Early detection of acute kidney injury after pediatric cardiac surgery

Predicting pediatric cardiac surgery-associated acute kidney injury using machine learning

BACKGROUND

Prediction of cardiac surgery-associated acute kidney injury (CS-AKI) in pediatric patients is crucial to improve outcomes and guide clinical decision-making. This study aimed to develop a supervised machine learning (ML) model for predicting moderate to severe CS-AKI at postoperative day 2 (POD2).

METHODS

This retrospective cohort study analyzed data from 402 pediatric patients who underwent cardiac surgery at a university-affiliated children's hospital, who were separated into an 80%-20% train-test split. The ML model utilized demographic, preoperative, intraoperative, and POD0 clinical and laboratory data to predict moderate to severe AKI categorized by Kidney Disease: Improving Global Outcomes (KDIGO) stage 2 or 3 at POD2. Input feature importance was assessed by SHapley Additive exPlanations (SHAP) values. Model performance was evaluated using accuracy, area under the receiver operating curve (AUROC), precision, recall, area under the precision-recall curve (AUPRC), F1-score, and Brier score.

RESULTS

Overall, 13.7% of children in the test set experienced moderate to severe AKI. The ML model achieved promising performance, with accuracy of 0.91 (95% CI: 0.82-1.00), AUROC of 0.88 (95% CI: 0.72-1.00), precision of 0.92 (95% CI: 0.70-1.00), recall of 0.63 (95% CI: 0.32-0.96), AUPRC of 0.81 (95% CI: 0.61-1.00), F1-score of 0.73 (95% CI: 0.46-0.99), and Brier score loss of 0.09 (95% CI: 0.00-0.17). The top ten most important features assessed by SHAP analyses in this model were preoperative serum creatinine, surgery duration, POD0 serum pH, POD0 lactate, cardiopulmonary bypass duration, POD0 vasoactive inotropic score, sex, POD0 hematocrit, preoperative weight, and POD0 serum creatinine.

CONCLUSIONS

A supervised ML model utilizing demographic, preoperative, intraoperative, and immediate postoperative clinical and laboratory data showed promising performance in predicting moderate to severe CS-AKI at POD2 in pediatric patients.

Immune urinary biomarkers predict infant cardiac surgery-associated acute kidney injury

BACKGROUND

Acute kidney injury (AKI) occurs frequently after infant cardiac surgery and is associated with poor outcomes, including mortality and prolonged length of stay. AKI mechanisms are poorly understood, limiting therapeutic targets. Emerging data implicates dysregulated immune activation in post-cardiac surgery AKI development. We sought to identify immune-mediated AKI biomarkers after infant cardiopulmonary bypass (CPB)-assisted cardiac surgery.

METHODS

A single-center prospective study of 126 infants less than 1 year old undergoing CPB-assisted surgery enrolled between 10/2017 and 6/2019. Urine samples were collected before CPB and at 6, 24, 48, and 72 h after surgery. Immune-mediated biomarkers were measured using commercial ELISA and Luminex™ multiplex kits. Based on subject age, neonatal KDIGO (< 1 month) or KDIGO criteria defined AKI. The Kruskal-Wallis rank test determined the relationship between urinary biomarker measurements and AKI.

RESULTS

A total of 35 infants (27%) developed AKI. AKI subjects were younger, underwent more complex surgery, and had longer CPB time. Subjects with AKI vs. those without AKI had higher median urinary chemokine 10 (C-X-C motif) ligand levels at 24, 48, and 72 h, respectively: 14.3 pg/ml vs. 5.3 pg/ml, 3.4 pg/ml vs. 0.8 pg/ml, and 1.15 pg/ml vs. 0.22 pg/ml (p < 0.05) post-CPB. At 6 h post-CPB, median vascular cell adhesion protein 1 (VCAM) levels (pg/mL) were higher among AKI subjects (491 pg/ml vs. 0 pg/ml, p = 0.04).

CONCLUSIONS

Urinary CXCL10 and VCAM are promising pro-inflammatory biomarkers for early AKI detection and may indicate eventual AKI therapeutic targets. A higher resolution version of the Graphical abstract is available as Supplementary information.

Predictors of early peritoneal dialysis initiation in newborns and young infants following cardiac surgery

OBJECTIVE

This single-centre, retrospective cohort study was conducted to investigate the predictors of early peritoneal dialysis initiation in newborns and young infants undergoing cardiac surgery.

