Introduction

Imaging methods represent an advance in medicine, highlighting the etiology of clinical manifestations. Dynamic intraoperative cholangiography (diOC) is a diagnostic imaging test of the biliary tract in which a contrast dye is injected into the bile duct and fluoroscopy is performed.

In 1931, Pablo Mirizzi first described this procedure. Its use was preferably systematic -not dynamic- during conventional cholecystectomy. Since the advent of the laparoscopic approach, the use of this procedure decreased, probably due to lack of training and to being considered unnecessary^1-4.

New forms of gallbladder traction and dissection of Calot’s triangle were described in 1991 and 1995, respectively. Both techniques reduced the incidence of surgical bile duct injury, and their authors recommended the use of dIOC^5,6. It plays a significant role in prevention of major and inadvertent bile duct injury, and in diagnosing endoluminal pathology^7-9. The aim of this study is to describe and assess the learning curve for performing a dynamic intraoperative cholangiography (dIOC) in different clinical situations and learning stages in a General Surgery residency.

Material and methods

The study period was from December 2017 to December 2018, in the Instituto Médico Platense (La Plata, Buenos Aires), tertiary level of health care center. It is a prospective, descriptive and observational study in which male and female patients >15 years of age with acute or chronic cholelithiasis and indication for laparoscopic cholecystectomy were included. Patients with acalculous gallbladder disease, conventional surgery, severe acute pancreatitis or cholangitis, and residual or relapsing cholelithiasis were excluded.

The resident learning curve for dIOC was assessed, and intraoperative findings and its impact on the procedure were described. Data were collected in a spreadsheet filled in at the end of each laparoscopic cholecystectomy (Lap Chole) and placed it in an urn; every four months, data were uploaded to an Excel spreadsheet.

Results

A total of 456 Lap Chole were performed; patients’ mean age was 46.9 (± 15.9); 164 were men and 292 women. dIOC was performed in 100% of the patients.

The average duration of Lap Chole was 54 min (± 21.3), with a median of 50 min. The average time of the dIOC was 6.7 ± 11.8 min, and the median was 3 min. In mean values, dIOC was 10% of the complete surgery, and 6.7% of the Lap Chole in 50% of the interventions. In the first semester, the duration of the complete surgery was 57 ± 20 min, and the median was 53 min; the dIOC time was 6.7 ± 8 min, and the median was 4 min; and the IOC/LC ratio was a median of 8.3. In the first semester, the duration of the complete surgery was 50 ± 2 min, and the median was 45 min; the dIOC time was 8 ± 16 min, and the median was 4 min; and the IOC/LC ratio was a median of 6.2.

Figure 1 shows the learning curve over 1 year; the red dots are the values of the IOC/LC ratio, and the light-blue dotted line shows the IOC/LC values with the model adjustment. There was a significant learning that reaches a plateau from the 10th month of the study onwards.

Figure 1 Learning Curve 

Table 1 shows the median ratio of cholangiography in the Lap Chole according to diagnoses, dIOC techniques, safety parameters and actors.

Table 1 Cholangiography ratio in Lap Chole according to diagnoses, IOC techniques, safety parameters and actors 

Significant differences between the IOC/LC medians were found, when the IOC/LC ratio was higher for complex surgeries, when the lumbar puncture set was used for dIOC, and during choledocholithiasis. (Table 1). When including the variables in a multivariate median regression, a significant learning in the different semesters is observed (Semester: b = -0.0245, p ˂ 0.001); complexity hinders the learning curve (Complexity: b = 0.1151, p ˂0.001), regardless of other variables. Considering the short, medium and long duration IOC/LC ratio according to quartiles 25% (IOC/ LC = 4%), 50% (IOC/LC = 6%), and 75% (IOC/LC = 9%) respectively, a positive correlation between what was learned and the operative time is observed.

The impact was (b = -0.0063, p = 0.054) for short surgeries; for medium, (b = -0.0245, p ˂ 0.001), and for long surgeries (b = -0.0386, p ˂ 0.001), regardless of the other variables included in the model. (Table 2).

Table 2 Impact of the learning variables on cholangiography, depending on the short, medium, and long IOC/LC ratio. 

The evolution of learning in two semesters shows that the greatest impacts were: when surgeries were complex (reduced the IOC/LC ratio from 32% of Lap Chole to 6%); with choledocolithiasis treatment (reduced from 19% to 6%); in the presence of cystic lithiasis (from 10% to 6%); in repeated cystic duct incision (16% to 6%); and when cyst dissection is repeated (13% to 6%) (Table 3).

Table 3 Evolution of learning according to diagnosis, dIOC techniques, safety parameters and actors (Residents and Surgeons). 

Discussion

The study design and the setting in which it was conducted make it possible to carry it out in other health centers. We have identified two points for discussion:

-The learning curve for performing dIOC.

-The identification of cystic lithiasis and choledocholithiasis.

Unlike the publication by Bresadola V. -who compared the resident’s and the staff surgeon’s performances-, our work focused on assessing the learning curve and its impact on the resident’s learning process during Lap Chole. Systematic dIOC was the parameter to measure the residents’ performance.¹⁶

The analysis of the results show that:

The dIOC time was reduced from 4 to 2 minutes from one semester to another in 50% of the interventions, for a median surgery duration of 45-55 minutes.

The IOC/LC ratio was reduced from 8.3 of the total procedures in the first semester to 6.2 in the second semester. It was observed that, despite multiple factors that could affect the IOC/LC ratio, learning evolved even when included in the multivariate analysis. The impact of learning improved in more complex surgeries; probably due to the gradual and progressive training of the residents^17,18.

The systematic use and consequent training in dIOC allowed the diagnosis and treatment of cystic lithiasis and choledocolithiasis in 100% of the cases, with no evidence of associated morbidity and mortality, consistent with the series reported by the Division of Surgery of Hospital Argerich¹².

We cannot confirm that the absence of surgical bile duct injury in our series is strictly associated to systematic dIOC, but we do agree with the publication of Alvarez F. about the difference in prognosis if the diagnosis of an inadvertent injury occurs in the postoperative period.¹⁹

Conclusion

Today, laparoscopic cholecistectomy is the most common procedure in the Division of General Surgery; Resident training is very important to perform this surgery with the highest safety parameters. Systematic intraoperative cholangiography is one of the pillars of safe cholecystectomy and provides training in the management of transcystic route with no change in postoperative morbidity and mortality.