About 20% of hospitalized patients with cirrhosis develop acute kidney injury (AKI), but preventive and treatment measures can have a real impact on outcomes, according to gastroenterologist Josh Levitsky, MD, MS, speaking during the Liver Meeting Digital Experience, held last November by the American Association for the Study of Liver Diseases.
“Kidney injury and hyponatremia are common, difficult to prevent, and detrimental to survival in hospitalized patients with end-stage liver disease,” he said. “Early prevention, recognition, and management strategies are really critical to optimize outcomes.”
Dr. Levitsky acknowledged that the overlap between aspects of AKI and cirrhosis can make diagnosis challenging. “Intrinsic renal failure due to structural renal disease or renal obstruction can overlap with functional renal failure, which is classically type 1 and type 2 HRS [hepatorenal syndrome], so it sometimes can be difficult to distinguish these etiologies,” he said.
New criteria developed by the International Club of Ascites and published in the April 2015 Journal of Hepatology aimed to better define AKI, chronic kidney disease, and acute-on-chronic kidney disease in patients with cirrhosis, who have different diagnostic criteria from those without cirrhosis. Under these criteria, there are three different stages of AKI.
“As you progress from AKI stage 1 to 2 to 3, the probability of survival at 90 days significantly declines,” Dr. Levitsky said. “Stage 1 would be an increase in serum creatinine by greater than or equal to 0.3 mg/dL or an increase in serum creatinine by greater than 50% to 100% from baseline, and this increases as the stage progresses to where there's worsening renal dysfunction, more increases from the baseline, or need for renal replacement therapy.”
Serum creatinine is not on its own a good measure of renal function, noted Dr. Levitsky, who is professor of surgery (organ transplantation), medicine (gastroenterology and hepatology), and medical education at Northwestern University Feinberg School of Medicine in Chicago. “Correlate it with eGFR [estimated glomerular filtration rate], based on age and other comorbidities,” he said.
There are also special considerations for interpreting this lab value in patients with cirrhosis. Elevated bilirubin levels can interfere with creatinine assays, an elevated creatinine level is more significant if muscle mass is low, and there is increased volume of distribution of creatinine in patients who also have anasarca, Dr. Levitsky said. “This drives down serum creatinine in patients with cirrhosis and really overestimates the GFR,” he said.
It's also important to look beyond the absolute numbers. “A large change in GFR can occur with just a little change in creatinine, so creatinine going from 0.5 to 0.9 [mg/dL] is a significant change compared to a creatinine going from 3.0 to 6.0 [mg/dL],” he said. “That is a concept that not everybody is familiar with, and we get less concerned about changes in creatinine at the lower levels when we really should be more concerned because it results in a significant difference in GFR, even when the creatinine is at a low level.”
Among hospitalized patients with cirrhosis who have AKI, two-thirds will have prerenal dysfunction and the remaining third will have intra-renal dysfunction, such as acute tubular necrosis or, more rarely, glomerulonephritis or obstruction, he said, citing a November 2008 review in Hepatology. Two-thirds of patients in the prerenal category will be volume-responsive, whereas one-third will be non-volume-responsive and meet diagnostic criteria for HRS type 1 or type 2, said Dr. Levitsky. “A hospitalized patient with cirrhosis with AKI, a lot of people think, ‘That's got to be HRS,’ when in fact, it's only a small percentage that actually have HRS.”
When evaluating AKI in a patient with cirrhosis, first focus on hydration. “The first thing to do is to look at the patient's volume status and give them back volume, usually by intravenous albumin, and discontinue their diuretic regimens so that you can adequately replete their intravascular volume,” he said.
Because a third of these patients will have some sort of acute intrinsic renal dysfunction, an abdominal ultrasound can look for any obstruction, “which is uncommon but worthy of evaluating,” Dr. Levitsky said. In addition, urinalysis and microscopy to look for casts are important to rule out glomerulonephritis, interstitial nephritis, or acute tubular necrosis, he said, adding that an infection workup is mandatory because treating infection early can help improve renal function.
