Early TIPS “should therefore be preferred to the current standard of care,” reported lead author Yong Lv, MD, of the Fourth Military Medical University in Xi’an, China, and colleagues, referring to standard pharmaceutical and endoscopic therapy.
“[The current standard] approach has improved patient outcomes,” the investigators wrote in. “However, up to 10%-20% of patients still experience treatment failure, requiring further intensive management. In such patients, [TIPS] is successful in achieving hemostasis in 90%-100% of patients. However, 6-week mortality remains high [35%-55%]. This is probably because the severity of the underlying liver disease has worsened and additional organ dysfunction may have occurred after several failed endoscopic therapy attempts.”
Previous studies have explored earlier intervention with TIPS; however, according to the investigators, these trials were inconclusive for various reasons. For example, uncovered stents and an out-of-date control therapy were employed in a trial by Monescillo et al., while a study by Garcia-Pagan et al. lacked a primary survival endpoint and has been criticized for selection bias. “Thus, whether early TIPS confers a survival benefit in a broader population remains to be assessed,” the investigators wrote.
To this end, the investigators screened 373 patients with advanced cirrhosis (Child-Pugh class B or C) and acute variceal bleeding. Of these, 132 were eligible for inclusion based on age, disease severity, willingness to participate, comorbidities, and other factors. Patients were randomized 2:1 to receive either early TIPS or standard therapy. Within 12 hours of hospital admission for the initial bleeding episode, all patients received vasoactive drugs or endoscopic band ligation and prophylactic antibiotics. Control patients continued vasoactive drugs for up to 5 days, followed by propranolol, which was titrated to reduce resting heart rate by 25% but not less than 55 beats per minute. Elective endoscopic band ligation was performed within 1-2 weeks of initial endoscopic treatment, then approximately every 2 weeks until variceal eradication, and additionally if varices reappeared. TIPS was allowed as rescue therapy. In contrast, patients in the TIPS group underwent the procedure with conscious sedation and local anesthesia within 72 hours of diagnostic endoscopy, followed by approximately 1 week of antibiotics and vasoactive drugs. TIPS revision with angioplasty or another stent placement was performed in the event of shunt dysfunction or reemergence of portal hypertensive complications. The final dataset contained 127 patients, as 3 were excluded after randomization because of exclusionary diagnoses, 1 withdrew consent, and 1 died before TIPS placement.
The primary endpoint was transplantation-free survival. The secondary endpoints were new or worsening ascites based on ultrasound score or sustained ascites necessitating paracentesis, failure to control bleeding or rebleeding, overt hepatic encephalopathy, other complications of portal hypertension, and adverse events.
After a median follow-up of 24 months, data analysis showed a survival benefit associated with early TIPS based on multiple measures. Out of 84 patients in the TIPS group, 15 (18%) died during follow-up, compared with 15 (33%) in the control group. Actuarial transplantation-free survival was also better with TIPS instead of standard therapy at 6 weeks (99% vs. 84%), 1 year (86% vs. 73%), and 2 years (79% vs. 64%). The hazard ratio for transplantation-free survival was 0.50 in favor of TIPS (P = .04). These survival advantages were maintained regardless of hepatitis B virus status or Child-Pugh/Model for End-Stage Liver Disease score.
Similarly to transplantation-free survival, patients treated with TIPS were more likely to be free of uncontrolled bleeding or rebleeding at 1 year (89% vs. 66%) and 2 years (86% vs. 57%). The associated hazard ratio for this outcome favored early TIPS (HR, 0.26; P less than .0001), and univariate and multivariate analysis confirmed an independent protective role. In further support of superiority over standard therapy, patients treated with TIPS were more likely than those in the control group to be free of new or worsening ascites at 1 year (89% vs. 57%) and 2 years (81% vs. 54%).
No significant intergroup differences were found for rates of overt hepatic encephalopathy, hepatic hydrothorax, hepatorenal syndrome, spontaneous bacterial peritonitis, hepatocellular carcinoma, serious adverse events, or nonserious adverse events. At 1 and 3 months, patients in the TIPS group had a slight increase of median bilirubin concentrations and median international normalized ratio; however, these values normalized after 6 months. A similar temporal pattern was observed in early TIPS patients with regard to median Model for End-Stage Liver Disease score.
“[The transplantation-free survival benefit of early TIPS] was probably related to better control of factors contributing to death, such as failure to control bleeding or rebleeding or new or worsening ascites, without increasing the frequency and severity of overt hepatic encephalopathy and other adverse events,” the investigators concluded. “This study provides direct evidence and greater confidence in the recommendations of current guidelines that early TIPS should be performed in high-risk patients without contraindications.
“Future studies addressing whether early TIPS can be equally recommended in Child-Pugh B and C patients are warranted,” the investigators added.
The study was funded by the National Key Technology R&D Program, Boost Program of Xijing Hospital, Optimized Overall Project of Shaanxi Province, and National Natural Science Foundation of China. The investigators reported no conflicts of interest.
SOURCE: Lv Y et al. Lancet Gastroenterol Hepatol. 2019 May 29.