From the AGA Journals

New definition of kidney injury is more predictive of mortality

View on the News

New definition of kidney injury predicts mortality

Renal dysfunction in patients with cirrhosis is often associated with sepsis. This combination involves a very high probability of death. Recently, the concept of acute kidney injury has been proposed to be extended also to renal failure occurring in patients with cirrhosis. AKI should overcome limitations associated with a fixed creatinine threshold, ensure rapid identification of renal dysfunction, and allow timely treatment in patients with hepatorenal syndrome. However, AKI should also overcome the skepticism of those who wish not to abandon previous definitions.

The recent paper of Dr. Wong and her colleagues explored the impact of AKI in 337 hospitalized patients with cirrhosis. Two-hundred eighty-seven patients had bacterial infection at admission, and 93 developed it during hospitalization. Overall, 68 patients died from multiorgan failure, whereas only 7% of patients without AKI died. Mortality ranged from 15% in patients who recovered from AKI to 80% in those who did not. Moreover, 76 patients (23%) developed a second infection, often associated with invasive procedures! An elevated Model for End-Stage Liver Disease score and a second infection were factors independently associated with AKI. Accordingly, the development of AKI in cirrhosis, even if reversible, was shown to be a strong predictor of short survival.

These findings show that, in cirrhosis, even small creatinine changes (0.3 mg) are clinically relevant, and that AKI is probably a hallmark of hemodynamic instability with a risk of multiorgan failure and death. The altered hemodynamics in patients with cirrhosis cause central hypovolemia. Aiming at protecting our patients from infection and AKI, we should also pay more attention to clinical procedures that raise serum creatinine level.

Dr. Francesco Salerno is in the department of internal medicine, at the Policlinico IRCCS San Donato, University of Milan (Italy); Dr. Vincenzo La Mura is with the Fondazione IRCCS Ca'Granda, in the department of gastroenterologia-1 of the Hospital Maggiore Policlinico, Milan. They reported no relevant financial conflicts.



The newly proposed consensus definition of acute kidney injury in patients with cirrhosis accurately predicts 30-day mortality and other adverse outcomes in this patient population much better than the current, more rigid definition would, according to a report in the December issue of Gastroenterology (doi:10.1053/j.gastro.2013.08.051).

In what they described as the largest prospective study of this topic to date, researchers found that the recently proposed, broader redefinition of acute kidney injury (AKI) correctly identified which patients were likely to die, develop severe complications such as organ failure, or require longer hospitalization, even when the AKI was transient and resolved completely after treatment.

Courtesy American Gastroenterological Association

More than half of the patients in this study who had episodes of AKI according to the new definition did not meet the criteria of the old definition. So using the new definition will help identify these high-risk patients at an earlier stage of renal dysfunction, "well before the stringent diagnostic criteria of [the old definition] are reached," when they will have a better treatment response, said Dr. Florence Wong of the division of gastroenterology, University of Toronto, and her associates.

The old definition of AKI required the presence of hepatorenal syndrome, with a serum creatinine level of greater than 2.5 mg/dL. This meant that patients with less severe renal dysfunction didn’t qualify and weren’t treated. But emerging evidence indicates that even mild degrees of renal dysfunction signal a poor prognosis, and that serum creatinine alone doesn’t accurately reflect renal dysfunction in advanced cirrhosis.

So the International Ascites Club and the Acute Dialysis Quality Initiative (ADQI) group proposed that acute kidney injury in cirrhosis should be redefined as an increase in serum creatinine level of 0.3 mg/dL or greater within 48 hours, or a 50% increase in serum creatinine level from a stable baseline reading within the previous 6 months, regardless of final serum creatinine level.

Dr. Wong and her colleagues assessed the new definition in a cohort of 337 cirrhotic patients treated during a 2-year period at 12 North American medical centers who were admitted with a bacterial infection (287 subjects) or who developed a bacterial infection during hospitalization (50 subjects). The most common infections were urinary tract infection (27% of patients), spontaneous bacterial peritonitis (21%), skin infection (14%), pneumonia (10%), and spontaneous bacteremia with no clear source of infection (9%).

Approximately half of these patients (49%) developed at least one episode of AKI during hospitalization. The 30-day mortality was significantly higher for those who developed AKI according to the new definition (34% mortality) than in those who did not (7% mortality), the investigators said.

Most patients who developed AKI had only a transient case, and their renal function completely recovered. Yet their subsequent mortality within 30 days was twice as high as that for patients who didn’t have any AKI.

The negative predictive value of the new definition of AKI was 93%, and the positive predictive value was 34%.

This study was supported in part by the National Institutes of Health, the National Institute of Diabetes and Digestive and Kidney Diseases, and the National Center for Research Resources. No financial conflicts of interest were reported.

Next Article: