key: cord-0874742-b454mk2f
authors: Van den Eynde, Jef; Salaets, Thomas; Louw, Jacoba J.; Herman, Jean; Breysem, Luc; Vlasselaers, Dirk; Desmet, Lars; Meyns, Bart; Budts, Werner; Gewillig, Marc; Mekahli, Djalila
title: Persistent Markers of Kidney Injury in Children Who Developed Acute Kidney Injury After Pediatric Cardiac Surgery: A Prospective Cohort Study
date: 2022-03-18
journal: J Am Heart Assoc
DOI: 10.1161/jaha.121.024266
sha: eb94bf719e13d8c9f440eaedebcd04a47bd257da
doc_id: 874742
cord_uid: b454mk2f

BACKGROUND: Acute kidney injury (AKI) after pediatric cardiac surgery is common. Longer‐term outcomes and the incidence of chronic kidney disease after AKI are not well‐known. METHODS AND RESULTS: All eligible children (aged <16 years) who had developed AKI following cardiac surgery at our tertiary referral hospital were prospectively invited for a formal kidney assessment ≈5 years after AKI, including measurements of estimated glomerular filtration rate, proteinuria, α(1)‐microglobulin, blood pressure, and kidney ultrasound. Longer‐term follow‐up data on kidney function were collected at the latest available visit. Among 571 patients who underwent surgery, AKI occurred in 113 (19.7%) over a 4‐year period. Fifteen of these (13.3%) died at a median of 31 days (interquartile range [IQR], 9–57) after surgery. A total of 66 patients participated in the kidney assessment at a median of 4.8 years (IQR, 3.9–5.7) after the index AKI episode. Thirty‐nine patients (59.1%) had at least 1 marker of kidney injury, including estimated glomerular filtration rate <90 mL/min per 1.73 m(2) in 9 (13.6%), proteinuria in 27 (40.9%), α(1)‐microglobinuria in 5 (7.6%), hypertension in 13 (19.7%), and abnormalities on kidney ultrasound in 9 (13.6%). Stages 1 to 5 chronic kidney disease were present in 18 (27.3%) patients. Patients with CKD were more likely to have an associated syndrome (55.6% versus 20.8%, P=0.015). At 13.1 years (IQR, 11.2–14.0) follow‐up, estimated glomerular filtration rate <90 mL/min per 1.73 m² was present in 18 of 49 patients (36.7%), suggesting an average estimated glomerular filtration rate decline rate of −1.81 mL/min per 1.73 m² per year. CONCLUSIONS: Children who developed AKI after pediatric cardiac surgery showed persistent markers of kidney injury. As chronic kidney disease is a risk factor for cardiovascular comorbidity, long‐term kidney follow‐up in this population is warranted.

cute kidney injury (AKI) in the immediate postoperative period is common, occurring in 30% to 60% of children undergoing pediatric cardiac surgery for congenital heart disease (CHD). A recent meta-analysis demonstrated that AKI after pediatric cardiac surgery was associated with increased shortterm morbidity and mortality, including higher rates of in-hospital mortality, need for kidney replacement therapy, and cardiac arrhythmias, as well as longer ventilation time and hospital length of stay. 1 Longer-term outcomes of AKI after pediatric cardiac surgery, however, are not well known. Adult studies have found robust associations of cardiac surgery-associated AKI with a 9-fold increased risk of developing chronic kidney disease (CKD) and a 2-fold increased risk of long-term mortality when compared with those without AKI. 2 In contrast, prior studies in children with CHD have been inconclusive with regard to the risk of persistent kidney injury and CKD and there is a scarcity of data in this population. [3] [4] [5] [6] [7] [8] While CHD was a lethal condition several decades ago, >90% of infants with CHD currently survive into adulthood. 9 The rising number of adults with CHD (ACHD) now accounts for two thirds of the CHD population. 10 As early mortality continues to decrease, it becomes more clear that CHD is never cured and results in multiple complications accumulated over the lifespan. 11 CKD is an important issue in the growing ACHD population, occurring in 30% to 50% of these patients 12 and being responsible for an excess burden in health care utilization. 13 Furthermore, as CKD represents an important cardiovascular risk factor, it needs to be managed and followed up as part of the standard care of this population. 14 Currently, AKI guidelines in children offer no recommendations on structured kidney follow-up. Therefore, this prospective cohort study investigated the mid-and long-term kidney consequences of AKI after pediatric cardiac surgery.

The data that support the findings of this study are available from the corresponding author on reasonable request.

This study conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the local ethical committee of the University Hospitals Leuven (project number S52045). All consecutive children (aged <16 years) with CHD who had developed AKI following cardiac surgery with extracorporeal circulation between December 2004 and December 2008 were identified through the local Pediatric Cardiology database and the hospitals electronic database. AKI was defined according to the Acute Kidney Injury Network (AKIN) classification 15 as an increase in serum creatinine ≥0.3 mg/dL or ≥50% within 48 hours after surgery. AKI was further subdivided into stage 1 (increase of ≥0.3 mg/dL or 50%-100%), stage 2 (increase of 100%-200%), and stage 3 (increase of ≥4 mg/dL with an acute increase of ≥0.5 mg/dL, or >200%, or requiring renal replacement therapy). AKIN was the leading classification for AKI at the time of protocol validation of this study (ie, June 2010). Demographic, perioperative, and postoperative data were retrieved from information available in their electronic medical records, as well as hospitalizations and outpatient consultations.

CHD was classified into the following types: intracardiac left-to-right shunts (atrial septal defect, ventricular septal defect, atrioventricular septal defect), obstructive left heart lesions (coarctation of the aorta, aortic stenosis, small left hearts, Shone complex, and other lesions with evolution to biventricular circulation), transposition of the great arteries ("simple" types, with or without ventricular septal defect), conotruncal What Are the Clinical Implications?

• Our findings provide incremental evidence that persistent markers of kidney injury are common after postoperative acute kidney injury in children with congenital heart disease at mid-and long-term follow-up. • Children may develop absolute or relative hyperfiltration in response to acute kidney injury, thus initially "masking" the injury; however, kidney injury progresses and eventually becomes evident, especially in those with syndromes or univentricular physiology. • As chronic kidney disease, proteinuria, and hypertension are important cardiovascular risk factors, structured follow-up of the kidneys in this vulnerable population is warranted and prevention of kidney disease should start in childhood to ensure optimal outcomes in the growing population of adults with congenital heart disease. 

Patients who had developed AKI after pediatric cardiac surgery and were still alive were contacted by mail and invited for a formal kidney assessment between June 2010 and December 2012. Patients who did not respond were contacted by telephone. Informed consent was obtained from the parents or caregivers of all patients. All assessments were performed on an outpatient basis on the same day as a planned cardiology consultation.

During the study visit, height, weight, and vital signs were collected and a full physical examination was performed. Blood pressure was measured at the upper extremity using an automated Dinamap monitor and cuff of appropriate size. The lowest of 3 readings taken 5 minutes apart was recorded. Hypertension was defined as systolic and/or diastolic blood pressure >95th percentile for sex, age, and height. 17 Children with previously diagnosed hypertension who were on antihypertensive therapy were also considered to have hypertension.

Kidney outcome was assessed by means of estimated glomerular filtration rate (eGFR), proteinuria, α 1 -microglobinuria, and kidney ultrasound. Serum creatinine was determined and eGFR was estimated using the 2009 Schwartz formula (eGFR=0.413×height

[cm]/creatinine [mg/dL]). 18 A clean catch or bagged urine sample was collected. Proteinuria was defined as a urine protein/creatinine ratio >0.2 g/g. α 1 -Microglobulinuria, an indicator of proximal tubular dysfunction, was defined as a urinary level of α 1microglobuline >12.5 mg/L. Kidney ultrasound with Doppler was performed to assess kidney length, aspect of kidney parenchyma, and kidney perfusion. Kidney length was standardized to height-matched normal ranges. 19 A kidney was considered small or large if its length was inferior or superior to 2 SDs compared with controls.

The primary objective of the study was to determine midterm kidney outcomes of AKI after pediatric cardiac surgery. CKD was defined according to Kidney Disease Improving Global Outcomes (KDIGO) guidelines: eGFR 90 mL/min to 100 mL/min per 1.73 m² with abnormalities on kidney ultrasound or biochemistry (stage 1), or eGFR <90 mL/min per 1.73 m² (stages 2-5). 20 Hyperfiltration was defined as an eGFR >140 mL/min per 1.73 m². A secondary objective was to evaluate the progressive decline in kidney function at long-term follow-up. To this end, the latest occasional eGFR measurement available from electronic medical records was evaluated. Most blood samples were obtained during routine checkups before catheterization. Samples taken at the time of intercurrent acute disease were excluded.

Continuous variables were checked for normality using Shapiro-Wilk test. Normally distributed variables are presented as mean±SD and were compared using Student t test; non-normally distributed variables are presented as median (interquartile range [IQR]) and were compared using Mann-Whitney U test. Categorical variables are expressed as frequency (percentage) and were compared with chi-square test. All analyses were performed using R Statistical Software (version 4.0.2 2020-06-22, Foundation for Statistical Computing). A 2-tailed P<0.05 was considered statistically significant.

A total of 640 cardiac operations in 571 children with CHD were performed at our tertiary referral hospital between December 2004 and December 2008. AKI occurred following 122 procedures (19.1%) in 113 patients (19.7%). The patients underwent the index operation at a median age of 79.0 days (IQR, 9.0-230 days) and 68 (60.2%) were male. Nine patients experienced a second AKI episode between this index operation and the kidney assessment visit. According to their highest AKIN classification, 64 patients (56.6%) had stage 1 AKI, 40 (35.4%) had stage 2 AKI, and 10 (8.8%) had stage 3 AKI. Renal replacement therapy was required in 2 patients (1.6%), one for a period of 45 days and another for 4 days. Fifteen patients (13.3%) died at a median of 31.0 days (IQR, 9.0-57.0 days) after surgery and at a median age of 93.0 days (IQR, 56.0-211 days). The cause of death was cardiorespiratory failure in 11 patients (80.0%) and severe infections in 3 patients (20.0%).

The 98 surviving patients were contacted to participate in a formal kidney assessment. Thirteen patients (13.3%) declined and 19 (19.4%) were lost to follow-up, mostly attributable to further follow-up abroad. Thus, a total of 66 children were included in the current study ( Figure 1 ). The study visits were performed at a median follow-up of 4.8 years (IQR, 3.9-5.7 years) after the index AKI episode. At that time, the median age was 5.6 years (IQR, 4.1-6.4 years). Patient demographics are presented in Table 1 . Patients included in the prospective cohort did not differ from those who were not included, apart from lower birth weight (3030±652 g versus 3381±614 g, P=0.012) and higher number of CHD surgeries (median, 2 [IQR, 1-2] versus 1 [IQR, 1-1], P=0.002) in the former.

Findings at the kidney assessment visit are presented in Table 2 . The patients had a mean length of 110±12. 

The mean eGFR at the formal assessment visit was 114±25 mL/min per 1.73 m². Hyperfiltration was observed in 8 children (12.1%). Eight children (12.1%) had a mildly decreased eGFR between 60 and 90 mL/min per 1.73 m 2 (CKD stage 2) and 1 child (1.5%) had an eGFR of 51 mL/ min per 1.73 m 2 (CKD stage 3a). Two of these patients had known kidney hypodysplasia and reduced kidney function since the neonatal period and their eGFR at the kidney assessment visit was comparable with their baseline eGFR before the index surgery. There were no differences in kidney function between AKIN stages (P=0.474). AKI indicates acute kidney injury; and CHD, congenital heart disease.

Proteinuria and α 1 -Microglobinuria Proteinuria was identified in 27 patients (40.9%). One child with known kidney hypodysplasia and reduced kidney function had overt proteinuria (2.1 g/g); the proteinuria was in the same range as before the AKI surgery. Proteinuria did not differ between AKIN stages (P=0.146). Urinary α 1 -microglobulin concentration was increased in 5 patients (7.6%), indicating tubular dysfunction. Three patients had associated proteinuria.

Hypertension was detected in 13 patients (19.6%): 9 had hypertension (including 3 patients who were taking antihypertensive treatment) and 4 were normotensive but taking antihypertensive medications for blood pressure control. There was no difference between AKIN stages with regard to hypertension (P=0.477).

Kidney ultrasound with Doppler detected abnormalities in 9 patients (13.6%). Five of these children had preexisting congenital anomalies of the kidney and urinary tract, which had already been documented on ultrasound during the neonatal period. Kidney intraparenchymal acceleration times and resistance indices were within normal ranges in all patients.

