ACUTE KIDNEY INJURY AFTER NEONATAL CARDIAC SURGERY: NEED TO IDENTIFY A CLINICALLY MEANINGFUL RENAL INJURY PHENOTYPE

Susan Nicolson MD, Jim DiNardo MD, Viviane Nasr MD, Lindsey Loveland Baptist, MD

ORIGINAL ARTICLE

Epidemiology of Acute Kidney Injury After Neonatal Cardiac Surgery:  A Report From the Multicenter Neonatal and Pediatric Heart and Renal Outcome Network; Alten JA, Cooper DS, Blinder JJ, Selewski DT, Tabbutt S, Sasaki J, Gaies MG, Bertrandt RA, Smith AH, Reichle G, Gist KM, Banerjee M, Zhang W, Hock KM, Borasino S:  Critical Care Medicine 2021 October1; 49(10): e941-e951. PMID 34166288

The Neonatal and Pediatric Heart and Renal Outcome Network (NEPHRON) conducted the 1st multi-center retrospective cohort study of 22,400 consecutive neonates in 22 PC4 sites between September 2015 and January 2018 in an attempt to describe the epidemiology, including perioperative predictors and associated outcomes, of cardiac surgery -associated acute kidney injury (CS-AKI). Data thought relevant to identifying CS-AKI were collected over the 1st six postoperative days (POD). Neonates with a preoperative serum creatine (SCr) > 1.5 mg/dl, receiving renal replacement therapy prior to surgery and/or requiring ECMO during the 1st 24 hours after surgery were excluded.

The median age at surgery was 7 days (interquartile range 5-11 days).  Single ventricle palliations occurred in 29% of the neonates and 70% were STAT Category 4 or 5.  Nearly one third of patients were admitted to the CICU with an open chest.  Multiple perioperative fluid management strategies were used including prophylactic peritoneal dialysis (PD) in 15%, diuretics in the 1st 24 postoperative hours in 52% and modified ultrafiltration (MUF) in 60%.

Neonatal modification of The Kidney Disease Improving Global Outcomes (KDIGO) criteria was used to identify and grade renal injury (see table).   

·       Varied greater than three-fold in rate (27%-86%) and severity (stage 3 – 0%-31%) across centers

·       Occurred in 54% of all neonates: 31% had stage 1, 14% stage 2, and 9% stage 3, with only stage 3 injury associated with hospital mortality

·       Occurred in both neonates who required CPB (59%) and those who did not (38%)

·       Preoperative enteral feeds and open sternum were associated with less injury, CPB was associated with increased injury, but duration of CPB was not associated with increased CS-AKI in the CPB cohort.

·        No association with CICU or hospital length of stay or duration of mechanical ventilation was found

Unfortunately, neither SCr nor UOP is sensitive or specific for AKI. Although SCr is the most widely used biomarker for kidney injury, it has limitations in neonates.  SCr in neonates is impacted by factors unrelated to renal injury: maternal creatinine in the 1st week of life, dilutional effect from fluid overload and concentration effect from excessive fluid removal.  SCr becomes elevated only after a significant decrease in glomerular filtration rate (GFR), and the increase may lag behind an insult by as much as 72 hours (1). Oliguria may occur independent of tubular dysfunction from prophylactic PD or use of peritoneal drains for passive drainage which may decrease UOP by providing an alternative route of fluid output.  Perceived oliguria may occur due to inaccurate quantification of UOP after removal of the Foley catheter. 

The data showed that more than half of the neonates experienced transient CS-AKI that peaked on POD 0-1. Variation in intraoperative and early postoperative fluid management (MUF, PD, diuretic use) may impact serum creatinine concentration and urine output on POD 0 and 1 leading to an early CS-AKI diagnosis, independent of true kidney injury.  This is supported by the fact the 35% of this cohort had elevated SCr without oliguria. Analyzing neonates with early transient CS-AKI with those with more persistent AKI may dilute the impact of CS-AKI on outcomes.

CS-AKI with both oliguria and an elevated SCr occurred infrequently.  The majority of neonates with persistent or late CS-AKI had oliguria.  CS-AKI with oliguria may be the clinically impactful phenotype as fluid overload in neonates is associated with significant morbidity and mortality independent of AKI (2) 40% of the CS-AKI cohort had oliguria/anuria with a normal SCr, consistent with the observation that diagnosing CS-AKI using SCr alone significantly underestimates the most clinically deleterious AKI (3)   

Not all serum creatinine elevation and/or oliguria are of equal clinical importance regarding their impact and associations with morbidity.  Other metrics such as time of onset, duration and/or rate of rise of SCr and/or oliguria may be more important than the peak staging of CS-AKI as defined in this study. The challenge that exists is early discrimination of findings that are a result of renal tissue injury, with significant impact on clinical outcomes rather than the transient changes of little clinical significance that existed in many of the neonates in this study. 

The modified KDIGO criteria fail to define a clinically meaningful injury phenotype in neonates following cardiac surgery which in turn may obscure its association with outcomes. The most promising urinary biomarkers for prediction of subsequent AKI are those that identify cell cycle arrest of kidney tubular epithelial cells (4). However, in the absence of a clinically available kidney-injury biomarker, there is need to create a neonatal CS-AKI definition that identifies neonates most at risk for associated morbidity and to identify initiatives to prevent and/or mitigate its negative impact on outcomes

References:

1.     Goren O, Matot I: Perioperative Acute Kidney Injury Br J Anaesth 2015: 115 (suppl2) ii3-II14.

2.     Mah KE, Hao S, Sutherland SM et al: Fluid Overload Independent of Acute Kidney Injury Predicts Poor Outcome in Neonates Following Congenital Heart Surgery Pediatr Nephrol 2018: 33, 511-520.

3.     Kaddourah A, Basu RK, Bagshaw SM, Goldstein SL: Aware Investigators: Epidemiology of Acute Kidney Injury in Critically Ill Children and Young Adults N Engl J Med 2017: 376(1): 11-20.

4.     Gumbert SD, Korf F, Jackson ML et al.: Perioperative Acute Kidney Injury.  Anesthesiology 2020: 132: 11-20

5.     Kellum JA et al. Acute Kidney Injury. Nat Rev Dis Primers 2021: 7,52.