University of Cambridge, UK
Published Date: 23/09/2021.
*Corresponding author: Aleena Haider, University of Cambridge, UK
Introduction: Percutaneous Cholecystostomy (PC) is a radiological intervention used in the management of high risk or critically ill patients with acute cholecystitis (AC)
Method: A retrospective study of outcomes following PC, including success rates, rates of res-olution of AC, complication rates, readmissions rates, post-procedure endoscopic retrograde cholangiopancreatography (ERCP) rates and rates of subsequent cholecystectomy.
Results: Our database identified 28 patients (14M;14F), median age 73 (range 40-93). 82% were ASA III/IV. Median follow-up was 2 (range 0-8) years. Imaging suggested AC in 61% and empyema in 39%. 86% were calculous. All procedures were USS-guided with 100% success. Resolution of AC occurred in 89.3%. Of three unresolved, there was 1 death day-1 post-PC (non-procedure related), 1 index cholecystectomy, 1 chronic complicated cholecystitis. 28.6% devel-oped complications, 2 major (1 late biliary peritonitis with subphrenic abscess, and 1 cholecys-to-cutaneous fistula with abdominal wall abscess), 17.9% had dislodged drains, and 14.3% had other minor complications. 20 (71.4%) patients had bile cultures taken, of which 60% were posi-tive. 17.9% patients were readmitted with AC, 1 had repeat PC. 21.4% had subsequent ERCP. 32.1% underwent subsequent cholecystectomy, of which there was 1 laparoscopic cholecystec-tomy (LC) index, 4 elective (3 LC, 1 open) and 4 emergencies (2 LC, 1 LC subtotal, 1 failed open with drain insertion). There were no procedure related mortalities although 39.3% patients died during the follow-up period, a reflection of their pre-existing multi-morbidity.
Conclusion: PC is both safe and effective with significant procedural success rates and resolu-tion rates. There are few major complications but significant morbidities including high rates of dislodged drains. There is a high readmission rate with further biliary disease and high rates of subsequent choledocholithiasis requiring subsequent ERCP. Only one-third of patients have subsequent cholecystectomy. Further RCTs are required to determine whether PC or LC is a su-perior option in high-risk patients.
Gallstone (GS) related diseases, of which AC is the most common, are some of the most fre-quently encountered acute surgical emergencies. 10-15% of the adult population in developed countries is affected by gallstones, with 20% of symptomatic patients presenting as AC . The incidence of gallstone disease increases with age , as do co-morbidities, making management of AC challenging.
AC can be calculous (approximately 90% cases) or acalculous. Acute Calculous Cholecystitis (ACC) results from gallstone impaction either at the neck of the gallbladder (GB) or in the cystic duct. This results in obstruction to the outflow of bile, with subsequent distension, oedema and inflammation of the GB wall. If the obstruction persists, bacterial superinfection occurs leading to AC, which can progress to empyema. Ischaemia and necrosis can also supervene, leading to a gangrenous GB. Subsequent perforation can lead to localized abscess formation or even peri-tonitis. Patients may become critically ill from sepsis and its sequelae if left untreated. Acute acalculous cholecystitis (AAC) occurs in absence of GS. It occurs in the critically ill and is thought to be related to either bile stasis or ischaemia of the GB.
Laparoscopic cholecystectomy (LC) under general anaesthetic is the standard treatment of ACC and is performed in the early acute phase of the disease, in patients fit to undergo the proce-dure. However, in high-risk critically ill patients where operative intervention poses a significant mortality risk, PC has an established role. PC can be used with two intentions. Firstly, it is most commonly used as a bridge or temporizing measure in AC, allowing elective planned cholecystectomy when the patient is more stable. Secondly, it can be used as a definitive treatment in multi-morbid patients who are unfit for surgery. Thirdly, in patients with AAC, it is the definitive procedure.
The Tokyo Guidelines recommend appropriate intervention for different grades of AC (see Table 1) . The severity of inflammation of the GB is associated with the difficulty of LC and carries in-creased risks of bile leak, common bile duct injury or conversion to an open procedure, espe-cially if LC is performed for severe AC beyond one week after the onset of symptoms. Grade I ACC (mild) is associated with no organ dysfunction and limited disease in the gallbladder, grade II (moderate) is associated with no organ dysfunction but extensive inflammatory gallbladder disease, making cholecystectomy difficult, whilst Grade III is associated with organ failure. Grade I patients are candidates for early LC; grade II patients could have either LC or PC (also called percutaneous transhepatic gallbladder drainage - PTGBD); immediate PC/PTGBD is strongly recommended for grade III patients.