Of the 66 patients who came to the kidney assessment visit, 39 (59.1%) had at least 1 marker of kidney injury (ie, reduced kidney function, proteinuria, α 1microglobinuria, abnormalities on ultrasound, and/or hypertension). Nineteen (28.8%) had 1 positive marker of kidney injury, 17 (25.8%) had 2 positive markers, and 3 (4.5%) had abnormalities in 3 markers. 

Between the formal kidney assessment and latest follow-up, 2 patients died: one 13.1-year-old patient at 7.5 years after the assessment attributable to sudden collapse after Potts shunt for severe pulmonary hypertension, and another 17.2-year-old patient at 10.0 years after the assessment following complications of COVID-19 in a failing Fontan circulation. In a total of 49 of the 66 patients originally included in the assessment, kidney function tests were obtained from the latest available visit after the formal assessment, at a median follow-up of 13.1 years (IQR, 11.2-14.0 years) after the index AKI episode. The median age was 13.6 years (IQR, 12.2-15.3 years). At that time, the mean eGFR had declined to 99±23 mL/min per 1.73 m² (P=0.001 compared with 114±25 mL/ min per 1.73 m² at the kidney assessment visit) (Figure 2A ). This suggests an average eGFR decline rate of −1.81 mL/min per 1.73 m² per year. If the decline would proceed at the same rate, it is predicted that 50% of our patients will have developed an eGFR <90 mL/min per 1.73 m² by early adulthood (at 18.1 years of follow-up), which corresponds to data in ACHD. 12 Hyperfiltration was observed in only 1 child Data are presented as frequency (percentage), mean±SD, or median (interquartile range). P value is given for the comparison between Acute Kidney Injury Network (AKIN) stages. BMI indicates body mass index; CAKUT, congenital anomalies of the kidney and urinary tract; and eGFR, estimated glomerular filtration rate.

(2.0%, P=0.047). A total of 11 of 49 patients (22.4%) who had normal kidney function at the assessment visit eventually developed kidney dysfunction ( Figure 2B ). Thus, 17 children (34.7%, P=0.008) had a mildly decreased eGFR between 60 mL/min per 1.73 m 2 and 90 mL/min per 1.73 m 2 (CKD stage 2) and Data are presented as frequency (percentage), mean±SD, or median (interquartile range). AKI indicates acute kidney injury; AKIN, Acute Kidney Injury Network; CAKUT, congenital anomalies of the kidney and urinary tract; CHD, congenital heart disease; CPB, cardiopulmonary bypass; ICU, intensive care unit; NA, not available; STAT, Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery; and TAPVR, totally anomalous pulmonary venous return.

*Standardized mean differences (SMDs) are presented for chronic kidney disease (CKD) compared with no CKD.

1 child (2.0%, P=0.832) had an eGFR of 42 mL/min per 1.73 m 2 (CKD stage 3b; this was the same child who had an eGFR of 51 mL/min per 1.73 m² at the kidney assessment visit). Apart from this one child, none of the patients were in nephrology follow-up.

In this cohort study, 66 children with CHD who had developed AKI after cardiac surgery underwent a formal kidney assessment 5 years after the index event ( Figure 3 ). Our findings at 5-year follow-up revealed that 59.1% of patients had at least 1 marker of kidney injury, including reduced kidney function (eGFR <90 mL/min per 1.73 m²) in 13.6%, proteinuria in 0.9%, α 1 -microglobinuria in 7.6%, hypertension in 19.7%, and abnormalities on kidney ultrasound in 13.6%. CKD stages 1 to 5, defined as eGFR <90 mL/min per 1. 

