Robot assistant versus human or another robot assistant in patients undergoing laparoscopic cholecystectomy.

Abstract

The role of a robotic assistant in laparoscopic cholecystectomy is controversial. While some trials have shown distinct advantages of a robotic assistant over a human assistant others have not, and it is unclear which robotic assistant is best. The aims of this review are to assess the benefits and harms of a robot assistant versus human assistant or versus another robot assistant in laparoscopic cholecystectomy, and to assess whether the robot can substitute the human assistant. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, and Science Citation Index Expanded (until February 2012) for identifying the randomised clinical trials. Only randomised clinical trials (irrespective of language, blinding, or publication status) comparing robot assistants versus human assistants in laparoscopic cholecystectomy were considered for the review. Randomised clinical trials comparing different types of robot assistants were also considered for the review. Two authors independently identified the trials for inclusion and independently extracted the data. We calculated the risk ratio (RR) or mean difference (MD) with 95% confidence interval (CI) using the fixed-effect and the random-effects models based on intention-to-treat analysis, when possible, using Review Manager 5. We included six trials with 560 patients. One trial involving 129 patients did not state the number of patients randomised to the two groups. In the remaining five trials 431 patients were randomised, 212 to the robot assistant group and 219 to the human assistant group. All the trials were at high risk of bias. Mortality and morbidity were reported in only one trial with 40 patients. There was no mortality or morbidity in either group. Mortality and morbidity were not reported in the remaining trials. Quality of life or the proportion of patients who were discharged as day-patient laparoscopic cholecystectomy patients were not reported in any trial. There was no significant difference in the proportion of patients who required conversion to open cholecystectomy (2 trials; 4/63 (weighted proportion 6.4%) in the robot assistant group versus 5/70 (7.1%) in the human assistant group; RR 0.90; 95% CI 0.25 to 3.20). There was no significant difference in the operating time between the two groups (4 trials; 324 patients; MD 5.00 minutes; 95% CI -0.55 to 10.54). In one trial, about one sixth of the laparoscopic cholecystectomies in which a robot assistant was used required temporary use of a human assistant. In another trial, there was no requirement for human assistants. One trial did not report this information. It appears that there was little or no requirement for human assistants in the other three trials. There were no randomised trials comparing one type of robot versus another type of robot. Robot assisted laparoscopic cholecystectomy does not seem to offer any significant advantages over human assisted laparoscopic cholecystectomy. However, all trials had a high risk of systematic errors or bias (that is, risk of overestimation of benefit and underestimation of harm). All trials were small, with few or no outcomes. Hence, the risk of random errors (that is, play of chance) is high. Further randomised trials with low risk of bias or random errors are needed.