Risk factors for acute renal failure requiring dialysis after surgery for congenital heart disease in children

Continuous renal replacement therapy in children after cardiac surgery

OBJECTIVE

The objective was to study the clinical course of children requiring continuous renal replacement therapy (CRRT) after cardiac surgery and to analyze the factors associated with mortality.

METHODS

A prospective observational study was performed that included all children requiring CRRT after cardiac surgery, comparing these patients with other critically ill children requiring CRRT. Univariate and multivariate analyses were performed to determine the influence of each factor on mortality.

RESULTS

Eighty-one (4.9%) of 1650 children undergoing cardiac surgery required CRRT; 65 of them (80.2%) presented multiorgan failure. Children starting CRRT after cardiac surgery had lower mean arterial pressure and lower urea and creatinine levels, and were more likely to require mechanical ventilation than other children on CRRT. The incidence of complications was similar. Cardiac surgery increased the probability of requiring CRRT for more than 14 days. Mortality was 43% in children receiving CRRT after cardiac surgery and 29% in other children (P = .05). Factors associated with mortality in the univariate analysis were age less than 12 months, weight less than 10 kg, higher Pediatric Risk of Mortality Score, hypotension, lower urea and creatinine on starting CRRT, and use of hemofiltration. In the multivariate analysis, the only factor associated with mortality was hypotension on starting CRRT (hazard ratio, 4.01; 95% confidence interval, 1.2-13.4; P = .024).

CONCLUSIONS

Although only a small percentage of children undergoing cardiac surgery required CRRT, mortality in these patients was high. Hypotension at the time of starting the technique was the only factor associated with a higher mortality.

Pediatric RIFLE for acute kidney injury diagnosis and prognosis for children undergoing cardiac surgery: a single-center prospective observational study

This study evaluated the performance of the pediatric RIFLE (pRIFLE) score for acute kidney injury (AKI) diagnosis and prognosis after pediatric cardiac surgery. It was a single-center prospective observational study developed in a pediatric cardiac intensive care unit (pCICU) of a tertiary children's hospital. The study enrolled 160 consecutive children younger than 1 year with congenital heart diseases and undergoing cardiac surgery with cardiopulmonary bypass. Of the 160 children, 50 (31 %) were neonates, and 20 (12 %) had a univentricular heart. Palliative surgery was performed for 53 patients (33 %). A diagnosis of AKI was determined for 90 patients (56 %), and 68 (42 %) of these patients achieved an "R" level of AKI severity, 17 patients (10 %) an "I" level, and 5 patients (3 %) an "F" level. Longer cross-clamp times (p = 0.045), a higher inotropic score (p = 0.02), and a higher Risk-Adjusted Classification for Congenital Heart Surgery score (p = 0.048) but not age (p = 0.27) correlated significantly with pRIFLE class severity. Patients classified with a higher pRIFLE score required a greater number of mechanical ventilation days (p = 0.03) and a longer pCICU stay (p = 0.045). Renal replacement therapy (RRT) was needed for 13 patients (8.1 %), with two patients receiving continuous hemofiltration, and 11 patients receiving peritoneal dialysis. At the start of dialysis, the distribution of RRT patients differed significantly within pRIFLE classes (p = 0.015). All deceased patients were classified as pRIFLE "I" or "F" (p = 0.0001). The findings showed that pRIFLE is easily and feasibly applied for pediatric patients with congenital heart disease. The pRIFLE classification showed that AKI incidence in pediatric cardiac surgery infants is high and associated with poorer outcomes.

Acute kidney injury according to pediatric RIFLE criteria is associated with negative outcomes after heart surgery in children

BACKGROUND

The aim of this study was to investigate the association between the occurrence of acute kidney injury (AKI) according to pediatric RIFLE (pRIFLE) criteria and adverse outcomes in children after heart surgery.

METHODS

Children undergoing heart surgery in a tertiary hospital in Southern Brazil were followed during their stay in the pediatric intensive care unit (PICU) or until death. The exposure variable was occurrence of AKI according to pRIFLE criteria which place AKI in three categories: R (risk), I (injury), and F (failure). The outcomes studied were death, length of mechanical ventilation (MV), and length of PICU stay.