A number of studies have examined kidney outcomes during follow-up of children who developed AKI after pediatric cardiac surgery. [3] [4] [5] [6] [7] Most of these have focused on the prevalence of CKD at variable lengths of followup and using different definitions. In a population of children undergoing heart transplantation, Hollander et al 4 reported a 6-and 12-month prevalence of CKD stages 3 to 5 (eGFR <60 mL/min per 1.73 m²) of 5% and 6% of patients, respectively. At a median of 1 year after AKI during the Fontan completion, CKD (defined in this study as eGFR <80 mL/min per 1.73 m²) was observed in 11% of the cohort by Esch et al. 5 In the multicenter TRIBE-AKI (Translational Research Investigating Biomarker Endpoints in AKI) study, 7 6% of children with AKI had CKD stages 2 to 5 (<90 mL/min per 1.73 m²) at 5-year follow-up. Using Danish regional population-based registries, Madsen et al 3 found that CKD stages 2 to 5 (eGFR <90 mL/min per 1.73 m²) occurred in 14% of patients at 5 years after AKI following surgery for CHD. Finally, Huynh et al 6 evaluated children 6 years after neonatal cardiac surgery and found CKD (defined as eGFR <90 mL/ min per 1.73 m² or albumin/creatinine ≥3 mg/mmol) in 9.5% of patients. The prevalence of CKD stages 2 to 5 (eGFR<90 mL/min per 1.73 m²) in our study (13.6%, 9 of 66 at 5-year follow-up) is in line with these prior investigations. In contrast, however, the comprehensive assessment in our study also allowed us to consider CKD stage 1 (eGFR 90-100 mL/min per 1.73 m² with abnormalities on kidney ultrasound or biochemistry), showing an even greater overall burden of CKD (27.3%, 18 of 66). Notably, a relatively high proportion of children with hyperfiltration was observed in our study (12.1%). Following the early landmark experimental studies by Brenner and colleagues, 21 hyperfiltration became recognized as a mechanism where the loss of functional nephrons leads to maladaptive hemodynamic changes, which increase glomerular capillary pressure and elevate single-nephron glomerular filtration rate. This has been observed in diabetes, solitary or remnant kidneys, and various forms of acquired kidney disease. Similarly, it has been suggested by Greenberg et al 7 that children may develop absolute or relative hyperfiltration in response to AKI, thus initially "masking" the injury. In support of this idea, Cooper et al 8 revealed that biomarkers of kidney injury such as interleukin 18 and liver-type fatty acidbinding protein remained elevated 7 years after AKI even in the absence of conventional evidence of CKD. This might explain why most previous studies (with length of follow-up of ≤7 years) in children with CHD could not reveal an association of AKI with the development of CKD, [5] [6] [7] [8] while studies in adults have established AKI as a risk factor for CKD with much shorter follow-up. 2 Further expanding on these observations, our study revealed a progressive decline in eGFR between 5 and 13 years of follow-up in our study. During this period, the group with hyperfiltration decreased (from 12.1% to 2.0%) and the group with CKD stages 2 to 5 increased (from 13.6% to 36.7%). Interestingly, the latter percentage is closer to the 30% to 50% reported in ACHD. 12 Taken together, these results suggest that the excess burden of CKD in ACHD might already start in childhood. While initially masked by hyperfiltration, kidney injury progresses and eventually becomes evident later (in adolescence or adulthood).

The prevalence of hypertension in our study (19.7%) was comparable to that found in the multicenter TRIBE-AKI study 7 (16.8%) and single-center FRAIL-AKI (Follow-Up Renal Assessment of Injury Long-Term After Acute Kidney Injury) study 8 (17.6%) . This is ≈5-fold higher than Van den Eynde et al AKI After Cardiac Surgery the prevalence in children of the same age in the general population (4.3%). 22 While hypertension is a common presentation of (re)coarctation of the aorta, 23 only 2 of 5 children with coarctation in our cohort were hypertensive. The cause of hypertension was most likely multifactorial, including enhanced sympathetic activity and upregulation of other neurohormonal pathways, 24 in addition to underlying CKD. Proteinuria has been previously reported in CHD and has been associated mainly with cyanotic CHD, Fontan circulation, shunts lesions complicated by pulmonary hypertension, and conditions with a systemic right ventricle. 25 The pathophysiology might include cyanosis, erythrocytosis, limited perfusion, and kidney congestion. Notably, albuminuria is associated with major adverse outcomes in ACHD. 26 We demonstrated that even α 1 -microglobulinuria is seen in 7.6%, which suggests an associated proximal tubular dysfunction. 27 The present study revealed that both abnormalities may already be present in childhood.

