González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 https://doi.org/10.1186/s13017-021-00368-x RESEARCH ARTICLE Open Access Mortality risk estimation in acute calculous cholecystitis: beyond the Tokyo Guidelines Ana María González-Castillo1,2* , Juan Sancho-Insenser1,2, Maite De Miguel-Palacio2, Josep-Ricard Morera-Casaponsa1, Estela Membrilla-Fernández1,2, María-José Pons-Fragero2, Miguel Pera-Román1,2 and Luis Grande-Posa1,2 Abstract Background: Acute calculous cholecystitis (ACC) is the second most frequent surgical condition in emergency departments. The recommended treatment is the early laparoscopic cholecystectomy; however, the Tokyo Guidelines (TG) advocate for different initial treatments in some subgroups of patients without a strong evidence that all patients will benefit from them. There is no clear consensus in the literature about who is the unfit patient for surgical treatment. The primary aim of the study is to identify the risk factors for mortality in ACC and compare them with Tokyo Guidelines (TG) classification. Methods: Retrospective unicentric cohort study of patients emergently admitted with and ACC during 1 January 2011 to 31 December 2016. The study comprised 963 patients. Primary outcome was the mortality after the diagnosis. A propensity score method was used to avoid confounding factors comparing surgical treatment and non-surgical treatment. Results: The overall mortality was 3.6%. Mortality was associated with older age (68 + IQR 27 vs. 83 + IQR 5.5; P = 0.001) and higher Charlson Comorbidity Index (3.5 + 5.3 vs. 0+2; P = 0.001). A logistic regression model isolated four mortality risk factors (ACME): chronic obstructive pulmonary disease (OR 4.66 95% CI 1.7–12.8 P = 0.001), dementia (OR 4.12; 95% CI 1.34–12.7, P = 0.001), age > 80 years (OR 1.12: 95% CI 1.02–1.21, P = 0.001) and the need of preoperative vasoactive amines (OR 9.9: 95% CI 3.5–28.3, P = 0.001) which predicted the mortality in a 92% of the patients. The receiver operating characteristic curve yielded an area of 88% significantly higher that 68% (P = 0.003) from the TG classification. When comparing subgroups selected using propensity score matching with the same morbidity and severity of ACC, mortality was higher in the non-surgical treatment group. (26.2% vs. 10.5%). Conclusions: Mortality was higher in ACC patients treated with non-surgical treatment. ACME identifies high-risk patients. The validation to ACME with a prospective multicenter study population could allow us to create a new alternative guideline to TG for treating ACC. Trial registration: Retrospectively registered and recorded in Clinical Trials. NCT04744441 Keywords: Acute cholecystitis, Acute calculous cholecystitis, Early cholecystectomy, High-risk patient, Delayed cholecystectomy, Percutaneous cholecystostomy, Non-surgical treatment, Mortality, Tokyo Guidelines, Charlson Comorbidity Index * Correspondence: 62187@parcdesalutmar.cat 1 Department of Surgery, Autonomous University of Barcelona, Passeig Marítim de la Barceloneta, 25-29, 08003 Barcelona, Spain 2 General Surgery Department, Hospital del Mar, Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Barcelona, Spain © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 2 of 10 Background Inclusion and exclusion criteria Early laparoscopic cholecystectomy is the recommended All patients were selected if they had acute cholecystitis treatment for acute calculous cholecystitis (ACC) [1], according to the Tokyo Guidelines of 2018 (TG18) and/ but mortality in severe and/or fragile patients have pro- or received a diagnosis of ACC in the Pathology report moted alternative initial non-surgical treatments (NST) (Table 1). The study case definition was a ‘pure acute such as percutaneous cholecystostomy (PC) or exclusive cholecystitis’; therefore, patients with any other con- antibiotic treatment. comitant diagnosis potentially influencing outcome Regardless of the treatment modality, ACC does not (postoperative cholecystitis, acute cholangitis, acute pan- have a high mortality rate, being 0.6% overall and 6% in creatitis, and post-endoscopic retrograde pancreato- severe cases [2–4] according to the highly influential cholangiography, or neoplasia) were excluded from the Tokyo Guidelines (TG). final analysis (Fig. 1). The optimal strategy for managing non-high-risk pa- tients with ACC is laparoscopic cholecystectomy (LC), Variables preferably carried out as soon as possible (2–3 days) [4– Primary data were available from a prospective database 6]. There is however an active debate about the optimal maintained in File Maker v.12 (Mountainview, CA, timing for the cholecystectomy [7, 8]. A WSES recent USA), which included basic demographic data, type of analysis recommends the LC within 7 days from hospital interventions, sex, total days of admission, and complica- admission and within 10 days from the onset of symp- tions. Every record was completed by browsing the elec- toms [1]. Nevertheless, despite the vast number of publi- tronic patient record, adding laboratory, and cations on ACC, a definite management strategy for microbiology data, as well as antibiotic therapy, duration extreme cases remains elusive. Small samples, hetero- of procedure, additional procedures, and grade of acute geneity in selection [9] and diagnosis, and assorted cholecystitis according to the TG18 diagnostic criteria methodology flaws combine to obscure the definition of (Table 1). high-risk patients who may benefit from NST. Preoperative comorbidities were assessed using the Using TG to dictate diagnosis, assessment, and treat- CCI [14] and surgical risk by ASA classification [19]. ment of ACC, it was expected that the vital prognosis The type of initial treatment was classified as surgical would be greatly improved. Nevertheless, the mortality treatment (ST; cholecystectomy either by laparoscopy or of ACC clusters in specific subgroups of patients where laparotomy) or NST, which was either percutaneous it is still significant. There is a crucial need to select the cholecystostomy (PC) or intravenous antibiotics alone. best strategy for these high-risk patients [10]. Regret- The main outcome measure was the mortality after tably, once the diagnostic of ACC is established, there is the diagnostic of ACC (30 days if the patient is dis- no worldwide consensus regarding who is a high-risk pa- charged from the hospital or at any time during the tient unfit for an urgent cholecystectomy. same admission if not discharged). In the quest to define the high-risk patient, a collec- tion of aggregate scores (APACHE [11], ASA [6, 12], P- POSSUM [12, 13], Charlson Comorbidity Index (CCI) Interventions [14], AAST [15], frailty score [16], and one or multiple All patients received intravenous antibiotic therapy as organ dysfunctions [17] have been employed to predict soon as the diagnosis was formulated, according to a mortality losing precision in the process and creating fixed protocol. confusion of who is the patient unfit for surgery. The Ultrasound-guided cholecystostomy was performed primary aim of the current study was to create a simpler percutaneously with an 8-Fr catheter (SKATER ™, Argon and effective set of variables to identify high-risk patients Medical Devices, Rochester, NY, USA) by either transhe- isolating the discrete risk factors that predict mortality patic or transperitoneal insertion, at the discretion of the after ACC to select the patients best suited for NST. radiologist. LC was performed according to the French technique Methods using 4 trocars. The content of the gallbladder was evac- This was a retrospective study carried out in a single uated by Veress needle puncture when necessary. center with a dedicated surgical emergency unit, from January 2011 to December 2016, in a Metropolitan Uni- Statistical analysis versity Hospital in Barcelona, Spain. The data were The normal distribution of the quantitative variables was reviewed and completed by 2018, and the analysis was assessed using the Kolmogorov–Smirnov test, which completed in March 2020. showed that none of the variables were normally distrib- The study candidates comprised 963 consecutive pa- uted; therefore, their values were expressed as median tients with a diagnosis of acute cholecystitis. and interquartile ranges. The Mann–Whitney U non- González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 3 of 10 Table 1 Differences between survivors and non-survivors in pre-treatment and post-treatment variables Variable Odds ratiob 95% IC P Gender (M vs. F) 0.898 0.41–1.97 0.791 c ASA score II 1.03 0.99–1.08 0.666 III 1.08 1.04–1.12 0.019 IV 1.21 1.04–1.42 0.001 Tokyo International Guidelines Severity Gradingd (TGSG) II (moderate) 3.19 0.38–26.78 0.258 III (severe) 10.3 1.36–77.7 0.005 TGSG PAFI < 300 0.39 0.042–3.58 0.390 TGSG Oliguria (diuresis < 0.5 mL/kg/h) 18.19 7.47–44.29 0.001 TGSG marked local inflammation 2.26 0.811–6.29 0.109 TGSG WBC > 18,000/mm3 3.73 1.63–8.52 0.001 TGSG PT-INR > 1.5 3.83 1.6–9 0.001 TGSG renal dysfunction (creatinine> 2 mg) 12.65 5.58–28.65 0.001 TGSG neurological dysfunction 1.65 0.67–4.06 0.269 TGSG cardiovascular dysfunction (amines) 18.94 7.44–48.18 0.001 TGSG Murphy’s sign 0.5 0.23–1.11 0.082 TGSG palpable tender mass in RUAQ 1.4 0.62–3.4 0.399 TGSG tachypnea (> 20 bpm) 9.46 4–22.4 0.001 TGSG duration of complaints > 72 h 2.15 0.96–4.82 0.056 SIRS 10.21 3.45–30.27 0.001 Exitus a Survivors a 6. P N = 26 (3.6) N = 699 (96.4) Charlson Comorbidity Index 3.50 (5.3) 0 (2) 0.001 Age (years) 83.0 (5.5) 68.0 (27) 0.001 Bilirubin (mg/dL) 1.64 (0.9) 0.86 (1.1) 0.002 Creatinine (gr/dL) 1.7 (2.2) 0.80 (0.4) 0.001 Alkaline phosphatase (UI/L) 142 (152) 93 (72) 0.001 Gamma-glutamyl-transpherase UI/L) 71 (71) 123 (239) 0.317 Glutamil oxaloacetic transaminase (UI/L) 57 (167) 27 (41) 0.005 PT-INR 1.43 (0.3) 1.19 (0.2) 0.001 Lactate (gr/dL) 2.20 (3) 1.5 (1.1) 0.008 Temperature (°C) 37.0 (1.7) 36.3 (1) 0.289 WBC > 1000/mm3 17 (13) 14 (7) 0.016 Partial oxygen pressure (mmHg) 98 (10) 99 (1) 0.010 CRP (gr/dL) 23.7 (22) 14 (26) 0.110 3 Platelets (1000/mm ) 162 (162) 210 (129) 0.210 Treatment N (%) OR 95% IC 7. P Cholecystectomy as first treatment 689 (95) 0.149 0.056–0.399 0.001 Cholecystectomy as last treatment 9 (1.2) 4.181 0.499–35.041 0.242 Laparotomy as initial approach 75 (10) 25.708 9.618–68.71 0.001 Converted laparoscopy 93 (13) 0.678 0.155–2.959 1.000 Additional procedures 164 (23) 3.087 1.28–7.40 0.015 Cholecystostomy as first treatment 9 (1.2) 1.643 0.247–10.95 0.627 Cholecystostomy as rescue treatment 4 (0.5) 1.800 0.154–20.99 0.535 González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 4 of 10 Table 1 Differences between survivors and non-survivors in pre-treatment and post-treatment variables (Continued) Antibiotics as primary treatment 27 (3.7) 0.609 0.091–4.056 0.627 Post-treatment Exitus Survivors 8. P N = 26 (3.6) N = 699 (96.4) Reoperation 0 (0) 2 (0.3) 1.000 Postoperative hospital stay (days) 13 (19) 3 (5) 0.001 Postoperative complications 26 (100) 287 (41) 0.001 Postoperative complications No 412 (59) (Clavien-Dindo) I 105 (15) II 88 (12.6) IIIa 45 (6.4) IIIb 17 (2.4) IVa 16 (2.3) IVb 15 (2.1) Severee 48 (7) PAFI Pa02/Fi02, WBC white cells blood count, PT-INR prothrombin time international normalized ratio, SIRS systemic inflammatory response syndrome ≥ 2 points, RUAQ right upper abdominal quadrant, bpm breaths per minute, CPR C-reactive protein a N (%) or median (IQR) b Odds ratio for mortality c ASA score 1 has been used to calculate the odds ratio for the remaining groups; ASA score V is not reported as there was only one patient in this group d Tokyo Guidelines Classification Year 2013 [18] TG I was used to calculate the odds ratio for the remaining groups e > Clavien-Dindo I parametric test was used to assess the significance of dif- (25%) received PC as primary approach. Subsequently, 4 ferences between means. (15%) patients of the only antibiotics initial treatment re- The association between qualitative variables was ceived PC due to unfavorable evolution, totaling 13 assessed with the chi-square test or Fisher’s exact test, as (36%) patients with PC, 5 (38%) of whom were finally required. The increased risk of an event associated with cholecystectomized owing to a worsening clinical course a variable was reported as the odds ratio (OR) and 95% (three of them were initially treated with a primary PC confidence interval (CI). and two were a secondary PC after a failure of exclusive Additionally, as this was a retrospective observational antibiotics treatment). Overall, 698 (96.3%) urgent chole- study and the treatment groups were markedly asym- cystectomies were performed. metric, we used the propensity score matching method The median age of patients was 69 years (interquartile [20] to select and compare two subgroups of patients range, 53–80 years), and the elderly (≥ 80 years) repre- evenly balanced by severity according to the TG18 cri- sented 26% of the total. teria and by comorbidity according to the CCI. The prevalence of diabetes mellitus and heart failure A model for predicting mortality was built using bino- was noteworthy, followed by kidney disease, peripheral mial logistic regression with stepwise progressive condi- vascular disease, chronic obstructive pulmonary disease tional entry and standard baseline conditions for (COPD), and acute myocardial infarction. Most of the admission and rejection of variables with significant dif- patients were classified as ASA II (52.3%) or ASA III ferences in the univariate analysis. The discrimination (32.8%) (Table 1). power of the model was assessed by receiver operating The grade of cholecystitis according to TG18 was mild characteristic (ROC) curves and was compared with the in 21%, moderate in 39%, and severe in 40% of patients. DeLong methods. The severity factors for each group are shown in Table 1. In 689 (95%) patients, ST was initially indicated, which Results was by laparotomy in 75 (11%), and laparoscopic in 623 This study was based on a group of 963 patients with (89%) of which 93 (13%) were converted to open acute cholecystitis, from whom 725 patients with pure cholecystectomy. ACC were selected (Fig. 1). Of these, 689 underwent ini- tial ST and 36 (5%) underwent NST. Among ST pa- Mortality tients, the median time from onset of symptoms to Overall, the mortality rate of the series was 3.6%. The surgery was 3 days (25–75IQR 2–5). Among the NST pa- mortality of each treatment type, including treatments tients, 27 (75%) initially received only antibiotics and 9 applied as rescue, is detailed in Fig. 1. González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 5 of 10 Fig. 1 Patient selection and patient flow diagram Legend: ACC: acute calculous cholecystitis. ST: surgi- cal treatment. NST: non-surgical treatment. PC: percu- taneous cholecystostomy. ATB: antibiotics The patients excluded from the analysis because they developed a non-pure ACC exhibited a wide variation in mortality rates (from 0 to 28%) depending on the precise etiology (Fig. 2). Patients who died were almost 20 years older, had higher ASA scores and CCI, and had the same distribu- tion of individual comorbidities (Table 1). Each discrete comorbidity factor of the CCI carried a disparate relative risk for mortality (OR from 1 to 12), as depicted in Fig. Fig. 2 Mortality depending on the precise etiology of ACC 3. González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 6 of 10 The mortality rate of patients with severe cholecystitis Predictive model (grade III of the TG18) was nine times greater than that A multivariate model predicting mortality, the acute of patients with mild cholecystitis. The clinical and bio- cholecystitis mortality estimation (ACME), retained a set logical variables of patients with associated ORs for mor- of four variables: COPD (OR 4.66; 95% CI 1.7–12.8; P = tality are detailed in Table 1. 