METHODS

There were fifty-seven newborns and young infants. All subjects received peritoneal dialysis catheter after completion of the cardiopulmonary bypass. Worsening post-operative (post-op) positive fluid balance and oliguria (<1 ml/kg/hour) despite furosemide were the clinical indications to start early peritoneal dialysis (peritoneal dialysis +). Demographic, clinical, and laboratory data were collected from the pre-operative, intra-operative, and immediately post-operative periods.

RESULTS

Baseline demographic data were indifferent except that peritoneal dialysis + group had more newborns. Pre-operative serum creatinine was higher for peritoneal dialysis + group (p = 0.025). Peritoneal dialysis + group had longer cardiopulmonary bypass time (p = 0.044), longer aorta cross-clamp time (p = 0.044), and less urine output during post-op 24 hours (p = 0.008). In the univariate logistic regression model, pre-op serum creatinine was significantly associated with higher odds of being in peritoneal dialysis + (p = 0.021) and post-op systolic blood pressure (p = 0.018) and post-op mean arterial pressure (p=0.001) were significantly associated with reduced odds of being in peritoneal dialysis + (p = 0.018 and p = 0.001, respectively). Post-op mean arterial pressure showed a statistically significant association adjusted odds ratio = 0.89, 95% confidence interval [0.81, 0.96], p = 0.004) with peritoneal dialysis + in multivariate analysis after adjusting for age at surgery.

CONCLUSIONS

In our single-centre cohort, pre-op serum creatinine, post-op systolic blood pressure, and mean arterial pressure demonstrated statistically significant association with peritoneal dialysis +. This finding may help to better risk stratify newborns and young infants for early peritoneal dialysis start following cardiac surgery.

Epidemiology data on the cost and outcomes associated with pediatric acute kidney injury

BACKGROUND

Hospitalized children with acute kidney injury (AKI) have not been extensively studied for clinical outcomes including hospital stay, the need for mechanical ventilation, mortality rates, and healthcare utilization. We hypothesize significant financial costs and increased morbidity and mortality associated with pediatric AKI.

METHODS

This is a retrospective study of pediatric patients (age ≤18 years) included in the Kids' Inpatient Database (KID) between January 1, 2016, and December 31, 2021. The results of the data analysis were utilized for comparative testing between the AKI and non-AKI cohorts.

RESULTS

The study included 4842 children [with AKI (n = 2424) and without AKI (n = 2418)]. The odds of mortality (p = 0.004) and mechanical ventilation (p < 0.001) were observed to be significantly higher among those with AKI as compared to those without AKI. Additionally, the median (IQR) duration of stay in the hospital (p < 0.001) and total cost (p < 0.001) were significantly higher among those with AKI vs. those without AKI.

CONCLUSIONS

AKI in children was associated with higher odds of mortality, longer duration of hospital stay, increased requirement of mechanical ventilation, and increased hospital expenditure. The scientific community can utilize this information to better understand the outcomes associated with this disease process in this patient population.

IMPACT

This article has thoroughly evaluated epidemiologic data associated with pediatric acute kidney injury (AKI) in hospitalized patients This study assesses mortality, hospital expenditure, and other factors to strengthen single-center and few multi-center studies and provides novel data regarding insurance and cost associated with pediatric AKI With increased knowledge of current epidemiology and risk factors, the scientific community can better understand prevention and outcomes in hospitalized children with AKI.

The Use of Renal Biomarkers in Pediatric Cardiac Patients With Acute Kidney Injury

Acute kidney injury (AKI) is a common and serious condition that occurs in approximately 30% to 50% of pediatric patients that undergo cardiac surgery. Currently used parameters to measure kidney function (serum creatinine and urine output) are often unreliable and delay the prediction of AKI, despite their adoption into clinical guidelines. Emerging evidence suggests that biomarkers such as neutrophil gelatinase-associated lipocalin, cystatin C, interleukin-18, kidney injury molecule 1, and liver-type fatty acid– binding protein may be useful in the identification and location of pediatric renal injury. Ontogeny-related changes in tubular function and nephrogenesis result in reference values that differ based on age and sex. In addition, changes in endogenous concentrations may result from factors such as cardiopulmonary bypass. The use of urine samples to measure renal biomarkers offers a significant advantage compared with routine blood sampling, especially in the neonatal patient population. Future research is warranted to determine age-dependent changes in AKI biomarkers and the relationship with pharmacokinetic clearance of commonly used medications in the postoperative cardiac patient.