One potential explanation for the presence of CKD in this population is persistent damage after AKI resulting from aortic clamping or hemodynamic imbalances during cardiopulmonary bypass. 28 However, we did not observe any differences in cardiopulmonary bypass or cross clamp time in those who developed CKD and those who did not. Although not measured in this study, cardiopulmonary bypass can lead to fluid shifts, increased blood vessel permeability, and depressed myocardial contractility, resulting in fluid overload. 29 The latter has been demonstrated to be a factor that may incite and worsen AKI, but can also independently affect morbidity and mortality. 30 Strikingly, there were also no differences in kidney outcomes between AKIN stages in our study, in contrast to adult studies where the risk of CKD increased in a graded fashion when AKI was more severe. 2 While the hyperfiltration theory In this cohort study, 66 children with congenital heart disease (CHD) who had developed acute kidney injury (AKI) after pediatric surgery underwent a formal kidney assessment 5 years after the index event. Our findings revealed that reduced kidney function (estimated glomerular filtration rate [eGFR] <90 mL/min per 1.73 m²) was present in 9 patients (13.6%), proteinuria in 27 patients (40.9%), α 1 -microglobinuria in 5 patients (7.6%), hypertension in 13 patients (19.7%), abnormalities on kidney ultrasound in 9 patients (13.6%), and chronic kidney disease (CKD) in 18 patients (27.3%). Occasional eGFR measurements obtained in 49 children 13 years after the index event revealed ongoing kidney function deterioration. These results suggest that persistent markers of kidney injury are common after postoperative AKI in children with CHD and warrant the initiation of structured kidney follow-up in this patient population.

Van den Eynde et al AKI After Cardiac Surgery might help explain why such relationships between AKI and CKD are only observed at an older age, it should also be considered that several other factors, apart from the index AKI episode, can contribute to the kidney phenotype in children with CHD. In our study, a significant association of CKD with syndromic diagnoses was revealed. It is known that certain mutations underlying syndromes such as trisomy 21 and 22q11.2 microdeletions might lead to both CHD and dysfunction of other organs. 31 Of note, 5 of 6 patients (83.3%) with trisomy 21 and 1 of 4 patients (25%) with 22q11.2 microdeletions in our cohort had CKD. Studies have identified genetic links between birth defects of the kidney and CHD. 32 Furthermore, CHD itself can alter hemodynamics and put pressure on the kidneys, both preoperatively (eg, ischemic hit in neonates with coarctation) and postoperatively (eg, venous congestion after Fontan completion). In the cohort by Huynh et al, 6 cyanosis postoperatively was the only independent predictor of kidney dysfunction, while Esch et al 5 reported that 25% of patients already had kidney dysfunction before Fontan completion. In addition, several studies have described a "cyanotic nephropathy" characterized by proteinuria. 33 Venous congestion, as seen in patients with univentricular hearts or those with pulmonary hypertension, atrial fibrillation, neurohormonal activation, endothelial dysfunction, and drug and procedure toxicity among others, have all been related to kidney dysfunction in both acquired and congenital cardiovascular diseases. 34 Univentricular heart defects were 2-fold more common among patients with CKD in our cohort (38.9% in patients with CKD versus 18.8% in patients with no CKD). Finally, also intermittent catheterizations (using potentially nephrotoxic contrast agents) and surgical interventions (other than the index surgery) might contribute to progressive kidney function decline.

The finding that persistent markers of kidney injury are common in patients who developed AKI after pediatric cardiac surgery does not necessarily imply that these observations are entirely attributable to the effect of AKI alone. This is an important consideration, because a causal relationship between AKI and CKD progression cannot be directly inferred from our study, as suggested above. In fact, prior studies in children have been unable to consistently demonstrate such a relationship in the pediatric population. [3] [4] [5] [6] [7] [8] There might be several reasons for why this is different from findings in adults, 2 including adaptive capacities of the kidneys in children (cfr. hyperfiltration) and the need for relatively longer follow-up to find a significant effect, or the possibility that other factors might play a more important role in driving kidney injury in these patients. Regardless of this discussion, our findings clearly demonstrate the large extent of kidney disease in this vulnerable population. Therefore, arguably more important than trying to tease out the exact contribution that AKI has in CKD progression, we should strive towards a better overall understanding of context wherein these patients find themselves at elevated risk of poor kidney health. AKI might only be one episodic piece of the puzzle that drives kidney injury in these patients, along with various other factors along the lifespan.

