Background: Cholelithiasis is a common surgical condition for general surgeons and is usually a teaching operation for surgical trainees. This study aimed to identify predictive factors for prolonged laparoscopic cholecystectomy (LC) time.
Methods: This is a retrospective observational study with 690 patients. Univariate and multivariate analyses were performed with operative time as the dependent factor and other parameters as independent factors.
Results: Inability to achieve the critical view of safety, choledocholithiasis, difficult abdominal access, intra-operative cholangiogram (IOC), gallstone spillage, intra-operative bile leak, and intra-operative blood loss were factors associated with prolonged LC operative time. The gender difference influenced operative time in the univariate analysis but not in the multivariate analysis. BMI (Body Mass Index) was not an influencing factor for the operative time in the univariate or multivariate analysis.
Conclusion: Choledocholithiasis and indications for IOC were pre-operative predictive factors for prolonged LC operative time in our study. Contrary to other studies, BMI did not influence operative time in LC.
Laparoscopic cholecystectomy, Operative time, BMI, Timing of surgery
LC: Laparoscopic Cholecystectomy; BMI: Body Mass Index; IOC: Intra-Operative Cholangiogram; ERCP: Endoscopic Retrograde Cholangiopancreatography; GSH: Groote Schuur Hospital; MPH: Mitchells Plain Hospital; VH: Victoria Hospital; NSH: New Somerset Hospital
Cholelithiasis is a common surgical condition for general surgeons, with a prevalence of up to 20% in the adult population in Western countries [1]. About 20% of individuals with cholelithiasis are symptomatic. Each year, up to 4% of those with symptomatic gallstones develop acute cholecystitis, the most common complication of gallstones. This incidence is even higher in the elderly population [2,3].
Early laparoscopic cholecystectomy (LC) is currently the gold standard of care for acute calculous cholecystitis. It remains one of the most frequently performed surgical procedures [4], favored over open cholecystectomy due to better post-operative outcomes [3]. However, severe acute cholecystitis with complex anatomy can still necessitate conversion to an open cholecystectomy. For over a decade, laparoscopic subtotal cholecystectomy has been the preferred option for complex cases with intricate anatomy during LC surgery [5].
In addition to laparoscopic appendectomies, LC is commonly a teaching operation for surgical trainees. However, operative time affects not only hospital costs but also correlates with the risk of post-operative systemic complications [6]. In a study conducted at Semmelweis University of Medicine in Hungary, the average operative time for LC improved from 78.3 minutes in 1994 to 53.4 minutes by the end of their "learning curve" in 2012 [7]. In contrast, a separate study in Japan and Taiwan reported an average LC operative time of 124 minutes between 2011 and 2013 [8]. Complex pathology and anatomical variants can increase both operative time and the risk of intraoperative complications, such as common bile duct injury [9]. Several factors are known to predict longer operative times in LC, including patients over the age of 55, pre-operative ERCP (endoscopic retrograde cholangiopancreatography), and a thickened gallbladder wall observed on imaging [10].
This study aims to identify the risk factors that influence LC operative time in our population and determine the average LC operative times across the Southern Metro of Cape Town.
Retrospective observational study.
From January 2019 to May 2022, all patients who underwent an intended laparoscopic cholecystectomy (LC) within the Cape Metro West region, resulting in either completion of LC or conversion to laparoscopic subtotal cholecystectomy, were entered into a database. We included all emergency and elective LC procedures performed on patients older than 18 years.
A total of 1,000 patients were recorded. However, only 690 patients with complete information were analyzed due to incomplete data capture.
Exclusion criteria:
Intended open cholecystectomy
Cholecystectomy as part of another operation (e.g., pancreaticoduodenectomy or trauma laparotomies)
The following data were collected: Age, gender, BMI, co-morbidities, timing of the procedure (elective or same admission), the hospital where the operations were performed, indications for LC (biliary colic; cholecystitis-confirmed clinically, biochemically, and radiologically; choledocholithiasis; gallstone pancreatitis; gallbladder polyps; or other gallstone complications), surgeon, assistant, whether a critical view of safety was achieved, the presence of an expected difficult gallbladder, difficult abdominal access, whether an intra-operative cholangiogram (IOC) was performed, gallstone spillage, bile leak, and estimated blood loss.
