Averting the legacy of kidney disease: focus on childhood

Machine Learning-Based Prediction of Acute Kidney Injury Following Pediatric Cardiac Surgery: Model Development and Validation Study

BACKGROUND

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a major complication following pediatric cardiac surgery, which is associated with increased morbidity and mortality. The early prediction of CSA-AKI before and immediately after surgery could significantly improve the implementation of preventive and therapeutic strategies during the perioperative periods. However, there is limited clinical information on how to identify pediatric patients at high risk of CSA-AKI.

OBJECTIVE

The study aims to develop and validate machine learning models to predict the development of CSA-AKI in the pediatric population.

METHODS

This retrospective cohort study enrolled patients aged 1 month to 18 years who underwent cardiac surgery with cardiopulmonary bypass at 3 medical centers of Central South University in China. CSA-AKI was defined according to the 2012 Kidney Disease: Improving Global Outcomes criteria. Feature selection was applied separately to 2 data sets: the preoperative data set and the combined preoperative and intraoperative data set. Multiple machine learning algorithms were tested, including K-nearest neighbor, naive Bayes, support vector machines, random forest, extreme gradient boosting (XGBoost), and neural networks. The best performing model was identified in cross-validation by using the area under the receiver operating characteristic curve (AUROC). Model interpretations were generated using the Shapley additive explanations (SHAP) method.

RESULTS

A total of 3278 patients from one of the centers were used for model derivation, while 585 patients from another 2 centers served as the external validation cohort. CSA-AKI occurred in 564 (17.2%) patients in the derivation cohort and 51 (8.7%) patients in the external validation cohort. Among the considered machine learning models, the XGBoost models achieved the best predictive performance in cross-validation. The AUROC of the XGBoost model using only the preoperative variables was 0.890 (95% CI 0.876-0.906) in the derivation cohort and 0.857 (95% CI 0.800-0.903) in the external validation cohort. When the intraoperative variables were included, the AUROC increased to 0.912 (95% CI 0.899-0.924) and 0.889 (95% CI 0.844-0.920) in the 2 cohorts, respectively. The SHAP method revealed that baseline serum creatinine level, perfusion time, body length, operation time, and intraoperative blood loss were the top 5 predictors of CSA-AKI.

CONCLUSIONS

The interpretable XGBoost models provide practical tools for the early prediction of CSA-AKI, which are valuable for risk stratification and perioperative management of pediatric patients undergoing cardiac surgery.

Proton pump inhibitors and the risk of hospital-acquired acute kidney injury in children

Background

To evaluate the association between use of proton pump inhibitor (PPI) and the risk of hospital-acquired acute kidney injury (HA-AKI) in hospitalized children.

Methods

We conducted a multicenter retrospective cohort study in hospitalized children aged 1 month to 18 years from 25 tertiary hospitals across China from 2013 to 2015. Patient-level data were obtained from the electronic hospitalization databases. AKI was defined and staged using the serum creatinine (SCr) data according to the Kidney Disease Improving Global Outcomes (KDIGO) criteria.

Results

Among 42,232 children analyzed, 11,496 (27.2%) used PPI, 1,760 (4.2%) used histamine 2 receptor antagonist (H2RA), and 3,514 (8.3%) had HA-AKI during hospitalization. Over 85% of PPIs were prescribed for prophylaxis of gastro-duodenal lesions in children. The use of PPI was associated with a significantly increased risk of HA-AKI compared with both non-users [odds ratio (OR), 1.37; 95% confidence interval (CI), 1.23–1.53)] and H2RA users (OR, 1.24; 95% CI, 1.01–1.52). The associations were consistent across children of different age range, gender, subtypes of PPIs and methods of administration. A larger effect was observed in children with chronic kidney disease (OR, 3.37; 95% CI, 2.46–4.62) and those needed intensive care (OR, 1.54; 95% CI, 1.33–1.78). The risk of HA-AKI was increased even within the recommended dosage range of PPI.

Conclusions

PPIs were widely used and associated with an increased risk of HA-AKI in hospitalized children in China.

Acute Kidney Injury among Hospitalized Children in China

BACKGROUND AND OBJECTIVES

High-quality epidemiologic data on AKI in children are particularly lacking in developing countries. This study aimed to assess the epidemiology and clinical correlates of AKI among hospitalized children in China.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

We performed a multicenter study, in a cohort of hospitalized children aged 1 month to 18 years, from 25 general and children's hospitals in China during 2013-2015. We obtained patient-level data from the electronic hospitalization information system and laboratory databases of all children who had at least two serum creatinine tests within any 7-day window during their first 30 days of hospitalization. We identified AKI events according to the creatinine criteria of Kidney Disease Improving Global Outcomes. The in-hospital outcomes of AKI, including mortality, kidney recovery, and length of stay, were assessed. We estimated the corresponding hazard ratios using a Cox proportional hazard model, with adjustment for age, sex, comorbidities, and clinical procedures.