Need for Structured Kidney Follow-Up in Children With CHD Except for one, none of the children in our cohort were seen or followed up by a nephrologist, despite sometimes clear kidney function deterioration in early adolescence. Detection and treatment of CKD, proteinuria, and hypertension in children is critical because these are risk factors for cardiovascular diseases and progressive kidney damage in adults. 35, 36 Patients with CHD are at especially high risk for long-term cardiac events, and the superposition of comorbidities such as coronary artery disease, diabetes, smoking, and hypertension in adulthood leads to an excess of health care utilization in ACHD. 13 This study, showing progressive kidney injury arising as early as in childhood, highlights the need for long-term and structured kidney follow-up after pediatric cardiac surgery in general and following AKI in particular. 37 During nephrology visits, patients and families should be educated about kidney dysfunction and its risk factors, lifestyle changes should be discussed, medications should be reviewed, and screening along with timely initiation of treatment should be pursued. The ideal window of opportunity is in the pediatric cardiology practice, ie, before any decline in kidney function, but optimization of the transition from pediatric cardiology to ACHD will be at least equally essential to ensure continuity of care. 38 With the emergence of novel, more sensitive biomarkers such as cystatin C and neutrophil gelatinase-associated lipocalin, detection and monitoring of kidney injury as well as personalized approaches to the prevention and treatment of kidney dysfunction will become available. 39, 40 Supporting the effectiveness of structured kidney follow-up, a study by Harel et al 41 demonstrated that mortality can be reduced if adult survivors of severe AKI are seen by a nephrologist within 90 days of discharge (8.4 versus 10.6 per 100 patient-years). Translating these insights into a coordinated care model, Ly et al 42 reported their piloting experience with the implementation of AKI clinics and similarly found reductions in rehospitalizations, morbidity, and mortality. Certainly, these studies pave the way for a more integrated approach towards AKI. In a similar manner, our present study proposes the need for a lifespan approach to kidney health for those who underwent cardiac surgery Van den Eynde et al AKI After Cardiac Surgery in childhood. This is in line with a growing understanding that CHD is never cured or reversed, but requires a lifetime of treatment and careful monitoring. 43 

It should be noted that this study was performed in a select population of patients with AKI after cardiac surgery with extracorporeal circulation, which is not necessarily representative of the whole CHD population. Only 25% of all people living with CHD undergo surgery in childhood, 44 and caution should be made in the extrapolation of our results. Of note, the number of CHD surgeries was higher in the patients included in the formal kidney assessment compared with those who were not. As a second limitation, we did not include a control group without AKI such that the exact contribution of AKI to kidney dysfunction could not be estimated; this was, however, not the main focus of the present study and has been investigated by other studies, as discussed above. [3] [4] [5] [6] [7] [8] Furthermore, no direct relationship between the kidney phenotype and the index surgery can be assumed, as most children underwent other surgeries, interventions, and cardiac catheterizations. The subgroup analyses should also be interpreted with caution, given limitations in statistical power.

This study revealed that AKI after pediatric cardiac surgery in children with CHD is common and that persistent and progressive markers of kidney injury are present in these children at mid-and long-term follow-up. Especially those with syndromes or univentricular physiology might be at risk. As the data on long-term kidney follow-up in children with CHD are scarce, no current recommendations are available regarding the follow-up of CKD, proteinuria, and hypertension. Based on our findings, we suggest a structured kidney follow-up in this vulnerable population and recommend that prevention of kidney disease should start in childhood to ensure optimal outcomes in the growing population of ACHD.

In-hospital outcomes of acute kidney injury after pediatric cardiac surgery: a metaanalysis

Chronic kidney disease after acute kidney injury: a systematic review and meta-analysis

Cardiac surgery in patients with congenital heart disease is associated with acute kidney injury and the risk of chronic kidney disease

Recovery from acute kidney injury and CKD following heart transplantation in children, adolescents, and young adults: a retrospective cohort study

Acute kidney injury after Fontan completion: risk factors and outcomes

Follow-up after neonatal heart disease repair: watch out for chronic kidney disease and hypertension!

Kidney outcomes 5 years after pediatric cardiac surgery the TRIBE-AKI study

Follow-up renal assessment of injury long-term after acute kidney injury (FRAIL-AKI)

Temporal trends in survival to adulthood among patients born with congenital heart disease from 1970 to 1992 in Belgium

Lifetime prevalence of congenital heart disease in the general population from

2020 ESC Guidelines for the management of adult congenital heart disease

Chronic kidney disease in congenital heart disease patients: a narrative review of evidence

Vonder Muhll IF, Majeed A. Comorbidity, healthcare utilisation and process of care measures in patients with congenital heart disease in the UK: crosssectional, population-based study with case-control analysis

Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization

Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury

An empirically based tool for analyzing mortality associated with congenital heart surgery

Summary of the fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents

Which creatinine and cystatin C equations can be reliably used in children?

Sonographic measurements and appearance of normal kidneys in children

Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline

The hyperfiltration theory: a paradigm shift in nephrology

Global prevalence of hypertension in children: a systematic review and metaanalysis

Diagnosis, imaging and clinical management of aortic coarctation

Neurohormonal activation and the chronic heart failure syndrome in adults with congenital heart disease

Serum and 24-hour urine analysis in adult cyanotic and noncyanotic congenital heart disease patients

Association of albuminuria with major adverse outcomes in adults with congenital heart disease results from the boston adult congenital heart biobank

Alpha-1-microglobulin: an indicator protein for renal tubular function

Higher total ultrafiltration volume during cardiopulmonary bypassassisted infant cardiac surgery is associated with acute kidney injury and fluid overload

The inflammatory response to cardiopulmonary bypass: part 1-mechanisms of pathogenesis

Fluid overload independent of acute kidney injury predicts poor outcomes in neonates following congenital heart surgery

Genomic frontiers in congenital heart disease

Genetic link between renal birth defects and congenital heart disease

Prevalence and associated factors of renal dysfunction and proteinuria in cyanotic congenital heart disease

Renal dysfunction in cardiovascular diseases and its consequences

Childhood to early-midlife systolic blood pressure trajectories: early-life predictors, effect modifiers, and adult cardiovascular outcomes

Advanced coronary and carotid arteriopathy in young adults with childhood-onset chronic renal failure

Is it time for a multi-specialty approach to cardio-renal dysfunction in children with cyanotic congenital heart disease?

Transition to adulthood and transfer to adult care of adolescents with congenital heart disease: a global consensus statement of the ESC Association of Cardiovascular Nursing and Allied Professions (ACNAP), the ESC Working Group on Adult Congenital Heart Disease (WG ACHD), the Association for European Paediatric and Congenital Cardiology (AEPC), the Pan-African Society of Cardiology (PASCAR), the Asia-Pacific Pediatric Cardiac Society (APPCS), the Inter-American Society of Cardiology (IASC), the Cardiac Society of Australia and New Zealand (CSANZ), the International Society for Adult Congenital Heart Disease (ISACHD), the World Heart Federation (WHF), the European Congenital Heart Disease Organisation (ECHDO), and the Global Alliance for Rheumatic and Congenital Hearts (Global ARCH)

Biomarkers in acute kidney injury

Strategies to prevent acute kidney injury after pediatric cardiac surgery a network meta-analysis

Nephrologist follow-up improves allcause mortality of severe acute kidney injury survivors

Characteristics and outcomes of survivors of critical illness and acute kidney injury followed in a pilot acute kidney injury clinic

Lifespan perspective on congenital heart disease research: JACC state-of-the-art review

Heart disease and stroke statistics-2021 update: a report from the

We would like to express our words of gratitude to all of the patients and their parents for partaking in this study, as well as the peripheral house doctors and pediatricians for collecting data. A special word of thanks to D. Thijs and secretaries A. Freys and A. Vloemans of pediatric cardiology and K. Rowan of pediatric nephrology for contacting patients and documenting data. J. Van den Eynde was supported by a Fellowship of the Belgian American Educational Foundation.Author contributions: Van den Eynde: conceptualization; data curation; formal analysis; investigation; methodology; project administration; software; validation; visualization; writing-original draft; and writing-review and editing. Salaets: conceptualization; data curation; investigation; supervision; and writing-review and editing. Louw: conceptualization; investigation; and supervision; writing-review and editing. Herman: conceptualization; investigation; supervision; and writing-review and editing. Breysem: conceptualization; investigation; supervision; and writing-review and editing. Vlasselaers: conceptualization; investigation; supervision; and writingreview and editing. Desmet: conceptualization; investigation; supervision; and writing-review and editing. Meyns: conceptualization; investigation; supervision; and writing-review and editing. Budts: conceptualization; investigation; supervision; and writing-review and editing. Gewillig: conceptualization; investigation; supervision; and writing-review and editing. Mekahli: conceptualization; investigation; supervision; and writing-review and editing.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no competing interests.