RESULTS

Eighty-five children were enrolled in the study. Of these, 47 (55.3 %) did not have AKI, while 22 (25.9 %), seven (8.2 %), and nine (10.6 %) were classified into pRIFLE categories R, I, and F, respectively. The incidence of death was 18.4 and 4.2 % in patients with and without AKI, respectively. Compared to children who did not develop AKI, the adjusted odds ratio for death was 1.05 [95 % confidence interval (CI) 0.09-11.11], 8.36 (95 % CI 1.32-52.63), and 7.85 (95 % CI 1.53-40.29) in the R, I, and F groups, respectively (p = 0.022). Duration of MV and of PICU stay were significantly higher in those children with AKI.

CONCLUSIONS

The occurrence of AKI according to pRIFLE criteria is associated to adverse outcomes in children after heart surgery.

Preoperative prediction of acute kidney injury--from clinical scores to biomarkers

Early acute kidney injury (AKI) diagnosis in critically ill children has been an important recent research focus because of the known association of AKI with poor outcomes and the requirement of early intervention to mitigate negative effects of AKI. In children having surgery, the preoperative period offers a unique opportunity to predict postoperative acute kidney injury (AKI), well before AKI occurs. Pediatric AKI epidemiologic studies have begun to identify which preoperative factors may predict development of postoperative cardiac surgery. Using these clinical risk factors, it may be possible to derive preoperative clinical risk scores and improve upon our ability to risk-stratify children into AKI treatment trials, pre-emptively provide conservative renal injury prevention strategies, and ultimately improve patient outcomes. Developing risk scores requires rigorous methodology and validation before widespread use. There is little information currently on the use of preoperative biological or physiological biomarkers to predict postoperative AKI, representing an important area of future research. This review will provide an overview of methodology of preoperative risk score development, discuss pediatric-specific issues around deriving such risk scores, including the combination of preoperative clinical and biologic biomarkers for AKI prediction, and suggest future research avenues.

Biomarkers of acute kidney injury in pediatric cardiac surgery

OBJECTIVES

Acute kidney injury (AKI) is a significant problem in children undergoing cardiopulmonary bypass (CPB). The aims of this study were to assess the diagnostic validity of serum CysC (sCysC), serum neutrophil gelatinase lipocalin (sNGAL), urine neutrophil gelatinase lipocalin (uNGAL), urine kidney injury molecule (uKIM)-1, and urine liver fatty acid-binding protein (uL-FABP) to predict AKI presence and severity in children undergoing CPB.

DESIGN AND METHODS

We performed a prospective single-center evaluation of sCysC, sNGAL, uNGAL, uKIM-1 and uL-FABP at 0, 2, 6, 24 and 48 h postoperatively in children undergoing CPB during cardiac surgery. AKI was defined as ≥25% decrease in the estimated creatinine clearance (eCCl) from pre-operative baseline at 48h after surgery.

RESULTS

Of the 112 patients, 18 patients (16.1%) developed AKI; four of them needed acute dialysis treatment and three AKI patients died. In the AKI compared to the non-AKI group, sCysC at 2h, and uNGAL and uL-FABP at 2-48 h were significantly increased, as well as CPB, aortic cross clamp time and length of hospital stay. Biomarkers increased with worsening AKI severity. At 2h after CPB the best accuracy for diagnosis of AKI had uL-FABP and sCysC with area under the receiver operator curve (AUC) of 0.89 and 0.73, respectively. At 6 and 24h after CPB the best AUC was found for uL-FABP (0.75 and 0.87 respectively) and for uNGAL (0.70 and 0.93, respectively).

CONCLUSIONS

sCysC, uNGAL and uL-FABP are reliable early predictors for AKI after CPB. By allowing earlier timing of injury and earlier intervention, they could improve AKI outcome.

Epidemiology of cardiac surgery-associated acute kidney injury in neonates: a retrospective study

BACKGROUND

Cardiac surgery is a known risk factor for acute kidney injury (AKI) in children. However, cardiac surgery-associated AKI (CS-AKI) in neonates has not been well studied. The objectives of this study were: (1) to describe the epidemiology of CS-AKI in neonates utilizing the Acute Kidney Injury Network (AKIN) definition, (2) to identify risk factors for neonatal CS-AKI, and (3) to determine if neonatal CS-AKI is associated with increased morbidity and mortality.

