Current status and therapeutic strategy of acute acalculous cholecystitis: Japanese nationwide survey in the era of the Tokyo guidelines

This study aimed to clarify the incidence, therapeutic modality, and prognosis of acute acalculous cholecystitis and to reveal its optimal treatment strategy.


| INTRODUCTION
][8][9][10] Early recognition and treatment of AAC have been advocated to prevent complications and mortality.On the other hand, several studies have reported that AAC could occur in healthy individuals with a prognosis similar to ACC, and most of such patients with AAC are outpatients rather than inpatients. 7,11,12As a matter of fact, the pathophysiology of AAC is not elucidated because its pathogenesis is complex and multifactorial, including anatomical abnormalities, gallbladder ischemia, bile stasis, clinically ill status, and so on. 2he Tokyo Guidelines (TG) were published in 2007 to standardize the diagnostic criteria and treatment strategy of AC and acute cholangitis, which has been revised two times to become more prevalent worldwide. 13,14In addition, the Tokyo Guidelines 2018 (TG18) afford a practical treatment algorithm for AC, though they have few descriptions about AAC. 14 Furthermore, although there had been several reports concerning AAC after the publication of the TG, there are no large-scale surveys that have identified and assessed AAC based on the diagnostic criteria and treatment strategy of the TG. 8,15hus, we performed a nationwide survey in order to investigate the incidence, therapeutic modality, and outcomes of AAC in Japan to clarify its prognostic factors and to reveal the optimal treatment strategy for AAC.

| METHODS
We performed a questionnaire survey of AAC as a project study of the Japanese Society for Abdominal Emergency Medicine (JSAEM) and retrospectively reviewed accumulated data from our institution and 41 board-certified hospitals in JSAEM (UMIN000047631).The questionnaire was framed to ask for detailed data on AAC cases and the number of AC patients who were treated from January 2018 to December 2020.Anonymous data collection was guaranteed by the study protocol, and this study was conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects by the Ministry of Health, Labor, and Welfare of Japan in 2014 and followed the principles in the 1964 Declaration of Helsinki and its later amendments.This study protocol and opt-out method of informed consent were also approved by the Institutional Review Board of Tohoku University (2021-1-863) and the committee of JSAEM (approval no.22-3).
Collected data on patient characteristics, diagnosis, and treatment are described below; patient characteristics included age, sex, height, bodyweight, comorbidity by the Charlson comorbidity index (CCI), use of anticoagulant, and American Society of Anesthesiologists physical status (ASA-PS); disease-related variables included date of onset, medical condition and comorbidity before treatment, body temperature, laboratory data (e.g., white blood cell (WBC), C-reactive protein (CRP), total bilirubin), and bile cultures; treatment-related factors included date of initial treatment and cholecystectomy, date of discharge, methods of treatment, the procedure of cholecystectomy, operation time, intraoperative blood loss, and need for conversion; post-treatment outcomes included prognosis, complications, mortality, length of hospital stay, and recurrence of AC within 90 days after discharge.
The diagnostic criteria and severity grading system for AC were according to the TG18, and patients were divided into Grade I, Grade II, and Grade III AC. 14 AAC was defined when no stones or sludge were found in the gallbladder on imaging studies and/or operative findings.7][18] In the analysis of outcomes based on the TG18 therapeutic strategy, patients were divided into two subgroups, namely those that conformed to the TG18 (CT) group

Conclusions:
The proportions of severity grade and mortality of acute acalculous cholecystitis were found to be similar to those of acute cholecystitis, and laparoscopic cholecystectomy is recommended as an effective treatment option.

