Acute Cholecystitis Optimal Timing for Early Cholecystectomy: a French Nationwide Study

Transcription

1 DOI 0.007/s x ORIGINAL ARTICLE Acute Cholecystitis Optimal Timing for Early Cholecystectomy: a French Nationwide Study Maxime Polo & Antoine Duclos 2,3,4 & Stéphanie Polazzi 2 & Cécile Payet 2 & Jean Christophe Lifante,2,3 & Eddy Cotte,2,3 & Xavier Barth 5 & Olivier Glehen,2,3 & Guillaume Passot,2,3 Received: 3 June 205 /Accepted: 2 August 205 # 205 The Society for Surgery of the Alimentary Tract Abstract Background The recommended treatment for acute calculous cholecystitis combines antibiotics and cholecystectomy. To reduce morbidity and mortality, guidelines recommend early cholecystectomy. However, the optimal timing for surgery on first admission remains controversial. This study aims to determine the best timing for cholecystectomy in patients presenting with acute calculous cholecystitis. Study Design The French national health-care database was analyzed to identify all patients undergoing cholecystectomy for acute cholecystitis during the same hospital stay between January 200 and December 203. Data regarding patients, procedures, and hospitals characteristics were collected. The relationship between surgery s timing and clinical outcome was evaluated by multiple logistic regressions. Results Overall, 42,452 patients from 507 hospitals were included in the study. Postoperative complications requiring invasive treatment occurred in 96 patients (2.3 %), and the mortality rate was. %. Adverse postoperative outcomes intensive care admission, reoperation, and postoperative sepsis were significantly lower when surgery was performed between days and 3 (3 3.3, , and %, respectively) when compared to surgery performed on the day of admission (5.6,.2, and 5.2 %, p<0.00) or from day 5 onward (4.5,, and 6.5 %, respectively; p<0.00). Mortality was also significantly lower in patients undergoing cholecystectomy between days and 3 after admission (0.8 %)when compared to patients operated on the day of admission or after day 3 (.4 % on day 0,.2 % on day 4, and.9 % from day 5: all p<0.00). Conclusion For patients with acute calculous cholecystitis, all efforts should be made to perform cholecystectomy within 3 days after hospital admission in order to decrease morbidity and mortality. Keywords Acute cholecystitis. Surgical timing. Outcome * Guillaume Passot guillaume.passot@chu-lyon.fr Department of General and Surgical Oncology, CH Lyon Sud, Hospices Civiles de Lyon, Pierre Bénite, France Pôle Information Médicale Évaluation Recherche, Hospices Civils de Lyon, F Lyon, France EMR 3738 Université Lyon, F Lyon, France Center for Surgery and Public Health, Brigham and Women s Hospital Harvard Medical School, Boston, MA, USA Department of General Surgery, Hospices Civils de Lyon, Hop Ed. Herriot, Lyon, France Introduction Gallstones are a frequent and usually asymptomatic disease in Western countries,,2 with gallstone-related complications occurring in to 4 % of patients. Acute cholecystitis is the most common of these complications 3,4 and can be fatal in the short term due to sepsis. 5 Following acute cholecystitis, the risk of developing another gallstone-related complication such as pancreatitis or cholangitis can reach 5 %. 6,7 Consequently, a cholecystectomy is traditionally performed after an initial presentation with acute cholecystitis, 8 with laparoscopic cholecystectomy as the gold standard for the surgical treatment of symptomatic cholelithiasis. 9,0 In recent years, it has been debated whether cholecystectomy should be performed during the initial hospital admission (early cholecystectomy) or more than 6 weeks

