New Kidney Injury Criteria Better Predict Cirrhosis Survival

Daniel M, Keller, PhD

April 07, 2011

April 7, 2011 (Berlin, Germany) — In a study of hospitalized patients with cirrhosis, the Acute Kidney Injury Network (AKIN) criteria for acute renal failure was an effective tool for predicting 90-day survival, a group of Spanish researchers announced here at the European Association for the Study of the Liver 46th Annual Meeting.

Dr. Pere Ginès

Renal failure in cirrhosis is currently defined as a serum creatinine level greater than 1.5 mg/dL, but senior author Pere Ginès, MD, chair of the liver unit in the Hospital Clinic in Barcelona and professor of medicine at the University of Barcelona in Spain, noted 2 shortcomings of this definition. First, 1.5 mg/dL represents a very low glomerular filtration rate (GFR), and second, it might be insensitive to changes in GFR because it does not take into account changes in serum creatinine levels.

Therefore, the researchers decided to assess the value of the AKIN criteria to predict outcome by performing the first prospective observational study of this patient population. The study involved 300 consecutive patients admitted to the hospital for complications of cirrhosis between July 2009 and December 2011. At baseline, patients were an average of 60 years of age, 60% were men, and 51% had alcoholic cirrhosis. Serum bilirubin was 4.6 ± 5.6 mg/dL, International Normalized Ratio was 1.6 ± 0.5, and serum creatinine was 1.3 ± 0.7 mg/dL. Patients were followed for 3 months.

The AKIN criteria for acute renal failure are an absolute increase in serum creatinine of 0.3 mg/dL or greater or a 50% or greater increase over baseline in a 48-hour period. The researchers stratified patients meeting these criteria into 2 groups: those with a peak serum creatinine level of greater than 1.5 mg/dL, and those with peak values of 1.5 mg/dL or less

During their hospitalization, 88 patients (29%) developed renal failure, according to AKIN criteria. At 3 months, 38% were alive, compared with 87% of the 212 patients without renal failure (P < .001). The 60 patients meeting the AKIN criteria and having a peak serum creatinine level of greater than 1.5 mg/dL (the current definition of acute renal failure) had a significantly lower survival rate than the 28 patients meeting AKIN criteria with peak creatinine levels of 1.5 mg/dL or less (29% vs 58%, respectively; P = .026).

Patients with elevated creatinine levels who did not meet the AKIN criteria (n = 30) had a 3-month survival rate of 80%. The investigators reported that multivariate analyses indicated that both the AKIN criteria and the current criteria for acute renal failure are independent predictors of survival, but that the AKIN criteria provide higher predictive power.

The researchers concluded that the development of renal failure that meets the AKIN criteria is common among patients hospitalized with cirrhosis, and leads to poor outcomes, "even when small increases in serum creatinine are considered." Unlike the older definition of renal failure, the AKIN criteria consider changes in creatinine levels.

"We combined 2 definitions — the new definition and the old definition — and we came up with a subclassification of patients that was associated with outcomes, so we think that this could potentially be a new way of defining renal failure in cirrhosis," Dr. Ginès predicted.

"We found that patients without renal failure have an excellent survival, which at 3 months was above 85%. Patients meeting the AKIN criteria but with a peak value of serum creatinine below 1.5 [mg/dL], which was below the former definition [of acute renal failure], had a survival of around 56%," he told Medscape Medical News. "Patients with the AKIN definition plus a peak value of serum creatinine above 1.5 [mg/dL] had a much worse prognosis. I think that the combination of the old and the new definition" is helping us to better determine the outcome of patients.

The study continues to enroll patients, and Dr. Ginès would like to see it replicated in other institutions involving at least 1000 patients. If it is validated, he said it could change practice "because you may treat the disease more aggressively if you know that prognosis is very poor," including putting the patient on a transplant waiting list.

Mauro Bernardi, MD, professor of internal medicine at the University of Bologna, Italy, who was not involved in the study, noted that the study addresses an important problem. He said that during a recent meeting of nephrologists and hepatologists to improve the diagnostic criteria of acute and chronic renal failure in cirrhosis, no consensus was reached.

"The problem we have in hepatology is related to the fact that these patients have muscle wasting. They do not produce creatinine," he said. So serum creatinine "is not an absolutely reliable marker of renal function in this specific context." He noted that using a serum creatinine reading of 1.5 mg/dL might be too rigid a criterion for renal failure in a patient not producing much creatinine, and would therefore not be a good indicator of renal function. Renal failure might actually be occurring with a reading below 1.5 mg/dL.

Applying the model used by the Spanish investigators might not solve the problem entirely, Dr. Bernardi said, but it looks useful. In the future, he said, it might be enlightening to look into using biomarkers of renal tubular damage to better diagnose acute renal failure and to more reliably stratify patients according to risks for negative outcomes.

Public funding supported the study. Dr. Ginès and Dr. Bernardi have disclosed no relevant financial relationships.

European Association for the Study of the Liver (EASL) 46th Annual Meeting. Presented March 31, 2011.