Safety and efficacy of continuous terlipressin infusion in HRS-AKI in a transplant population

Abstract

Hepatorenal syndrome-acute kidney injury (HRS-AKI) is associated with significant morbidity and mortality. While liver transplantation is the definitive treatment, continuous terlipressin infusion for HRS-AKI may provide benefit and, as such, was assessed in a population composed of candidates for liver transplant (LT). Fifty hospitalized LT-eligible patients with HRS-AKI received a single bolus followed by continuous terlipressin infusion. Acute-on-chronic liver failure grade 3, serum creatinine (SCr)>5.0 mg/dL, or Model for End-Stage Liver Disease (MELD) ≥35 were exclusions. Fifty hospitalized patients who received midodrine and octreotide or norepinephrine for HRS-AKI served as a historical comparator cohort. Complete response (CR) was defined as a ≥30% decrease in SCr with end-of-treatment (EOT) SCr≤1.5, partial response as a ≥30% decrease in SCr with EOT SCr>1.5, and nonresponse as a <30% decrease in SCr. CR rate was significantly higher in the terlipressin cohort compared to the historical cohort (64% vs. 16%, p <0.001). Survival, while numerically higher in those who received terlipressin, was statistically similar (D30: 94% vs. 82%, p =0.12; D90: 78% vs. 68%, p =0.37). Renal replacement therapy (RRT) was more common among terlipressin NR than CR and PR (70% vs. 3% vs. 13%, p < 0.001). EOT MELD and SCr were significantly lower within terlipressin cohort (MELD: 19 vs. 25, SCr: 1.4 vs. 2.1 mg/dL, p <0.001). Sixteen of 40 terlipressin-treated patients received LT-alone (terlipressin CR in 10/16). One patient on terlipressin had a hypoxic respiratory failure that responded to diuretics; one possibly had drug-related rash. With continuous terlipressin infusion, a CR rate of 64% was observed with a favorable safety profile. Terlipressin use was associated with lower EOT MELD and SCr than the historical midodrine and octreotide/norepinephrine cohort; LT-alone was accomplished in a high proportion of complete terlipressin responders.

Graphical abstract

FIGURE 1

Response rates, etiology of liver disease, and AKI stage among terlipressin cohort. (A) Overall response rates to terlipressin in the prospective cohort and M&O/NorEpi in the historical cohort. (p<0.001, Fisher exact test). (B) Etiology of liver disease and response to treatment among terlipressin cohort (p=0.51, Fisher exact test). (C) AKI stage and response to treatment among terlipressin cohort (p=0.40, Fisher exact test). Abbreviations: AKI, acute kidney injury; CR, complete response; M&O, midodrine and octreotide; NorEpi, norepinephrine; NR, nonresponse; PR, partial response.

FIGURE 2

Survival, renal replacement therapy (RRT), and liver transplant (LT) by day 90 of follow-up. (A) Cumulative incidence of death with LT and use of RRT as competing risks by terlipressin response group (p=0.13). (B) Need for RRT by terlipressin response group (p<0.001, RRT-free survival with death and LT as competing risks). (C) Rate of transplantation (LT-alone or simultaneous liver and kidney transplant) by terlipressin response group (p=0.20, transplant-free survival with death as competing risk). Abbreviation: NO, nonresponse group.

FIGURE 3

Median MELD and MELD-Na at baseline* and EOT.^^a (A) Median MELD scores. Levels of significance: *p=0.33, ^p<0.001 (Wilcoxon rank sum test). (B) Median MELD-Na scores. Levels of significance: *p=0.24, ^p=0.02 (Wilcoxon rank sum test). ^aBaseline MELD/MELD-Na missing for 3 terlipressin and 7 historical patients. EOT MELD/MELD-Na missing for 5 terlipressinn and 13 historic patients. EOT scores not significantly compared for response groups as groups were defined by EOT creatinine. Abbreviations: CR, complete response; EOT, end of treatment; MELD, Model for End-Stage Liver Disease; M&O, midodrine and octreotide; NorEpi, norepinephrine; NR, nonresponse; PR, partial response.