The Clinical Research On Preoperative Ultrasound In Predicting Difficult Laparoscopic Cholecystectomy

Objective: According to preoperative ultrasonic examination index acquired inpatients undergoing laparoscopic cholecystectomy, the prospective study aboutpreoperative predicting the difficult degree of laparoscopic cholecystectomy was carriedout, then an objective、simple and accurate scoring system for predicting the difficultdegree of laparoscopic cholecystectomy basing on these parameters analysis and treatedwith scientific statistics had been established to instruct the performer choose appropriatepatients of laparoscopic cholecystectomy before operating, reduce the blindness, avoidvarious of the occurrence of complications and reduce the rates of conversion formlaparoscopic cholecystectomy to open surgery.Methods:1078hospitalized patients according with indication of laparoscopiccholecystectomy had been treated with laparoscopic cholecystectomy from May2010toMay2012in the department of general surgery of the Shanghai Six People’s AffiliatedHospital of Shanghai Jiao Tong University.1078consecutive patients undergoinglaparoscopic cholecystectomy were randomly divided into two groups:960patients intraining group,118patients in testing group. The first phase of study (Scoring system wasestablished.): With self-controlled trial, preoperative ultrasonic examination parametersincluding size of gallbladder、thickness of gallbladder wall、impacted stone in neck ofgallbladder、number and may diameter of stone、adhesions in umbilicus or in Calot’striangle or in funds of gallbladder were measured and collected; postoperative variablesincluding operating time、bleeding loss、conversion form laparoscopic cholecystectomy toopen surgery、placing drainage-tubes、postoperative complication、number of postoperativeday staying in hospital、pain of incision、gastrointestinal reaction and anal exhaust werealso collected. According to nine items of evaluative index registered postoperatively,standard for the difficult degree of laparoscopic cholecystectomy was determined.According to the difficult degree of960cases from May2010to January2012in thedepartment of general surgery of the Shanghai Six People’s Affiliated Hospital of ShanghaiJiao Tong University, training samples were divided into easy group and difficult group. T test and Chi-square test of univariate analysis were applied to choose statisticallysignificant preoperative ultrasonic variables. Then multivariate analysis with Logisticregression was performed for these risk factors. Ultimately the risk factors for difficultlaparoscopic cholecystectomy were determined. Each risk factor was given a scoreaccording to its coefficient of the mathematical function established with analysis oflogistic regression. A scoring model for predicting the difficult lever of laparoscopiccholecystectomy was then established. The comprehensive scores were calculated for allthe960cases. These cases were grouped from low to high score. The ratio of difficultlaparoscopic cholecystectomy and conversion from laparoscopic cholecystectomy to opensurgery in these groups of different score were compared between each adjoining group.(Receiver-operator Characteristic Curve, ROC) area under the curve had been calculatedafter ROC was drawn. ROC was then applied to evaluate the effectiveness of the scoringmodel. A score as the cut off value between difficult laparoscopic cholecystectomy andeasy laparoscopic cholecystectomy was chosen to obtain the optimum sensitivity andspecificity. According to the scoring system for evaluating classification standards ofdifficult laparoscopic cholecystectomy, technical grading standard for surgeon wasproposed. The second phase of study (Scoring system was proved.): The scoring systemwas applied to calculate comprehensive score of118testing samples from February2012to May2012in general surgery department of Shanghai Six People’s Affiliated Hospital ofShanghai Jiao Tong University. According to evaluating possibility of prospectivemisjudgment, the value of the scoring system was evaluated.Results:1. The811of960cases were diagnosed as gallbladder stone by preoperativeultrasonography, the801of960cases were confirmed by postoperative pathology, whichwas consistent with the clinical diagnosis. The rate of diagnostic accordance forgallbladder stone was98.76%. The149of960cases were diagnosed as gallbladder polypsby preoperative ultrasound. The149cases were totally confirmed by postoperativepathology, which was consistent with the clinical diagnosis. The rate of diagnosticaccordance for gallbladder polyps was100%.The globally diagnostic accuracy ofgallbladder benign disease by preoperative ultrasonography was99%.2. The variables of preoperative ultrasonic parameters with statistically significanteffect were size of gallbladder≥50cm2, thickness of gallbladder wall≥4mm, the maxdiameter of stone≥2cm, impacted stone in neck of gallbladder, and adhesions in Calot’s triangle or in funds of gallbladder. The six items of preoperative ultrasonic index were therisk factors for difficult laparoscopic cholecystectomy.3. According to result of logistic regression, preoperative risk factor (X) predictingfunction P of probability of difficult lever laparoscopic cholecystectomy was thenestablished.