METHODS

This was a retrospective study involving 122 neonates (≤28 days) undergoing cardiac surgery from 2006 to 2009. Neonates with and without AKI were identified using serum creatinine (SCr) and urine output (UO) data.

RESULTS

Cardiac surgery-AKI occurred in 76 (62 %) neonates, of whom 22 (29 %) were AKIN stage 1, 19 (25 %) were stage 2, and 35 (46 %) were stage 3. AKI mostly occurred early as 75 % of patients achieved their maximal AKIN stage within the first 48 h post-operatively. In the multivariate analysis, cardiopulmonary bypass duration of ≥120 min was independently associated with AKI [odds ratio (OR) 2.53, 95 % confidence interval (CI) 1.03-6.30]. Severe AKI (AKIN stage 3) was independently associated with mortality (OR 6.70, 95 % CI 1.08-41.50) and a longer stay in the pediatric intensive care unit (hazard ratio 9.09, 95 % CI 1.35-60.95). The majority of severe AKI cases (65 %) were identified with AKIN UO criteria alone without significant rises in SCr.

CONCLUSIONS

Cardiac surgery-AKI is common in neonates when the AKIN definition is utilized and is associated with higher morbidity and mortality, especially in those with more severe AKI.

Is serum cystatin C an early predictor for acute kidney injury following cardiopulmonary bypass surgery in infants and young children?

Acute kidney injury (AKI) is a potential complication following cardiopulmonary bypass (CPB) surgery in infants and young children with congenital heart disease (CHD). The current pilot study evaluates the predictive value of serum cystatin C (CysC) for AKI after pediatric CPB surgery. We prospectively enrolled 43 children with CHD (30 males and 13 females) who underwent CPB surgery. They were aged 3 years or younger. Serum CysC was measured at baseline and at 6 hours, 12 hours, 24 hours, and 48 hours after initiating CBP. Twenty-one (48.8%) patients developed AKI after cardiac surgery. Children who developed AKI had more complex cardiac surgical procedures (based on the Risk Adjustment for Congenital Heart Surgery 1 [RACHS-1] category), a longer CPB duration, and a longer aortic clamping time (ACT). Serum concentrations of CysC postoperatively demonstrated an initial decline at 6 hours, a subsequent increase at 12 hours, and stabilized at 24-48 hours. The best predictive performance was achieved 6 hours after CPB with an area under receiver operating characteristic curve (AUC) value of 0.69. The optimal cut-off value was 0.47 mg/L (sensitivity, 85.7%; specificity, 50%). By contrast, the RACHS-1 category, CPB duration, and ACT had higher AUC values at 0.735, 0.791, and 0.731, respectively. In conclusion, serum CysC had limited predictive capacity and was not superior to clinical risk factors for the early detection of AKI after CPB surgery in this pilot study, which focused on infants and young children.

Comparison of urinary biomarkers for early detection of acute kidney injury after cardiopulmonary bypass surgery in infants and young children

Acute kidney injury (AKI) is a potential complication for children with congenital heart disease (CHD) after cardiopulmonary bypass (CPB) surgery. This study was designed to investigate and compare the predictive values of urinary biomarkers for AKI after CPB surgery in infants and young children and to determine the optimal timing of testing and the cutoff value for each biomarker. The study prospectively enrolled 58 CHD children 3 years of age or younger who were undergoing CPB surgery. Urinary neutrophil gelatinase-associated lipocalin (NGAL), interleukin-18 (IL-18), microalbumin (MA), N-acetyl-ß-D-glucosaminidase (NAG), α1-microglobulin (α1-MG), and creatinine (UCr) were measured at baseline and at various time points after surgery. Children who experienced AKI had more complex cardiac surgical procedures as evaluated by Risk Adjustment for Congenital Heart Surgery 1 (RACHS-1), longer CPB and aortic clamping times, and worse clinical outcomes than those who did not. In the AKI group, all five urinary biomarkers increased substantially and peaked at 4 h after surgery. In contrast, in the non-AKI group, they increased slightly or had no significant changes during the first 24 h. All the biomarkers had the best predictive performances at 4 h after surgery. At this time point, NAG had the minimum area under the curve (AUC) (0.747), which was significantly lower than that of the others (AUC, 0.82-0.85; P < 0.05). The optimal cutoff value of each biomarker was 290 ng/mg UCr for NAGL, 1,477 pg/mg UCr for IL-18, 400 mg/g UCr for MA, 225 U/g UCr for NAG, and 290 mg/g UCr for α1-MG. In conclusion, urinary NGAL, IL-18, MA, and α1-MG had similar predictive performances for the early detection of AKI after CPB surgery in infants and young children.

