Neutrophil Gelatinase-Associated Lipocalin as Early Predictor of Acute Kidney Injury After Cardiac Surgery in Adults With Chronic Kidney Failure Andrea Perrotti, MD, Guillaume Miltgen, MD, Albin Chevet-Noel, MD, Camille Durst, MD, Dewi Vernerey, MS, Karine Bardonnet, MD, Siamak Davani, MD, PhD, and Sidney Chocron, MD, PhD Departments of Thoracic and Cardio-Vascular Surgery, Biochemistry, and Pharmacology, and Methodology and Quality of Life in Oncology unit, University Hospital Jean Minjoz, Besanc¸on, France
Background. To assess the utility of neutrophil gelatinase-associated lipocalin (NGAL) as an early marker of acute kidney injury (AKI) occurring after cardiac surgery in patients with prior chronic kidney failure. Methods. Patients with preoperative creatinine clearance 60 mL minL1 1.73 mL2 or less according to the Cockcroft-Gault formula and scheduled to undergo cardiac surgery were eligible for inclusion. The AKI was defined as an increase in plasma creatinine greater than 50% over preoperative values. Threshold values of NGAL predictive of AKI were determined using receiver operating characteristic curve analysis, and predictive value of NGAL for AKI was evaluated by logistic regression. Results. Over a 1-year inclusion period, 166 patients were included. At 6 hours post-surgery, hypertension, occurrence of at least 1 postoperative complication, and
NGAL greater than 155ng/mL were shown to be independent predictors of AKI. NGAL greater than 155 ng/mL at 6 hours was associated with an odds ratio for risk of postoperative AKI of 7.1 [2.7 to 18]. On average, diagnosis of postoperative AKI was made 20 hours earlier using NGAL at 6 hours post-surgery as compared with a diagnosis based on a 50% increase in creatinine over baseline. The threshold for NGAL of 155 ng/mL at 6 hours had a sensitivity of 79% and specificity of 58% for the diagnosis of AKI. Conclusions. Earlier diagnosis of AKI post-surgery based on NGAL assessment makes it possible to initiate appropriate therapy at an earlier stage in this high-risk patient population.
A
in children [6] and adults [7, 8]. NGAL is released in response to ischemia of the proximal tubules of the kidney, and NGAL values early after surgery could predict the duration and severity of AKI. In addition, NGAL can independently predict deteriorating renal function and could therefore be useful even in the context of preexisting renal failure. A novel marker that could identify AKI after surgery at an earlier stage than currently possible with monitoring of creatinine plasma levels would make it possible to initiate appropriate therapy more quickly in these patients. To the best of our knowledge, no study to date has investigated the utility of early NGAL assessment to predict the occurrence of AKI post-cardiac surgery in patients suffering from chronic renal failure prior to surgery. Therefore, we aimed to assess the predictive ability of plasma NGAL levels to identify deteriorations in renal function after cardiac surgery in patients with preexisting renal failure.
