Sc. (Director, Statistics and Data Management Division, The George Institute for Global Health, Sydney, Australia), for invaluable statistical assistance.
†††European System for Cardiac Operative Risk Evaluation. Available at: http://www.euroscore.org/calc.html. Accessed November 11, 2011.
European System for Cardiac Operative Risk Evaluation. Available at: http://www.euroscore.org/calc.html. Accessed November 11, 2011.×
References
Tonelli, M, Wiebe, N, Culleton, B, House, A, Rabbat, C, Fok, M, McAlister, F, Garg, AX Chronic kidney disease and mortality risk: A systematic review.. J Am Soc Nephrol. (2006). 17 2034–47 [Article] [PubMed]
Matsushita, K, van der Velde, M, Astor, BC, Woodward, M, Levey, AS, de Jong, PE, Coresh, J, Gansevoort, RT Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: A collaborative meta-analysis.. Lancet. (2010). 375 2073–81 [Article] [PubMed]
van der Velde, M, Matsushita, K, Coresh, J, Astor, BC, Woodward, M, Levey, A, de Jong, P, Gansevoort, RT, Levey, AS, de Jong, PE, El-Nahas, M, Eckardt, KU, Kasiske, BL, Ninomiya, T, Chalmers, J, Macmahon, S, Tonelli, M, Hemmelgarn, B, Sacks, F, Curhan, G, Collins, AJ, Li, S, Chen, SC, Hawaii Cohort, KP, Lee, BJ, Ishani, A, Neaton, J, Svendsen, K, Mann, JF, Yusuf, S, Teo, KK, Gao, P, Nelson, RG, Knowler, WC, Bilo, HJ, Joosten, H, Kleefstra, N, Groenier, KH, Auguste, P, Veldhuis, K, Wang, Y, Camarata, L, Thomas, B, Manley, T Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts.. Kidney Int. (2011). 79 1341–52 [Article] [PubMed]
Nashef, SA, Roques, F, Michel, P, Gauducheau, E, Lemeshow, S, Salamon, R European system for cardiac operative risk evaluation (EuroSCORE).. Eur J Cardiothorac Surg. (1999). 16 9–13 [Article] [PubMed]
Lee, TH, Marcantonio, ER, Mangione, CM, Thomas, EJ, Polanczyk, CA, Cook, EF, Sugarbaker, DJ, Donaldson, MC, Poss, R, Ho, KK, Ludwig, LE, Pedan, A, Goldman, L Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery.. Circulation. (1999). 100 1043–9 [Article] [PubMed]
Gupta, PK, Gupta, H, Sundaram, A, Kaushik, M, Fang, X, Miller, WJ, Esterbrooks, DJ, Hunter, CB, Pipinos, II, Johanning, JM, Lynch, TG, Forse, RA, Mohiuddin, SM, Mooss, AN Development and validation of a risk calculator for prediction of cardiac risk after surgery.. Circulation. (2011). 124 381–7 [Article] [PubMed]
Detsky, AS, Abrams, HB, Forbath, N, Scott, JG, Hilliard, JR Cardiac assessment for patients undergoing noncardiac surgery. A multifactorial clinical risk index.. Arch Intern Med. (1986). 146 2131–4 [Article] [PubMed]
Goldman, L, Caldera, DL, Nussbaum, SR, Southwick, FS, Krogstad, D, Murray, B, Burke, DS, O’Malley, TA, Goroll, AH, Caplan, CH, Nolan, J, Carabello, B, Slater, EE Multifactorial index of cardiac risk in noncardiac surgical procedures.. N Engl J Med. (1977). 297 845–50 [Article] [PubMed]
Fleisher, LA, Beckman, JA, Brown, KA, Calkins, H, Chaikof, EL, Fleischmann, KE, Freeman, WK, Froehlich, JB, Kasper, EK, Kersten, JR, Riegel, B, Robb, JF, Smith, SCJr, Jacobs, AK, Adams, CD, Anderson, JL, Antman, EM, Buller, CE, Creager, MA, Ettinger, SM, Faxon, DP, Fuster, V, Halperin, JL, Hiratzka, LF, Hunt, SA, Lytle, BW, Nishimura, R, Ornato, JP, Page, RL, Tarkington, LG, Yancy, CW ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery) developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery.. J Am Coll Cardiol. (2007). 50 e159–241 [Article] [PubMed]
Duncan, L, Heathcote, J, Djurdjev, O, Levin, A Screening for renal disease using serum creatinine: Who are we missing?. Nephrol Dial Transplant. (2001). 16 1042–6 [Article] [PubMed]
Swedko, PJ, Clark, HD, Paramsothy, K, Akbari, A Serum creatinine is an inadequate screening test for renal failure in elderly patients.. Arch Intern Med. (2003). 163 356–60 [Article] [PubMed]
National Kidney Foundation, K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification.. Am J Kidney Dis. (2002). 39 S1–266 [PubMed]
Levey, AS, Bosch, JP, Lewis, JB, Greene, T, Rogers, N, Roth, D A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group.. Ann Intern Med. (1999). 130 461–70 [Article] [PubMed]
Levey, AS, Stevens, LA, Schmid, CH, Zhang, YL, Castro, AF 3rd, Feldman, HI, Kusek, JW, Eggers, P, Van Lente, F, Greene, T, Coresh, J A new equation to estimate glomerular filtration rate.. Ann Int Med. (2009). 150 604–12 [Article] [PubMed]
Go, AS, Chertow, GM, Fan, D, McCulloch, CE, Hsu, CY Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.. N Engl J Med. (2004). 351 1296–305 [Article] [PubMed]
Cooper, WA, O’Brien, SM, Thourani, VH, Guyton, RA, Bridges, CR, Szczech, LA, Petersen, R, Peterson, ED Impact of renal dysfunction on outcomes of coronary artery bypass surgery: Results from the Society of Thoracic Surgeons National Adult Cardiac Database.. Circulation. (2006). 113 1063–70 [Article] [PubMed]
Hillis, GS, Croal, BL, Buchan, KG, El-Shafei, H, Gibson, G, Jeffrey, RR, Millar, CG, Prescott, GJ, Cuthbertson, BH Renal function and outcome from coronary artery bypass grafting: Impact on mortality after a 2.3-year follow-up.. Circulation. (2006). 113 1056–62 [Article] [PubMed]
Gibson, PH, Croal, BL, Cuthbertson, BH, Chiwara, M, Scott, AE, Buchan, KG, El-Shafei, H, Gibson, G, Jeffrey, RR, Hillis, GS The relationship between renal function and outcome from heart valve surgery.. Am Heart J. (2008). 156 893–9 [Article] [PubMed]
Cockcroft, DW, Gault, MH Prediction of creatinine clearance from serum creatinine.. Nephron. (1976). 16 31–41 [Article] [PubMed]
Brosius, FC 3rd, Hostetter, TH, Kelepouris, E, Mitsnefes, MM, Moe, SM, Moore, MA, Pennathur, S, Smith, GL, Wilson, PW Detection of chronic kidney disease in patients with or at increased risk of cardiovascular disease: A science advisory from the American Heart Association Kidney And Cardiovascular Disease Council; the Councils on High Blood Pressure Research, Cardiovascular Disease in the Young, and Epidemiology and Prevention; and the Quality of Care and Outcomes Research Interdisciplinary Working Group: Developed in collaboration with the National Kidney Foundation.. Circulation. (2006). 114 1083–7 [PubMed]
Stroup, DF, Berlin, JA, Morton, SC, Olkin, I, Williamson, GD, Rennie, D, Moher, D, Becker, BJ, Sipe, TA, Thacker, SB Meta-analysis of observational studies in epidemiology: A proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.. JAMA. (2000). 283 2008–12 [Article] [PubMed]
Hayden, JADC, Cote, PDCP, Bombardier, CMD Evaluation of the quality of prognosis studies in systematic reviews.. Ann Int Med. (2006). 144 427–37 [Article] [PubMed]
Higgins, JP, Thompson, SG, Deeks, JJ, Altman, DG Measuring inconsistency in meta-analyses.. BMJ. (2003). 327 557–60 [Article] [PubMed]
Song, F, Sheldon, TA, Sutton, AJ, Abrams, KR, Jones, DR Methods for exploring heterogeneity in meta-analysis.. Eval Health Prof. (2001). 24 126–51 [Article] [PubMed]
Aveline, C, Leroux, A, Vautier, P, Cognet, F, Le Hetet, H, Bonnet, F [Risk factors for renal dysfunction after total hip arthroplasty].. Ann Fr Anesth Reanim. (2009). 28 728–34 [Article] [PubMed]
Azizzadeh, A, Sanchez, LA, Miller, CC 3rd, Marine, L, Rubin, BG, Safi, HJ, Huynh, TT, Parodi, JC, Sicard, GA Glomerular filtration rate is a predictor of mortality after endovascular abdominal aortic aneurysm repair.. J Vasc Surg. (2006). 43 14–8 [Article] [PubMed]
Bakken, AM, Protack, CD, Saad, WE, Hart, JP, Rhodes, JM, Waldman, DL, Davies, MG Impact of chronic kidney disease on outcomes of superficial femoral artery endoluminal interventions.. Ann Vasc Surg. (2009). 23 560–8 [Article] [PubMed]
Browner, WS, Li, J, Mangano, DT In-hospital and long-term mortality in male veterans following noncardiac surgery. The Study of Perioperative Ischemia Research Group.. JAMA. (1992). 268 228–32 [Article] [PubMed]
Chonchol, MB, Aboyans, V, Lacroix, P, Smits, G, Berl, T, Laskar, M Long-term outcomes after coronary artery bypass grafting: Preoperative kidney function is prognostic.. J Thorac Cardiovasc Surg. (2007). 134 683–9 [Article] [PubMed]
Del Duca, D, Iqbal, S, Rahme, E, Goldberg, P, de Varennes, B Renal failure after cardiac surgery: Timing of cardiac catheterization and other perioperative risk factors.. Ann Thorac Surg. (2007). 84 1264–71 [Article] [PubMed]
Huynh, TT, van Eps, RG, Miller, CC 3rd, Villa, MA, Estrera, AL, Azizzadeh, A, Porat, EE, Goodrick, JS, Safi, HJ Glomerular filtration rate is superior to serum creatinine for prediction of mortality after thoracoabdominal aortic surgery.. J Vasc Surg. (2005). 42 206–12 [Article] [PubMed]
Thakar, CV, Worley, S, Arrigain, S, Yared, JP, Paganini, EP Influence of renal dysfunction on mortality after cardiac surgery: Modifying effect of preoperative renal function.. Kidney Int. (2005). 67 1112–9 [Article] [PubMed]
Estrera, AL, Miller, CC3rd, Madisetty, J, Bourgeois, S, Azizzadeh, A, Villa, MA, Safi, HJ Ascending and transverse aortic arch repair: The impact of glomerular filtration rate on mortality.. Ann Surg. (2008). 247 524–9 [Article] [PubMed]
Foot, CL, Chinthamuneedi, M, Fraser, JF, Smith, SE, Fayers, T, Tesar, P, Mullany, DV The association between preoperative eGFR and outcomes in cardiac surgical patients.. Crit Care Resusc. (2009). 11 184–90 [PubMed]
Haddad, F, Greenberg, RK, Walker, E, Nally, J, O’Neill, S, Kolin, G, Lyden, SP, Clair, D, Sarac, T, Ouriel, K Fenestrated endovascular grafting: The renal side of the story.. J Vasc Surg. (2005). 41 181–90 [Article] [PubMed]
Holzmann, MJ, Ahnve, S, Hammar, N, Jörgensen, L, Klerdal, K, Pehrsson, K, Ivert, T Creatinine clearance and risk of early mortality in patients undergoing coronary artery bypass grafting.. J Thorac Cardiovasc Surg. (2005). 130 746–52 [Article] [PubMed]
Holzmann, MJ, Hammar, N, Ahnve, S, Nordqvist, T, Pehrsson, K, Ivert, T Renal insufficiency and long-term mortality and incidence of myocardial infarction in patients undergoing coronary artery bypass grafting.. Eur Heart J. (2007). 28 865–71 [Article] [PubMed]
Ibáñez, J, Riera, M, Saez de Ibarra, JI, Carrillo, A, Fernández, R, Herrero, J, Fiol, M, Bonnin, O Effect of preoperative mild renal dysfunction on mortality and morbidity following valve cardiac surgery.. Interact Cardiovasc Thorac Surg. (2007). 6 748–52 [Article] [PubMed]
Kangasniemi, OP, Mahar, MA, Rasinaho, E, Satomaa, A, Tiozzo, V, Lepojärvi, M, Biancari, F Impact of estimated glomerular filtration rate on the 15-year outcome after coronary artery bypass surgery.. Eur J Cardiothorac Surg. (2008). 33 198–202 [Article] [PubMed]
Karkouti, K, Wijeysundera, DN, Yau, TM, Callum, JL, Cheng, DC, Crowther, M, Dupuis, JY, Fremes, SE, Kent, B, Laflamme, C, Lamy, A, Legare, JF, Mazer, CD, McCluskey, SA, Rubens, FD, Sawchuk, C, Beattie, WS Acute kidney injury after cardiac surgery: Focus on modifiable risk factors.. Circulation. (2009). 119 495–502 [Article] [PubMed]
Kertai, MD, Boersma, E, Bax, JJ, van den Meiracker, AH, van Urk, H, Roelandt, JR, Poldermans, D Comparison between serum creatinine and creatinine clearance for the prediction of postoperative mortality in patients undergoing major vascular surgery.. Clin Nephrol. (2003). 59 17–23 [Article] [PubMed]
Ledoux, D, Monchi, M, Chapelle, JP, Damas, P Cystatin C blood level as a risk factor for death after heart surgery.. Eur Heart J. (2007). 28 1848–53 [Article] [PubMed]
Loef, BG, Epema, AH, Smilde, TD, Henning, RH, Ebels, T, Navis, G, Stegeman, CA Immediate postoperative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival.. J Am Soc Nephrol. (2005). 16 195–200 [Article] [PubMed]
Lok, CE, Austin, PC, Wang, H, Tu, JV Impact of renal insufficiency on short- and long-term outcomes after cardiac surgery.. Am Heart J. (2004). 148 430–8 [Article] [PubMed]
Marrocco-Trischitta, MM, Melissano, G, Kahlberg, A, Calori, G, Setacci, F, Chiesa, R Chronic kidney disease classification stratifies mortality risk after elective stent graft repair of the thoracic aorta.. J Vasc Surg. (2009). 49 296–301 [Article] [PubMed]
Miller, CC3rd, Grimm, JC, Estrera, AL, Azizzadeh, A, Coogan, SM, Walkes, JC, Safi, HJ Postoperative renal function preservation with nonischemic femoral arterial cannulation for thoracoabdominal aortic repair.. J Vasc Surg. (2010). 51 38–42 [Article] [PubMed]
Noyez, L, Plesiewicz, I, Verheugt, FW Estimated creatinine clearance instead of plasma creatinine level as prognostic test for postoperative renal function in patients undergoing coronary artery bypass surgery.. Eur J Cardiothorac Surg. (2006). 29 461–5 [Article] [PubMed]
O’Hare, AFJ, Reiber, G, Rodriguez, R, Daley, J, Khuri, S, Henderson, W, Johansen, K Postoperative mortality after nontraumatic lower extremity amputation in patients with renal insufficiency.. J Am Soc Nephrol. (2004). 15 427–34 [Article] [PubMed]
O’Hare, AM, Feinglass, J, Sidawy, AN, Bacchetti, P, Rodriguez, RA, Daley, J, Khuri, S, Henderson, WG, Johansen, KL Impact of renal insufficiency on short-term morbidity and mortality after lower extremity revascularization: Data from the Department of Veterans Affairs’ National Surgical Quality Improvement Program.. J Am Soc Nephrol. (2003). 14 1287–95 [Article] [PubMed]
Owens, CD, Ho, KJ, Kim, S, Schanzer, A, Lin, J, Matros, E, Belkin, M, Conte, MS Refinement of survival prediction in patients undergoing lower extremity bypass surgery: Stratification by chronic kidney disease classification.. J Vasc Surg. (2007). 45 944–52 [Article] [PubMed]
Palomba, H, de Castro, I, Neto, AL, Lage, S, Yu, L Acute kidney injury prediction following elective cardiac surgery: AKICS Score.. Kidney Int. (2007). 72 624–31 [Article] [PubMed]
Pearce, JD, Edwards, MS, Stafford, JM, Deonanan, JK, Davis, RP, Corriere, MA, Crutchley, TA, Hansen, KJ Open repair of aortic aneurysms involving the renal vessels.. Ann Vasc Surg. (2007). 21 676–86 [Article] [PubMed]
