Epidemiology of participants in a prospective cohort study on pediatric acute kidney injury in South Korea
Article information
Abstract
Purpose
Acute kidney injury (AKI) is a common and life-threatening complication in hospitalized patients, leading to increased hospital stays and higher mortality rates. In South Korea, prospective studies that comprehensively describe the epidemiology of AKI in this population are lacking. This study aimed to evaluate the demographic characteristics of AKI in children treated at a tertiary care center in South Korea.
Methods
This prospective cohort study included children who were diagnosed with AKI at a single tertiary center between February 2016 and July 2021.
Results
Of the 781 enrolled patients, 763 were analyzed. The cohort had a mean age of 6.72 years, and 442 patients (57.9%) were male. Hemato-oncologic diseases (202 patients, 26.5%) were the most common underlying condition, followed by preterm infants (128 patients, 16.8%) and kidney disease (108 patients, 14.2%). Nephrotoxic agent exposure (165 patients, 21.6%) represented the leading cause of AKI, followed by infection (135 patients, 17.7%), severe illness (127 patients, 16.6%), and dehydration (121 patients, 15.9%). Patients frequently presented with AKI stage 1 (363 patients, 47.6%). In patients without underlying conditions (96 patients), stage 3 AKI occurred most frequently (40 patients, 41.7%) with a greater tendency for AKI progression (25 patients, 26.0%). However, this group also experienced a higher recovery rate (81 patients, 84.4%).
Conclusions
This study offers a comprehensive understanding of pediatric AKI in Korea. These findings highlight the complexity of pediatric AKI and underscore the significance of tailored management strategies based on patient characteristics and AKI etiologies.
Introduction
Acute kidney injury (AKI) is a frequent complication in patients with critical illness, leading to poorer clinical outcomes. Its occurrence is correlated with longer hospital stays and higher mortality rates [1]. Moreover, AKI is a well-recognized risk factor for hypertension, cardiovascular disease, and chronic kidney disease (CKD) development [2]. Several large-scale studies have provided valuable insights into the epidemiology and outcomes of neonatal and pediatric AKI. The multicenter Assessment of Worldwide Acute Kidney Injury Epidemiology in Neonates (AWAKEN) study has emphasized that neonatal AKI is both common and independently associated with higher mortality and longer hospital stays [3,4]. Similarly, the Assessment of Worldwide Acute Kidney Injury, Renal Angina, and Epidemiology (AWARE) study, which focused on pediatric intensive care unit (ICU) admissions, has revealed the association between AKI and poor outcomes, including increased mortality [5]. Additionally, studies including the Translational Research Investigating Biomarker Endpoints in AKI (TRIBE-AKI), Follow-Up Renal Assessment of Injury Long-Term After AKI (FRAIL-AKI), and Assessment, Serial Evaluation, and Subsequent Sequelae in AKI (ASSESS-AKI) have investigated AKI following cardiac surgery. These three studies revealed no differences in renal outcomes or hypertension incidence at 4- to 7-year follow-up between children with and without cardiac surgery-associated AKI [6-8]. However, these studies mainly focused on selected patient populations. A recent retrospective study has reported that AKI survivors are at a 2- to 4-fold higher risk of major adverse kidney events, hypertension, and subsequent AKI than matched hospitalized comparators [1]. However, prospective studies for general hospitalized patients are limited.
In South Korea, a paucity of prospective studies that comprehensively describe the epidemiology of AKI in this population exists. Therefore, this prospective cohort study aimed to evaluate the demographic characteristics of AKI in children treated at a tertiary care center in South Korea.
Methods
Study participants
We included children aged <18 years who were diagnosed with AKI at a single tertiary center from February 2016 to July 2021. The diagnosis of AKI was defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [9]. The following were the inclusion criteria: (1) an elevation in serum creatinine level of more than 1.5-fold the baseline value, an absolute increase of ≥0.3 mg/dL, or a urine output of <0.5 mL/kg/hr for at least 6 hours; and (2) provision of informed consent by the caregiver and/or the child. The following were the exclusion criteria: (1) lack of informed consent; (2) insufficient availability of clinical data; and (3) history of kidney transplantation or nephrectomy before the AKI event.
