Abstract
BACKGROUND & AIMS: Fluid resuscitation strategy in cirrhosis and septic shock remains unclear, especially in the setting of cirrhotic cardiomyopathy (CCM). In a single-center propensity-matched analysis (discovery cohort; group 1) and in a subsequent multicenter validation cohort (group 2), we compared two pragmatic point-of-care ultrasound (POCUS)-guided fluid resuscitation strategies in cirrhosis and septic shock. Patients received a crystalloid challenge to assess fluid responsiveness, followed by either balanced salt solution (BSS) with 20% albumin vs. BSS alone. Outcomes of interest were sepsis-related circulatory failure (need for fluids and additional vasopressor support) and all-cause mortality at 28 and 90 days. METHODS: POCUS was performed at time 0h, 24 h, 48 h, 72 h, and repeated as clinically needed in patients with cirrhosis and septic shock with mean arterial pressure (MAP) <65 mmHg. Resuscitation targets were as follows: MAP ≥ 65 mmHg, inferior vena cava (IVC) parameters of adequate volume resuscitation (IVC diameter between 1.8 and 2 cm; IVC collapsibility index: 20-30%), lactate clearance, optimized cardiac index, and lung profile. RESULTS: One hundred seventy patients (age: 51.1 ±9.3 years, 47.1% alcohol-associated cirrhosis, model for end-stage liver disease with sodium [MELD-Na]: 27.4 ± 4.3) were in group 1 and group 2 had 126 patients (age: 50.8 ± 11.4 years, alcohol-associated cirrhosis: 42.3%, MELD-Na: 27.6 ± 4.2). The 28-day mortality rate was similar with either resuscitation strategy: 36.5% vs. 31.8% (P = 0.314) in the discovery cohort and 35.9% vs. 30.6% (P = 0.330) in the validation cohort, respectively. At 90 days, mortality rates were 40% vs. 32.9% (P = 0.213) and 40.6% vs. 32.3% (P = 0.215), respectively. The total infused BSS was lower in the 20% albumin vs. BSS arm in both the discovery (1595 ± 315 mL vs. 1948 ± 318 mL, P < 0.001) and validation (1503.9 ± 249.8 mL vs. 2116.9 ± 380.6 mL, P < 0.001) cohorts, respectively. CCM was diagnosed in 75 (44.1%) and 43 (34.1%) in group 1 and group 2, respectively, and was predictive of circulatory failure (adjusted hazard ratio; aHR: 2.3, 95% [confidence interval]CI: 1.5-3.7, P < 0.001 & aHR: 2.1, 95%CI: 1.2-3.9, P = 0.009) and mortality (aHR: 2.9, 95% CI:1.7-4.9, P < 0.001 & aHR: 2.4, 95%CI: 1.2-4.9, P = 0.011), independent of fluid strategy, baseline MAP, and baseline MELD-Na. CONCLUSION: In patients with cirrhosis and septic shock, mortality is influenced by the presence of CCM and not by crystalloid/albumin resuscitation strategy.