Our LC technique was based on the four-trocar method. The first goal was to achieve the critical view of safety before transecting the cystic duct and artery. If the critical view of safety could not be obtained, subtotal cholecystectomy or alternative methods of identifying the cystic duct were employed.
Surgical residents, registrars, and medical officers were classified as "Junior." The majority of surgeries performed by "Juniors" were assisted by a "Senior" (general surgeon, hepatobiliary fellow, or hepatobiliary subspecialist). Some surgeries were performed exclusively by "Juniors" without direct assistance from a "Senior," but a "Senior" was either present in the operating theatre, unscrubbed, or readily available if needed.
Data were analyzed using the statistical software GraphPad Prism 9.5.0. Operative time was used as the dependent variable. We first conducted a univariate analysis using Pearson correlation, Student's t-test, Chi-square test, and ANOVA. Multivariate analysis was then performed using multiple linear regression models, with operative time as the dependent variable. All variables with a p-value < 0.05 in the univariate analysis were included as independent factors in the regression model.
We analyzed a total of 690 patients, with a mean age of 44 years (range: 18-82) and a mean BMI of 32.6 (range: 16.4-58.8). The majority of patients were female (87%). The operations were performed across four hospitals: Groote Schuur Hospital (GSH), the tertiary referral center, and three secondary hospitals with general surgery specialists: Mitchells Plain Hospital (MPH), Victoria Hospital (VH), and New Somerset Hospital (NSH).
Univariate analysis (Table 1) identified several factors with a statistically significant influence on operative time: Gender, timing of the procedure, hospitals, surgical indications, surgeons, achievement of the critical view of safety, expected difficult gallbladder (Table 2 defines "expected difficult gallbladder"), difficult abdominal access, IOC, gallstone spillage, intra-operative bile leak, and estimated blood loss.
Table 1: Univariate analysis. View Table 1
Table 2: Expected difficult Gallbladder (if one or more of the following). View Table 2
Unexpectedly, BMI was not a statistically significant factor influencing operative time (p = 0.2212). Further ANOVA analysis (Table 3) comparing operative times across different BMI categories also failed to show statistically significant differences between BMI groups. Additionally, ANOVA tests were conducted to compare operative times among different age groups (Table 4), hospitals (Table 1), and operative indications (Table 1).
Table 3: ANOVA test for different BMI grading groups. View Table 3
Table 4: ANOVA test for different age groups. View Table 4
Variables with statistically significant results (p < 0.05) in the univariate analysis were selected for further examination using multiple linear regression models (Table 5). The multiple regression analysis identified seven factors that significantly influenced operative time: Critical view of safety, choledocholithiasis as a surgical indication, difficult abdominal access, IOC, gallstone spillage, intra-operative bile leak, and estimated blood loss.
Table 5: Multivariate analysis model. View Table 5
Of the 298 intra-operative bile leaks recorded, 267 were from the gallbladder, eight were due to common bile duct (CBD) injuries, and the remainder were from the cystic duct.
In one case of severe intra-operative blood loss (estimated at 1 liter), extensive adhesions of the omentum, transverse colon, and duodenum to the gallbladder were observed. Despite this, the procedure was successfully completed laparoscopically.
Our study's mean operative time was comparable to those reported by Yokoe, Zdichavsky, Enami, and Souadka, et al. [8,11-13] we identified several factors that influenced LC operative time. Some of these factors were statistically significant in the univariate analysis but not in the multivariate analysis.
Of the 1,000 patients in our database, 133 were under the age of 30. Initially, we believed that more young people in our population were suffering from gallstone diseases, but we later encountered a study from Pakistan that reported 48% of their cholecystectomy patients were also under 30 [14]. This trend of symptomatic gallstones at a younger age may be attributed to the increasing prevalence of obesity, physical inactivity, and early pregnancy [15].
In our univariate analysis, male patients had a longer mean operative time (15 minutes longer than females) with a significant p-value (p = 0.0003). However, this significance disappeared in the multivariate analysis (p = 0.3228) when other factors were accounted for. Unlike the study conducted by Alqahtani, et al. [16] we found that gender may not be a significant factor influencing LC operative time.
Despite the common belief that higher BMI correlates with increased operative difficulty in LC, our study did not support this. Most of our patients (85%) had a BMI > 25, with 19% classified as morbidly obese (BMI > 40). The highest BMI recorded in our study was 58.8. Contrary to the findings of Zdichavsky and Enami, et al. [11,12] BMI was not a statistically significant factor for LC operative time in our study. Neither univariate analysis nor ANOVA showed any significant correlation between BMI and operative time.
