Mortality Risk Estimation in Acute Calculous Cholecystitis: Beyond the Tokyo Guidelines
Mortality Risk Estimation in Acute Calculous Cholecystitis: Beyond the Tokyo Guidelines
Acute calculous cholecystitis (ACC) is the second most frequent surgical condition in emergency departments, the complication rate of ACC is 20-55%, and the mortality rate is 0.5-15% in recent series. The Tokyo Guidelines (TG) advocate for different initial treatments of ACC with no clear evidence that all patients will benefit from them.
The objective of the study is to identify the risk factors for mortality in ACC and compare them with TG classification.
It is a retrospective cohort study conducted from January 2011 to December 2016 in a single center with a dedicated surgical emergency unit in a Metropolitan University Hospital in Barcelona, Spain. The analysis of the data was finished in March 2020.
The study candidates comprised 963 consecutive patients with a diagnosis of ACC according to the TG18 and/or received a diagnosis of ACC in the Pathology report in those that an emergent cholecystectomy was performed.
The study case definition was a 'Pure Acute Cholecystitis' (pure ACC); therefore, patients with any other concomitant diagnosis potentially influencing outcome (Postoperative cholecystitis, Acute Cholangitis, Acute Pancreatitis, Incidental Cholecystectomy, Acalculous Cholecystitis, Chronic Cholecystitis/Persistent Colic, Post-endoscopic retrograde pancreato-cholangiography, or Neoplasia) were excluded from the final analysis.
Variables:
Primary data were available from a prospective database maintained in File Maker v.12 (Mountainview, CA, USA), which included basic demographic data, type of interventions, sex, days of admission, and complications. Every record was completed by browsing the electronic patient record, adding laboratory and microbiology data, as well as antibiotic therapy, duration of procedure, additional procedures, and grade of acute cholecystitis according to the TG18 diagnostic criteria.
Preoperative comorbidities were assessed using the Charlson Comorbidity Index and surgical risk by ASA classification. The type of initial treatment was classified as Surgical Treatment (Cholecystectomy either by laparoscopy or laparotomy) or Non-Surgical Treatment, which was either percutaneous cholecystostomy or intravenous antibiotics alone.
The main outcome measure was the mortality after the diagnostic of ACC. In the patients that were discharged, 30 days after the diagnosis, if the patients was not discharged in 30 days, at any time during the same admission.
Interventions:
All patients received intravenous antibiotic therapy from the moment the diagnosis was formulated, according to a fixed protocol.
Ultrasound-guided cholecystostomy was performed percutaneously with an 8-Fr catheter (SKATER ™, Argon Medical Devices, Rochester, NY, USA) by either transhepatic or transperitoneal insertion, at the discretion of the radiologist.
Laparoscopic Cholecystectomy was performed according to the French technique using 4 trochars. The content of the gallbladder was evacuated by Veress needle puncture when necessary.
Statistical Analysis:
The normal distribution of the quantitative variables was assessed using the Kolmogorov-Smirnov test, which showed that none of the variables were normally distributed; therefore, their values were expressed as median and interquartile ranges. The Mann-Whitney U non-parametric test was used to assess the significance of differences between means.
The association between qualitative variables was assessed with the chi-square test or Fisher's exact test, as required. The increased risk of an event associated with a variable was reported as the odds ratio (OR) and 95% confidence interval (CI).
As this was a retrospective observational study and the treatment groups were markedly asymmetric, we used the propensity score matching method to select and compare two subgroups of patients evenly balanced by severity according to the TG18 criteria and by comorbidity according to the Charlson Comorbidity Index.
A model for predicting mortality was built using binomial logistic regression with stepwise progressive conditional entry and standard baseline conditions for admission and rejection of variables with significant differences in the univariate analysis. The discrimination power of the model was assessed by receiver operating characteristic (ROC) curves and was compared with the DeLong method.
Legal and Ethical considerations This study was approved by the clinical research ethical committee of the Hospital del Mar and was classified as a non-clinical trial.
Inclusion Criteria: