Related Experiment Video
Updated: Apr 30, 2026
A Novel Dual-Modal Deep Learning Approach for Real-Time Removal of Hepatic Fluorescence in Indocyanine Green-Guided Laparoscopic Cholecystectomy
Published on: April 17, 2026
A comprehensive predictive scoring method for difficult laparoscopic cholecystectomy
Mittalgodu Anantha Krishna Murthy Vivek1, Alfred Joseph Augustine1, Ranjith Rao1
1Department of Surgery, Kasturba Medical College, Manipal University, Mangalore, Karnataka - 575 001, India.
Context:
Laparoscopic cholecystectomy (LC) is the gold standard cholecystectomy. LC is the most common difficult laparoscopic surgery performed by surgeons today. The factors leading to difficult laparoscopic cholecystectomy can be predicted.
Aims:
To develop a scoring method that predicts difficult laparoscopic cholecystectomy.
Settings And Design:
Bidirectional prospective study in a medical college setup.
Materials And Methods:
Following approval from the institutional ethical committee, cases from the three associated hospitals in a medical college setup, were collected using a detailed proforma stating the parameters of difficulty in laparoscopic cholecystectomy. Study period was between May 10 and June 12. Preoperative, sonographic and intraoperative criteria were considered.
Statistical Analysis Used:
Chi Square test and Receiver Operater Curve (ROC) analysis.
Results:
Total 323 patients were included. On analysis, elderly patients, males, recurrent cholecystitis, obese patients, previous surgery, patients who needed preoperative Endoscopic retrograde cholangiopancreatography (ERCP), abnormal serum hepatic and pancreatic enzyme profiles, distended or contracted gall bladder, intra-peritoneal adhesions, structural anomalies or distortions and the presence of a cirrhotic liver on ultrasonography (USG) were identified as predictors of difficult LC. A scoring system tested against the same sample proved to be effective. A ROC analysis was done with area under receiver operator curve of 0.956. A score above 9 was considered difficult with sensitivity of 85% and specificity of 97.8%.
Conclusions:
This study demonstrates that a scoring system predicting the difficulty in LC is feasible. There is scope for further refinement to make the same less cumbersome and easier to handle. Further studies are warranted in this direction.