Biomarkers of acute kidney injury in pediatric cardiac surgery

Acute kidney injury (AKI) is characterized by a sudden decrease in kidney function. Children with congenital heart disease are a special group at risk of developing AKI. We performed a systematic review of the literature to search for studies reporting the usefulness of novel urine, serum, and plasma biomarkers in the diagnosis and progression of AKI and their association with clinical outcomes in children undergoing pediatric cardiac surgery. In thirty studies, we analyzed the capacity to predict AKI and poor outcomes of five biomarkers: Cystatin C, Neutrophil gelatinase-associated lipocalin, Interleukin-18, Kidney injury molecule-1, and Liver fatty acid-binding protein. In conclusion, we suggest the need for further meta-analyses with the availability of additional studies.

Early biochemical markers in the assessment of acute kidney injury in children after cardiac surgery

Our aim was to evaluate biochemical markers in plasma (NGAL, CysC) and urine (NGAL, KIM-1) in children's early onset of acute kidney injury after congenital heart defect surgery using cardiopulmonary bypass. This study prospectively included 100 children with congenital heart defects who developed AKI. Patients with acute kidney injury had significantly higher CysC levels 6 and 12 h after cardiac surgery and plasma NGAL levels 2 and 6 h after cardiac surgery. The best predictive properties for the development of acute kidney injury are the combination (+CysCpl or +NGALu) after 12 h and a combination (+CysCpl and +NGALu) 6 and 24 h after cardiac surgery. We showed that plasma CysC and urinary NGAL could reliably predict the development of acute kidney injury. Measurement of early biochemical markers in plasma and urine, individually and combination, may predict the development of cardiac surgery-associated acute kidney injury in children.

Daily Urinary Sodium Excretion Monitoring in Critical Care Setting: A Simple Method for an Early Detection of Acute Kidney Injury

Introduction:

Making an early diagnosis of acute kidney injury (AKI) is crucial. Classical biomarkers are not capable of early detection of AKI, but novel biomarkers that do have this capability are expensive and not universally available. This prospective study attempts to mitigate these limitations through the evaluation of daily urine analysis on patient admitted to a critical care unit in order to detect early AKI.

Methods:

Daily urinary indices were measured on every patient admitted to the intensive care unit (ICU) from the time of admission until his/her discharge from the ICU or death. This renal monitoring consisted of daily blood and spot morning urine samples in order to measure creatinine, urea, sodium, chloride and potassium in order to calculate the fractional excretion of sodium (FENa), chloride, urea and potassium. The data collected on these patients in the previous days was analyzed to determine whether or not there was a significant statistical difference in the urinary indices one day before the clinical diagnosis of AKI (day – 1) and 2 days before the diagnosis (day – 2). The statistical test applied was a single rank test, using as a limit of significance a value of P < 0.05.

Results:

Of the 203 patients included, 61 developed AKI. A statistical significant difference was documented only in the value of urinary sodium (UNa) and FENa between day-1 (one day before AKI clinical diagnosis) and day-2 (two days before AKI clinical diagnosis).

Conclusion:

Daily monitoring of UNa and FENa detected a significant change in their basal values 24 hours before clinical diagnosis of AKI was made.

Variation in clinical usefulness of biomarkers of acute kidney injury in young children undergoing cardiac surgery

BACKGROUND

Acute kidney injury (AKI) is one of the most significant postoperative complications of pediatric cardiac surgery. Because serum creatinine has limitations as a diagnostic marker of AKI, new biomarkers including neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and interleukin-18 (IL-18) are being evaluated to overcome these limitations and detect AKI at an early stage after cardiac surgery.

PURPOSE

This study aimed to investigate the clinical usefulness of these biomarkers in young children.

METHODS

Thirty patients with congenital heart diseases who underwent cardiac surgery using cardiopulmonary bypass (CPB) were selected, and their urine and blood samples were collected at baseline and 6, 24, and 48 hours after surgery. Serum creatinine and blood urea nitrogen levels as well as NGAL, KIM-1, and IL-18 levels in urine samples were measured, and clinical parameters were evaluated.

RESULTS

Of the 30 patients, 12 developed AKI within 48 hours after cardiac surgery. In the AKI group, 8 of 12 (66.6%) met AKI criteria after 24 hours, and urine KIM-1/creatinine (Cr) level (with adjustment of urine creatinine) peaked at 24 hours with significant difference from baseline level. Additionally, urine KIM-1/Cr level in the AKI group was significantly higher than in the non-AKI group at 6 hours. However, urine NGAL/Cr and IL-18/Cr levels showed no specific trend with time for 48 hours after cardiac surgery.