K E Y W O R D S
acalculous cholecystitis, acute cholecystitis, cholecystostomy, laparoscopic cholecystectomy, Tokyo guidelines and those that deviated from the TG18 (DT) group.In Grade I and II patients, those who did not undergo upfront cholecystectomy with ASA-PS ≤2 and age-adjusted CCI ≤5 were allocated to the DT group, and the others were defined as the CT group.As for Grade III, patients who underwent straightforward cholecystectomy in the state of ASA-PS ≥3 or age-adjusted CCI ≥4 were assigned to the DT group, and the others were allotted to the CT group.Duration of treatment was set as the period from initial treatment to discharge.An academic hospital was defined as an institution that is linked to a medical school or university performing medical education and research, and the others were defined as general hospitals.

| Statistical analysis
Descriptive statistics are expressed as mean ± standard deviation for normally distributed continuous variables, the median with an interquartile range for non-normally distributed variables, numbered as percentages for categorical variables.The conformity of data to normal distribution was evaluated with the Shapiro-Wilk test.Statistical significance was determined by the unpaired t-test or Mann-Whitney U test for continuous variables, and discrete variables were evaluated using the Chi-square test or Fisher's exact probability test.A multivariate analysis was conducted using logistic regression analysis including variables with statistical significance in the univariate analysis, and the cutoff value of continuous variables was determined using a receiver operating characteristic curve and the Youden index.All statistical tests were two-sided, and statistical significance was defined as p < .05.All analyses were performed using JMP Pro 16 (SAS Institute Inc., Cary, NC, USA).

| Patient characteristics, treatments, and outcomes of AAC
Between January 2018 and December 2020, 6136 patients with AC were treated in the participating institutions and 434 patients diagnosed as AAC were enrolled in this study.Among these patients, two were excluded because of a lack of detailed data in one and the possibility of cholangitis ascertained after registration in the other.Eventually, the detailed data of 432 patients with AAC, which made up 7.04% of AC, were collected in this study.More than half of the patients experienced upfront cholecystectomy.Cholecystostomy as an initial intervention was performed in 105 patients (24.3%) and subsequent cholecystectomy was performed in 57 patients.A total of 77 patients underwent medical treatment alone.In the entire cohort, 195 patients experienced bile culture, and 117 were diagnosed with positive bile culture.The most commonly cultured microorganism from bile was Escherichia coli (22.6%), followed by Klebsiella spp.(21.0%) and Enterococcus spp.(7.2%) (Table S1).After treatment, 19 patients suffered from sepsis and 11 patients died in hospital.

| Risk factors of in-hospital death after treatment
Patient characteristics and backgrounds were compared between patients who died in the hospital and those who survived (Table S2).In this analysis, deceased patients had significantly worse background factors such as higher ASA-PS (p < .001),higher ageadjusted CCI (p < .001),and higher grade of the TG18 severity (p = .011);however, age, sex, body mass index (BMI), and laboratory data were comparable to surviving patients.Further investigation was performed with regard to patients' comorbidity and condition before treatment (Table 2).In the univariate analysis, the rate of myocardial infarction/congestive heart failure, cerebrovascular disease/dementia, hemiplegia, and moderate to severe liver disease in the nonsurvivor group was significantly higher than that in the survivor group.As for the patients' condition, deceased patients were significantly associated with postsurgical status, during parenteral nutrition, during ICU stay, neurological dysfunction, and hepatic dysfunction.In the multivariate analysis, myocardial infarction/ congestive heart failure was found to be the only independent risk factor of in-hospital death in patients with AAC.

| Comparative studies among treatment modalities
Patient characteristics and treatment outcomes between surgical and non-surgical treatment were compared (Table 3).There were significant differences noted in the background between the groups including age, BMI, ASA-PS, age-adjusted CCI, the TG18 severity grade, and body temperature before treatment.As for treatment outcomes, patients in the nonsurgery group had a significantly longer duration of treatment and a higher rate of in-hospital death than the surgery group, although they were treated significantly earlier after onset and underwent blood transfusion less frequently.There were only five patients (4.0%) who experienced recurrence after nonsurgical treatment, although the nonsurgery group had a significantly higher rate of recurrence compared to the surgery group.
In the subgroup analysis, we subsequently compared background factors and perioperative outcomes of the patients with upfront cholecystectomy to those with cholecystectomy after cholecystostomy (Table S3).In the background factors, we found that straightforward cholecystectomy had a significantly lower grade in the TG18 severity (p = .013)and higher CRP level at onset (p = .01).However, intra-and postoperative outcomes were comparable between the two groups.On the other hand, in the comparison between open and laparoscopic cholecystectomy (LC) shown in Table 4, LC was significantly associated with better postoperative outcomes, such as lower intraoperative blood loss, lower rate of blood transfusion, less infectious complications, lower sepsis rate, shorter duration of postoperative hospital stay, and lower mortality.As for background factors, patients in the LC group were significantly younger and had lower ASA-PS, lower age-adjusted CCI, lower grade of the TG18 severity, and lower level of CRP and total bilirubin before treatment compared to those in the open cholecystectomy group.

| Therapeutic outcomes according to the TG18 strategy
To determine whether the TG18 strategy is an appropriate treatment for AAC, we compared post-treatment outcomes between the CT and DT groups (Figure 1).In the patients with Grade I and II AAC, 339 patients (85.1%) were allocated to the CT group.To alleviate differences in the backgrounds, we compared perioperative outcomes between the CT and DT groups in the patients confined to ASA-PS of ≤2 and age-adjusted CCI of ≤5 (Table 5).In this analysis, patients in the DT group had lower BMI and ASA-PS compared to the CT group.As for the post-treatment course, although morbidity and mortality were similar, the duration of treatment in the DT group was significantly longer than that in the CT group.On the other hand, in the case of Grade III, a comparative study between the CT and DT groups restricted to the patients with ASA-PS of ≥3 or age-adjusted CCI of ≥4 was implemented (Table 6), and six patients (9.5%) with ASA-PS of ≤2 and age-adjusted CCI of ≤3 were excluded from this analysis.Patients with negative predictive factors proposed by the TG18, that is, neurological dysfunction, respiratory dysfunction, and total bilirubin of ≥2 mg/ dL, were all included in the DT group. 14We found that the DT group was significantly associated with younger age, lower ASA-PS, and lower CRP level, although posttreatment outcomes were comparable to the CT group except for a higher blood transfusion rate.

| DISCUSSION
In this study, we retrospectively reviewed the present status of AAC in Japan through a questionnaire survey focusing on the background factors, therapeutic modalities, T A B L E 2 Uni-and multivariate analyses of comorbidity and condition before treatment between survival and nonsurvival patients.and prognosis.The present study is the first survey to elucidate the real-world data of AAC according to the TG and the major findings of this study are as follows: (1) AAC accounts for 7.04% of AC and its post-treatment mortality is 2.5%; (2) myocardial infarction/congestive heart failure is the only independent risk factor for in-hospital death; and

