Advances in the management of acute cholecystitis

Abstract The diagnosis and management of acute cholecystitis (AC) continues to evolve. Among the most common surgically treated conditions in the USA, appropriate diagnosis and management of AC require astute clinical judgment and operative skill. Useful diagnostic and grading systems have been developed, most notably the Tokyo guidelines, but some recent clinical validation studies have questioned their generalizability to the US population. The timing of surgical intervention is another area that requires further investigation. US surgeons traditionally pursue laparoscopic cholecystectomy (LC) for AC patients with symptoms onset <72 hours, but for patients with symptoms over 72 hours, surgeons often elect to treat the patients with antibiotics and delay LC for 4‐6 weeks to permit the inflammation to subside. This practice has recently been called into question, as there are data suggesting that LC even for AC patients with over 72 hours of symptoms confers decreased morbidity, shorter length of stay, and reduced overall healthcare costs. Finally, the role of percutaneous cholecystostomy (PC) needs to be better defined. Traditional role of PC is a temporizing measure for patients who are poor surgical candidates. However, there are data suggesting that in AC patients with organ failure, PC patients suffered higher mortality and readmission rates when compared with a propensity‐matched LC cohort. Beyond diagnosis, the surgical management of AC can be remarkably challenging. All surgeons need to be familiar with best‐practice surgical techniques, adjunct intra‐operative imaging, and bail‐out options when performing LC.

known as acute cholecystitis (AC). Approximately 1%-2% of individuals with gallstones become symptomatic each year. 2,4 Of those with symptomatic gallstones, 10% will develop AC. 5 In people under 50 years of age, women are three times more likely than men to develop AC. 6  pared to OC, LC resulted in an increased likelihood of same-day discharge from the hospital (91% vs 70%), reduced morbidity (16% vs 36%), and lower unadjusted mortality (0.4% vs 3%). 9 Furthermore, the conversion rate from LC to OC was 9.5%. Interestingly, LC cases that were converted to OC still had lower morbidity and mortality than cases that were initiated as OC, suggesting that early LC should be the treatment of choice for AC.
Not all results are consistent with this data. A 2-year prospective multicenter survey of over 1000 patients in Belgium, including all centers, revealed that LC and OC approaches were employed in 93.2% and 6.8% of patients, respectively. 10 Independent predictive factors of an initial OC approach included history of upper abdominal surgery, age over 70 years, surgeons with more than 10 years of experience, and gangrenous cholecystitis. The conversion rate from LC to OC was 11.4%. Bile duct injuries, a devastating complication, occurred in 2.7% of the OC group and 1.1% of the LC group. 10 However, in those patients whose operation was started laparoscopically but who were converted to open, 13.7% suffered some form of biliary complication. These results suggest that operation for AC can still be associated with a significant complication rate and that we need to continue to evaluate our approach to the difficult cholecystectomy. There continue to be several areas of significant controversy (Table 1).

| D IAG NOS IS AND CL A SS IFI C ATI ON
Accurate diagnosis of cholecystitis requires a multifactorial, systematic approach that involves a detailed history, physical exam, serologic tests, and imaging. The 2007 Tokyo Guideline (TG07) provided a system of diagnostic criteria and severity grading scale for cholecystitis. 11 Subsequent studies revealed that the TG07 guidelines had a sensitivity and specificity of 85% and 50%, respectively. 12 The suboptimal specificity prompted a revision of the TG07 to include local and systemic signs of inflammation, as well as imaging findings. These new diagnostic criteria resulted in the 2013 Tokyo Guidelines (TG13), with improved sensitivity and specificity of 91% and 97%, respectively. 13 Since the establishment of TG13, a review of 216 articles, including 19 randomized controlled trials, showed that the severity grading accurately predicted mortality, 14 length of hospitalization, and laparotomy conversion rates. 15,16 Given these findings, the decision was made to decline further revisions to TG13 in the updated Tokyo Guideline 2018 (TG18).
The TG18 diagnostic criteria for AC include three components:  16 Clinically relevant AC severity grading Although the Tokyo Guidelines have provided a severity grading system, its applicability in US populations have been questioned. 18,19 Timing of surgical intervention AC patients with over 72 h of symptoms may benefit from early LC as opposed to delayed LC in 4-6 wks. 30 Indications for PC In grade III AC patients, a propensity-matched cohort study showed that those who received PC and interval LC did worse than those who did not receive PC. 19 Optimal  19 In light of these findings, the authors suggested the need for further critical evaluation and possible refinement of the Tokyo Guidelines.
The Tokyo Guidelines 2018 has since been revised to recommend that grade III AC can be effectively managed with early LC at advanced institutions with specialized surgeons. 20

| TIMING OF SURG I C AL INTERVENTI ON
The relationship between AC outcomes and the timing of surgical intervention has been the subject of ongoing study. Initial studies concluded that early LC for AC was associated with a higher conversion rate, more complications, and longer surgery times. 21  Conversion rate to open surgery and overall complication rates did not differ significantly between the two groups. 26,27 Likewise, it has been shown that of those patients with AC who are managed nonsurgically, 9.7%-23% fail treatment and undergo emergency LC, 27,28 which is associated with significantly higher mortality, morbidity, and conversion rate than elective LC. 29 At least one recent trial randomizing patients to less than or more than 72 hours between onset of symptoms and LC found no significant differences in outcomes. 30 We need further studies comparing delayed LC (>72 hours) to intentional "cool down" with operation in 4-6 weeks.