cute kidney injury (AKI) is one of the most frequent complications after cardiac surgery [1]. The incidence of postoperative AKI in cardiac surgery ranges from 5% to 50% depending on preoperative renal status, and the criteria used to define AKI [2, 3]. Regardless of whether renal function was normal before surgery or whether there is a deterioration of previous chronic renal failure, AKI after surgery is associated with increased morbidity and perioperative mortality in-hospital [4]. In routine practice, renal function is monitored by regular assessment of plasma creatinine levels. However, there is a time delay between the onset of renal impairment and the resulting telltale increase in blood creatinine levels. Thus, by the time creatinine levels are found to be elevated, renal function has already been impaired for several days [5]. Recent studies have underlined the promising properties of neutrophil gelatinase-associated lipocalin (NGAL), a protein normally expressed at very low plasma levels, as an early predictor of AKI after cardiac surgery
Accepted for publication Oct 3, 2014. Address correspondence to Dr Chocron, Department of Thoracic and Cardio-Vascular Surgery, EA3920, University Hospital Jean Minjoz, Blvd Fleming, 25000 Besanc¸on, France; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
Material and Methods We performed this prospective study in a single university teaching hospital that is a referral center for a population basin of approximately 1.2 million inhabitants 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.10.011
2
PERROTTI ET AL NGAL AND AKI
(region of Franche-Comte in eastern France). This study was approved by the local ethics committee (CPP EST II, registered under the number 10/544) and was registered with the ClinicalTrials.gov database under the number NCT01227122. All participants provided written informed consent. The inclusion criteria for the study were adult (age >18 years) patients with preoperative creatinine clearance 60 mL min1 1.73 m2 or less according to the Cockcroft Gault formula scheduled to undergo cardiac surgery, and who provided written informed consent. Exclusion criteria were the following: emergency surgery; injection of contrast medium in the 3 days preceding surgery or intravenous iodinated contrast medium injection within the first 24 hours post-surgery; patients with preoperative blood creatinine levels greater than 300 mmol/L; patients undergoing hemodialysis; patients with ongoing inflammation, infection or cancer; pregnancy; patient refusal; adults under legal protection.
Neutrophil Gelatinase-Associated Lipocalin Assessments NGAL was assessed from blood samples taken at 5 time points: namely at induction of general anesthesia (baseline NGAL); at 15 minutes after interruption of extracorporeal circulation; and at 6, 12, and 24 hours after the end of surgery. Blood samples were centrifuged at 2,250 250 g for 13 2 minutes at room temperature (18 to 25 Celsius), aliquoted in 0.5 mL tubes (Eppendorf, Le Pecq, France), and frozen at 20 C for later analysis. Tests were performed using a specific kit, namely Triage NGAL test (Biosite Inverness Medical, San Diego, CA). The measurable range is from 15 to 1,300 ng/mL. Serum creatinine and creatinine clearance according to the Cockcroft-Gault formula were assessed at 6 and 12 hours after surgery, and then once daily until the third postoperative day.
Definition of Acute Kidney Injury Acute kidney injury was defined as an increase in blood creatinine levels of at least 50% over preoperative values. Patients were classified into the following 2 groups accordingly: group 1, patients with a postoperative increase in creatinine less than 50% over baseline (no-AKI group); and group 2, patients with a postoperative increase in creatinine ( 50%) over baseline (AKI group).
Correlations The correlation between baseline NGAL and baseline creatinine clearance value and to serum creatinine level was tested. In the AKI group, the correlation between the ratio of preoperative NGAL to NGAL at 6 hours and the severity of the subsequent AKI—defined as the ratio of the lowest postoperative creatinine clearance during follow-up (considered as nil for patients who had hemofiltration)—to preoperative creatinine clearance was also tested.
Statistical Analysis Quantitative data are described as mean standard deviation or median interquartile range for non-normally
Ann Thorac Surg 2015;-:-–-
distributed variables. Qualitative data are described as number (percentage). Preoperative, perioperative, and postoperative characteristics were compared using the Mann-Whitney and c2 or Fisher exact test, as appropriate. Correlations were evaluated by the Pearson correlation coefficient. We considered the strength of a correlation as weak for correlation coefficients ranging from 0.1 to 0.3, moderate for 0.3 to 0.6, and strong for correlation coefficients greater than 0.6. To identify a threshold value of NGAL predictive of AKI, receiver operating characteristic (ROC) curve analysis was used, and the sensitivity and specificity of the cutoff values were calculated. Independent predictors of postoperative AKI were identified by logistic regression. All variables with a p value less than 0.10 by univariate analysis were included in the model. All analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). A p value of less than 0.05 was considered statistically significant.
Results Between November 1, 2012 and November 27, 2013, 819 patients underwent cardiac surgery in our institution. Among the 213 patients who met inclusion criteria, 8 refused to participate and 39 had too many missing values to be analyzed. Therefore 166 patients were included in the study. Preoperative renal failure was moderate (creatinine clearance from 30 to 60 mL min1 1.73 m2) in most patients (149, 90%), and severe (creatinine clearance from 15 to 29 mL min1 1.73 m2) in 17 patients (10%). The baseline characteristics of the study population are shown in Table 1. There was no significant difference between groups at baseline except for hypertension and preoperative creatinine clearance. There was a borderline significant difference in the duration of cardiopulmonary bypass (CPB) between groups (p ¼ 0.05). In the AKI group, average time to 50% increase in creatinine was 25.6 21.7 hours. Although statistically significant (p < 0.01) the correlation between preoperative NGAL values and (1) preoperative creatinine clearance and (2) preoperative creatinine values was only moderate (r ¼ 0.34 and 0.42, respectively). In the AKI group, the ratio of preoperative NGAL to NGAL at 6 hours was not correlated with the severity of subsequent AKI (r ¼ 0.02, p ¼ 0.9). Table 2 details the postoperative complications in both groups. There was a higher overall rate of complications in the AKI group. There were 6 postoperative deaths (4%), 1 in the no AKI group from acute respiratory distress syndrome, and 5 in the AKI group; namely, 1 from acute respiratory distress syndrome, 3 from mesenteric ischemia, and 1 from heart failure. The threshold NGAL values at each time point, as determined by ROC curve analysis, are indicated in Figure 1. At 15 minutes, sensitivity was 54% and specificity was 71%; at 6 hours, these values were 79% and 58%, respectively. By multivariate analysis (Table 3) at 6 hours a history of hypertension (odds ratio [OR] ¼ 3.2 [1.2 to 8.9]),
Ann Thorac Surg 2015;-:-–-
3
PERROTTI ET AL NGAL AND AKI
Table 1. Baseline Characteristics of the Study Population Patient Characteristics Preoperative Age (years) Male sex (%) BMI (kg/m2) Diabetes (%) Dyslipidemia (%) History of hypertension (%) Family history of CAD (%) PAD (%) LVEF Diuretics presurgery (%) Mean preoperative creatinine clearance Perioperative data CABG on-pump (%) CABG off-pump (%) Isolated aortic valve replacement (%) Other surgeryb (%) CPB duration (minutes) Duration of aortic clamping (minutes) Postoperative data Need of vasopressive medications (%) Antihypertensive agents (%) Time to extubation (hours) Blood loss at 24 hours (mL) a
p Value
Background. To assess the utility of neutrophil gelatinase-associated lipocalin (NGAL) as an early marker of acute kidney injury (AKI) occurring after cardiac surgery in patients with prior chronic kidney failure. Methods. Patients with preoperative creatinine clearance 60 mL minL1 1.73 mL2 or less according to the Cockcroft-Gault formula and scheduled to undergo cardiac surgery were eligible for inclusion. The AKI was defined as an increase in plasma creatinine greater than 50% over preoperative values. Threshold values of NGAL predictive of AKI were determined using receiver operating characteristic curve analysis, and predictive value of NGAL for AKI was evaluated by logistic regression. Results. Over a 1-year inclusion period, 166 patients were included. At 6 hours post-surgery, hypertension, occurrence of at least 1 postoperative complication, and
NGAL greater than 155ng/mL were shown to be independent predictors of AKI. NGAL greater than 155 ng/mL at 6 hours was associated with an odds ratio for risk of postoperative AKI of 7.1 [2.7 to 18]. On average, diagnosis of postoperative AKI was made 20 hours earlier using NGAL at 6 hours post-surgery as compared with a diagnosis based on a 50% increase in creatinine over baseline. The threshold for NGAL of 155 ng/mL at 6 hours had a sensitivity of 79% and specificity of 58% for the diagnosis of AKI. Conclusions. Earlier diagnosis of AKI post-surgery based on NGAL assessment makes it possible to initiate appropriate therapy at an earlier stage in this high-risk patient population.