Remote ischemic preconditioning in cyanosed neonates undergoing cardiopulmonary bypass: a randomized controlled trial

OBJECTIVE

The myocardial protective effect of remote ischemic preconditioning has been demonstrated in heterogeneous groups of patients undergoing cardiac surgery. No studies have examined this technique in neonates. The present study was performed to examine the remote ischemic preconditioning efficacy in this high-risk patient group.

METHODS

A preliminary, randomized, controlled trial was conducted to investigate whether remote ischemic preconditioning in cyanosed neonates undergoing cardiac surgery confers protection against cardiopulmonary bypass. Two groups of neonates undergoing cardiac surgery were recruited for the present study: patients with transposition of the great arteries undergoing the arterial switch procedure and patients with hypoplastic left heart syndrome undergoing the Norwood procedure. The subjects were randomized to the remote ischemic preconditioning or sham control groups. Remote ischemic preconditioning was induced by four 5-minute cycles of lower limb ischemia and reperfusion using a blood pressure cuff. Troponin I and the biomarkers for renal and cerebral injury were measured pre- and postoperatively.

RESULTS

A total of 39 neonates were recruited-20 with transposition of the great arteries and 19 with hypoplastic left heart syndrome. Of the 39 neonates, 20 were randomized to remote ischemic preconditioning and 19 to the sham control group. The baseline demographics appeared similar between the randomized groups. The cardiopulmonary bypass and crossclamp times were not significantly different between the 2 groups. The troponin I levels were not significantly different at 6 hours after cardiopulmonary bypass nor were the postoperative inotrope requirements. Markers of renal (neutrophil gelatinase-associated lipocalin) and cerebral injury (S100b, neuron-specific enolase) were not significantly different between the 2 groups.

CONCLUSIONS

Our data suggest that remote ischemic preconditioning in hypoxic neonates undergoing cardiopulmonary bypass surgery does not provide myocardial, renal, or neuronal protection. Additional studies are needed to examine the relationships among developmental age, hypoxia, and the molecular mechanisms of ischemic preconditioning.

Acute kidney injury after surgery for congenital heart disease

BACKGROUND

The RIFLE criteria (risk, injury, failure, loss, and end-stage kidney disease) have been used to assess acute kidney injury (AKI) in various populations of critically ill children. There are limited reports of AKI using RIFLE criteria in large pediatric populations undergoing congenital heart disease surgery.

METHODS

Records of patients 18 years and younger who underwent surgery for congenital heart disease between January 2006 and November 2009 were reviewed. The RIFLE score was determined for each patient postoperatively. Multivariate logistic regression analyses were performed to determine risk factors for AKI and the association with clinical outcomes, with subanalyses of patients 1 month of age or younger.

RESULTS

Data for 458 patients (median age, 7.6 months) were collected and analyzed. Evidence of AKI was demonstrated in 234 patients (51%), the vast majority of whom recovered within 48 hours. Younger age, higher RACHS-1 (risk-adjusted classification for congenital heart surgery) category, and longer cardiopulmonary bypass time were associated with development of AKI. Acute kidney injury was associated with longer duration of ventilation and lengths of intensive care unit and hospital stay. Incidence of AKI in patients 1 month of age or younger was 60.9%, of which more than half required greater than 72 hours to recover. In patients 1 month of age or younger, use of cardiopulmonary bypass, lower preoperative serum creatinine, and higher preoperative blood urea nitrogen were associated with AKI, and AKI was the only factor associated with longer intensive care unit and hospital lengths of stay.

CONCLUSIONS

Incidence of AKI based on RIFLE criteria in patients undergoing congenital heart disease surgery is higher than previously reported. Risk factors include age 1 month or younger and use of cardiopulmonary bypass. Acute kidney injury is associated with longer lengths of stay.

Early elevation of cardiac troponin I is predictive of short-term outcome in neonates and infants with coronary anomalies or reduced ventricular mass undergoing cardiac surgery

OBJECTIVE

The present study aimed to assess the usefulness of routine monitoring of cardiac troponin I concentrations within 24 hours of surgery (cTn-I<24h) in neonates and infants undergoing cardiac surgery.

METHODS

The added predictive ability of a high peak cTn-I<24h (within the upper quintile per procedure) for a composite outcome, including 30-day mortality and severe morbidity, was assessed retrospectively. The predicted risk for the composite outcome was estimated from a logistic regression model including preoperative and intraoperative variables. Adding a high peak cTn-I<24h to the risk model resulted in reclassification of the predicted risk. It also allowed quantification of the improvement in reclassification and discrimination by the difference between c-indexes, the Net Reclassification and the Integrated Discrimination Indexes (NRI and IDI).

RESULTS

Overall, 1023 consecutive patients were included. Adding a high peak cTn-I<24h to the model resulted in no improvement in reclassification or discrimination in the overall population (difference between c-indexes: 0.011 [-0.004 to 0.029], NRI = 0.06, P = .22, IDI = 0.02, P = .06), except in a subgroup of patients undergoing the arterial switch operation with or without ventricular septal defect closure and/or aortic arc repair, anomalous origin of the left coronary artery from the pulmonary artery repair, truncus arteriosus repair, Norwood procedure, and Sano modification, in whom NRI = 0.23 (P = .005) and IDI = 0.05 (P < .001).

CONCLUSIONS

Patients with coronary anomalies and patients with reduced ventricular mass should benefit from the routine monitoring of cTn-I concentrations after surgery for congenital cardiac disease.

Aprotinin, transfusions, and kidney injury in neonates and infants undergoing cardiac surgery

BACKGROUND

A significantly increased risk of acute kidney injury (AKI) with the prophylactic use of aprotinin has been reported in adults undergoing cardiac surgery, but not in children. Blood product transfusions have also been shown to carry an independent risk of AKI. The present study assessed associations between AKI, aprotinin, and transfusions in neonates and infants undergoing cardiac surgery.

METHODS

All neonates and infants undergoing surgery with cardiopulmonary bypass over a 42 month period, before and after the withdrawal of aprotinin, were included retrospectively. AKI was assessed by the Acute-Kidney-Injury-Network classifications. A propensity score was used to balance treated and untreated groups.

RESULTS

Three hundred and ninety patients received aprotinin and 568 patients did not. Inverse probability of treatment weighting resulted in good balance between groups for baseline and surgical characteristics. Controls underwent surgery with smaller bypass circuits and fewer transfusions. After adjustment for the use of miniaturized circuits and for the year of surgery, no significant association between the incidence of AKI, dialysis, and aprotinin was noted. Red blood cell transfusions were associated with an increased risk of AKI and dialysis: odds ratios (ORs) 1.64 (1.12-2.41) and 2.07 (1.13-3.73), respectively; as were fresh frozen plasma transfusions, ORs 2.28 (1.68-3.09) and 3.11 (1.95-4.97), respectively. Platelet transfusions were associated with an increased risk of dialysis: OR 2.20 (1.21-4.01).

CONCLUSIONS

Blood product transfusions, but not the prophylactic use of aprotinin, are significantly associated with AKI after cardiac surgery in neonates and infants.

Evaluation of a new tool for morbidity assessment in congenital cardiac surgery

BACKGROUND

To date, no instrument to assess morbidity in congenital cardiac surgery has been validated. In the Aristotle system, morbidity is accounted for by a subjective assessment of length of intensive care unit stay. A previously published Morbidity Index (MI), still in development, has been derived from objective data. The present study aims to assess the feasibility and utility of the MI at a single institution and its association with the Aristotle Comprehensive Complexity (ACC) score.

METHODS

Patients undergoing congenital cardiac surgery at our institution were enrolled retrospectively. The MI was calculated from the nonweighted sum of its components. The ability of the ACC to predict the components was estimated.