0.001), dementia (OR 4.12; 95% CI 1.34–12.7; P = Patients with an initial NST experienced a mortality 0.001), age > 80 years (OR 1.12; 95% CI 1.02–1.21; P = six times higher than those with initial ST. Differences 0.001), and preoperative vasoactive amines (OR 9.9; 95% in mortality between patients initially ascribed to exclu- CI 3.5–28.3; P = 0.001), which accurately predicted mor- sive antibiotics or PC were not significant (15% vs. 22%; tality in 92% of cases. The ROC curve yielded an area P = 0.62). The open cholecystectomy approach was under the curve of 88%, well above the 68% (P = 0.003) followed by a 20-fold higher mortality than the laparo- from the TG18 classification (Fig. 4). scopic cholecystectomy approach (20% vs. 1%; P = 0.001). However, the laparoscopic converted procedures Discussion had similar mortality than non-converted cholecystecto- A distinctive feature of this study is the strict selection mies (2.2% vs 3.1%; P = 1). Patients who switched from of patients. Some 238 patients were discarded to obtain NST to ST (n = 9) had an overall mortality of 11%. a homogeneous cohort of pure AAC [21, 22]. Many source papers used by the TG18 suffer from multiple etiologies pooling [3, 21–23] or only exclude the chronic Propensity score matching: comparison among cholecystitis subgroup [24], undoubtedly being a source subgroups of similar severity of statistical noise. Some reports on ACC perform some From the 689 patients with ST, a subgroup of 36 pa- intents of refined patient selection [25], but a strict ap- tients with ACC was pair-matched for identical severity proach has not been used before, and we believe it is (TG18) to the 36 patients with NST using the propensity essential. score matching method. Mortality in the NST group was The current study reports on the experience meticu- twice that of the ST group, although this difference was lously registered in a surgical emergency unit where not statistically significant owing in part to the low NST has not been adopted as recommended by TG18. prevalence (Table 2). In fact, we opted for NST in 5% of patients, whereas Fig. 3 Charlson Comorbidity Index components as risk for mortality in acute cholecystitis González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 7 of 10 Table 2 Differences between iST and iNST with the PSM analysis Variable iSTa(19) iNSTa (19) P Complications 63.2% (12) 63.2% (12) 0.631 Severe complicationsb 26.3% (5) 42.1% (8) 0.248 Postoperative complications (Clavien-Dindo) No 36.8% (7) 36.8% (7) Grade I 10.5% (2) 5.3% (1) Grade II 15.8% (3) 5.3% (1) Grade IIIa 10.5% (2) 10.5% (2) Grade IIIb 5.3% (1) 10.5% (2) Grade IV 10.5% (2) 5.3% (1) Grade V 10.5% (2) 26.5% (5) a % (patients). iST initial surgical treatment, iNST initial non-surgical treatment b Clavien-Dindo classification > IIIb 40% of the series were classified as TG grade III. This at- 0.6 to 13.5% [23, 30, 32, 33]. Nevertheless, a fifth of the titude has been adopted by a significant number of 26 patients who died in our series did so after the stan- groups with ample clinical experience in ACC manage- dardized 30 days to calculate postoperative mortality. ment [15, 26–29]. The mortality rate not including these patients would In the current series, the distributions of TG18 sever- drop to 2.8%. Furthermore, if chronic cholecystitis and ity classification were clearly more slanted toward sever- post-endoscopic retrograde cholangiopancreatography ity than the majority of other series of non-selected cases were included in the total count, as is routinely ACC where the severe proportion oscillates between 5 done in many series, the mortality would have been and 19% [23, 29–31]. This plethora of severe is mainly wrongly reported at 1.8%, missing almost half of the de- due to higher proportions of kidney and neurological ceased due to complications after ACC was diagnosed dysfunction, and/or international normalized ratio ≥ 1.5 and treated. In contrast, our series did not include acal- among our patients. culous or postoperative cholecystitis; two etiologies with very low prevalence but higher mortality and morbidity Mortality [18]. According to the TG18, the accepted mortality for acute Advanced age, high ASA score, and CCI are almost cholecystitis should be < 1% [4]. In the current study, universally cited as mortality risk factors, not only for the 30-day mortality was substantially higher (3.6%). ACC but for any emergency procedure [23, 34, 35]. Large multicenter studies reported a mortality rate from However, not every discrete comorbidity factor of the CCI carries the same relative risk for mortality, their ORs ranging from 1 to 12, as depicted in Fig. 3. That is why we chose to “deconstruct” the CCI and calculate the contribution