2 after the onset of symptoms (delayed cholecystectomy). Several recent studies have shown a shorter length of hospital stay in the early group. 4 Furthermore, a population-based analysis found a lower risk of major bile duct injury with early surgery. 5 In 2007, international consensus developed the Tokyo Guidelines for diagnosis and management of cholecystitis, which recommended an early cholecystectomy for patients presenting with acute calculous cholecystitis. 6 These guidelines were revised in 203 with early surgical treatment remaining the gold standard 7 but which can be adjusted according to cholecystitis severity (recommendation A). 6 However, even if early cholecystectomy is strongly recommended, the exact timing of surgery remains unclear. A recent multicenter randomized trial concluded that laparoscopic cholecystectomy performed within 24 h of hospital admission was superior to the delayed approach in morbidity and cost. 8 A population-based analysis by Banz et al. in 20 compared clinical outcomes after laparoscopic cholecystectomy for acute cholecystitis performed at various time points after hospital admission. According to this study, laparoscopic cholecystectomy has better outcomes if performed within 48 h after hospital admission. 9 In 202, Brooks et al. advocated immediate cholecystectomy for patients with acute cholecystitis to reduce both length of hospital stay and postoperative morbidity. 20 As laparoscopic cholecystectomy is a relatively safe procedure with a mortality rate of 0.5 to 2 %, 2 no study to date has been large enough to compare serious events such as mortality in patients undergoing early cholecystectomy. The aim of our study was to determine the most appropriate timing for cholecystectomy for minimizing major complications using a large population analysis via the French nationwide health-care database. Methods Data and Population A longitudinal analysis of the French nationwide hospital database was designed to assess the optimal timing (days after hospital admission) for early cholecystectomy in patients with acute calculous cholecystitis performed according to international guidelines. The French nationwide hospital database in acute care includes information with acceptable validity about all inpatient stays that have occurred in every French public and private hospital. 22 Standard discharge abstracts for each of these hospitalizations contain compulsory demographic information; primary and secondary diagnoses using the International Classification of Diseases, 0th revision (ICD-0 codes); emergency status; intensive care unit (ICU) admission; as well as procedural codes associated with the care provided (Appendix ). Within this cohort, all adults aged 8 years or older who underwent cholecystectomy for acute calculous cholecystitis within the same hospital stay in France from January 200 to December 203 were identified. Patients surgically treated after 5 days from hospital admission (delayed cholecystectomy) were excluded. The following data were considered in analysis: age, gender, year of hospital discharge, number of comorbidities, open or laparoscopic cholecystectomy, day of surgery after hospital admission, and hospital type (public, private, or teaching hospital). We then evaluated the relationship between surgical timing (delay between admission and cholecystectomy) and clinical outcomes including postoperative mortality, ICU admission, revision surgery, acts of interventional radiology or endoscopy performed on biliary tract after surgery, and postoperative sepsis (patient safety indicators 3) 23 during the hospital stay. Codes pertaining to endoscopy or interventional radiology have been taken into account only from postoperative day. Statistical Analysis Categorical variables were presented using absolute and relative frequencies. Continuous variables were described using mean and standard deviation. Cholecystectomy timings were separated into six categories (days 0,, 2, 3, 4, and 5 or more) and compared using the chi-squared (χ 2 )test. In order to explore each outcome, we ran four multilevel logistic regressions, considering postoperative death, admission in ICU, revision surgery, and postoperative sepsis, taking into account the clustering effect of patients within certain hospitals. 24 Cholecystectomy timing was the main predictor entered with cubic polynomials to examine non-linear association. In order to control for potential confounders, patient characteristics (e.g., sex, age, Elixhauser comorbidities (Appendix 2) and cholecystectomy procedure codes) were selected a priori as clinically important covariates, as well as the year of hospital discharge to consider secular trends and coding variations. Additionally, outcomes were adjusted for hospital status to account for case mix variations across institutions. All tests were two-tailed, and p<0.05 was considered as statistically significant. Data manipulation and analyses were performed using the Statistical Analysis System (version 9.2; SAS Institute Inc., Cary, NC). Results From 200 to 203, 42,452 patients with acute calculous cholecystitis who underwent early cholecystectomy in 507

3 French hospitals were included in our study. Population characteristics are summarized in Table. Mean age at the time of surgery was 60. years (SD 8.6) with 9,957 (47 %) patients being male. Surgery was performed in public hospitals for 26,674 patients (62.8 %). Mean hospital stay was 7.8 days (SD 6.6), and overall mortality was. %. Reoperation was necessary for only 309 patients (0.7 %). Regarding cholecystectomy timing, more patients (4,750, 34.7 %) had surgery on day of hospital admission than any other day (Fig. ). Table 2 summarizes patient characteristics and clinical outcomes according to the timing of cholecystectomy. Laparoscopic cholecystectomy was performed in 35,302 patients (83.2 %). Patients operated on within the first day of admission were significantly younger and had less comorbidity than patients undergoing surgery later on. There was a statistically significant association between the time of surgery and the rate of death (p<0.00), which was lower following cholecystectomy on day, 2, or 3 after admission (0.9, 0.8, and.0 %, respectively) compared to the day of admission (.4 %) or after day 3 (.2 % at day 4 and.9 % on/after day 5). The rate of endoscopic procedures did not significantly vary depending on the timing of cholecystectomy. Patients who underwent surgery on the day of admission had a significantly higher risk Table Population characteristics N=42,452 Percent Men 9, Age (years) mean (SD) 60. (8.6) No. comorbidities a mean (SD) 0.9 (.3) Procedure Laparotomy Laparoscopy 35, Cholecystectomy timing (days) mean (SD) 2.3 (2.2) Year of hospital discharge , , , Status Public or private non-for-profit 26, Teaching Private for profit 0, Postoperative outcome Death 47. Intensive care Revision surgery Postoperative sepsis Endoscopic acts Interventional radiology % of admission to intensive care than patients operated later. Intensive care admission, reoperation, and postoperative sepsis were significantly lower when surgery was performed between days and 3 (3 3.3, , and %, respectively) than following surgery on the day of admission (5.6,.2, and 5.2 %, respectively) or on/after day 5 (4.5,, and 6.5 %, respectively) (p<0.00). The rate of interventional radiology procedures is also much higher if surgery was performed on/after day 5 (0.4 %) compared with surgery performed earlier (0., 0.2, 0.2, 0., and 0.2 % at days 0,, 2, 3, and 4, respectively; p<0.05). Table 3 reports mortality according to cholecystectomy timing and preoperative ICU admission. For patients admitted to an ICU preoperatively, more underwent surgery on the day of admission than any other day (n=53, 45.7 %) and demonstrated a non-significant decrease in mortality with increasing delay between admission and operation (2.4, 7.0, and 6.3 % for surgery at day 0, day 3, and day >4, respectively, p=0.53). Conversely, the mortality of patients not admitted to an ICU preoperatively was significantly higher if surgery was prolonged (0.87 % at day 0, 0.82 % between days and day 3, and.6 % starting from day 4, p<0.00). Figure 2 reports the predicted outcome for mortality (a), intensive care admission (b), reoperation (c), and postoperative sepsis (d) according to the timing of cholecystectomy. For each outcome, surgical timing was a significant predictor per cubic polynomial analysis (p<0.05). Each metric was high following surgery on the day of admission, then decreased and stabilized between the first and third days after admission, and then increased again in the following days. Discussion Timing of cholecystectomy (Days) Fig. Distribution of cholecystectomy timings This study of over 40,000 patients the largest populationbased study to date analyzing the timing of early cholecystectomy for acute calculous cholecystitis according to

4 Table 2 Patients characteristics according to cholecystectomy timing Cholecystectomy timing (days) p value Number of patients , Gender (%) <0.00 Men Women Age (%) <0.00 <45 years years years years Comorbidities a (%) <0.00 Procedure (%) <0.00 Laparotomy Laparoscopy Year of hospital discharge (%) < Status (%) <0.00 Public or private non-for-profit Teaching Private for profit Death (%) <0.00 Intensive care (%) <0.00 Reoperation (%) <0.00 Postoperative sepsis (%) <0.00 Endoscopy (%) Interventional radiology (%) international guidelines found that surgery within 3 days after a patient s admission was ideal. The risk of mortality was significantly higher for patients who underwent surgical treatment after day 3 post admission, including a twofold risk when performing surgery on or after day 5 compared to patients treated on day. The rates of reoperation, postoperative sepsis, and intensive care admission are also significantly higher after day 3. These findings are congruent with several recent studies reporting similar benefit in length of hospital stay, cost, conversion rates, and postoperative complications. 6, 20 Our study also showed that cholecystectomy within the first 24 h was associated with greater morbidity and mortality, which contrasts with several recent studies This is likely Table 3 Mortality according to cholecystectomy timing and preoperative ICU admission Cholecystectomy timing (days) p value Number of patients admitted in ICU preoperatively 53 (45.7 %) 35 (44.4 %) 36 (0. %) Deaths among patients admitted in ICU preoperatively 33 (2.4 %) 23 (7 %) 3 (6.3 %) Deaths among patients not admitted in ICU preoperatively 47 (0.87 %) 232 (0.82 %) 33 (.6 %) <0.00

5 (a) Mortality (%) (b) Intensive care (%) (c) (%) surgery Revision Timing of cholecystectomy (day) Timing of cholecystectomy (day) Sepsis (%) Timing of cholecystectomy (day) Timing of cholecystectomy (day) Fig. 2 Predicted outcomes death (a), intensive care (b), revision surgery (c), and postoperative sepsis (d) according to the timing of cholecystectomy. Gender, age, comorbidities, procedure, year, and hospital status were considered for adjustment (d) due to selection bias, as our day-of-surgery cohorts did not control for the severity of cholecystitis. In general, patients operated on the day of admission were younger, had less comorbidity, and would therefore have a lower surgical risk. Consequently, the higher than expected morbidity and mortality for this group can be explained by its inclusion of extremely ill patients suffering from septic sequelae who required an emergency cholecystectomy. This assertion is supported by a subgroup analysis of patients requiring preoperative ICU admission, which found that the majority (45.7 %) underwent surgery on day of admission (D0) with a very high subsequent mortality rate (2.4 %). Conversely, patients who underwent surgery on day 0 without requiring preoperative ICU admission demonstrate similar mortalitycomparedtothoseoperatedonbetweendays and 3. Accordingly, the management of patients undergoing cholecystectomy on the day of admission should probably split into separate groups that reflect the preoperative risk incurred by ICU admission. According to the recently revised Tokyo Guidelines, acute cholecystitis cases were classified into three grades of severity: grade III (severe) acute cholecystitis associated with organ dysfunction, grade II (moderate) acute cholecystitis associated with local inflammation which renders surgical intervention difficult, and grade I (mild) acute cholecystitis which does not meet the criteria of grade III or grade II acute cholecystitis. 6 Grade I would consist of patients with mild or moderate acute cholecystitis for whom operation within 24 h can be absolutely licit in agreement with recent studies 8 20 and our own subgroup analysis of patients who were not admitted to an ICU preoperatively. The second group would consist of patients with severe acute cholecystitis and a poor clinical condition. According to Tokyo Guidelines, for patients with the most severe cholecystitis (grade III), initial medical management is recommended with antibiotics, correction of hydroelectrolytic disorders, and percutaneous cholecystostomy. A 204 study by Atar et al. supports this strategy by showing that percutaneous cholecystostomy is a safe and effective procedure in critically ill patients during the acute phase of cholecystitis that also improves surgical outcomes. 25 The fact remains, however, that our database does not permit stratification of patients into groups based on Tokyo Guidelines. The main strength of our study is the large cohort of patients and the variety of sites from which data were collected, thus avoiding potential bias due to differences in clinical practice between hospitals or experience variability between surgeons and allowing the demonstration of significant differences for rare events such as mortality. Indeed, recent large analyses reported on this topic have

6 found better outcomes if operation was performed within 48 h after admission but none were large enough to show significant differences in mortality. 9,20 In the treatment of biliary complications, endoscopic procedures concern either the insertion of a biliary stent or the removal of common bile duct stones to treat bile duct injuries, bile leakage, or residual common bile duct stones. In our study, no significant correlation was found between the number of endoscopic procedures and the timing of cholecystectomy. This may reflect a shortcoming in the medical record regarding these procedures, especially in light of the direct relationship between cholecystectomy timing and bile duct injury recently described by De Mestral et al.. 5 The rate of postoperative interventional radiology procedures was significantly higher if cholecystectomy was performed starting on/ after day 5 (p=0.042). The rarity of such interventions and, to a lesser extent, the exclusion of procedures performed on the same day as cholecystectomy prevent the discovery of differences between other groups. We believe, however, that any potential differences in endoscopy or interventional radiology are minimal in scale and clinical importance to those herein described for morbidity and mortality. The main risk of bias was related to corpus data. In order to avoid systematic errors and missing data, we focused on outcomes considered to have acceptable validity and easily identifiable coding within ICD-0: acute calculous cholecystitis, mortality, revision surgery, ICU admission, and postoperative sepsis. These data are encoded by medical practitioners based on clinical, laboratory, radiological, and peroperative findings according to the ICD-0. Furthermore, several studies have demonstrated the validity and quality of such data extracted from the French nationwide hospital database. 22,26,27 For postoperative sepsis in particular, we used the patient safety indicator (PSI), which is an internationally recognized metric for this type of data. The codes for cholecystectomy are similarly robust, as they are extracted from the French common classification of medical procedures (CCAM), which is analogous to the Current Procedural Terminology (CPT) system used in American health care. Another limitation of this study was the unit used for determining cholecystectomy timing, namely the date of rather than hours passed since admission due to the absence of timestamps for admissions or procedures in the French database. This also explains our inability to describe surgical outcomes in terms of the onset of symptoms or for operations performed at different times of day, the latter being especially noteworthy in light of recent reports of an increased risk of complications with nighttime laparoscopic cholecystectomy. 28 This limitation does, however, remove potential recall bias from variable patient accounts. According to Tokyo Guidelines, the timing of cholecystectomy should be measured from the onset of symptoms. Due to the retrospective nature of this study, the delay between surgery and onset of symptoms could not be evaluated, and we used the time between hospital admissions and operation. However, access to care is easy in France, with a hospital bed/patients ratio of around 7./000 and a hospital/patient ratio as high as 4.2/00, 000. Therefore, delay between admission and operation might reflect delay between onsets of symptoms and operation. Finally, our exclusion of patients who received a cholecystectomy at a different admission or hospital but still within 5 days after initial admission and omission of perioperative medical therapies such as antibiotic therapy prior and after hospital admission represent inherent selection biases and may influence results. However in France, guidelines are well established concerning antibiotic therapy for cholecystitis and we believe any potential variance on either account is mitigated by the power and multicenter design of our study and does not limit our present interpretation of these results. Conclusion For patients presenting with acute calculous cholecystitis and without contraindication to surgery, all efforts should be made to perform cholecystectomy within 3 days after hospital admission. In the case of severely ill patients with septic complications from cholecystitis, conservative therapy should be employed and focus on supportive care and resuscitation prior to cholecystectomy. Acknowledgments The authors greatly acknowledge Sophie Polo and Peter Lundberg for their participation in the manuscript edition. The authors report no conflict of interest, sources of funding for research and publication, or previous presentation or publication of this material. Appendix Table 4 Codes relative to surgery (a), endoscopic (b), and interventional radiology (c) procedures Codes (a) HMCA006 HMFA007 ZCJA002 ZCQA00 HMFC004 ZCQC002 CODES (b) HMGE002 HMLE002 Entitled Choledocojejunostomy (laparotomy) Cholecystectomy (laparotomy) Evacuation of abdominal fluid collections (laparotomy) Exploratory laparotomy Cholecystectomy (laparoscopy) Exploratory laparoscopy Entitled Removal of common bile duct stones (oesogastroduodenoscopy) Establishment of a biliary stent (oesogastroduodenoscopy)

7 Table 4 (continued) HMPE00 CODES (c) HLHH002 HLHJ004 HMJH004 HMJH006 ZZJH00 ZZJH003 ZZJH005 ZZJH006 ZZJH007 ZZJJ004 ZZJJ007 ZZJJ008 ZZJJ03 Appendix 2 Elixhauser comorbidities include congestive heart failure, cardiac arrhythmias, valvular disease, pulmonary circulation disorders, peripheral vascular disorders, hypertension uncomplicated/complicated, paralysis, other neurological disorders, chronic pulmonary disease, diabetes uncomplicated/complicated, hypothyroidism, renal failure, liver disease, peptic ulcer disease excluding bleeding, AIDS/HIV, lymphoma, metastatic cancer, solid tumor without metastasis, rheumatoid arthritis/ collagen vascular diseases, coagulopathy, obesity, weight loss, fluid and electrolyte disorders, blood loss anemia, deficiency anemia, alcohol abuse, drug abuse, psychoses, and depression. References Endoscopic biliary sphincterotomy (oesogastroduodenoscopy) Entitled Ct-guided percutaneous drainage of liver fluid collection US-guided percutaneous drainage of liver fluid collection Ct-guided percutaneous transhepatic biliary drainage US- and/or radiologic-guided percutaneous transhepatic biliary drainage Ct-guided paracentesis drainage of abdominal fluid collection Ct-guided percutaneous drainage of abdominal fluid collection Ct-guided percutaneous drainage of several abdominal fluid collections Radiologic-guided percutaneous drainage of several abdominal fluid collections Radiologic-guided percutaneous drainage of abdominal fluid collection US-guided percutaneous drainage of several abdominal fluid collections US-guided percutaneous drainage of abdominal wall fluid collection US-guided percutaneous drainage of abdominal fluid collection US-guided paracentesis drainage of abdominal fluid collection. Jensen KH, Jørgensen T. Incidence of gallstones in a Danish population. Gastroenterology. 99 Mar;00(3): Bates T, Harrison M, Lowe D, et al. Longitudinal study of gall stone prevalence at necropsy. Gut. 992 Jan ;33(): Friedman GD. Natural history of asymptomatic and symptomatic gallstones. Am J Surg. 993 Apr;65(4): Halldestam I, Enell E-L, Kullman E, Borch K. Development of symptoms and complications in individuals with asymptomatic gallstones. Br J Surg Jun;9(6): Ausania F, Guzman Suarez S, Alvarez Garcia H, et al. Gallbladder perforation: morbidity, mortality and preoperative risk prediction. Surg Endosc. 