CONCLUSION

It is suggested that urine KIM-1/Cr concentration could be considered a good biomarker for early AKI prediction after open cardiac surgery using CPB in young children with congenital heart diseases.

Novel Urinary Biomarkers for Acute Kidney Injury and Prediction of Clinical Outcomes After Pediatric Cardiac Surgery

Acute kidney injury (AKI) is a serious complication of pediatric cardiac surgery, with high morbidity and mortality. We aimed to evaluate the perioperative risk factors for AKI, and the validity of novel diagnostic urinary biomarkers after pediatric cardiac surgery. We analyzed 103 consecutive pediatric patients (≤ 18 years old), who underwent cardiac surgery. AKI was defined by ≥ 50% increase in serum creatinine levels from baseline. Urinary liver-type fatty acid binding protein (L-FABP) and neutrophil gelatinase-associated lipocalin (NGAL) were measured postoperatively at the intensive care unit (ICU) admission, subsequently at 4, 12, and 24 h. Areas under the receiver-operating characteristic curves (AUC) were calculated at each assessment time. AKI had developed in 47 patients (45.6%) by the second postoperative day. Univentricular status, aortic cross-clamping time, and intraoperative fluid balance were independently associated with AKI (p = 0.02, 0.01 and 0.01, respectively). Urinary L-FABP and NGAL were significantly higher in the AKI group at each point (p < 0.05). The predictive abilities of both biomarkers (AUC = 0.78-0.90) at ICU admission and 4 h after were especially high. The patients with L-FABP greater than the cutoff value at ICU admission and 4 h after ICU admission had significantly longer intubation and hospitalization periods (p < 0.05). Those with elevated NGAL levels at admission, and 4 h and 24 h after ICU admission, had significantly longer intubation, ICU stay, and hospitalization (p < 0.05). L-FABP and NGAL can be useful biomarkers for detecting early AKI after pediatric cardiac surgery and predicting adverse clinical outcomes.

Near-Infrared-Based Cerebral Oximetry for Prediction of Severe Acute Kidney Injury in Critically Ill Children After Cardiac Surgery

Supplemental Digital Content is available in the text.

Cerebral oximetry by near-infrared spectroscopy is used frequently in critically ill children but guidelines on its use for decision making in the PICU are lacking. We investigated cerebral near-infrared spectroscopy oximetry in its ability to predict severe acute kidney injury after pediatric cardiac surgery and assessed its additional predictive value to routinely collected data.

Biomarkers of Acute Kidney Injury after Cardiac Surgery: A Narrative Review

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a major and serious complication in patients undergoing cardiac surgery and is independently associated with perioperative mortality and mortality. Therapeutic intervention aiming at reversing kidney dysfunction seems disappointing across multiple settings. Consequently, attention has shifted from treatment to prevention and early detection. The Kidney Disease: Improving Global Outcomes (KDIGO) guidelines have unified diagnostic standards mainly based on the serum creatinine (Scr) level or urine output, but neither marker is kidney specific. Efforts have been made to identify novel biomarkers with high sensitivity and specificity. The diagnostic capabilities of neutrophil gelatinase-associated lipocalin (NGAL) and G1 cell cycle arrest biomarker as biomarkers have been confirmed in a large number of clinical trials. The utility of biomarkers of cardiac function and inflammation has been validated in clinical studies. Aiming to offer valuable information for further research, we summarize the progress in defining current markers relevant to CSA-AKI in the last three years.

Preoperative levels of urinary uromodulin predict acute kidney injury after pediatric cardiopulmonary bypass surgery

BACKGROUND

Acute kidney injury (AKI) is a common complication of cardiopulmonary bypass surgery (CPB) in children. Several promising postoperative AKI biomarkers have been identified, but no preoperative biomarkers are available. We evaluated the association of urinary uromodulin (uUMOD) with postoperative AKI.

METHODS

One hundred and one children undergoing CPB were enrolled. Urine was collected prior to CPB, and AKI was defined as ≧50% increase in serum creatinine from preoperative baseline within 48 h of surgery.