Table 1: Tokyo Guidelines 2013: TG13 flowchart for the management of acute cholangitis and cholecystitis .
PC was initially described in 1867 but was first performed under USS guidance in 1980. It is a minimally invasive radiological procedure that involves placement of a drainage catheter into the lumen of the GB, under aseptic conditions. It is performed under image guidance via either the trocar or modified Seldinger technique. Indications of PC include calculous or acalculous cholecystitis, cholangitis, and biliary obstruction. The majority of patients undergoing PC are generally unfit for a cholecystectomy at initial presentation (Figure 1) .
Figure 1: Transhepatic vs transperitoneal percutaneous cholecystostomy .
There are 2 approaches, trans-hepatic and trans-peritoneal, each with its own advantages and disadvantages. With the trans-hepatic route the catheter is passed through the liver (extra-peritoneal) into the gallbladder. This theoretically not only reduces the risk of bile peritonitis but gives greater stability to the catheter. If the gallbladder is significantly distended, the trans-peritoneal route becomes more feasible. Since the trans-peritoneal approach carries the risk of bile peritonitis however, the trans-hepatic route is generally preferred (Figure 2). Ultrasound-guided cholecystostomy with an 8-French drainage catheter (white arrow) placed into the gallbladder using the trocar tech-nique (same patient with Figure 1). White asterisk indicates posterior acoustic shadowing from gallstones .
Figure 2: Ultrasound-guided cholecystostomy with an 8-French drainage catheter (white arrow) placed into the gallbladder using the trocar technique (same patient with Fig. 1). White asterisk indicates posterior acoustic shadowing from gallstones .
There are usually no absolute contraindications to PC, but relative contraindications include co-agulopathy (needs correction), ascites, GB full of stones preventing access, and GB tumor. Complications associated with PC either are immediate or can occur within days. These include haemorrhage, bile leak/bile peritonitis, pneumothorax, bowel perforation (usually via the trans-peritoneal route), colonization of the GB, infection/abscesses and catheter displace-ment/migration (most common).
The main aim of this study was to evaluate outcomes following PC at our Health Board. Primary outcomes were success rates, rate of resolution of AC, complications, readmission rates and subsequent cholecystectomy performed. Secondary outcomes were bile cultures, repeat proce-dures and other procedures performed.
A total of 28 patients were identified on WelshPAS (Digital Patient Record System) as having undergone PC between 2011 and August 2020. Patient data was extracted for retrospective analysis using Welsh Clinical Portal and patient medical notes. Data collected included patient demographics, ASA grades, calculous vs acalculous cholecystitis, procedural success rates, reso-lution rates, complications, bile cultures taken (and organisms cultured), re-admissions rates, further procedures and subsequent cholecystectomy.
The database identified 28 patients (14 male and 14 female) with an age range of 40 to 93 years (median age 73). 82% of patients were of an ASA grade of III or IV (18 patients grade III, 5 patients grade IV). The median patient follow-up was 2 years (ranging from 0 - 8 years). Ultra-sound and/or CT imaging indicated severe AC in 61% of patients (with 86% being calculous) and AC with associated empyema in 39% of patients (Table 2).
All cholecystostomies were ultrasound-guided with a 100% success rate. Resolution of acute cholecystitis occurred in 25 (89.3%) patients. Of the three unresolved patients, there was one chronic complicated cholecystitis, one index cholecystectomy and one death day-1 post chole-cystostomy (due to a cardiac cause). 28.6% of patients developed complications due to the PC, some of whom had more than one complication. Two major complications included one late biliary peritonitis (including subphrenic abscess), and one cholecysto-cutaneous fistula with abdominal wall abscess. Both these patients had sur-gical intervention and recovered . (17.9%) patients had dislodged drains (with one patient re-quiring a repeat procedure within 1 week), whilst 3 other complications were blocked drain, excessive granulation of drain site and abdominal wall cellulitis (Table 3).
