Total number of procedures performed for acute cholecystitis from 2002 to 2007.
Procedure type as a percentage of all procedures performed. CCT indicates cholecystostomy tube placement; CONV, conversion from LC to OC; LC, laparoscopic cholecystectomy; and OC, open cholecystectomy.
Detailed findings. A, Procedure type by year as a percentage of all procedures performed. B, Trends for open cholecystectomy (OC), conversion from laparoscopic cholecystectomy (LC) to OC (CONV), and cholecystostomy tube placement (CCT).
Management of the cohort. Missing data are the result of patients lost to follow-up. CCT indicates cholecystostomy tube placement; CONV, conversion from LC to OC; LC, laparoscopic cholecystectomies; and OC, open cholecystectomies.
Timing of eventual cholecystectomy among patients who underwent cholecystostomy tube placement (CCT).
Wiseman JT, Sharuk MN, Singla A, Cahan M, Litwin DEM, Tseng JF, Shah SA. Surgical Management of Acute Cholecystitis at a Tertiary Care Center in the Modern Era. Arch Surg. 2010;145(5):439-444. doi:10.1001/archsurg.2010.54
The advent of laparoscopy has changed the paradigm of surgical training and care delivery for the treatment of patients with acute cholecystitis (AC).
Retrospective data collection and analysis.
Hospital admissions with a primary diagnosis of AC at a tertiary care center from January 1, 2002, to January 1, 2007.
During the study period, 923 patients were admitted with a primary diagnosis of AC. One hundred fourteen patients were excluded from the study because of missing data, medical management, incomplete operative notes or documents, or metastatic gastrointestinal cancer.
Main Outcome Measures
Patient demographics, preoperative morbidity, procedures (medical and surgical), and postoperative outcomes were statistically analyzed using χ2 test, t test, and analysis of variance.
Eight hundred nine patients (87.6%) with a primary diagnosis of AC underwent surgery by 44 surgeons. Procedures included 663 laparoscopic cholecystectomies (LCs) (82.0%), 9 open cholecystectomies (1.1%), 51 conversions from LC to open cholecystectomy (6.3%), and 86 cholecystostomy tube placements (10.6%). During the study period, cholecystostomy tube placements increased, while open cholecystectomies and conversions from LC to open cholecystectomy decreased (P < .05). Laparoscopic cholecystectomy was associated with significantly better outcomes, including shorter postsurgical stay (2.2 vs 6.3 days for other modalities) and fewer complications (8.5% vs 17.0%).
Based on 5-year results from a tertiary care center, LC was performed with a low conversion rate to open surgery and was associated with decreased morbidity and mortality compared with other surgical modalities to treat AC. Our data confirm the benefits and widespread use of LC in the modern era, reflecting changes in the training paradigm and learning curve for laparoscopy.
Acute cholecystitis (AC) is a significant health care burden in the United States. The standard treatment, laparoscopic cholecystectomy (LC), is performed by most practicing surgeons today.1 This procedure was initially considered unsafe and harmful in the setting of acute inflammation of the gallbladder, but LC is now the most common procedure performed for AC in the United States, with acceptable results.1- 4 Although LC is the standard therapy for AC, it is not the only treatment option. Certain indications, including (but not limited to) comorbidities, age, and intraoperative conditions, might call for a treatment other than LC. Alternatives to treating AC include open cholecystectomy (OC), cholecystostomy tube (CCT) placement, and conversion from LC to OC (CONV).
Since the advent of laparoscopy, there has been a changing paradigm in surgical training and development of acute care surgical services, especially in the setting of AC. Reacting to trends, the surgical management of AC is dynamic. National findings have shown that LC is performed with increasing frequency to treat AC, with reduced mortality, hospital stay, and cost compared with OC.2 Current assessment is lacking about the safety and efficacy of AC treatment in a tertiary care hospital setting involving many surgeons across a wide spectrum of training and expertise. Therefore, we aimed to describe modern trends in the surgical management of patients with AC at a tertiary care center in the United States.
We created a database containing all patients (N = 923) who were admitted at UMass Memorial Medical Center with a primary diagnosis of AC from January 1, 2002, to January 1, 2007. The medical center is a component of the largest health care system in central Massachusetts, and its nonprofit integrated design provides all levels of care, from primary to tertiary. The study was reviewed by the University of Massachusetts Institutional Review Board and was exempted from board oversight because no personal identifiers for patients or surgeons were used among the registry data.