“What comes of this is basically excluding intrinsic and postrenal, and then you've got prerenal dysfunction, and if they are not responsive to fluid repletion, then the diagnosis should be hepatorenal syndrome, either type 1 or type 2, depending on the acuity,” said Dr. Levitsky.
Diagnosis of type 1 HRS, defined as a significant change in serum creatinine level or diminishment in GFR in a short period of time (two to three weeks), can be complicated by diuretic treatment, fluid loss, or volume expansion with saline rather than with albumin, which can leak into the interstitium, he noted. “To really make the diagnosis, you need recent serum creatinine values,” Dr. Levitsky said. “So if that's not available, you have to make an assumption, if the increase is really acute.”
For treatment of type 1 HRS, randomized controlled trials of albumin and vasoconstrictors have shown that octreotide plus midodrine is about half as effective as terlipressin or norepinephrine, according to a review published in the November 2017 Kidney International.
“It's very apparent that terlipressin and norepinephrine are likely the best therapies to provide splanchnic vasoconstriction and increase the perfusion to the kidney,” Dr. Levitsky said. “There are more recent studies for terlipressin and norepinephrine, and we hope to get drug approval in the United States soon for terlipressin to be able to use it in a hospitalized patient who's not in the intensive care unit.”
Evaluation of hyponatremia is also important, since patients often present with hyponatremia alongside HRS, he noted. “We know that hyponatremia along with AKI is an independent risk factor for mortality in patients with liver failure . . . so assessment of volume status is critical,” Dr. Levitsky said. “The vast majority of these patients follow along the pathway of hypervolemic hyponatremia and have low urine sodium, which goes along with either prerenal [dysfunction] or prerenal with hepatorenal syndrome.”
In general, mild and asymptomatic hyponatremia does not require rapid correction, he said, “But if it's symptomatic, it needs to be corrected more abruptly.” If hyponatremia is chronic but symptomatic, then it's best to be more conservative with fluid replacement, Dr. Levitsky said. “Vaptans, which were of significant interest in this patient population, don't work well in this setting of cirrhosis. They increase mortality and the risk of liver injury,” he noted.
Of course, preventing AKI is preferable to treating it. Dr. Levitsky offered some tips for doing so in hospitalized patients with cirrhosis.
“Antibiotic prophylaxis in variceal bleeding prevents AKI, giving albumin with large-volume paracentesis (over 5 L), close monitoring of diuretic therapy (monitoring patients' electrolytes, urine output, and weight), making sure that lactulose is not given in too high of quantities to create significant volume loss from diarrhea,” he listed.
In some cases of refractory ascites, placement of a transjugular intrahepatic portosystemic shunt (TIPS) can also prevent AKI. “Or, if there's resolution of acute kidney injury, a TIPS performed can maybe prevent the next acute kidney injury event,” Dr. Levitsky added.
For patients with AKI, even the early stages, he also advises avoiding NSAIDs and IV contrast when possible. In addition, in patients with spontaneous bacterial peritonitis (SBP), albumin at the time of diagnosis and on days one and three can help prevent AKI, he said.
Looking to the future of AKI and cirrhosis, research offers hope that biomarkers may help direct care. A study published in the August 2014 Hepatology that looked at biomarkers including neutrophil gelatinase-associated lipocalin and interleukin-18 “generally showed that the more biomarkers that patients had in the urine, the more likelihood that they had acute tubular necrosis,” whereas those who had no elevated biomarkers more often had prerenal dysfunction. “I think these are exciting in terms of prognosticating and helping manage patients with acute kidney injury,” Dr. Levitsky said.
His own group's research has shown that two biomarkers (osteopontin and tissue inhibitor matrix metalloproteinase 1), along with age and diabetes, can predict future renal function in patients who are going to undergo liver transplant. The validation study, published online in September 2019 by Hepatology, could assist in the choice between a liver transplant alone versus a simultaneous liver-kidney transplant, according to Dr. Levitsky. “We're looking at further studies to test this in real time,” he said.