RESULTS

The MI increased more than length of stay across procedures with increasing complexity. Renal failure requiring long-term dialysis was not observed. A requirement for temporary dialysis, which is not an MI component, correlated well with the MI (r = 0.51, p < 0.001). The ACC was a good predictor for most components of the MI: length of intensive care unit stay 3 days or more, time to extubation, postoperative extracorporeal assistance, and, for temporary dialysis, with areas under the receiver operating characteristics curve of 0.67 (0.65, 0.70), 0.71 (0.68, 0.74) , 0.84 (0.75, 0.91), and 0.83 (0.77, 0.87) respectively.

CONCLUSIONS

Implementation of the MI in the Aristotle system will improve prediction of complications and long lengths of stay, making it a better morbidity indicator than length of intensive care unit stay. Requirement for temporary dialysis may be considered as an MI component.

Failure of remote ischemic preconditioning to reduce the risk of postoperative acute kidney injury in children undergoing operation for complex congenital heart disease: a randomized single-center study

OBJECTIVE

The objective of this study was to evaluate whether remote ischemic preconditioning can protect kidney function in children undergoing operation for complex congenital heart disease.

METHODS

Children (n = 113) aged 0 to 15 years admitted for complex congenital heart disease were randomly allocated according to age to remote ischemic preconditioning and control groups. After exclusion of 8 patients, we conducted the analysis on 105 patients (remote ischemic preconditioning group, n = 54; control group, n = 51). Before surgery, remote ischemic preconditioning was performed as 4 cycles of 5 minutes of ischemia by inflating a cuff around a leg to 40 mm Hg above the systolic pressure. End points were development of acute kidney injury, initiation of dialysis, plasma creatinine, estimated glomerular filtration rate, plasma cystatin C, plasma and urinary neutrophil gelatinase-associated lipocalin, and urinary output. Secondary end points included postoperative blood pressure, inotropic score, and mortality, as well as morbidity reflected by reoperation and stays in the intensive care unit and hospital.

RESULTS

Overall, 57 of the children (54%) had acute kidney injury develop, with 27 (50%) in the remote ischemic preconditioning group and 30 (59%) in the control group (P > .2). Remote ischemic preconditioning was not associated with improvement in either any of the renal biomarkers or any of the secondary end points.

CONCLUSIONS

We found no evidence that remote ischemic preconditioning provided protection of kidney function in children undergoing operation for complex congenital heart disease.

Incidence, risk factors, and outcomes of acute kidney injury after pediatric cardiac surgery: a prospective multicenter study

OBJECTIVE

To determine the incidence, severity, and risk factors of acute kidney injury in children undergoing cardiac surgery for congenital heart defects.

DESIGN

Prospective observational multicenter cohort study.

SETTING

Three pediatric intensive care units at academic centers.

PATIENTS

Three hundred eleven children between the ages of 1 month and 18 yrs undergoing pediatric cardiac surgery.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Acute kidney injury was defined as a ≥50% increase in serum creatinine from the preoperative value. Secondary outcomes were length of mechanical ventilation, length of intensive care unit and hospital stays, acute dialysis, and in-hospital mortality. The cohort had an average age of 3.8 yrs and was 45% women and mostly white (82%). One-third had prior cardiothoracic surgery, 91% of the surgeries were elective, and almost all patients required cardiopulmonary bypass. Acute kidney injury occurred in 42% (130 patients) within 3 days after surgery. Children ≥2 yrs old and <13 yrs old had a 72% lower likelihood of acute kidney injury (adjusted odds ratio: 0.28, 95% confidence interval: 0.16, 0.48), and patients 13 yrs and older had 70% lower likelihood of acute kidney injury (adjusted odds ratio: 0.30, 95% confidence interval: 0.10, 0.88) compared to patients <2 yrs old. Longer cardiopulmonary bypass time was linearly and independently associated with acute kidney injury. The development of acute kidney injury was independently associated with prolonged ventilation and with increased length of hospital stay.

CONCLUSIONS

Acute kidney injury is common after pediatric cardiac surgery and is associated with prolonged mechanical ventilation and increased hospital stay. Cardiopulmonary bypass time and age were independently associated with acute kidney injury risk. Cardiopulmonary bypass time may be a marker for case complexity.