20 (71.4%) patients had bile cultures taken, with 70% of these being positive. Gram-negative organisms were cultured in 8 patients (mainly coliform) whilst gram-positive organisms (3 En-terococcus, 1 Aerococus, 1 Streptococcus viridans) were cultured in 7 patients (Table 4). 17.9% of patients were readmitted with further acute cholecystitis, with one patient having a repeat PC one week after the primary procedure, and another patient having a second PC pro-cedure 2 years after the first. 21.4% of patients had subsequent endoscopic retrograde cholan-gio-pancreatography (ERCP). 32.1% of patients underwent subsequent cholecystectomy. Of these, one was an index procedure, four were elective procedures (3 laparoscopic, 1 open), and four were emergency procedures (2 laparoscopic, 1 laparoscopic subtotal, and one failed open with open cholecystostomy tube drain inserted) (Table 5 & 6). While there were no procedure-related mortalities, 11 patients (39.3%) died during the follow-up period due to pre-existing morbidities as reflected by their ASA classifications.
PC is a minimally invasive procedure that can be lifesaving for patients who are critically ill from severe AC. PC leads to resolution of AC in most cases, with high success rates and minimal mortality. It is reliable, cost effective and can be easily performed. There are associated significant morbidities however, and it does not deal with the primary source of the problem, which is gallstones.
A systematic review conducted by Winbladh et al in 2007  analysed the safety and efficacy of PC in elderly and critically ill patients. It reported a success rate of 91% in patients with con-firmed ACC and a procedure related mortality of 0.4%. The overall complication rate was low (6.2%). Our overall complication rate was much higher than this (28.6% of patients), although the majority of these were minor complications.
Bundy et al reported technical success and resolution rates of 100% . Other studies report reso-lution rates of around 90%. Our study had similar outcomes. Furtado et al showed a 29% rate of subsequent cholecystectomy,  similar to our study. The study also found that although PC was a life-saving procedure, there was significant associated morbidity, with 44% rate of choledocho-lithiasis, 27% rate tube dislodgment, and 23% rate postoperative abscess. Our results were also comparable to this.
There have been few RCTs comparing PC to emergency cholecystectomy (EC). A multi-centre RCT Netherlands  very interestingly showed LC as superior to PC drainage in treatment of high-risk patients with ACC. It demonstrated no difference in mortality between LC and PC (3% vs 9%, P=0.27) in high-risk patients with ACC. However, LC had a significantly lower major compli-cation rate than PC (12% vs 65%, P<0.001). Recurrent biliary disease occurred more often in the PC group compared to the LC group (53% vs 5%, P<0.001). In this RCT, LC not only reduced rate of major complications but also need for re-intervention for recurrent biliary disease.
Our study failed to correlate with the high mortality rate for PC in severe disease compared to the RCT, although we do acknowledge that our study was limited by low patient numbers. Our rate of major complications was also significantly lower that in this study. However, our study did show significant rates of readmissions with biliary disease as well as a high rate of chole-docholithiasis requiring ERCP.
A study by Schlottmann et al in 2018  used a retrospective population base analysis of over 200,000 elderly patients (7516 PC vs 193,399 cholecystectomy). This study showed that there was a higher incidence of post-procedural morbidity and mortality in the PC group compared to the cholecystectomy group and concluded that elderly patients with AC should undergo chole-cystectomy. However, in 2011 Melloul et al  found that although PC and EC were both effective in the resolution of AAC sepsis, EC was associated with a higher procedure-related mortality and conversion rate and concluded that PC remains a valuable intervention.
There were some obvious limitations with our study, such as low patient numbers even though this represented a ten-year review. Additionally, patients were not necessarily graded as per Tokyo Guidelines for severity since the study period pre-dates the publication of the guidelines. As most patients were ASA III/IV, we presume they all fell into AAC grades II and III.
Our study demonstrated that PC is both safe and effective in the treatment of severe acute cholecystitis. PC was associated with a high procedural success rate and high-resolution rate, with no procedure related mortalities. There were few major complications, however a significant rate of overall complications. We found PC to be associated with significant readmission rates with further biliary disease and high rates of choledocoholithiasis requiring subsequent ERCP. Only one third of patients had subsequent cholecystectomy. In view of significant overall complication rates and recurrent biliary disease, further RCTs need to be conducted in order to determine whether LC or PC is a superior intervention in these high-risk patients. PC, however, remains an invaluable treatment option.