Each record in the database represents a single hospital admission and discharge. Individual records were reviewed, and pertinent history, demographics, laboratory findings, and outcomes were recorded. Follow-up data were recorded if available. Patients with a primary diagnosis of AC were identified based on International Classification of Diseases, Ninth Revision, Clinical Modification codes 574.0, 574.3, 574.6, 574.8, 575.0, or 575.12. These included patients admitted with AC who were discharged and returned for an elective cholecystectomy and patients who underwent cholecystectomy at the same admission. From this cohort, patients with a primary procedure code of cholecystectomy or related procedures were identified based on procedures performed (51.23 or 51.24 for LC, 51.22 for OC, V64.4 or V64.41 for CONV, and 51.01, 51.02, or 51.03 for CCT placement). All operative notes were reviewed to confirm intent and operative strategy. This ensured that the cohort was robust. We defined CONV as an operation begun laparoscopically but switched during surgery to an open procedure. Partial cholecystectomy procedures (n = 9) were included in the LC group, as these were performed laparoscopically and involved similar surgical anatomy. One hundred fourteen patients with missing data, medical management, incomplete operative notes or documents, or metastatic gastrointestinal cancer were excluded from the study.
Demographic and operative characteristics of patients were captured within our database. Age was maintained as a continuous variable. Race/ethnicity was categorized as white, black, American Indian or Alaskan native, Asian, Hispanic, or unknown. Primary insurance type was divided into the following 4 groups: Medicare, Medicaid, private, self-pay, or free care. Comorbidities were accounted for by scanning each patient's medical record for 10 conditions considered true comorbid conditions associated with adverse outcomes among hospitalized patients. Comorbidities included previous abdominal surgery, transfer from an intensive care unit, atrial fibrillation, coronary artery disease, previous myocardial infarction, chronic renal insufficiency, congestive heart failure, hypertension, cirrhosis, and chronic obstructive pulmonary disease. Based on these comorbidities, we created a preoperative morbidity score by adding the number of comorbidities for each patient. Diagnostic imaging and relevant findings were recorded for each patient. Laboratory values on admission included white blood cell count and total bilirubin, alkaline phosphatase, aspartate aminotransferase, and albumin levels. Intraoperative procedures, including intraoperative cholangiography and drain placement, were recorded from the operative note. Anatomic pathology reports were examined to confirm the diagnosis of AC.
In-hospital mortality was identified as any death following surgery during the hospital stay or within the first 30 days. Postoperative complications were obtained by scanning patient notes for surgically related issues. Any relevant complications were recorded in the following categories: bile leak, bile duct injury, retained stone, bleeding, wound infection, urinary tract infection, pneumonia, subhepatic fluid collection, myocardial infarction, and mortality. Each patient was assigned a postoperative complication score, which was determined by adding the number of postoperative complications for each patient.
Continuous variables were recorded as the mean (SD). Commercially available programs (SAS 9.1.3; SAS Institute, Cary, North Carolina; and Excel; Microsoft Corporation, Redmond, Washington) were used to analyze data. Statistical analyses using χ2 test, t test, analysis of variance, means test, and frequency test were conducted for all variables in the patient cohort. Patient characteristics and outcomes were analyzed after stratification by admission year and by operation performed.
Among 923 patients admitted with AC, 809 underwent surgery by 44 surgeons from January 1, 2002, to January 1, 2007, at UMass Memorial Medical Center. Forty-one patients in the cohort (5.1%) had previous nonsurgical admission for biliary colic or for biliary-related illness at UMass Memorial Medical Center. Table 1 summarizes the demographics of the cohort. All patients were assessed for the presence of 10 comorbidities listed in Table 1 before admission for AC. Hypertension was present in 307 (37.9%). The mean number of comorbidities per patient in our cohort was 1.1 (1.4). The maximum number of comorbidities was 7 (n = 3).
The number of surgical procedures performed among patients with AC increased slightly from 2002 (n = 104) to 2003 (n = 149) (Figure 1). From 2003 to 2007, the total number of procedures per year remained stable (range, 132-149). Laparoscopic cholecystectomy was the most frequently performed procedure from 2002 to 2007. Over 5 years, LC composed 82.0% of all procedures performed, CCT placement, 10.6%; CONV, 6.3%; and OC 1.1% (Figure 2). During the study period, CCTs increased, while OCs and CONVs decreased (P < .05). Conversions from LC to OC decreased in frequency from 11.1% of all procedures in 2002 to 3.4% in 2007, while CCTs increased from 4.2% of all procedures in 2002 to 15.0% in 2007 (Figure 3). Intraoperative cholangiography was performed in 43.0% of procedures and peritoneal drain placement in 38.2%.