AKI and eGFR definitions
AKI staging was determined according to the KDIGO criteria [9]:
• Stage 1: a 1.5- to 1.9-fold increase in the serum creatinine level from baseline or a urine output of <0.5 mL/kg/hr for 6 to 12 hours
• Stage 2: a 2.0- to 2.9-fold increase in the serum creatinine level from baseline or a urine output of <0.5 mL/kg/hr for ≥12 hours
• Stage 3: a 3.0-fold increase in the serum creatinine level, initiation of kidney replacement therapy (KRT), a urine output of <0.3 mL/kg/hr for ≥24 hours, or anuria for ≥12 hours
The estimated glomerular filtration rate (eGFR) was calculated using the Schwartz equation as follows: eGFR (mL/min/1.73 m²)=(0.413×height in cm)/serum creatinine in mg/dL [10].
Data collection
Data were prospectively collected during AKI diagnosis. Baseline characteristics, including demographic information, underlying diseases, and laboratory values, were recorded for all enrolled patients. Underlying diseases were categorized as follows: no underlying disease, hemato-oncologic disease, preterm infant, kidney disease, cardiologic disease, CKD, liver disease, diabetes mellitus, pulmonary disease, and other diseases. For statistical analysis, these categories were consolidated into the following six groups: no underlying disease, hemato-oncologic disease, CKD/kidney disease, preterm infant, cardiologic disease, and others. The AKI etiology was determined on the basis of clinical and laboratory data and classified into the following categories: nephrotoxic agent exposure, infection, severe illness, dehydration, circulation failure, noncardiac surgery, cardiac surgery, contrast agent exposure, and unknown causes. Severe illness was characterized by the recurrence of an underlying condition, including nephrotic syndrome, or the diagnosis of leukemia, glomerulonephritis, atypical hemolytic uremic syndrome, or thrombotic microangiopathy. Nephrotoxic agents identified in our cohort included renin-angiotensin-system blocking agents, diuretics, calcineurin inhibitors, and chemotherapeutic agents such as cisplatin, ifosfamide, and methotrexate, as well as broad-spectrum antibiotics (e.g., piperacillin–tazobactam, aminoglycosides, vancomycin), antiviral agents (e.g., acyclovir, ganciclovir), and antifungals (e.g., liposomal amphotericin B). For statistical analysis, these conditions were further classified into the following five categories: dehydration/circulation failure/severe illness, nephrotoxic or contrast agent exposure, infection, surgery, and others.
Additional data, including the need for KRT, AKI aggravation, kidney function recovery, and ICU admission during AKI diagnosis were collected. The point of AKI aggravation was defined as the period when the eGFR was at its lowest or when KRT was initiated, whichever occurred first. The point of AKI recovery was defined as the period when the eGFR peaked following its lowest point during the AKI course, assessed up to a follow-up period of 3 months.
Statistical analysis
Descriptive statistics were used to summarize the patient demographics and baseline characteristics. Continuous variables were expressed as means±standard deviations, whereas categorical variables were expressed as frequencies and percentages. For comparisons across multiple groups, analysis of variance or the Kruskal-Wallis test was used. Categorical variables were compared using the chi-square test or Fisher exact test, as appropriate. A P-value of <0.05 was considered statistically significant. All statistical analyses were performed using the SAS statistical software (SAS system for Windows, version 9.4; SAS Institute).
Results
Baseline characteristics of the patients
Overall, 781 pediatric patients were initially enrolled in this prospective cohort. However, 18 patients were excluded on the basis of the following criteria: eight patients had undergone nephrectomy, one patient had received a kidney transplant before AKI onset, and nine patients did not meet the AKI diagnostic criteria. After applying these exclusions, 763 patients remained in the final analysis.
Table 1 summarizes the baseline characteristics of the study population. The cohort had a mean age of 6.72 years, and 442 patients (57.9%) were male. Hemato-oncologic diseases (202 patients, 26.5%) were the most common underlying condition, followed by preterm infants (128 patients, 16.8%) and kidney disease (108 patients, 14.2%). Nephrotoxic agent exposure (165 patients, 21.6%) represented the leading cause of AKI, followed by infection (135 patients, 17.7%), severe illness (127 patients, 16.6%), and dehydration (121 patients, 15.9%). The mean eGFR at baseline was 109.2 mL/min/1.73 m2, which decreased to 50.7 mL/min/1.73 m2 during AKI diagnosis. Stage 1 was the predominant stage of AKI observed. The baseline characteristics categorized by AKI etiology and age group are presented in Supplementary Tables 1 and 2 (available online), respectively. Given the clinical heterogeneity within the severe illness group, a subgroup analysis was conducted to compare leukemia-related and non-leukemia-related severe illness, as shown in Supplementary Table 3 (available online). No significant differences were observed in baseline eGFR or AKI stage distribution between these two subgroups.
Baseline characteristics of patients enrolled in cohort
Baseline characteristics of the enrolled patients according to the underlying disease
The baseline characteristics of the study population categorized by underlying disease are summarized in Table 2. Age, AKI etiology, eGFR at baseline, eGFR during AKI diagnosis, AKI stage, AKI aggravation, AKI recovery, and ICU admission showed significant differences (all P<0.001) across the underlying disease groups.
Baseline characteristics of patients enrolled in cohort according to underlying etiology
Children with CKD/kidney disease were older (mean age, 9.59 years) than those in the other groups. Among those without underlying conditions, infection (50 patients, 52.1%) was the most common AKI etiology. In contrast, among children with hemato-oncologic disease, the leading AKI etiology was nephrotoxic or contrast agent exposure (83 patients, 41.1%). In patients with CKD/kidney disease and preterm infants, the most common AKI etiologies were dehydration, circulation failure, and severe illness (101 patients, 68.2%).
Significant differences in the eGFR were observed across the groups, both at baseline (P<0.001) and during AKI diagnosis (P<0.001). Patients without underlying conditions most frequently exhibited stage 3 AKI (40 patients, 41.7%).
Patients without underlying conditions were more likely to experience AKI aggravation (25 patients, 26.0%), while their recovery rate was also high (81 patients, 84.4%). Preterm infants more frequently experienced ICU admission (10 patients, 7.81%). Significant differences in sex distribution or KRT requirement were not noted across the groups. In the preterm infant group, AKI events occurred both during the neonatal ICU (NICU) stay and after discharge, and an additional subgroup analysis comparing these two groups was conducted (Supplementary Table 4, available online). Although overall AKI etiologies did not significantly differ between the subgroups, AKI related to nephrotoxic agents tended to occur more frequently during the NICU stay.
Baseline characteristics of the enrolled patients according to the AKI stage
Table 3 presents the baseline characteristics of the children categorized by AKI stage. Significant differences in age (P<0.001), underlying disease (P<0.001), AKI etiology (P=0.004), eGFR during AKI diagnosis (P<0.001), AKI aggravation (P<0.001), and AKI recovery (P<0.001) were observed across AKI stages.
Baseline characteristics of patients enrolled in cohort according to AKI stage
The AKI stage 1 group had a higher age of participants (mean, 7.92 years old). Furthermore, the AKI stage 1 group exhibited the most frequent occurrence of hemato-oncologic disease (131 patients, 36.1%). The AKI stage 2 group had the largest proportion of preterm infants (54 patients, 25.2%), whereas the AKI stage 3 group had a predominance of patients with CKD/kidney disease (47 patients, 25.3%). The most common AKI etiologies in all stages included dehydration, circulation failure, and severe illness. In the AKI stage 3 group, 24.2% (45 patients) experienced AKI aggravation, but this group also showed a high recovery rate of 78.5% (146 patients).
Discussion
We here investigated various baseline characteristics influenced by factors, including underlying diseases and AKI stage, in a cohort of pediatric patients with AKI. Hemato-oncologic diseases were the most common underlying condition, and nephrotoxic agent exposure was the leading AKI etiology. The most frequently observed AKI stage was stage 1.
AKI etiologies vary across studies, reflecting differences in healthcare settings and patient populations [11]. In several high-income settings, cardiopulmonary bypass, sepsis, heart failure, organ transplantation, tumor lysis syndrome, and nephrotoxic agent exposure constitute the leading causes of AKI [12]. However, as this study was conducted at a single tertiary care center with a dedicated hemato-oncology unit, the most common preexisting condition was hemato-oncologic diseases, with nephrotoxic agent exposure being the primary AKI etiology. This finding is consistent with those of previous studies indicating that nephrotoxic agent exposure can significantly cause AKI in children with hemato-oncologic disorders [13,14]. In our cohort, the most common cause of AKI in children without underlying conditions was infection. However, previous studies have reported that children with conditions, including diabetes mellitus, organ transplantation, human immunodeficiency virus, urinary tract anomalies, and malnutrition, exhibit a higher prevalence of sepsis-related AKI [15]. In preterm infants, the most frequent AKI etiologies encompassed dehydration, circulatory failure, and severe illness. Previous meta-analyses have identified perinatal asphyxia, sepsis, patent ductus arteriosus, necrotizing enterocolitis, and nephrotoxic medications as major risk factors for neonatal AKI [16]. These results underscore the diverse etiologies of AKI in children with different underlying conditions, stressing the significance of tailored monitoring and management to address the primary causes of AKI in each patient population.
In our cohort, the majority of patients presented with stage 1 AKI, which agrees with most studies on pediatric AKI. The AWARE study reported that 26.9% of children with critical illness developed AKI within 1 week of pediatric ICU admission, with 11.6% experiencing stage 2 or 3 AKI [5]. Similarly, the AWAKEN study revealed a 29.9% overall AKI incidence in neonatal ICUs, with stages 1, 2, and 3 occurring in 13.9%, 7.1%, and 8.9% of the cases, respectively [3]. However, a single-center observational study reported stage 3 as the most prevalent (44.7%), followed by stages 2 (31.0%) and 1 (24.2%) [17]. In our cohort, the distribution of AKI stages varied according to the underlying conditions. Patients without preexisting diseases were more likely to present with stage 3 AKI, whereas those with hemato-oncologic disorders or kidney disease were predominantly diagnosed with stage 1. This pattern may be attributed to delayed recognition and referral in previously healthy children, who are less likely to undergo routine kidney function monitoring and are often referred after unsuccessful management at primary or secondary healthcare facilities. In contrast, children with chronic underlying conditions are typically monitored more closely, and clinicians maintain a higher index of suspicion for AKI, facilitating earlier detection and diagnosis [17].
The progression and recovery patterns of AKI significantly vary across different patient groups and AKI etiologies, with significant implications for clinical management and outcomes. Preterm infants exhibited markedly high kidney function recovery rates, consistent with previous research indicating that all preterm infants recovered normal kidney function before hospital discharge following AKI [18]. Patients with stage 3 AKI most frequently exhibited AKI aggravation; however, in our study, this group also demonstrated the highest recovery rates. Moreover, children without underlying conditions showed the highest rates of both AKI aggravation and recovery. The higher incidence of stage 3 AKI in patients without preexisting conditions may contribute to this finding, likely reflecting the predominance of reversible etiologies such as infection. However, severe AKI is a significant risk factor for progression to acute kidney disease, defined as any abnormality in kidney structure or function spanning <3 months following AKI [19]. To better understand AKI outcomes, further analysis is warranted, particularly by adjusting for potential confounding factors that may impact disease progression and recovery patterns.
Despite the valuable insights provided by this study, several limitations should be acknowledged. First, as this study was conducted at a single tertiary center, the results may not be generalizable to all pediatric populations, particularly those in different geographic regions or healthcare settings. Second, the ability to assess the prevalence and characteristics of AKI across all hospitalized patients was hindered by the absence of a comparative cohort of hospitalized children without AKI. Lastly, the AKI aggravation and recovery-associated factors were not adjusted for potential confounders, as only their distribution was analyzed.
Notwithstanding its limitations, this study holds significance as a prospective cohort that encompasses general hospitalized pediatric patients rather than focusing on high-risk groups, offering a more comprehensive understanding of pediatric AKI in Korea. These findings highlight the intricacy of pediatric AKI and emphasize the significance of personalized management approaches considering patient-specific factors and AKI etiologies. Future studies should prioritize investigating long-term renal outcomes, integrating more comprehensive data on proteinuria, hypertension, and other relevant risk factors.
Supplementary Material
Supplementary Tables 1 to 4 can be found via https://doi.org/10.3339/ckd.25.014.
Baseline characteristics of patients enrolled in cohort according to AKI etiology
Baseline characteristics of patients enrolled in cohort according to age group
Comparison of clinical characteristics between leukemia-related and non-leukemia-related severe illness AKI
Comparison of preterm infant AKI occurring during NICU stay versus after discharge
Notes
Ethical statements
The Institutional Review Board of Seoul National University Hospital approved this study (IRB number: 1602-010-739). Before study participation, written informed consent was obtained from patients’ caregivers or the patients themselves, depending on the age.
Conflicts of interest
Hee Gyung Kang is an editorial board member of the journal but was not involved in the peer reviewer selection, evaluation, or decision process of this article. There are no other potential conflicts of interest relevant to this article.
Funding
This study was supported by a grant from the Seoul National University Hospital (SNUH) Research Fund (grant no. 2520160020).
Acknowledgments
The authors thank all the subjects for their participation in this study.
Author contributions
Conceptualization: YHA, HGK
Data curation: NC
Formal analysis: NP, NC, JK
Investigation: NP, NC, JK
Methodology: YHA, NC, JK
Writing-original draft: NC
Writing-review & editing: NC, HGK, NP, JK, YHA
All authors have read and approved the final manuscript.
Data availability statement
All data used for analysis are shown in the tables in this article. Data sharing is applicable to this article if requested by other investigations for replicating the results.