We explored the possibility that higher BMI patients were predominantly operated on by "Seniors," potentially neutralizing the effect of BMI on operative time. However, further analysis found no statistically significant difference (p = 0.1377) in the ratio of "Junior" versus "Senior" surgeons across BMI groups.
It is possible that BMI must be far above 40 to significantly influence LC operative time. Since only 17 patients in our study had a BMI greater than 50, the small sample size likely diluted the results for the class III obesity group. As a result, we were unable to detect a significant difference in operative time among different BMI groups. Nonetheless, it is difficult to dismiss the common challenges faced in morbidly obese patients, such as hepatomegaly, fatty liver changes, and intrahepatic gallbladders, which are generally expected to lengthen operative time.
While the timing of surgery showed a significant p-value (p = 0.0017) in the univariate analysis, this significance was lost in the multivariate analysis. This is likely because most biliary colic cases were elective, effectively selecting themselves out, suggesting that the surgical indication, rather than the timing of the procedure, had a greater impact on operative time in the multivariate model.
Groote Schuur Hospital, our tertiary referral center, had a longer average operative time (10 minutes longer than the secondary hospitals). This may be attributed to the complexity of cases, such as those requiring intra-operative ERCP. However, the significance of this finding was only present in the univariate analysis and did not persist in the multivariate analysis.
Among the different indications for LC, choledocholithiasis was associated with the longest operative time, while biliary colic had the shortest. This is likely due to the inflammation caused by pre-operative ERCP attempts, which can complicate the anatomy around the liver, common bile duct (CBD), and duodenum. The need for IOC or on-table ERCP in cases of choledocholithiasis may have also contributed to the longer operative times.
Previous studies have shown that a surgeon's level of experience is a predictive factor for operative time [11-13,17]. However, in our study, the mean operative time for cases performed by "Seniors" was 12 minutes longer than those performed by "Juniors." This finding, which contrasts with other studies, may be explained by the fact that simpler cases were typically managed by junior surgeons. While more complex cases, in which juniors encountered difficulties, were taken over by seniors and subsequently recorded under seniors’ names. This trend of longer operative times for "Seniors" was also observed in Zdichavsky's, et al. study [11].
Magsood reported that the presence of a "Junior" assistant resulted in longer operative times [18], likely due to intra-operative teaching. However, our study found no such correlation. In our setting, the combination of a "Senior" surgeon with a "Senior" assistant typically occurred in complex or challenging LC cases, which may explain why the assistant's level of experience did not influence operative time.
The multivariate analysis identified two parameters with the highest coefficient values and lowest p-values: critical view of safety (coefficient -20.05, p < 0.0001) and estimated blood loss (coefficient 31.62, p < 0.0001). The most significant factors prolonging operative time were the inability to achieve the critical view of safety and substantial intra-operative blood loss. Most of the variables with significant p-values in the multivariate analysis were consequences of a difficult gallbladder rather than causative factors, with the exception of choledocholithiasis and the indication for IOC, which were predictive factors identified preoperatively.
The quality of data collection in our study could have been improved. Of the initial 1,000 patients, 310 had incomplete information in the database, and retrieving the missing data from hard-copy files at various hospitals proved impossible. As this was a retrospective study, there is an inherent risk of selection bias and reporting errors.
In our study, choledocholithiasis and the indication for IOC were two pre-operative predictive factors that prolonged LC operative time by 15 and 19 minutes, respectively. Intra-operative factors associated with prolonged LC operative time included an inability to achieve the critical view of safety, difficult abdominal access, gallstone spillage, intraoperative bile leak, and significant blood loss.
Unlike findings from other studies, BMI did not appear to influence LC operative time, showing no statistical significance in both univariate and multivariate analyses. Similarly, while gender showed a correlation in the univariate analysis, this correlation disappeared when other factors were considered in the multivariate analysis.
Nothing to disclose.
The authors confirm contribution to the paper as follows: study conception and design: Hung-Jou Chang, Christo Kloppers; data collection: Hung-Jou Chang, Caitlin Moodley; analysis and interpretation of results: Hung-Jou Chang; draft manuscript preparation: Hung-Jou Chang. All authors reviewed the results and approved the final version of the manuscript.