RESULTS

A total of 19,908 (20%) patients with AKI were identified among 101,836 pediatric inpatients, of which 7220 (7%) were community acquired and 12,688 (13%) were hospital acquired. Up to 96% of these AKI events were not diagnosed on the discharge records. The cumulative incidence of AKI in infants (28%) was twice that in adolescents (12%). The profiles of risk factors differed between community-acquired and hospital-acquired AKI and varied with age. Diarrhea and sepsis were the top risk factors for community-acquired AKI, each contributing 6% of the risk. Congenital heart disease/cardiac surgery was the major risk factor for hospital-acquired AKI, contributing to 19% of cases. Exposure to nephrotoxic drugs, mostly nonsteroidal anti-inflammatory drugs and proton pump inhibitors, was common in hospitalized children and was associated with a higher risk of AKI. Death occurred in 842 out of 19,908 patients (4%) with AKI versus 450 out of 81,478 children (0.5%) without AKI. The risk of in-hospital death was higher among children with severe AKI, shock, and respiratory failure. Pediatric AKI was associated with longer hospital stay and higher daily cost, even after adjustment for covariates.

CONCLUSIONS

Pediatric AKI is common and is substantially underdiagnosed in China.

A New Criterion for Pediatric AKI Based on the Reference Change Value of Serum Creatinine

BACKGROUND

Current definitions of AKI do not take into account serum creatinine's high variability in children.

METHODS

We analyzed data from 156,075 hospitalized children with at least two creatinine tests within 30 days. We estimated reference change value (RCV) of creatinine on the basis of age and initial creatinine level in children without kidney disease or known AKI risk, and we used these data to develop a model for detecting pediatric AKI on the basis of RCV of creatinine. We defined pediatric AKI according to pediatric reference change value optimized for AKI in children (pROCK) as creatinine increase beyond RCV of creatinine, which was estimated as the greater of 20 μmol/L or 30% of the initial creatinine level.

RESULTS

Of 102,817 children with at least two serum creatinine tests within 7 days, 5432 (5.3%) had AKI as defined by pROCK compared with 15,647 (15.2%) and 10,446 (10.2%) as defined by pediatric RIFLE (pRIFLE) and Kidney Disease Improving Global Outcomes (KDIGO), respectively. Children with pROCK-defined AKI had significantly increased risk of death (hazard ratio, 3.56; 95% confidence interval, 3.15 to 4.04) compared with those without AKI. About 66% of patients with pRIFLE-defined AKI and 51% of patients with KDIGO-defined AKI, mostly children with initial creatinine level of <30 μmol/L, were reclassified as non-AKI by pROCK, and mortality risk in these children was comparable with risk in those without AKI by all definitions.

CONCLUSIONS

pROCK criterion improves detection of "true" AKI in children compared with earlier definitions that may lead to pediatric AKI overdiagnosis.

History of Childhood Kidney Disease and Risk of Adult End-Stage Renal Disease

BACKGROUND

The long-term risk associated with childhood kidney disease that had not progressed to chronic kidney disease in childhood is unclear. We aimed to estimate the risk of future end-stage renal disease (ESRD) among adolescents who had normal renal function and a history of childhood kidney disease.

METHODS

We conducted a nationwide, population-based, historical cohort study of 1,521,501 Israeli adolescents who were examined before compulsory military service in 1967 through 1997; data were linked to the Israeli ESRD registry. Kidney diseases in childhood included congenital anomalies of the kidney and urinary tract, pyelonephritis, and glomerular disease; all participants included in the primary analysis had normal renal function and no hypertension in adolescence. Cox proportional-hazards models were used to estimate the hazard ratio for ESRD associated with a history of childhood kidney disease.

RESULTS

During 30 years of follow-up, ESRD developed in 2490 persons. A history of any childhood kidney disease was associated with a hazard ratio for ESRD of 4.19 (95% confidence interval [CI], 3.52 to 4.99). The associations between each diagnosis of kidney disease in childhood (congenital anomalies of the kidney and urinary tract, pyelonephritis, and glomerular disease) and the risk of ESRD in adulthood were similar in magnitude (multivariable-adjusted hazard ratios of 5.19 [95% CI, 3.41 to 7.90], 4.03 [95% CI, 3.16 to 5.14], and 3.85 [95% CI, 2.77 to 5.36], respectively). A history of kidney disease in childhood was associated with younger age at the onset of ESRD (hazard ratio for ESRD among adults <40 years of age, 10.40 [95% CI, 7.96 to 13.59]).

CONCLUSIONS

A history of clinically evident kidney disease in childhood, even if renal function was apparently normal in adolescence, was associated with a significantly increased risk of ESRD, which suggests that kidney injury or structural abnormality in childhood has long-term consequences.