GUEST EDITORIAL

Acute Kidney Injury in the CICU – Silent but Deadly

 

Dr. Gist

Katja M Gist, DO, MScS

Dr. Penk

Jamie S. Penk, MD

Dr. Kwiatkowski
David M. Kwiatkowski, MD
Dr. Cooper

David Cooper, MD, MPH

Katja M Gist, DO, MScS
Pediatric Cardiologist and Intensivist
Children’s Hospital Colorado 
Assistant Professor of Pediatrics 
University of Colorado, Anschutz Medical Campus
School of Medicine, Department of Pediatrics

Jamie S. Penk, MD
Cardiac Intensivist
Northwestern University Feinberg School of Medicine
Division of Cardiology
Ann & Robert H. Lurie Children’s Hospital of Chicago

David M. Kwiatkowski, MD
Cardiac Intensivist
Assistant Professor
Stanford University School of Medicine

David Cooper, MD, MPH
Treasurer, PCICS
Medical Director, Cardiac Intensive Care Unit
Medical Director, Cardiac Anesthesia Recovery Unit
Co-Director, Center for Acute Care Nephrology
The Heart Institute
Cincinnati Children's Hospital Medical Center

Cincinnati, Ohio

Acute kidney injury (AKI) is common after congenital cardiac surgery, with an incidence ranging from 20-60%, and is known to confer an increased risk for short- and long-term adverse outcomes. Infants, children and adults who develop AKI after cardiopulmonary bypass (CPB) have at least five times greater odds of dying compared to similar patients without AKI. It is now widely recognized that AKI is a systemic disease that predisposes patients to a variety of other complications for which renal replacement therapy is not helpful. These systemic complications such as increased infection risk (including sepsis), respiratory failure and heart failure are responsible for the high rate of death in patients with established AKI.

This paradigm suggests that reducing the morbidity and mortality of AKI will require prevention and management of these complications. To this end, existing diagnostic modalities render an imprecise definition of AKI. For example, serum creatinine during the first few days of life reflects maternal levels; thus, there is inherent difficulty in defining a baseline for determination of the magnitude of rise in post-operative creatinine and thus possible misclassification of tubular injury. The co-existence of fluid overload may mask AKI diagnosis because serum creatinine is diluted. Correction of creatinine for the degree of fluid overload leads to more sensitive detection of AKI and has strengthened the association of cardiac surgery associated AKI with poor outcomes. Biomarkers of actual renal injury may afford a more accurate and precise description of tubular injury. Despite the widespread literature on the use of AKI biomarkers that identify tubular injury, only two are currently available for clinical use: neutrophil gelatinase associated lipocalin (NGAL) and the product of tissue inhibitor matrix metalloproteinase-2 and insulin like growth factor binding protein-7, (TIMP-2*IGFBP-7), also known as Nephrocheck®. Combining markers of functional (serum creatinine and urine output) and structural (NGAL and TIMP-2*IGFBP-7) injury for prediction of AKI development and its duration would be advantageous.  In fact, where elevation of creatinine without an increase in the structural marker predicts a short duration but the opposite predicts a long duration has been validated in the pediatric cardiac surgery population.

There are many studies assessing the timing and predictive performance of both NGAL and TIMP-2*IGFBP-7 after pediatric cardiac surgery, but incorporation into clinical practice is limited because of the availability of testing platforms within clinical laboratories. Thus, any major initiatives to reduce AKI incidence and poor outcomes have been limited. Simple quality improvement initiatives [i.e. NINJA (Nephrotoxic Injury Negated by Just in time Action)] that avoid fluid overload and nephrotoxic medication exposures to at risk patients are likely to prove more beneficial when biomarkers are readily available for use.

Biomarkers offer an exciting opportunity to detect AKI prior to change in creatinine. However, until biomarkers become universally available, other methods of significant AKI identification are needed.  The furosemide stress test (FST) is a potential way to use a commonly available drug to asses renal function. The FST has been shown to predict the development of AKI.  It has also been shown to improve risk stratification of severe AKI when used in conjunction with biomarkers as well as to predict the need for renal replacement therapy (RRT).

The FST is a functional assessment of the kidney first validated in adult critical illness.  A dose of furosemide is administered and hourly urine output for the first two and/or six hours is used to assess the kidney's response to that dose. Low urine output as a response to furosemide is deemed a failure. In pediatrics, two retrospective single-center studies of infants who underwent cardiac surgery showed an association between lower urine output after the first dose of furosemide and subsequent development of AKI. Moreover, a recent multicenter retrospective study demonstrated a similar association between a decreased responsiveness to furosemide and the development of AKI in older children (median age 6.3 months) after cardiac surgery.  Given these findings, the FST holds promise for clinical use. Prospective studies should be performed to help characterize the risk profiles of various patient populations (i.e. single vs biventricular physiology or neonates vs children). Importantly, the FST is a cheap, easy and currently available tool which can alert clinicians to risk of AKI which may then allow time for treatment to prevent progression of AKI.