Factors
(3) cholecystectomy, especially a laparoscopic approach, is preferable when a patient with AAC is acceptable for surgery.The 7.04% occurrence rate of AAC in this study was similar to that reported in the previous series using similar criteria for AAC. 7,12However, several studies reported a higher incidence of AAC than that in the present study. 9,11ost of the studies used identical diagnostic criteria for AAC to this study, such as no stones or sludges in the gallbladder on imaging or pathological findings, although the diagnostic criteria of AC were different among studies. 6,11,15The divergence in the incidence of AAC might be attributed to the various criteria of AC.Therefore, we used the TG, which is now widely accepted, as the criteria to diagnose AC.We obtained a similar AAC occurrence rate to previous studies where AC was diagnosed pathologically, accordingly the TG18 diagnostic criteria could be equal to the pathological diagnosis. 7,12koe et al. reported the distribution of the TG severity grade in their Japan-Taiwan collaborative study of AC and showed 36.4%,46.0%, and 17.6% of patients were diagnosed as Grade I, II, and III, respectively. 19This severity grade is equivalent to that of AAC patients in the present study.In addition, they also reported almost the same level of mortality from AC as that of AAC shown in this study.These results coincide with those reported in another study. 9Therefore, it seems that the high mortality of AAC shown in previous studies was overestimated and there could be selection and publication bias. 6,8,103][4][5] We assume that the incidence and distribution of the severity grade of AAC would be identical to those of AC from these results, but further studies with a larger number of cases will be required to verify our results.
In the comparison between patients with in-hospital death and those who survived, there were no significant differences in age, sex, BMI, or laboratory data at the initial presentation.However, the nonsurvivor group was significantly associated with higher ASA-PS, higher age-adjusted CCI, and higher grade of the TG18 severity, which suggested that the in-hospital death of AAC patients was strongly affected by the comorbidity and poor condition rather than the severity of inflammation.Univariate analysis showed that the rate of neurological and hepatic dysfunction was significantly higher in the nonsurvivor group, which seems to be similar to the results reported by Endo and his colleagues. 16Although multivariate analysis showed different results, this difference in independent risk factors for mortality is thought to be due to differences in the patient background and small sample size.We should keep in mind that severe comorbidity and an increasing number of organ dysfunctions could heavily influence post-treatment outcomes, especially in AAC patients. 16,19he appropriate treatment strategy for AAC is still controversial.Ueno et al. showed that patients who underwent early LC for AAC had comparable intra-and postoperative outcomes with those for ACC, and recommended early laparoscopic surgery for AAC. 20Conversely, Abbas et al. reported that conservative treatment including antibiotics and gallbladder drainage (GBD) was the mainstay of AAC management. 10In our study, patients who underwent cholecystectomy had significantly shorter durations of treatment and lower mortality compared to those who did not, although the nonsurgery group was significantly associated with a more severe condition.Moreover, patients who experienced LC had better postoperative outcomes than those treated with open cholecystectomy.As for ACC, the CHOCOLATE trial demonstrated that LC compared with percutaneous cholecystostomy reduced the rate of major complications, utilization of healthcare resources, and cost in high-risk AC patients. 21Based on the findings of the present study and the abovementioned randomized control study, we recommend a cholecystectomy, especially LC, in cases with AAC, when the patient is in condition to withstand surgery.In contrast, GBD before surgery was proved to have no influence on the postoperative outcomes of AAC in this study.The TG18 recommended primary GBD followed by delayed cholecystectomy in patients unfit for early surgery, and a large retrospective study revealed that cholecystostomy could reduce mortality after cholecystectomy in Grade III AC. 14,16 Therefore, we think that the determination to  perform cholecystectomy or GBD followed by cholecystectomy should be individualized and depend on the patient's condition also in the case of AAC.Interestingly, the recurrence rate of AAC following a nonsurgical procedure was 4.0% in this study.Noh et al. evaluated the clinical outcomes of cholecystostomy in 271 patients with AAC and reported a recurrence rate of 2.3%. 15Gu and colleagues also reported that the recurrence rate after conservative treatment for AAC was significantly lower than that for ACC. 9 Conversely, Anderson et al. reported that cholecystostomy led to no survival benefit in AAC patients in their largest cohort study. 8heoretically, AAC does not recur after the resolution of inflammation and GBD could be a definitive treatment for AAC since AAC is provoked via complex mechanisms unlike ACC and is not induced by obstruction of the cystic duct. 2,22Although we now recommend surgical treatment for ACC, the effect of nonsurgical treatment for AAC remains to be fully elucidated.
We also elucidated the spectrum of microorganisms isolated from bile cultures in patients with AAC (Table S1).AAC can occur aseptically, and few studies have reported the microorganisms involved in AAC.However, knowledge of the microorganisms is important for the selection of appropriate antimicrobial agents for AAC, especially for empirical therapy.In this analysis, 60% of AAC patients who underwent bile culture were diagnosed as positive, and the cultured microorganisms were identical to those isolated from AC patients in the previous report. 23Therefore, antibiotics suitable for the treatment of AC could be available without differentiating between AAC and ACC.These results would support our findings that the rate of infectious complications and sepsis in patients with nonsurgical treatment was equivalent to that in patients with surgical treatment (Table 3).
It would be necessary to determine the treatment strategy using the comorbid status such as ASA-PS and CCI which were reported to be associated with mortality of AAC, therefore we evaluated the efficacy of the TG18 flow chart in the treatment for AAC. 10 The TG18 proposed a CCI of 6 or greater and ASA-PS of 3 or greater as surgical risk factors in Grade I and II AC, which required cholecystectomy following general supportive care. 14In our study, more than 80% of patients with Grade I and II AAC followed the TG18 strategy.Patients who deviated from the TG18 had significantly longer treatment times compared to those who conformed, which suggests the superiority of following the TG18 flowchart in Grades I and II even in cases with AAC.As for severity Grade III, the TG18 indicated CCI of 4 or greater and ASA-PS of 3 or greater as a high risk for cholecystectomy and recommended elective cholecystectomy after improvement of the acute illness.However, more than 90% of Grade III AAC patients in this study had ASA-PS of ≥3 or age-adjusted CCI of ≥4, namely more than 90% of patients were proposed not to undergo straightforward cholecystectomy according to the TG18 flow chart.On the contrary, post-treatment outcomes in the DT group were similar to those in the CT group except for a higher blood transfusion rate.Kohga et al. reported that LC could be performed even in Grade III AC as safely as in Grade I and II. 24As assessments of the intraoperative difficulty and surgical techniques in LC have progressed since the publication of the TG18, there might be some patients who benefitted from up-front cholecystectomy even though they had ASA-PS of ≥3 or age-adjusted CCI of ≥4 in the cases with Grade III AAC. 25 However, two deaths were actually recorded in the DT group and the DT group had significantly younger age, lower ASA-PS, and lower CRP levels than the CT group.Being able to perform surgery is not the same as being able to complete the procedure safely.Therefore, careful judgment by experienced surgeons is required when considering surgical treatment for patients with grade III AAC.
The limitations of this study chiefly pertain to its retrospective design.In addition, this study did not compare AAC with ACC directly, so we could not make distinctive findings of the characteristics and treatment outcomes of AAC from this study.There is also a possibility of underdiagnosis of ACC because more than a quarter of the cases were diagnosed only by imaging studies.Furthermore, mainly surgical units responded to our questionnaires, although patient data were collected from various institutions encompassing the whole area in Japan, which leaves considerable room for selection bias.
In conclusion, the present study clarified that the proportion with a severity grade and mortality from AAC are similar to those of AC, and LC is recommended for the treatment of AAC if a patient is eligible for surgery.The TG18 strategy was useful for the treatment of severity Grade I and II AAC.However, even in the cases with Grade III AAC, upfront cholecystectomy could be beneficial in some patients.Our results are the first step to developing an understanding and improving the treatment for AAC, and further comparative studies between AAC and ACC are needed to evaluate these results provided in the present study.

T A B L E 5
Comparative analysis between patients who conformed and those who deviated from the Tokyo Guidelines 2018 in Grade I and II acute acalculous cholecystitis.

Table 1
were enrolled from general hospitals, but the rate of AAC in AC patients was similar between academic and general hospitals (7.13% vs. 7.02%; p = .663).A total of 98 patients developed AAC during hospitalization, of which 10 were in the intensive care unit (ICU) and 88 were in hospital wards, and 28 patients developed it during parenteral nutrition.
Abbreviations: ASA-PS, American Society of Anesthesiologists physical status; CCI, Charlson comorbidity index; ICU, intensive care unit.patients Flow chart of 432 patients enrolled in the study.In Grade I and II, those who did not undergo upfront cholecystectomy with American Society of Anesthesiologists physical status (ASA-PS) ≤2 and age-adjusted Charlson comorbidity index (CCI) ≤5 (DT group) and those who did in the same condition (CT group) were compared.As for Grade III acute acalculous cholecystitis, patients who underwent upfront cholecystectomy with ASA-PS ≥3 or age-adjusted CCI ≥4 (DT group) and those who did not in the same status (CT group) were compared.TG18, Tokyo Guidelines 2018.
Note: Values with parenthesis indicate the median and interquartile ranges unless indicated otherwise.Bold type represents a significant difference.Abbreviations: ASA-PS, American Society of Anesthesiologists physical status, CCI, Charlson comorbidity index, TG, Tokyo guidelines. a Data expressed as numbers and percentages.a Data expressed as the medians and interquartile ranges.
Values with parenthesis indicate numbers and percentages unless indicated otherwise.Bold represents significant difference.Abbreviations: ASA-PS, American Society of Anesthesiologists physical status; CCI, Charlson comorbidity index; CT, conformed to the Tokyo Gideline 2018; DT, deviated from the Tokyo Gidelines 2018.Comparative analysis between patients who conformed to and deviated from the Tokyo Guidelines 2018 in Grade III acute acalculous cholecystitis.
Note:a Data are expressed as median and interquartile ranges.T A B L E 6