| THE D IFFI CULT CHOLEC YS TEC TOMY
It would be remiss to not include a brief overview of the evolving surgical techniques in addressing the difficult cholecystectomy which most commonly is associated with AC. Patients with technically difficult cholecystectomies are at significantly higher risk for conversion to OC and are at higher risk for biliary duct injury (BDI). 10 Risk factors that predict a difficult operation include symptoms >72 hours, WBC count greater than 18 000/mm 3  and (c) no more than two structures should be seen entering the GB. 33  to an open procedure is likely indicated. However, before conversion, thoughtful judgement is needed to determine whether an open approach will significantly facilitate the dissection.
All surgeons need to have bail-out options in their surgical armamentarium when the CVS cannot be achieved. Resecting the GB from the "dome down" is an option, although it is not without significant risk. If only the dome of the GB is exposed, operative cholecystostomy tube placement may be appropriate. If the hepatocystic triangle cannot be safely defined, the surgeon may perform a subtotal cholecystectomy, leaving the posterior wall on the liver. 36 Usually, a minimum of 2 cm GB neck is preserved, and impacted stones are removed. The neck can be either left open (fenestrating) or oversewn (reconstituting), and a drain is left in the GB fossa.

| PERCUTANEOUS CHOLEC YS TOS TOMY
Postoperative mortality rates in LC for high-risk patients such as the elderly or critically ill have been estimated at 5%-30%. 37 Among these patients, PC has been a preferred alternative as this procedure decreases postoperative mortality rates in high-risk patients to 10%-12%. 37 It is important to note that PC can be a technically difficult procedure with potentially high conversion rates. In one study, PC within 2 days of symptom onset had an 8.3% conversion rate, whereas PC between 3 and 6 days from symptoms onset had a 33.3% conversion rate. 38 Similarly, PC done within 2 days of ad-  41 There clearly is a subset of patients more optimally treated with PC ( Figure 1).
Despite the benefits of PC, the decision to pursue PC should be carefully assessed on an individual basis. In a propensity-matched cohort of Medicare patients with grade III AC, patients with PC for GB drainage had worse short-and long-term outcomes compared to those without PC. Specifically, those with tube placement had significantly higher 30-day, 90-day, and 2-year readmission and mortality rates compared to those without tube placement. 19 They were also less likely to undergo cholecystectomy in the following 2 years after hospitalization, had longer hospital stays and more complications. 19 One recent randomized trial from the Netherlands in patients with an APACHE II score of >7 was abandoned after they found significantly higher reintervention and morbidity in the PC group. 42 These conflicting results imply that more specific studies need to be conducted on the precise indications for PC. After PC, the optimal timing to remove the GB is controversial.
One study that looked at early surgery (<72 hours) and delayed surgery (>72 hours) after PC reported higher incidence of postoperative complications and longer operation time for the early surgery group, although those with early surgery had shorter hospital stays. 44 No difference in conversion rate between the two groups was observed. Another study reported early surgery after PC (<72 hours) had higher bleeding and longer operating times than delayed surgery (>5 days). 45 Yet another reported no difference between early surgery (<10 days) and delayed surgery (>10 days) regarding complication rates, operating time, conversion rates, and total hospital stay. 46 Thus, the precise timing of surgery after PC needs to be more thoroughly studied. Most surgeon in the USA will wait an interval of at least 4-6 weeks.

| SUMMARY AND CON CLUS I ON
Although AC is one of the most commonly treated surgical conditions, its diagnosis and management are complex and nuanced.
There continue to be controversies surrounding its diagnosis and classification, the optimal approach based on the timing of operation in relation to the onset and magnitude of symptoms, and the appropriate use of PC. Operatively, LC can range from straightforward cases to some of the most challenging operations in abdominal surgery. Thus, any surgeon performing LC needs to be well-versed in the various intra-operative techniques and bail-out options.

D I SCLOS U R E
Conflicts of interest: the authors declare no conflicts of interest for this article.
F I G U R E 1 A 76-year-old man presented with minimal abdominal pain, nausea, emesis, and night sweats. Ultrasound showed cholelithiasis with thickened gall bladder (GB) walls (A) and 2.8 × 2.6 × 2 cm hypoechoic collection in the right hepatic lobe (B). Computed tomography scan showed thickened and irregular GB walls, pericholecystic stranding, and intraluminal membranes, suggestive of gangrenous cholecystitis (C,D). Fluid collection adjacent to GB (red arrows) is suggestive of pericholecystic abscess.