In pediatric cardiac critical care, maintenance of fluid balance is one those mundane conversations that quintessentially exists in every patient round and handoff, with goals casually chosen and variably followed. However, evidence is mounting that fluid balance is one of the few modifiable risk factors that exist in critical care management and deserves a more concentrated focus. Recent studies have demonstrated that fluid overload in pediatric patients after cardiac surgery is an independent risk factor for higher mortality and greater morbidity including longer duration of mechanical ventilation and hospital stay. Although fluid overload has been considered a transient bystander of acute kidney injury, studies have demonstrated that its deleterious associations exist even when seen in the absence of AKI or when analyzed with similar degrees of AKI. Furthermore, fluid overload may be masking the true incidence of AKI, as our primary biomarker, creatinine, is subject to dilutional effects. To better understand the effects of fluid overload in the pediatric postoperative population, the 22-center Neonatal and Pediatric Heart Renal Outcomes Network (NEPHRON) has collectively evaluated over 2,200 infants after cardiac surgery. Results are expected to better demonstrate associations with outcomes and to demonstrate center-specific practice variation, relative to fluid overload, that may be associated with better or worse outcomes.

Despite the recognition of the importance of establishing a negative fluid balance in those with fluid overload, maladaptive physiologic responses cause some cardiac patients to be resistant to diuretic medications. This is especially a problem among patients with heart failure. In postoperative patients, especially in those where fluid restriction fails to prevent fluid overload in the setting of being refractory to diuretics, RRT is often needed.  In infants after cardiac surgery, this is most commonly accomplished using peritoneal dialysis. There is tremendous center-specific practice variation with respect to peritoneal dialysis use despite the fact that multiple retrospective and prospective studies have demonstrated improved outcomes with rare adverse events. Among infants with postoperative oliguria, peritoneal dialysis leads to less fluid overload, shorter duration of mechanical ventilation and shorter ICU stays and an association with lower mortality.

In older pediatric patients, RRT is often accomplished using continuous veno-venous hemodialysis (CVVH).  However, recent interest has focused on providing better options to smaller patients either through invention of miniaturized devices, or adaptations of existing devices.  Although there are no FDA approved dialysis membranes for patients weighing <20kg, the Baxter Healthcare Corporation is currently evaluating the smaller HF20 membrane for the existing Prismaflex® system to provide dialysis in this population. Otherwise, hemodialysis in the smallest patients is being performed off-label.  Meanwhile, others are developing experience using the adult ultrafiltration device, the Aquadex FlexFlow System®, for isolated fluid removal without clearance. In particular adult populations, this device has proven to be an effective method of fluid removal, inspiring the use in some pediatric populations given the device’s smaller venous access and extracorporeal blood volume. Some groups even perform post-market modifications to utilize the device to deliver full dialysis in neonates as small as 2kg. Elsewhere, Claudio Ronco and his colleagues in Italy have developed, and are leading the evaluation of, a miniaturized dialysis machine named the Cardio-Renal Pediatric Dialysis Emergency Machine (CARPEDIEM™). This device is the first primarily designed for neonates (for use in neonates as small as 2.5kg). Although in modern practice, few practical commercially available devices for treatment of fluid overload exist for our smallest patients, the past decade has brought incredible academic energy to developing solutions, which hopefully will be available soon.

It previously was assumed that patients with a single episode of AKI would recover kidney function without long-term consequence. However, during the last decade, epidemiologic data from critically ill children and adults suggest that AKI survivors are at considerable risk of developing chronic kidney disease (CKD). Adults who experienced AKI have a ninefold increased risk of developing CKD, a threefold increased risk of developing end-stage kidney disease, and a twofold increased risk of long-term mortality risk as compared to those without AKI. In follow-up studies of pediatric AKI, the incidence of CKD ranges from 27%-67%.  However, the implications of post-CPB AKI are less clear. Utilizing a standardized AKI definition/classification system, in a critically ill pediatric population (including post-operative cardiac patients) revealed a 10% incidence of CKD one to three years following AKI and almost half were considered at risk of CKD development. 

There is also mounting data about the long-term consequences of AKI in patients with congenital heart disease.  A single center study of patients with hypoplastic left heart syndrome and its variants describe the incidence and significance of AKI throughout the three stages of palliation.  AKI is common among these high-risk patients affecting 58-75% of patients after each surgical stage. Significantly, the group determined that the sequelae of severe AKI as a neonate are substantial, predisposing patients to death or need for ECMO after Stage 1 palliation, and despite normalization of creatinine, risk of severe AKI after Stage 2 is elevated in those affected as a neonate.  A study in adults with congenital heart disease demonstrated that 50% of young adults with CHD have impaired glomerular filtration rate (GFR), even those with “simple” defects, and this was associated with mortality.

As in AKI, there have been obstacles to assessing a potential AKI to CKD link, including the lack of a standard AKI definition and reliance on serum creatinine as the primary biomarker to detect and diagnose AKI and CKD.  As discussed above, novel urinary biomarkers of kidney injury measured at different times post-CPB have been able to predict AKI.  The potential of these biomarkers to identify patients with CKD has also been explored.  In a study of patients seven years following CPB, without conventional evidence of CKD, there was evidence of renal urinary biomarker elevation consistent with ongoing subclinical injury.  The potential application of biomarkers in the detection and management of CKD in pediatric heart disease patients warrant further exploration.

The impact of AKI and fluid overload, whether together or separately, on pediatric cardiac patients is clear with important short- and long-term consequences.  Further study is warranted to prevent, diagnose and treat AKI as well as fluid overload.  Likewise, the importance of following AKI survivors throughout adulthood to understand and treat the long-term implications of AKI/CKD is paramount.

Back to top