RESULTS

Forty-seven patients (47%) developed AKI, and 92% of participants in the lowest quartile of preoperative uUMOD concentrations developed AKI compared with 8% in the highest quartile. Patients with preoperative uUMOD levels in the lowest quartile had 132.3× increased risk of postoperative AKI versus the highest quartile. Raw uUMOD levels were significantly lower in patients with AKI vs. no AKI. Significance was unchanged after correcting uUMOD levels for urinary creatinine. Receiver operating characteristic analysis showed preoperative uUMOD strongly predicted postoperative AKI, with area under the curve (AUC) 0.90. Stepwise logistic regression analysis revealed a model combining uUMOD, and bypass time predicted AKI at p<0.001. Neither Risk Adjustment for Congenital Heart Surgery 1 (RACHS) score nor age improved the model's ability to predict AKI. Independent analysis demonstrated that while bypass time was associated with AKI, the predictive ability of bypass time (AUC 0.77) was less than that of preoperative uUMOD levels (AUC 0.9).

CONCLUSIONS

Children with lowest preoperative levels of uUMOD have greatly increased risk of AKI post-CPB. If uUMOD were used to risk-stratify patients undergoing CPB, clinical measures could be taken to minimize AKI development.

Urinary biomarkers of cell cycle arrest are delayed predictors of acute kidney injury after pediatric cardiopulmonary bypass

BACKGROUND

Several novel biomarkers that predict acute kidney injury (AKI) have recently been proposed. We have evaluated the sequential patterns of biomarker elevation after pediatric cardiopulmonary bypass (CPB) and determined their diagnostic accuracy.

METHODS

We measured the ability of neutrophil gelatinase-associated lipocalin (NGAL), interleukin-18 (IL-18), liver type fatty-acid binding protein (L-FABP), kidney injury molecule-1 (KIM-1), tissue inhibitor of metalloproteinase-2 (TIMP-2), and insulin-like growth factor binding protein 7 (IGFBP7), to predict AKI (≥50% increase in serum creatinine from baseline). Areas under the receiver-operator characteristic curves (AUCs) were calculated for each biomarker and for various biomarker combinations at multiple time points after CPB.

RESULTS

Of 150 patients examined, AKI had developed in 50 patients by 24 h after CPB, with an elevated NGAL concentration first noted at 2 h post-CPB, increases in IL-18, L-FABP, and the product of TIMP-2 and IGFBP7 first noted at 6 h, and an elevated KIM-1 level noted at 12 h. At each time point, urine NGAL remained the marker with the highest predictive ability (AUC > 0.9). The addition of any other biomarker did not increase the predictive accuracy of NGAL alone at 2 and 6 h. At 12 h, when compared to NGAL alone, the combination of NGAL, IL-18, and TIMP2 improved the AUC for AKI prediction (from 0.938 to 0.973).

CONCLUSIONS

While urine NGAL remains a superior stand-alone test at the 2 and 6 h time points after pediatric CPB, a panel of carefully selected biomarkers may prove optimal at later time points.

Current Biochemical Monitoring and Risk Management of Immunosuppressive Therapy after Transplantation

Immunosuppressive drugs play a crucial role in the inhibition of immune reaction and prevention of graft rejection aswell as in the pharmacotherapy of autoimmune disorders. Effective immunosuppression should provide an adequate safety profile and improve treatment outcomes and the patients' quality of life. High-risk transplant recipients may be identified, but a definitive prediction model has still not been recognized. Therapeutic drug monitoring (TDM) for immunosuppressive drugs is an essential, but at the same time insufficient tool due to low predictability of drug exposition and marked pharmacokinetic variability. Parallel therapeutic, biochemical and clinical monitoring may successfully optimize and individualize therapy for transplanted recipients, providing optimal medical outcomes. Modern pharmacotherapy management should include new biomarkers with better sensitivity and specificity that can identify early cell damage. The aim of this study was to point out the importance of finding new biomarkers that would enable early detection of adverse drug events and cell damage in organ transplant recipients. We wanted to confirm the importance of routine biochemical monitoring in improving the safety of immunosuppressive treatment.

Pediatric acute kidney injury: prevalence, impact and management challenges

The incidence of pediatric acute kidney injury (AKI) is increasing globally, as are the associated morbidities and mortality. A recent standardization of the definition of AKI has allowed for a more accurate assessment of the epidemiology of pediatric AKI. Recent advances in leveraging electronic medical health record systems have allowed for real-time risk stratification and prevention of pediatric AKI in the hospital setting. Newly developed and validated clinical scores have improved our ability to predict AKI and provide a rational context for biomarker utilization in hospitalized children. Novel non-invasive diagnostic and predictive biomarkers have been launched globally to improve our ability to diagnose and predict AKI and its adverse outcomes as well as recovery. This review summarizes the most current literature, focusing on the epidemiology, management, and early diagnostic strategies in pediatric AKI.