Avijit Roy1 , Madhumita Mukhopadhyay2 , Shuvankar Mukherjee3 , Quazi Mehfuzur Rahman4
BACKGROUND Though Laparoscopic Cholecystectomy (LC), is considered as gold standard treatment for cholelithiasis, it is hard to predict whether it will be easy or a difficult one. At present no widely accepted scoring system can predict the difficulty and the level of difficulty in LC. Our study was done to formulate a scoring system for pre-operative prediction of difficult LC and identify those variables which contribute to difficult LC. METHODS Patients having symptomatic gall stone disease of both sexes of any age admitted under Department of Surgery in a tertiary level teaching hospital in Kolkata who subsequently underwent LC by a single experienced surgeon with > 5 years of experience in performing LC from June 2010 to May 2011 (1 year) were included in the study whereas cases of LC converted to open procedure for instrument failure were excluded from the study. We used eleven variables [age, sex, previous admission with biliary symptoms, BMI, abdominal surgical scar, palpable gall bladder, thick walled gall bladder, pericholecystic oedema, contracted / intrahepatic gall bladder, number of gall stones, stone impaction at GB neck] with relevant weightage of scoring to each. RESULTS The scoring system we used was a pre-validated one on Indian population with mild modification. Out of 108 patients, the scoring system predicted easy LC for 74, and difficult for 34. Per operatively, 68 patients had an easy and 40 a difficult LC. Univariate analysis showed h / o hospitalisation with biliary symptoms, GB wall thickness, pericholecystic oedema and stone impaction at GB neck were statistically significant predictors of difficult LC. Intra-hepatic and / or contracted GB had 100 % association with difficult LC. Multivariate analysis revealed h / o hospitalization with biliary symptoms, GB wall thickness and stone impaction at GB neck had significant contribution in difficult LC. Sensitivity of the scoring system was 95.59 % and 77.5 %, specificity 77.5 % and 95.59 %, positive predictive values 87.84 % and 91.12 % and negative predictive values 91.12 % and 87.84 % respectively for easy and difficult predictions. Discordance between preoperative prediction and per-operative observations was statistically insignificant [McNemar χ2 = 2.083 p = 0.01] which adds to strength of validity of the scoring system. CONCLUSIONS This scoring system successfully predicts difficult LC preoperatively. The study also identified the factors (h / o hospitalization with biliary symptoms, thick walled GB and stone impaction at GB neck) causing difficult LC.