Available laboratory values were collected for the cohort, including white blood cell count and bilirubin, alkaline phosphatase, aspartate aminotransferase, and albumin levels. Mean elevated values included the following: 11 900 (4800)/μL white blood cell count in 799 patients, 1.6 (2.1) mg/dL bilirubin level in 794 patients, and 88.9 (166.6) U/L aspartate aminotransferase level in 794 patients (to convert white blood cell count to ×109/L, multiply by 0.001; to convert bilirubin level to micromoles per liter, multiply by 17.104; and to convert aspartate aminotransferase level to microkatals per liter, multiply by 0.0167). The mean alkaline phosphatase level was in the normal range at 112.0 (108.9) U/L in 794 patients, as was the mean albumin level at 3.7 (1.9) g/dL in 91 patients (to convert alkaline phosphatase level to microkatals per liter, multiply by 0.0167; to convert albumin level to grams per liter, multiply by 10).
Available diagnostic imaging results were recorded for 781 patients. Ultrasonography was the diagnostic tool of choice and was used in 79.0% of patients, while computed tomography was used in 17.0% of patients. Radiologist readings of diagnostic images documented AC in 51.9% of patients. Pathologic reports on gallbladder specimens obtained at cholecystectomy confirmed AC in 97.0% of patients.
Table 2 compares patient demographics and outcomes for all 4 procedures. The LC group was younger and was healthier before admission (mean preoperative morbidity score, 0.8 vs 2.1 for other modalities). In addition, the LC group had significantly better outcomes, including shorter postsurgical stay (2.2 vs 6.3 days for other modalities) and fewer complications (8.5% vs 17.0% for other modalities).
Following discharge from the hospital, 10 patients were readmitted because of complications (eg, retained stones) related to their surgery. Complications were similar among patients who underwent intraoperative cholangiography during cholecystectomy and those who did not.
Five postoperative deaths (4 in the CCT group and 1 in the LC group) occurred among the cohort; these patients had a mean of 2.6 comorbidities and a mean age of 76 years. Among the entire cohort, 31 patients (3.8%) had retained a stone, and 21 patients (2.6%) had a bile leak. No myocardial infarction or bile duct injury occurred. Frequencies of other postoperative complications among the cohort are as follows: bleeding in 6 patients (0.7%), wound infection in 3 (0.4%), urinary tract infection in 5 (0.6%), pneumonia in 14 (1.7%), and subhepatic fluid collection in 3 (0.4%).
Eighty-six patients underwent CCT placement as their initial primary treatment of AC. As shown in Figure 4, surgeons placed 31 CCTs, while radiologists placed 55. More than half of the CCT group had at least 3 comorbidities compared with approximately 30% in the OC and CONV groups. Within a mean of 80 days after CCT placement, 47 patients underwent interval cholecystectomy (39 were LCs) (Figure 5). The most common postoperative complications in the CCT group were bile leakage (8 patients), pneumonia (3 patients), and a retained stone (2 patients). Eighteen patients had at least 1 complication related to their CCT, including tube replacement, recurrent pain, and tube occlusion.
Our database findings based on a retrospective study of AC surgical management at a tertiary care center confirm the known benefits of LC in the modern era. The laparoscopic approach is favored in complicated and uncomplicated cases, likely reflecting change in the paradigm of surgical training and in the learning curve for laparoscopy. Laparoscopic cholecystectomy was performed in 82.0% of patients with AC treated surgically at our center by 44 different surgeons. The use of OC as an initial operative intervention decreased over the study period. This trend correlated with increased CCT placement in difficult cases rather than OC or CONV. The surgical management of AC is safe and warrants continued investigation and specialization, as the acute process continues to be a major health care burden.
Laparoscopic cholecystectomy is the procedure of choice in the management of patients admitted with AC. Single-center and population-based studies4- 12 documented the benefits of LC in the management of various gallbladder diseases, including AC. A study1 using the Nationwide Inpatient Sample found that LCs increased from 83% in 1998 to 93% in 2005 relative to OCs. After adjusting for differences in demographics and patient selection in that study, a 4.6-fold higher risk of death was associated with OC vs LC. Despite the obvious benefits of laparoscopy, many surgeons advocate the open approach for safety. Possibly because of improved training and laparoscopic skill sets, this misconception may have dissipated. Our center has a wide array of surgeons with varying degrees of laparoscopic experience and training. We were surprised to find that most procedures studied herein were performed laparoscopically, with excellent success. The rates of myocardial infarction and bile duct injury were 0%, which is lower than those reported in other large series.3,13 The operative mortality was 0.6%.
