A retrospective review of percutaneous transhepatic gallbladder drainage for non-responsive or otherwise non-operable acute cholecystitis in a tertiary hospital

Abstract

Background: Percutaneous transhepatic gallbladder drainage (PTGBD) is described in the 2018 Tokyo Guidelines as an alternative to laparoscopic cholecystectomy (LC) for sepsis control for acute cholecystitis (AC). In South Africa, patients often present late or with recurrent AC. LC may be possible, but this is dictated by limited emergent access to operating theatres. The safety and efficacy of PTGBD has not been described in South Africa. There is no universal consensus on the optimal timing of interval cholecystectomy following PTGBD. The aim is to demonstrate the outcomes of PTGBD in patients with AC, not suitable for LC or not responding to antimicrobials. Materials and Methods: A retrospective review of radiology records was performed of patients who underwent PTGBD for AC in Groote Schuur Hospital, Cape Town, over a three- year period between May 2013 and July 2016. Patients with PTGBD for malignancy or acalculous cholecystitis were excluded. Technical success (correct placement of tube in the gallbladder), clinical response, procedure- related morbidity and mortality were recorded. Interval LC parameters were investigated. Results: 37 patients had PTGBD, with technical success and clinical improvement in 29 (78.38%). Malposition (3/37) was the most common complication (8.11%). Two patients required emergency surgery (5.4%), while one tube was dislodged. Mean tube placement duration was 36.83 days (Range 1- 211). 16 patients (43.24%) went on to have LC. Eight required conversion to open surgery (50%). Four had subtotal cholecystectomy (25%). Average surgical time was 135 minutes (Range 60-300). There were no procedure- related mortalities. Eight patients (21.62%) died in the 90- day period following tube insertion. Conclusions: PTGBD is safe for high- risk patients with AC, with high technical success and low complication rate. Subsequent LC should be performed, but is usually challenging. The requirement for PTGBD may predict disease process associated with poor outcome.