Factors associated with acute kidney injury or failure in children undergoing cardiopulmonary bypass: a case-controlled study

Acute kidney injury (AKI) is a serious complication that occurs commonly following cardiopulmonary bypass (CPB) in infants and children. Underlying risk factors for AKI remain unclear, given changes in CPB practices during recent years. This retrospective, case-control study examined the relationships between patient, perioperative factors, AKI, and kidney failure in children who underwent CPB.

METHODS

Cohorts of children with and without AKI were identified from the cardiac perfusion and nephrology consult databases. Demographic, perioperative, and postoperative outcome data were extracted from the databases and from medical records. Children were stratified into groups based on the Acute Dialysis Quality Initiative's RIFLE definitions for acute kidney risk or injury (AKI-RI) and kidney failure.

RESULTS

The study groups included 308 controls (no AKI-RI or failure), 161 with AKI-RI, and 89 with failure. Young age, preoperative need for mechanical ventilation, milrinone, or gentamicin; intraoperative use of milrinone and furosemide; durations of CPB and anesthesia; multiple cross-clamp and transfusion of blood products were significantly associated with AKI or failure. Young age, perioperative use of milrinone, multiple cross-clamps, extracorporeal membrane oxygenation, cardiac failure, neurological complications, sepsis, and failure significantly increased the odds of mortality.

CONCLUSION

This study identified multiple perioperative risk factors for AKI-RI, failure, and mortality in children undergoing CPB. In addition to commonly known risk factors, perioperative use of milrinone, particularly in young infants, and furosemide were independently predictive of poor renal outcomes in this sample. Findings suggest a need for the development of protocols aimed at renal protection in specific at risk patients.

Predictors of long intensive care unit stay following cardiac surgery in children

OBJECTIVE

Prolonged length of stay in intensive care units after congenital heart disease surgery is associated with poor outcome, places a considerable burden on the financial resources of hospitals, and is an organizational challenge as well. This research discusses the impact of perioperative factors on prolonged stay in intensive care units.

METHODS

This is a retrospective study examining the determinants of prolonged intensive care length of stay in 693 children after cardiac surgery. Univariate and multivariate analyses were performed for an intensive care unit stay over 3 and over 14 days.

RESULTS

Neonatal age, preoperative mechanical ventilation and preoperative myocardial dysfunction, complexity and duration of procedures, as well as postoperative complications (low cardiac output syndrome, bleeding, re-operation, acute kidney injury, sepsis, respiratory insufficiency, pulmonary hypertension, pneumothorax, postoperative cardiac arrest, pneumonia, and delayed sternum closure) prolong intensive care unit hospitalization over 3 days. Patients with acute kidney injury requiring renal replacement therapy, pneumothorax, pulmonary hypertension, need for re-operation during the same admission, and myocardial dysfunction prior to surgery are at high risk of intensive care unit stay over 14 days.

CONCLUSIONS

Some patients with a risk of prolonged hospitalization may be identified preoperatively, the others just after the operation. Optimizing preoperative status and aggressive treatment of complications may have significant influence on the duration of hospitalization in intensive care units. The knowledge of risk factors may facilitate organizational procedures and rational bed management.

A small post-operative rise in serum creatinine predicts acute kidney injury in children undergoing cardiac surgery

To predict development of acute kidney injury and its outcome we retrospectively studied children having cardiac surgery. Acute kidney injury (AKI) was defined using the serum creatinine criteria of the pediatric Risk Injury Failure Loss End-Stage (pRIFLE) kidney disease definition. We tested whether a small rise (less than 50%) in creatinine on post-operative days 1 or 2 could predict a greater than 50% increase in serum creatinine within 48 h in 390 children. AKI occurred in 36% of patients, mostly in the first 4 post-operative days. Using logistic regression, significant independent risk factors for AKI were bypass time, longer vasopressor use, and a tendency for younger age. Using Cox regression, AKI was independently associated with longer intensive care unit stay and duration of ventilation. Patients whose serum creatinine did not increase on post-operative days 1 or 2 were unlikely to develop AKI (negative predictive values of 87 and 98%, respectively). Percentage serum creatinine rise on post-operative day 1 predicted AKI within 48 h (area under the curve=0.65). Our study shows that AKI after pediatric heart surgery is common and is a risk factor for poorer outcome. Small post-operative increases in serum creatinine may assist in the early prediction of AKI.