of each component of the specific popula- tion of ACC patients. Identifying the high-risk patient Regarding reducing mortality, surgeons have been strug- gling to identify patients too frail and/or too severely af- fected, who will be better served without surgery. In this context, we label this patient as a high-risk patient. The standardization and use of therapeutic algorithms that include the preoperative assessment of surgical risk mor- tality rate of patients admitted with a diagnosis of acute cholecystitis is essential. Hypothetically, in some high- risk patients at admission, their general condition may be improved with medical treatment, and a risk evalu- ation carried out 24–48 h after admission could reassign the patient to the group of early ST. Some authors consider high-risk patients as those with Fig. 4 ROC curve for mortality: ACME vs TG18 failure of at least one organ or multiple organ González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 8 of 10 dysfunctions [36]. Patients with an ASA III or IV score Treatment modalities have an expected postoperative mortality of 5–27%, are Patients with initial NST experienced a mortality six considered high risk for cholecystectomy [37], and have times higher than those initially selected for initial ST. higher morbidity [23, 38]. In contrast, in a prospective Differences in mortality between patients initially as- observational study, González-Muñoz et al. found that cribed to exclusive antibiotic therapy or cholecystostomy patients with ASA > II and only medical treatment had a were not significant. Patients who switched from NST to mortality of 17%, whereas those operated upon early had ST (n = 9) had an overall mortality of 11%. Likewise, a no mortality [13, 24] massive observational prospective study by Endo et al. In the current series, nearly 40% of patients would compared four strategies (antibiotic therapy, cholecys- have been labeled as high risk. However, we believe that tostomy either as a definitive treatment or as a bridge to we should not exclude them from early surgery. surgery, and early cholecystectomy). Their results sup- Age alone appears in some studies as an independent ported early cholecystectomy over any other strategy, surgical risk [39]. However, age by itself does not in- with or without cholecystostomy [38]. ST is the first op- crease mortality [40]. Decreased functional reserve cap- tion in the management of ACC crystallized in the acity in addition to comorbidities, usually but not always CHOCOLATE trial, which encouraged early laparo- linked to age, are the main risk factors for mortality [39]. scopic cholecystectomy over cholecystostomy [41]. Consequently, although age per se is not regarded as a The complexity of the course of treatment, with sub- contraindication for early cholecystectomy, it has been groups of NST patients requiring rescue cholecystost- widely recognized that the use of frailty and surgical risk omy and those in turn that finally required rescue scores could contribute to achieving the best clinical cholecystectomy is reflected in Fig. 1. In every non- judgment in elderly people [40]. randomized trial, there are patients migrating from the There is no consensus on which of the already avail- four initial treatment groups (to be analyzed by able surgical risk scores better predicts postoperative intention-to-treat) to treatments finally received (to be mortality in acute cholecystitis. APACHE II > 15 [41], analyzed per protocol). This circumstance undoubtedly CCI > 6 [23], and P-POSSUM > 40 [12, 13] have all been contributed to the long period of uncertainty surround- used to support ST. None of them were specifically cre- ing the best therapy option for high-risk patients with ated for ACC; therefore, none of them are particularly ACC. advantageous over the others, and none of them offer an outstanding prediction ability. A variety of frailty scores Propensity score matching: comparison among have been widely used to predict surgical outcomes in subgroups of similar severity vast register-based studies [42]. Few studies have This technique has been used occasionally in the chole- assessed the influence of frailty on outcomes for patients cystectomy series [43, 44]. In the current study, when with CAL. Fagenson et al. selected the modified frailty comparing high-risk subgroups with the same preopera- index to successfully predict both mortality (AUC = tive morbidity and ACC severity selected by PSM, the 0.83) and Clavien-Dindo grade IV complications (AUC mortality was higher in the NST group than in the ST = 0.73) in a NSQIP-based study. Both ACME and frailty group (26.2% vs. 10.5%), probably reflecting that we re- coincide in including COPD and dementia as a strong served the NST for patients with more severe ACC. predictors for mortality [16]. We found that the individual weight of each variable Limitations of the study included in the CCI was quite different when studied in The retrospective nature of this study is undoubtedly its a cohort of ACC patients (Fig. 3). main limitation. In contrast, the limited application of NST makes this cohort valid to determine risk factors for postoperative complications but less solvent when Acute cholecystitis mortality estimation determining the best strategy for severe ACC in the By selecting variables with a greater impact on the uni- common range for a series of consecutive patients in a variate analysis, we devised the acute cholecystitis mor- single center [23, 41]. tality estimation (ACME). The model retained a set of The patients in this study had more advanced disease, four variables which accurately predicted mortality bet- the severe ACC is being overrepresented in comparison ter than the TG classification (Fig. 4). Using a similar ap- with most other series which can bias the conclusions proach, Fagenson et al. chose the modified frailty index toward a higher mortality estimation. However, we be- to identify patients with the worst prognosis after chole- lieve that this population composition makes this ana- cystectomy. They also opted for extracting the more sig- lysis more robust in assessing risk factors for mortality. nificant CCI components and ended up with a very The follow-up was limited to 30 days after the inter- similar AUC of 83% for predicting mortality [16]. vention or until the patient's discharge. More extensive González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 9 of 10 monitoring would likely discover a greater number of Consent for publication complications [13]. Nevertheless, we reported mortality Not applicable directly related to the ACC taking place beyond the 30th Competing interests day of admission, which is routinely missing from other The authors declare that they have no competing interests. reports circumscribed to 30-day mortality. Received: 9 March 2021 Accepted: 28 April 2021 Conclusion Laparoscopic cholecystectomy is the recommended References treatment for acute calculous cholecystitis, but not all 1. Pisano M, Allievi N, Gurusamy K, Borzellino G, Cimbanassi S, Boerna D, et al. 2020 World Society of Emergency Surgery updated guidelines for the patients benefit from initial surgical treatment. diagnosis and treatment of acute calculus cholecystitis. World J Emerg Surg. The Tokyo Guidelines classification would probably be 2020;15(1):1–26. https://doi.org/10.1186/s13017-020-00336-x. as effective using only 2 groups: mild (mild and moder- 2. Shaffer EA. Epidemiology and risk factors for gallstone disease: has the paradigm changed in the 21st century? Curr Gastroenterol Rep. 2005;7(2): ate) and severe ACC. 132–40. https://doi.org/10.1007/s11894-005-0051-8. Not all the factors of Charlson Comorbidity Index 3. Yokoe M, Takada T, Hwang TL, Endo I, Akazawa K, Miura F, et al. Validation carry the same risk of mortality in ACC. A new simpli- of TG13 severity grading in acute cholecystitis: Japan-Taiwan collaborative study for acute cholecystitis. J Hepatobiliary Pancreat Sci. 2017;24(6):338–45. fied, highly predictive model of mortality (ACME) in- https://doi.org/10.1002/jhbp.457. cluded a different set of variables that included age > 80 4. Okamoto K, Suzuki K, Takada T, Strasberg SM, Asbun HJ, Endo I, et al. Tokyo years, COPD, dementia, and administration of amines in Guidelines 2018: flowchart for the management of acute cholecystitis. J Hepatobiliary Pancreat Sci. 2018;25(1):55–72. https://doi.org/10.1002/jhbp.51 the preoperative period. 6. In line with the principal aim of the study, the mortal- 5. Zafar SN, Obirieze A, Adesibikan B, Cornwell EE, Fullum TM, Tran DD. ity risk score ACME could promptly identify the high- Optimal time for early laparoscopic cholecystectomy for acute cholecystitis. JAMA Surg. 2015;150(2):129–36. https://doi.org/10.1001/jamasurg.2014.2339. risk patient with ACC in our population. Its validation 6. Gutt CN, Encke J, Köninger J, Harnoss JC, Weigand K, Kipfmüller K, et al. comparing it with the TG in a prospective multicenter Acute cholecystitis: early versus delayed cholecystectomy, a multicenter study is now mandatory. randomized trial (ACDC Study, NCT00447304). Ann Surg. 2013;258(3):385–91. https://doi.org/10.1097/SLA.0b013e3182a1599b. Abbreviations 7. Kao LS, Ball CG, Chaudhury PK. Evidence-based reviews in surgery. early ACC: Acute calculous cholecystitis; ST: Surgical treatment; NST: Non-surgical cholecystectomy for cholecystitis. Ann Surg. 2018;268(6):940–2. https://doi. treatment; PC: Percutaneous cholecystostomy; ATB: Antibiotics; CCI: Charlson org/10.1097/sla.0000000000002867. Comorbidity Index; ASA: American Society Anesthesiologist score; TG: Tokyo 8. Macafee DA, Humes DJ, Bouliotis G, Beckingham IJ, Whynes DK, Lobo DN. Guidelines; OR: Odds ratio; CI: Confidence interval; ROC: Receiver operating Prospective randomized trial using cost-utility analysis of early versus characteristic; IQR: Interquartile range; COPD: Chronic obstructive pulmonary delayed laparoscopic cholecystectomy for acute gallbladder disease. Br J disease; AMI: Acute myocardial infarction Surg. 2009;96(9):1031–40. https://doi.org/10.1002/bjs.6685. 9. Campanile FC, Catena F, Coccolini F, Lotti M, Piazzalunga D, Pisano M, et al. The need for new “patient-related” guidelines for the treatment of acute Acknowledgements cholecystitis. World J Emerg Surg. 2011;6(1):2–4. https://doi.org/10.1186/174 Not applicable 9-7922-6-44. 