(Y): P=e~y／(1+e~y)Y=β0+β1X1+β2X2+βiXi=﹣6.966+2.624×size of gallbladder+3.177×thickness ofgallbladder wall+1.812×impacted stone in neck of gallbladder+1.855×max diameter ofstone+2.071×adhesions in Calot’s triangle+2.375×adhesions in funds of gallbladder.(risk factor=l, no risk factor=0)According to feature of logistic regression, each risk factor was given a score basingon the coefficient in front of each risk factor in the function of probability of difficultlaparoscopic cholecystectomy: size of gallbladder≥50cm2(yes26, not0), thickness ofgallbladder wall≥4mm (yes32, not0), max diameter of stone≥2cm (yes19, not0),impacted stone in neck of gallbladder (yes18, not0), adhesions in Calot’s triangle (yes21,not0), adhesions in funds of gallbladder (yes24, not0).4. The ROC was drawn after the comprehensive scores of960cases have beencalculated and collected. The ROC curve revealed that area under the curve (AUC) was0.939, the standard error of the area was0.001, and the scoring model forecast difficultlaparoscopic cholecystectomy statistically significant (P=0.000). The higher score theyhad, the more possibility of difficult laparoscopic cholecystectomy they had. The95％confidence interval of the area was0.937~0.942excluded0.5. The same conclusion can bededuced．When the score of64points was considered as diagnostic sector, the best pair ofsensitivity and specificity of87.7%and91.2%could be available.5. The960cases basing from low to high score were roughly divided into four groups.(More than115points does not exist.)The ideally easy group: from0to26points; Therelatively easy group: from27to52points; The relatively difficult group: from53to78points; The especially difficult group: more than79points. The higher score the grouphad，the higher ratio of difficult laparoscopic cholecystectomy and conversion formlaparoscopic cholecystectomy to open surgery the different group had. The differentnumber of difficult laparoscopic cholecystectomy between each adjoining group hadsignificantly statistical difference (P<0.05), which stated the distinction between four groups reasonably. According to the scoring system for evaluating classification standardsof difficult laparoscopic cholecystectomy, technical grading standard for operation wasproposed. ClassⅠ: from0to26points; Class Ⅱ: from27to52points; Class Ⅲ: from53to78points; Class Ⅳ: more than79points. Weconducted analysis and evaluation forsurgeon performing laparoscopic cholecystectomy, then put forward the technical gradingstandard for them: ClassⅠ: suitable for surgeon who was beginners with surgicalexperience of laparoscopic cholecystectomy less than50cases; Class Ⅱ: suitable forsurgeon who had surgical experience of laparoscopic cholecystectomy from50to300cases; Class Ⅲ: suitable for surgeon who had surgical experience of laparoscopiccholecystectomy from300to1000cases; Class Ⅳ: suitable for surgeon who had surgicalexperience of laparoscopic cholecystectomy more than1000cases.6. Prospective misjudgment evaluated probability of118testing samples. The difficultsituation of the predictive operation and the actual operation was compared through thescoring system. The5of118cases were misjudged, rate of misjudgment was about4.2%.Among118cases, the rate of accordance for ideally easy group was98%, the rate ofaccordance for relatively easy group was94%, the rate of accordance for relativelydifficult group was88%, the rate of accordance for especially difficult group was100%,the rate of overall accordance was95.8%, which reached the level of satisfaction.Conclusions:①Ultrasonography has high accuracy for diagnosis of gallbladderbenign diseases, which is the first choice of preoperative imaging examination method forlaparoscopic cholecystectomy.②A scoring system established basing preoperativeultrasonic imaging data for forecasting probabilities of difficult laparoscopiccholecystectomy is able to predict difficult degree of laparoscopic cholecystectomyobjectively and accurately, which has important clinical value for guiding performerchoose laparoscopic cholecystectomy cases, preventing or reducing the incidence ofcomplications and reducing the rate of conversion form laparoscopic cholecystectomy toopen surgery.