204 Aug 27; 6. Gurusamy KS, Koti R, Fusai G, Davidson BR. Early versus delayed laparoscopic cholecystectomy for uncomplicated biliary colic. Cochrane Database Syst Rev. 203;6:CD Lawrentschuk N, Hewitt PM, Pritchard MG. Elective laparoscopic cholecystectomy: implications of prolonged waiting times for surgery. ANZ J Surg Nov;73(): Strasberg SM, Clavien PA. Overview of therapeutic modalities for the treatment of gallstone diseases. Am J Surg. 993 Apr;65(4): Ferreres AR, Asbun HJ. Technical aspects of cholecystectomy. Surg Clin North Am. 204 Apr;94(2): Martin IG, Holdsworth PJ, Asker J, et al. Laparoscopic cholecystectomy as a routine procedure for gallstones: results of an Ballcomers^ policy. Br J Surg. 992 Aug;79(8): Germanos S, Gourgiotis S, Kocher HM. Clinical update: early surgery for acute cholecystitis. Lancet May 26;369(9575): Gurusamy K, Samraj K, Gluud C, et al. Meta-analysis of randomized controlled trials on the safety and effectiveness of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Br J Surg. 200 Feb;97(2): Lau H, Lo CY, Patil NG, Yuen WK. Early versus delayed-interval laparoscopic cholecystectomy for acute cholecystitis: a metaanalysis. Surg Endosc Jan;20(): Siddiqui T, MacDonald A, Chong PS, Jenkins JT. Early versus delayed laparoscopic cholecystectomy for acute cholecystitis: a meta-analysis of randomized clinical trials. Am J Surg Jan;95(): De Mestral C, Rotstein OD, Laupacis A, et al. Comparative operative outcomes of early and delayed cholecystectomy for acute cholecystitis: a population-based propensity score analysis. Ann Surg. 204 Jan;259(): Yamashita Y, Takada T, Kawarada Y, et al. Surgical treatment of patients with acute cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg. 2007;4(): Miura F, Takada T, Strasberg SM, et al. TG3 flowchart for the management of acute cholangitis and cholecystitis. J Hepato- Biliary-Pancreat Sci. 203 Jan;20(): Gutt CN, Encke J, Köninger J, et al. Acute cholecystitis: early versus delayed cholecystectomy, a multicenter randomized trial (ACDC study, NCT ). Ann Surg. 203 Sep;258(3): Banz V, Gsponer T, Candinas D, Güller U. Population-based analysis of 43 patients with acute cholecystitis: defining the optimal time-point for laparoscopic cholecystectomy. Ann Surg. 20 Dec;254(6): Brooks KR, Scarborough JE, Vaslef SN, Shapiro ML. No need to wait: an analysis of the timing of cholecystectomy during admission for acute cholecystitis using the American College of Surgeons National Surgical Quality Improvement Program database. J Trauma Acute Care Surg. 203 Jan;74():67 73; Kimura Y, Takada T, Kawarada Y, et al. Definitions, pathophysiology, and epidemiology of acute cholangitis and cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg. 2007;4(): Chantry AA, Deneux-Tharaux C, Cans C, et al. Hospital discharge data can be used for monitoring procedures and intensive care related to severe maternal morbidity. J Clin Epidemiol. 20 Sep;64(9): McDonald K, Romano P, Geppert J, et al. Measures of Patient Safety Based on Hospital Administrative Data - The Patient

8 Safety Indicators. Technical Review 5 (Prepared by the University of California San Francisco-Stanford Evidence-based Practice Center under Contract No ). AHRQ Publication No Rockville, MD: Agency for Healthcare Research and Quality (Accessed January 8, 204, at gov/downloads/pub/evidence/pdf/psi/psi.pdf). 24. McCulloch CE, Searle SR, Neuhaus JM. Generalized linear and mixed models, 2nd Edition. Wiley, Atar E, Bachar GN, Berlin S, et al. Percutaneous cholecystostomy in critically ill patients with acute cholecystitis: complications and late outcome. Clin Radiol. 204 Jun;69(6):e Duclos A, Polazzi S, Lipsitz SR, et al. Temporal variation in surgical mortality within French hospitals. Med Care. 203 Dec;5(2): Couris CM, Polazzi S, Olive F, et al. Breast cancer incidence using administrative data: correction with sensitivity and specificity. J Clin Epidemiol Jun;62(6): Phatak UR, Chan WM, Lew DF, et al. Is nighttime the right time? Risk of complications after laparoscopic cholecystectomy at night. J A m Coll Surg. 204 Oct;29(4):78 24.