10. Pisano M, Campanile FC. Acute calculus cholecystitis: commentary on Tokyo Authors’ contributions Guidelines 2018. J Hepatobiliary Pancreat Sci. 2018;25(3):E3–4. https://doi. Ana María González-Castillo: conceptualization, database curation and org/10.1002/jhbp.535. Accessed 2011. validation, patient selection, and writing and reviewing the manuscript. Juan 11. Kortram K, van Ramshorst B, Bollen TL, Besselink MGH, Gouma DJ, Karsten T, Sancho-Insenser: conceptualization, methodology and advance statistics, and et al. Acute cholecystitis in high risk surgical patients: Percutaneous review manuscript. Ricard-Josep Morera-Casaponsa: database curation, basic cholecystostomy versus laparoscopic cholecystectomy (CHOCOLATE trial): statistics, and table creation. Maite De Miguel-Palacio: database curation and Study protocol for a randomized controlled trial. Trials. 2012;13(1):1–7. patient selection. Estela Membrilla-Fernandez: review manuscript and data- https://doi.org/10.1186/1745-6215-13-7. base curation. Maria-José Pons-Fragero: database creation and initial patient 12. The Royal College of Surgeons of England/Department of Health. The selection. Miguel Pera-Román: review manuscript. Higher Risk General Surgical Patient: Towards Improved Care for a Luis Grande-Posa: conceptualization and review manuscript. All authors read Forgotten Group. London; 2011. https://www.rcseng.ac.uk/library-and- and approved the final manuscript. publications/rcs-publications/docs/the-higher-risk-general-surgical-patient/. 13. González-Muñoz JI, Franch-Arcas G, Angoso-Clavijo M, Sánchez-Hernández Funding M, García-Plaza A, Caraballo-Angeli M, et al. Risk-adjusted treatment This study did not receive any public or private funding. selection and outcome of patients with acute cholecystitis. Langenbeck’s Arch Surg. 2017;402(4):607–14. https://doi.org/10.1007/s00423-016-1508-y. Availability of data and materials 14. Charlson ME, Pompei P, Ales KL, MacKenzie R, MacKenzie CR. A new The datasets generated and analyzed during the current study are available method of classifying prognostic comorbidity in longitudinal studies: from the corresponding author on reasonable request, but the results are Development and validation. J Chronic Dis. 1987;40(5):373–83. https://doi. going to be able to consult in ClinicalTrials.gov for a year. org/10.1016/0021-9681(87)90171-8. 15. Hernández M, Murphy B, Aho JM, et al. Validation of the AAST EGS acute Declarations cholecystitis grade and comparison with the Tokyo guidelines. Surgery. 2018;163(4):739–46. https://doi.org/10.1016/j.surg.2017.10.041. Ethics approval and consent to participate 16. Fagenson AM, Powers BD, Zorbas KA, Karhadkar S, Karachristos A, di Carlo A, This study was approved by the clinical research ethical committee of the et al. Frailty predicts morbidity and mortality after laparoscopic Hospital del Mar and was classified as a non-clinical trial. Approval Number: cholecystectomy for acute cholecystitis: an ACS-NSQIP cohort analysis. J 2016/6973/I. Board Name: CEIm-PSMAR. Affiliation: Institut Hospital del Mar Gastrointest Surg. 2020;24(4):1–9. https://doi.org/10.1007/s11605-020-04 d’Investigacions Mèdiques. Email: ceic-psmar@imim.es 570-1. González-Castillo et al. World Journal of Emergency Surgery (2021) 16:24 Page 10 of 10 17. Resio BJ, Chiu AS, Zhang Y, Pei KY. Characterization of high mortality 36. Yokoe M, Takada T, Strasberg SM, Solomkin JS, Mayumi T, Gomi H, et al. probability operations at national surgical quality improvement program TG13 diagnostic criteria and severity grading of acute cholecystitis (with hospitals. JAMA Surg. 2020;155(1):85–8. https://doi.org/10.1001/jamasurg.201 videos). J Hepatobiliary Pancreat Sci. 2013;20(1):35–46. https://doi.org/10.1 9.3750. 007/s00534-012-0568-9. 18. Treinen C, Lomelin D, Krause C, Goede M, Oleynikov D. Acute acalculous 37. Feigal DW, Blaisdell FW. The estimation of surgical risk. Med Clin North Am. cholecystitis in the critically ill: risk factors and surgical strategies. 1979;63(6):1131–43. https://doi.org/10.1016/S0025-7125(16)31631-5. Langenbeck’s Arch Surg. 2015;400(4):421–7. https://doi.org/10.1007/s00423- 38. Anderloni A, Buda A, Vieceli F, Khashab MA, Hassan C, Repici A. Endoscopic 014-1267-6. ultrasound-guided transmural stenting for gallbladder drainage in high-risk 19. Mayhew D, Mendonca V, Murthy BVS. A review of ASA physical status – patients with acute cholecystitis: a systematic review and pooled analysis. historical perspectives and modern developments. Anaesthesia. 2019;74(3): Surg Endosc. 2016;30(12):5200–8. https://doi.org/10.1007/s00464-016-4894-x. 373–9. https://doi.org/10.1111/anae.14569. 39. Saunders DI, Murray D, Pichel AC, Varley S, Peden CJ. Variations in mortality 20. Austin PC. An introduction to propensity score methods for reducing the after emergency laparotomy: The first report of the UK emergency effects of confounding in observational studies. Multivariate Behav Res. laparotomy network. Br J Anaesth. 2012;109(3):368–75. https://doi.org/10.1 2011;46(3):399–424. https://doi.org/10.1080/00273171.2011.568786. 093/bja/aes165. 