CVVH was performed in lieu of intermittent HD when subjects required intravenous vasopressor support and/or were considered hemodynamically unstable by the treating nephrologist. Subjects were classified as “listed” for liver transplant if they were registered on the United Network for Organ Sharing waiting list (before or after the start of RRT) at the study’s liver transplant center.Īll subjects received either intermittent HD or continuous veno-venous hemofiltration (CVVH) as their modality of RRT. Subjects were classified as “not listed” if they were evaluated and rejected for liver transplant listing, or if they were felt not to be appropriate for transplant evaluation by treating clinicians (due to critical illness, ongoing alcohol use, etc.). Subjects were excluded if they were lost to follow-up after discharge ( n=3).ĭefinitions and causes of cirrhosis were on the basis of clinical determination by the treating hepatologist. MELD score and Chronic Liver Failure Consortium Acute-on-Chronic Liver Failure (CLIF-C ACLF) score were calculated at the time of RRT initiation ( 40, 41). All potential subjects underwent review of the electronic medical record to confirm accuracy of diagnostic codes and medical history (see Definitions). Hemodialysis (HD), CRRT treatments, and clinical diagnoses were identified using the Current Procedural Terminology, 4th Edition and International Classification of Diseases-9/10 codes (see Supplemental Table 1). Data were complete for each variable, except as noted in table/figure footnotes. Follow-up data were obtained via electronic medical record and database review and, if needed, by search of online obituaries and Social Security Death Index. All subjects were treated using local standard of care without intervention from the study team.
Data were identified using a centralized clinical data collection warehouse designed for research and quality improvement purposes ( 37– 39).
We performed a retrospective cohort study of all inpatients with cirrhosis and AKI requiring RRT for either HRS or ATN between 20 in a network of five acute care hospitals (Partners Healthcare, Massachusetts), including one liver transplant center (Massachusetts General Hospital, Boston, MA), where all transplant listing evaluation was done for this population.
#Meld score 24 prognosis update#
To provide an update in this growing group, we aimed to compare outcomes after initiation of RRT in cirrhosis, while stratifying by transplant listing status and adjusting for cause of AKI. Over time, advancements in the medical management of cirrhosis have led to allograft allocation at higher Model for End-Stage Liver Disease (MELD) scores, thus creating a larger population of noncandidates with indications for RRT ( 36). Although RRT usually serves as a bridge to liver transplantation among those listed, literature is sparse regarding the use of RRT in those not listed for transplantation, particularly in HRS. The available literature is limited by small sample sizes ( 22– 24), older studies that do not reflect modern practice patterns ( 25– 27), or inclusion of only liver transplantation candidates/recipients ( 23, 24, 28– 35). Portal hypertension and splanchnic vasodilation result in decreased effective circulating volume, ascites formation, and low mean arterial pressure ( 19– 21), all of which are barriers to adequate volume management and often necessitate transfer to the intensive care unit for support with intravenous vasopressors and continuous RRT (CRRT) ( 18).Ĭurrently, there are few studies to guide clinical judgment of which patients with severe cirrhosis are optimal candidates for RRT.
In addition to the typical acute complications of RRT, such as intradialytic hypotension ( 9– 12), increased risk of cardiac events ( 13– 15), and complications related to venous access ( 16, 17), there are additional physiologic challenges inherent in those with cirrhosis ( 18). Those who require RRT represent the most morbid subgroup of the cirrhotic population ( 8). Not unexpectedly, health care costs in this setting are high ( 5– 7). These AKI subtypes carry a dismal prognosis, with 40% requiring RRT and 60% dying by 90 days ( 3, 4). About one-third of this AKI improves with administration of albumin or crystalloid, leaving two-thirds of these patients with persistent AKI, such as hepatorenal syndrome (HRS) or acute tubular necrosis (ATN) ( 1– 4). AKI is a common and morbid complication of cirrhosis, occurring in up to 50% of those with decompensated cirrhosis ( 1, 2).