Debate about the optimal treatment for AC is no longer as emotionally charged as it was 10 years ago. Current controversy surrounds the options in patients with AC who are extremely ill. During the past 5 years, our center has seen a trend toward CCT placement instead of other options such as OC. Further studies are ongoing to determine the effectiveness of CCT treatment, including the rate of eventual cholecystectomy, surgical vs radiologic placement, and cost-effectiveness. To date, a sole study14 compared CCT placement with conservative management among high-risk patients, concluding that it was a suitable alternative to nonoperative management. At our center, we were surprised to find so few CONVs (6.3% of patients) and OCs (1.1% of patients) as primary treatment for AC. This is even more significant given the range and diversity of 44 surgeons in a single department of surgery. Cholecystostomy tube placement essentially replaced OC in the past 2 years among most surgeons in the department.
Breakdown of the CCT group led to some notable observations. More than half of the patients eventually underwent cholecystectomy (mostly LC), many of whom had their CCT placed by a surgeon. This likely reflects more surgical input from the onset compared with the setting in which a radiologist places the CCT at the request of a practitioner. A surgical input setting is the most likely scenario for the management of CCT placement, outpatient follow-up, and eventual planning for surgery. Compared with radiologists, it is also possible that patients in whom CCTs are placed by surgeons are healthier. The time to eventual cholecystectomy was protracted (mean, 80 days) among the CCT group, affirming the complexity of managing these tubes and determining the timing for surgery. The need for eventual cholecystectomy among high-risk patients is subject to debate, as removal of the CCT may be a simple treatment in the short term, with the gallbladder decompressed. Further studies examining cost-effectiveness, patient satisfaction, and prospective outcomes are under way. Because few patients were in the CCT group, we were unable to retrospectively determine factors that may predict the use or benefit of tube placement over CONV or OC.
There are several limitations that must be considered when interpreting the results of this study. Most important, this was a retrospective study, and true surgical intent cannot be accurately assessed. We are initiating a prospective study to determine surgical intent and expectations for patients with AC. Although we tracked medical records for patients beyond their hospital stay, if available, we were unable to consider complications that may have occurred and been treated at other hospitals. For example, if a patient had a postoperative bile leak and sought treatment at an outside hospital after discharge, this information would be missing from this data set. In addition, we cannot reliably confirm whether patients underwent a procedure for AC during their initial hospitalization or whether it was performed later.
In summary, this study confirms the benefits of LC for AC in the modern era across a wide range of surgeons in a single department of surgery. Increased CCT placement compared with CONV or OC represents another change beyond the use of LC in the management of these patients. The laparoscopic approach is favored in complicated and uncomplicated settings of AC, possibly reflecting changes in the training paradigm and learning curve for laparoscopy.
Correspondence: Shimul A. Shah, MD, Department of Surgery, Surgery Outcomes Analysis and Research, University of Massachusetts Medical School, 55 Lake Ave N, Room S6-432, Worcester, MA 01655 (firstname.lastname@example.org).
Accepted for Publication: December 14, 2009.
Author Contributions: Mr Wiseman and Ms Sharuk contributed equally to this article. Study concept and design: Wiseman, Singla, and Shah. Acquisition of data: Wiseman, Sharuk, Singla, and Shah. Analysis and interpretation of data: Wiseman, Sharuk, Singla, Cahan, Litwin, Tseng, and Shah. Drafting of the manuscript: Wiseman and Shah. Critical revision of the manuscript for important intellectual content: Wiseman, Sharuk, Singla, Cahan, Litwin, Tseng, and Shah. Statistical analysis: Wiseman, Singla, Tseng, and Shah. Obtained funding: Shah. Administrative, technical, and material support: Shah. Study supervision: Wiseman, Cahan, Tseng, and Shah.
Financial Disclosures: None reported.
Funding/Support: This study was supported by the Worcester Foundation for Biomedical Research (Dr Shah).
Previous Presentation: This study was presented at the 90th Annual Meeting of the New England Surgical Society; September 13, 2009; Newport, Rhode Island; and is published after peer review and revision.