21. Yokoe M, Takada T, Hwang TL, Endo I, Akazawa K, Miura F, et al. Descriptive 40. Pisano M, Ceresoli M, Cimbanassi S, Gurusamy K, Coccolini F, Borzellino G, review of acute cholecystitis: Japan-Taiwan collaborative epidemiological et al. 2017 WSES and SICG guidelines on acute calcolous cholecystitis in study. J Hepatobiliary Pancreat Sci. 2017;24(6):319–28. https://doi.org/10.1 elderly population. World J Emerg Surg. 2019;14(1):10. https://doi.org/10.11 002/jhbp.450. 86/s13017-019-0224-7. 22. Naidu K, Beenen E, Gananadha S, Mosse C. The yield of fever, inflammatory 41. Loozen CS, Van Santvoort HC, Van Duijvendijk P, et al. Laparoscopic markers and ultrasound in the diagnosis of acute cholecystitis: a validation cholecystectomy versus percutaneous catheter drainage for acute of the 2013 Tokyo Guidelines. World J Surg. 2016;40(12):2892–7. https://doi. cholecystitis in high risk patients (CHOCOLATE): Multicentre randomised org/10.1007/s00268-016-3660-5. clinical trial. BMJ. 2018;363:k3965. https://doi.org/10.1136/bmj.k3965. 23. Endo I, Takada T, Hwang TL, Akazawa K, Mori R, Miura F, et al. 42. Castillo-Angeles M, Cooper Z, Jarman MP, Sturgeon D, Salim A, Havens JM. Optimal treatment strategy for acute cholecystitis based on predictive Association of frailty with morbidity and mortality in emergency general factors: Japan-Taiwan multicenter cohort study. J Hepatobiliary surgery by procedural risk level. JAMA Surg. 2021;156(1):68–74. https://doi. Pancreat Sci. 2017;24(6):346–61. https://doi.org/10.1002/jhbp.456. org/10.1001/jamasurg.2020.5397. 24. Yokoe M, Takada T, Strasberg SM, Solomkin JS, Mayumi T, Gomi H, et al. 43. Takemoto Y, Abe T, Amano H, Hanada K, Fujikuni N, Yoshida M, et al. New diagnostic criteria and severity assessment of acute cholecystitis in Propensity score-matching analysis of the efficacy of late cholecystectomy revised Tokyo guidelines. J Hepatobiliary Pancreat Sci. 2012;19(5):578–85. for acute cholecystitis. Am J Surg. 2017;214(2):262–6. https://doi.org/10.101 https://doi.org/10.1007/s00534-012-0548-0. 6/J.AMJSURG.2017.01.015. 25. Lin D, Wu S, Fan Y, Ke C. Comparison of laparoscopic cholecystectomy and 44. Asai K, Watanabe M, Kusachi S, Matsukiyo H, Saito T, Ishii T, et al. Evaluating delayed laparoscopic cholecystectomy in aged acute calculous cholecystitis: the timing of laparoscopic cholecystectomy for acute cholecystitis in an a cohort study. Surg Endosc. 2020;34(7):2994–3001. https://doi.org/10.1007/ experienced center based on propensity score matching. Asian J Endosc s00464-019-07091-4. Surg. 2017;10(2):166–72. https://doi.org/10.1111/ases.12353. 26. Bekki T, Abe T, Amano H, Hanada K, Kobayashi T, Noriyuki T, et al. Validation of the Tokyo guideline 2018 treatment proposal for acute cholecystitis from Publisher’s Note a single-center retrospective analysis. Asian J Endosc Surg. 2020(1):1–7. Springer Nature remains neutral with regard to jurisdictional claims in https://doi.org/10.1111/ases.12801. published maps and institutional affiliations. 27. Abe T, Amano H, Hanada K, Bekki T, Minami T, Yonehara S, et al. Efficacy and safety of early cholecystectomy for comorbid acute cholecystitis and acute cholangitis: Retrospective cohort study. Ann Med Surg. 2019;38:8–12. https://doi.org/10.1016/j.amsu.2018.10.031. 28. Pisano M, Ceresoli M, Allegri A, et al. Single center retrospective analysis of early vs. delayed treatment in acute calculous cholecystitis: application of a clinical pathway and an economic analysis. Ulus Travma Acil Cerrahi Derg. 2015;21(5):373–9. https://doi.org/10.5505/tjtes.2015.62679. 29. Amirthalingam V, Low JK, Woon W, Shelat V. Tokyo Guidelines 2013 may be too restrictive and patients with moderate and severe acute cholecystitis can be managed by early cholecystectomy too. Surg Endosc Other Interv Tech. 2017;31(7):2892–900. https://doi.org/10.1007/s00464-016-5300-4. 30. Lee S-W, Yang SS, Chang C-S, Yeh H-J. Impact of the Tokyo guidelines on the management of patients with acute calculous cholecystitis. J Gastroenterol Hepatol. 2009;24(12):1857–61. https://doi.org/10.1111/j.1440-1 746.2009.05923.x. 31. Joseph B, Jehan F, Dacey M, et al. Evaluating the relevance of the 2013 Tokyo Guidelines for the diagnosis and management of cholecystitis. J Am Coll Surg. 2018;227(1):38–43.e1. https://doi.org/10.1016/j.jamcollsurg.2018.02.016. 32. Lo C-M, Liu C-L, Fan S-T, Glas F, Lai EC, Wong J. Prospective randomized study of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Ann Surg. 1998;227(4):461–7. https://doi.org/10.1097/ 00000658-199804000-00001. 33. Rice CP, Vaishnavi KB, Schaeffer AB, et al. Operative complications and economic outcomes of cholecystectomy for acute cholecystitis. World J Gastroenterol. 2019;25(48):6916–27. https://doi.org/10.3748/wjg.v25.i48.6916. 34. Strasberg SM. Acute Calculous Cholecystitis. N Engl J Med. 2008;359(3):325. https://doi.org/10.1056/NEJMx080023. 35. Bonaventura A, Leale I, Carbone F, Liberale L, Dallegri F, Montecucco F, et al. Pre-surgery age-adjusted Charlson Comorbidity Index is associated with worse outcomes in acute cholecystitis. Dig Liver Dis. 2019;51(6):858–63. https://doi.org/10.1016/j.dld.2018.10.002.