Perez-Valdivieso, JR, Monedero, P, Vives, M, Garcia-Fernandez, N, Bes-Rastrollo, M Cardiac-surgery associated acute kidney injury requiring renal replacement therapy. A Spanish retrospective case-cohort study.. BMC Nephrol. (2009). 10 27 [Article] [PubMed]
Powell, RJ, Roddy, SP, Meier, GH, Gusberg, RJ, Conte, MS, Sumpio, BE Effect of renal insufficiency on outcome following infrarenal aortic surgery.. Am J Surg. (1997). 174 126–30 [Article] [PubMed]
Sidawy, AN, Aidinian, G, Johnson, ON3rd, White, PW, DeZee, KJ, Henderson, WG Effect of chronic renal insufficiency on outcomes of carotid endarterectomy.. J Vasc Surg. (2008). 48 1423–30 [Article] [PubMed]
van de Wal, RM, van Brussel, BL, Voors, AA, Smilde, TD, Kelder, JC, van Swieten, HA, van Gilst, WH, van Veldhuisen, DJ, Plokker, HW Mild preoperative renal dysfunction as a predictor of long-term clinical outcome after coronary bypass surgery.. J Thorac Cardiovasc Surg. (2005). 129 330–5 [Article] [PubMed]
van Gameren, M, Klieverik, LM, Struijs, A, Venema, AC, Kappetein, AP, Bogers, AJ, Takkenberg, JJ Impact of the definition of renal dysfunction on EuroSCORE performance.. J Cardiovasc Surg (Torino). (2009). 50 703–9 [PubMed]
Walter, J, Mortasawi, A, Arnrich, B, Albert, A, Frerichs, I, Rosendahl, U, Ennker, J Creatinine clearance versus serum creatinine as a risk factor in cardiac surgery.. BMC Surg. (2003). 3 4 [Article] [PubMed]
Wang, F, Dupuis, JY, Nathan, H, Williams, K An analysis of the association between preoperative renal dysfunction and outcome in cardiac surgery: Estimated creatinine clearance or plasma creatinine level as measures of renal function.. Chest. (2003). 124 1852–62 [Article] [PubMed]
Welten, GM, Schouten, O, Chonchol, M, Hoeks, SE, Feringa, HH, Bax, JJ, Dunkelgrün, M, van Gestel, YR, van Domburg, RT, Poldermans, D Temporary worsening of renal function after aortic surgery is associated with higher long-term mortality.. Am J Kidney Dis. (2007). 50 219–28 [Article] [PubMed]
Wijeysundera, DN, Karkouti, K, Beattie, WS, Rao, V, Ivanov, J Improving the identification of patients at risk of postoperative renal failure after cardiac surgery.. Anesthesiology. (2006). 104 65–72 [Article] [PubMed]
Yu, HY, Li, JY, Sun, S, Hung, KY, Wang, JL, Chen, YS, Wang, SS, Lin, FY Late dialysis rate for coronary artery bypass grafting patients with moderate-to-severe renal impairment: Comparison between off-pump and conventional method.. Eur J Cardiothorac Surg. (2008). 33 364–9 [Article] [PubMed]
Zakeri, R, Freemantle, N, Barnett, V, Lipkin, GW, Bonser, RS, Graham, TR, Rooney, SJ, Wilson, IC, Cramb, R, Keogh, BE, Pagano, D Relation between mild renal dysfunction and outcomes after coronary artery bypass grafting.. Circulation. (2005). 1129 Suppl I270–5 [PubMed]
Lin, Y, Zheng, Z, Li, Y, Yuan, X, Hou, J, Zhang, S, Fan, H, Wang, Y, Li, W, Hu, S Impact of renal dysfunction on long-term survival after isolated coronary artery bypass surgery.. Ann Thorac Surg. (2009). 87 1079–84 [Article] [PubMed]
Welten, GM, Chonchol, M, Hoeks, SE, Schouten, O, Dunkelgrün, M, van Gestel, YR, Goei, D, Bax, JJ, van Domburg, RT, Poldermans, D Statin therapy is associated with improved outcomes in vascular surgery patients with renal impairment.. Am Heart J. (2007). 154 954–61 [Article] [PubMed]
