Does a Spoonful of Insulin Make the Acute Kidney Injury Go Down?*
This study uses data from an existing randomized trial to answer an important question: Does postoperative glycemic control affect the development of AKI after pediatric cardiopulmonary bypass (CPB)? The avoidance of hyperglycemia has been postulated to limit AKI in several adult studies (2–4) but has failed to demonstrate an association in others (5, 6). Similar to previous reports, this study confirms that younger age, more complex surgery and longer CPB times are associated with the development of AKI (7, 8), but importantly, those aspects are largely out of provider control. On the other hand, glycemic control is a modifiable strategy that could potentially influence patient outcomes.
This large well-designed trial found that postoperative glucose control did not prevent development of AKI in the pediatric cardiac patient. Their conclusions, however, are somewhat limited by the timing of intervention. Initiation of tight glucose control was not commenced until postoperative care and therefore not active during the main period of renal insult, namely CPB (9). To that point, the study notes that the rate of intraoperative hyperglycemia was higher in patients who later developed AKI. Therefore, while the study can assert with reasonable certainty that there is no evidence that tight glycemic control in the postoperative period prevents AKI, the study cannot entirely eliminate hyperglycemia as a risk factor. The authors and this journal should be commended for advancing this negative study, a challenge in itself.
As important as that conclusion is, however, it is hardly the most interesting finding of this study. This two-center trial occurred at two premier pediatric cardiovascular centers, the University of Michigan (UM) and Boston Children’s Hospital (BCH), but the occurrence rate of AKI was vastly divergent. Despite a relatively similar patient cohort and similar surgical procedures performed, the occurrence rate of AKI and moderate/severe AKI at UM was almost five times greater (66% vs 15%; p < 0.001 and 51% vs 9%; p = 0.01). At first glance one might think outcomes at BCH are superior, but the story is not that simple. The occurrence rate of AKI at UM is only somewhat higher than that published in many similar studies including those mentioned in the trial (7, 8, 10), whereas the occurrence rate at BCH is remarkably lower. Of the many studies cited by Blinder, only one study (also from BCH) reported an occurrence rate of AKI less than 26% (11).
The authors did an outstanding job of teasing out institutional differences and found two important disparities in perioperative care. First, most patients at UM received modified ultrafiltration (MUF) with surgery while the practice was rare at BCH, and second, most patients at BCH received steroids with surgery, which was rare at UM. Logically, the authors posit that steroids may be protective, or that MUF may increase the risk of AKI (or at least an elevation in creatinine due to prerenal azotemia). Given collinearity with study center, these factors could not be evaluated.
Inflammation is one of the most established mechanisms of AKI development post CPB (12).