A
in children [6] and adults [7, 8]. NGAL is released in response to ischemia of the proximal tubules of the kidney, and NGAL values early after surgery could predict the duration and severity of AKI. In addition, NGAL can independently predict deteriorating renal function and could therefore be useful even in the context of preexisting renal failure. A novel marker that could identify AKI after surgery at an earlier stage than currently possible with monitoring of creatinine plasma levels would make it possible to initiate appropriate therapy more quickly in these patients. To the best of our knowledge, no study to date has investigated the utility of early NGAL assessment to predict the occurrence of AKI post-cardiac surgery in patients suffering from chronic renal failure prior to surgery. Therefore, we aimed to assess the predictive ability of plasma NGAL levels to identify deteriorations in renal function after cardiac surgery in patients with preexisting renal failure.
cute kidney injury (AKI) is one of the most frequent complications after cardiac surgery [1]. The incidence of postoperative AKI in cardiac surgery ranges from 5% to 50% depending on preoperative renal status, and the criteria used to define AKI [2, 3]. Regardless of whether renal function was normal before surgery or whether there is a deterioration of previous chronic renal failure, AKI after surgery is associated with increased morbidity and perioperative mortality in-hospital [4]. In routine practice, renal function is monitored by regular assessment of plasma creatinine levels. However, there is a time delay between the onset of renal impairment and the resulting telltale increase in blood creatinine levels. Thus, by the time creatinine levels are found to be elevated, renal function has already been impaired for several days [5]. Recent studies have underlined the promising properties of neutrophil gelatinase-associated lipocalin (NGAL), a protein normally expressed at very low plasma levels, as an early predictor of AKI after cardiac surgery
Accepted for publication Oct 3, 2014. Address correspondence to Dr Chocron, Department of Thoracic and Cardio-Vascular Surgery, EA3920, University Hospital Jean Minjoz, Blvd Fleming, 25000 Besanc¸on, France; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
Material and Methods We performed this prospective study in a single university teaching hospital that is a referral center for a population basin of approximately 1.2 million inhabitants 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2014.10.011
2
PERROTTI ET AL NGAL AND AKI
(region of Franche-Comte in eastern France). This study was approved by the local ethics committee (CPP EST II, registered under the number 10/544) and was registered with the ClinicalTrials.gov database under the number NCT01227122. All participants provided written informed consent. The inclusion criteria for the study were adult (age >18 years) patients with preoperative creatinine clearance 60 mL min1 1.73 m2 or less according to the Cockcroft Gault formula scheduled to undergo cardiac surgery, and who provided written informed consent. Exclusion criteria were the following: emergency surgery; injection of contrast medium in the 3 days preceding surgery or intravenous iodinated contrast medium injection within the first 24 hours post-surgery; patients with preoperative blood creatinine levels greater than 300 mmol/L; patients undergoing hemodialysis; patients with ongoing inflammation, infection or cancer; pregnancy; patient refusal; adults under legal protection.
Neutrophil Gelatinase-Associated Lipocalin Assessments NGAL was assessed from blood samples taken at 5 time points: namely at induction of general anesthesia (baseline NGAL); at 15 minutes after interruption of extracorporeal circulation; and at 6, 12, and 24 hours after the end of surgery. Blood samples were centrifuged at 2,250 250 g for 13 2 minutes at room temperature (18 to 25 Celsius), aliquoted in 0.5 mL tubes (Eppendorf, Le Pecq, France), and frozen at 20 C for later analysis. Tests were performed using a specific kit, namely Triage NGAL test (Biosite Inverness Medical, San Diego, CA). The measurable range is from 15 to 1,300 ng/mL. Serum creatinine and creatinine clearance according to the Cockcroft-Gault formula were assessed at 6 and 12 hours after surgery, and then once daily until the third postoperative day.
Definition of Acute Kidney Injury Acute kidney injury was defined as an increase in blood creatinine levels of at least 50% over preoperative values. Patients were classified into the following 2 groups accordingly: group 1, patients with a postoperative increase in creatinine less than 50% over baseline (no-AKI group); and group 2, patients with a postoperative increase in creatinine ( 50%) over baseline (AKI group).
Correlations The correlation between baseline NGAL and baseline creatinine clearance value and to serum creatinine level was tested. In the AKI group, the correlation between the ratio of preoperative NGAL to NGAL at 6 hours and the severity of the subsequent AKI—defined as the ratio of the lowest postoperative creatinine clearance during follow-up (considered as nil for patients who had hemofiltration)—to preoperative creatinine clearance was also tested.
Statistical Analysis Quantitative data are described as mean standard deviation or median interquartile range for non-normally
Ann Thorac Surg 2015;-:-–-
distributed variables. Qualitative data are described as number (percentage). Preoperative, perioperative, and postoperative characteristics were compared using the Mann-Whitney and c2 or Fisher exact test, as appropriate. Correlations were evaluated by the Pearson correlation coefficient. We considered the strength of a correlation as weak for correlation coefficients ranging from 0.1 to 0.3, moderate for 0.3 to 0.6, and strong for correlation coefficients greater than 0.6. To identify a threshold value of NGAL predictive of AKI, receiver operating characteristic (ROC) curve analysis was used, and the sensitivity and specificity of the cutoff values were calculated. Independent predictors of postoperative AKI were identified by logistic regression. All variables with a p value less than 0.10 by univariate analysis were included in the model. All analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). A p value of less than 0.05 was considered statistically significant.
Results Between November 1, 2012 and November 27, 2013, 819 patients underwent cardiac surgery in our institution. Among the 213 patients who met inclusion criteria, 8 refused to participate and 39 had too many missing values to be analyzed. Therefore 166 patients were included in the study. Preoperative renal failure was moderate (creatinine clearance from 30 to 60 mL min1 1.73 m2) in most patients (149, 90%), and severe (creatinine clearance from 15 to 29 mL min1 1.73 m2) in 17 patients (10%). The baseline characteristics of the study population are shown in Table 1. There was no significant difference between groups at baseline except for hypertension and preoperative creatinine clearance. There was a borderline significant difference in the duration of cardiopulmonary bypass (CPB) between groups (p ¼ 0.05). In the AKI group, average time to 50% increase in creatinine was 25.6 21.7 hours. Although statistically significant (p < 0.01) the correlation between preoperative NGAL values and (1) preoperative creatinine clearance and (2) preoperative creatinine values was only moderate (r ¼ 0.34 and 0.42, respectively). In the AKI group, the ratio of preoperative NGAL to NGAL at 6 hours was not correlated with the severity of subsequent AKI (r ¼ 0.02, p ¼ 0.9). Table 2 details the postoperative complications in both groups. There was a higher overall rate of complications in the AKI group. There were 6 postoperative deaths (4%), 1 in the no AKI group from acute respiratory distress syndrome, and 5 in the AKI group; namely, 1 from acute respiratory distress syndrome, 3 from mesenteric ischemia, and 1 from heart failure. The threshold NGAL values at each time point, as determined by ROC curve analysis, are indicated in Figure 1. At 15 minutes, sensitivity was 54% and specificity was 71%; at 6 hours, these values were 79% and 58%, respectively. By multivariate analysis (Table 3) at 6 hours a history of hypertension (odds ratio [OR] ¼ 3.2 [1.2 to 8.9]),
Ann Thorac Surg 2015;-:-–-
3
PERROTTI ET AL NGAL AND AKI
Table 1. Baseline Characteristics of the Study Population Patient Characteristics Preoperative Age (years) Male sex (%) BMI (kg/m2) Diabetes (%) Dyslipidemia (%) History of hypertension (%) Family history of CAD (%) PAD (%) LVEF Diuretics presurgery (%) Mean preoperative creatinine clearance Perioperative data CABG on-pump (%) CABG off-pump (%) Isolated aortic valve replacement (%) Other surgeryb (%) CPB duration (minutes) Duration of aortic clamping (minutes) Postoperative data Need of vasopressive medications (%) Antihypertensive agents (%) Time to extubation (hours) Blood loss at 24 hours (mL) a
p Value