Wijeysundera, DN, Karkouti, K, Dupuis, JY, Rao, V, Chan, CT, Granton, JT, Beattie, WS Derivation and validation of a simplified predictive index for renal replacement therapy after cardiac surgery.. JAMA. (2007). 297 1801–9 [Article] [PubMed]
Huang, TM, Wu, VC, Young, GH, Lin, YF, Shiao, CC, Wu, PC, Li, WY, Yu, HY, Hu, FC, Lin, JW, Chen, YS, Lin, YH, Wang, SS, Hsu, RB, Chang, FC, Chou, NK, Chu, TS, Yeh, YC, Tsai, PR, Huang, JW, Lin, SL, Chen, YM, Ko, WJ, Wu, KD Preoperative proteinuria predicts adverse renal outcomes after coronary artery bypass grafting.. J Am Soc Nephrol. (2011). 22 56–63 [Article]
Grubb, A, Nyman, U, Björk, J, Lindström, V, Rippe, B, Sterner, G, Christensson, A Simple cystatin C-based prediction equations for glomerular filtration rate compared with the modification of diet in renal disease prediction equation for adults and the Schwartz and the Counahan-Barratt prediction equations for children.. Clin Chem. (2005). 51 1420–31 [Article] [PubMed]
Mathew, A, Devereaux, PJ, O’Hare, A, Tonelli, M, Thiessen-Philbrook, H, Nevis, IF, Iansavichus, AV, Garg, AX Chronic kidney disease and postoperative mortality: A systematic review and meta-analysis.. Kidney Int. (2008). 73 1069–81 [Article] [PubMed]
Kalantar-Zadeh, K, Amin, AN Toward more accurate detection and risk stratification of chronic kidney disease.. JAMA. (2012). 307 1976–7 [Article] [PubMed]
Levey, AS, Coresh, J, Balk, E, Kausz, AT, Levin, A, Steffes, MW, Hogg, RJ, Perrone, RD, Lau, J, Eknoyan, G National Kidney Foundation practice guidelines for chronic kidney disease: Evaluation, classification, and stratification.. Ann Int Med. (2003). 139 137–47 [Article] [PubMed]
Hillis, GS, Cuthbertson, BH, Croal, BL Renal function, revascularization and risk.. Eur Heart J. (2007). 28 782–4 [Article] [PubMed]
Levey, AS, Coresh, J Chronic kidney disease.. Lancet. (2012). 379 165–80 [Article] [PubMed]
Tonelli, M, Bohm, C, Pandeya, S, Gill, J, Levin, A, Kiberd, BA Cardiac risk factors and the use of cardioprotective medications in patients with chronic renal insufficiency.. Am J Kidney Dis. (2001). 37 484–9 [Article] [PubMed]
Palmer, SC, Craig, JC, Navaneethan, SD, Tonelli, M, Pellegrini, F, Strippoli, GF Benefits and harms of statin therapy for persons with chronic kidney disease: A systematic review and meta-analysis.. Ann Intern Med. (2012). 157 263–75 [Article] [PubMed]
Palmer, SC, Di Micco, L, Razavian, M, Craig, JC, Perkovic, V, Pellegrini, F, Copetti, M, Graziano, G, Tognoni, G, Jardine, M, Webster, A, Nicolucci, A, Zoungas, S, Strippoli, GF Effects of antiplatelet therapy on mortality and cardiovascular and bleeding outcomes in persons with chronic kidney disease: A systematic review and meta-analysis.. Ann Intern Med. (2012). 156 445–59 [Article] [PubMed]
Schrier, RW Role of diminished renal function in cardiovascular mortality: Marker or pathogenetic factor?. J Am Coll Cardiol. (2006). 47 1–8 [Article] [PubMed]
Le Manach, Y, Ibanez Esteves, C, Bertrand, M, Goarin, JP, Fléron, MH, Coriat, P, Koskas, F, Riou, B, Landais, P Impact of preoperative statin therapy on adverse postoperative outcomes in patients undergoing vascular surgery.. Anesthesiology. (2011). 114 98–104 [Article] [PubMed]
Wijeysundera, DN, Beattie, WS, Rao, V, Granton, JT, Chan, CT N-acetylcysteine for preventing acute kidney injury in cardiac surgery patients with pre-existing moderate renal insufficiency.. Can J Anaesth. (2007). 54 872–81 [Article] [PubMed]
Appendix 1. Postsurgical Renal Failure Systematic Review: Quality Assessment Tool
Observational Studies
  1. How did the study define renal injury/damage or failure, and in what time frame?

    • a.Yes
    • b.No
  2. Did the study define a particular adverse outcome, and in what time frame? E.g., AMI at 30 days.

    • a.Yes
    • b.No
  3. Did the study define a specific endpoint, e.g., mortality, dialysis dependence?

    • a.Yes
    • b.No
  4. Which fluid was used to sample biomarkers?

    • a.Urine
    • b.Serum
    • c.Both
  5. Was there a series of biomarkers taken before surgery?

    • a.Yes
    • b.No
  6. Was this in comparison to a standard measurement (i.e., creatinine)?

    • a.Yes
    • b.No
  7. Did the study include people with existing renal disease?

    • a.Yes
    • b.No
  8. 8. Was this accounted for/adjusted for?

    • a.Yes
    • b.No
  9. Were other preexisting comorbidities, e.g., cardiovascular disease accounted for?

    • a.Yes
    • b.No
  10. 10. Did the study draw comparison with a current predictive clinical model?

    • a.Yes
    • b.No
Fig. 1.
Flow diagram of study selection.
Flow diagram of study selection.
Fig. 1.
Flow diagram of study selection.
×
Fig. 2.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
Fig. 2.
Risk of death for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2 compared with ≥60 ml·min·1.73 m−2 at both short-term and long-term follow-up.
×
Fig. 3.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 3.
Relative risk of all-cause mortality within 30 days of surgery in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Fig. 4.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 4.
Relative risk of long-term all-cause mortality in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Fig. 5.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
Fig. 5.
All-cause mortality within 30 days of surgery. Events categorized according to preoperative estimated glomerular filtration rate (eGRF). A total of eight cohort studies are represented. Risks are expressed as proportions (e.g., 0.10 = 10% increased risk) and the area of the circle is proportional to the sample size. The center line models estimated risk from adjusted data of preoperative estimated glomerular filtration rate categorized into Chronic Kidney Disease by stage, with dotted lines representing 95% confidence intervals. Regression equation logit(p) = −1.8701 to 0.0238 *eGFR.
×
Fig. 6.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
Fig. 6.
Risk of death at short-term follow-up with stepwise reduction in creatinine clearance by 10 ml·min·1.73 m−2.
×
Fig. 7.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
Fig. 7.
Risk of acute renal injury or adverse cardiovascular event for patients with estimated glomerular filtration rate <60 ml·min·1.73 m−2.
×
Fig. 8.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
Fig. 8.
Relative risk of postoperative acute kidney injury in patients with baseline estimated glomerular filtration rate (eGFR) <60 ml·min·1.73 m−2 according to selected subgroups. MDRD = Modified Diet in Renal Disease Study.
×
Fig. 9.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
Fig. 9.
Comparison of creatinine and estimated glomerular filtration rate (eGFR) in prediction of death within 30 days of surgery.
×
Table 1.
Characteristics of Included Studies
Characteristics of Included Studies×
Characteristics of Included Studies
Table 1.
Characteristics of Included Studies
Characteristics of Included Studies×
×
Appendix 1, Table 2.
Methodological Quality of Articles in Systematic Review
Methodological Quality of Articles in Systematic Review×
Methodological Quality of Articles in Systematic Review
Appendix 1, Table 2.
Methodological Quality of Articles in Systematic Review
Methodological Quality of Articles in Systematic Review×
×