Abstract

Background: Non-selective-beta-blockers (NSBB) are main stay of therapy in portal hypertension management. Apart from direct portal pressure reduction, NSBB’s modulate inflammatory cascade, which could be beneficial in patients with acute decompensation (AD). We therefore aimed to evaluate effect of NSBB on 28-day mortality and markers of systemic inflammation in a propensity score matched (PSM) cohort of AD patients requiring intensive care unit(ICU) admission.

Methods: Patients were recruited from registry of AD patients requiring ICU admission. Out of the total 445 patients, 109 patients were on NSBB before admission (NSBB-group) who were PSM for pre-admission Childs-Turcotte-Pugh(CTP) score. 106 patients not on NSBB, served as the control-group. The on-admission to ICU parameters, markers of systemic inflammation; white cell count (WCC), neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), C-reactive protein (CRP) and 28-days mortality were compared by standard statistical tests.

Results: After PSM, no difference was observed in aetiology of cirrhosis, or precipitating event for AD. Pre-admission creatinine, bilirubin, INR and haemoglobin were similar between the groups. Pre-admission WCC and NLR was lower in NSBB-group, whereas CTP(median;7) and MELD(median:12) were similar. On admission to ICU, NSBB-group had lower heart rate(p<0.001), platelets(p=0.010), WCC(p=0.018), NLR(p=0.008) and CRP(p= 0.013). Significantly more community acquired bacterial infections(p=0.004), renal(p=0.003) and liver(p=0.034) failure and higher grades of acute-on-chronic-liver-failure(ACLF)(p=0.005) were seen in non-NSBB-group. Significantly lower 28-day(p=0.001) and 90-day(p<0.001) mortality was seen in NSBB-group (Figure 1). Univariate and multivariable analysis for 28-days mortality showed that while ACLF at presentation and community acquired bacterial infection were independent negative predictors, prior-NSBB use was positive predictors of survival.

Conclusion: Prior-NSBB use blunts inflammatory response favourably and is associated with improved 28 and 90-day mortality in critically-ill patients with acute decompensation (Figure 2).