Por Mihai Paduraru
This meta-analysis is focused on assessing the safety and efficacy of laparoscopic surgery within a fast track program for colorectal resection patients by comparing this approach to open surgery. In order to evaluate the value of this study we need to consider the validity of the methodology; the clinical importance of the results; and applicability of the findings.
The rigour of the methodology in this study is variable. The search strategy is lacking in certain details. One example is the search was initially over a 25 year time period but the RCTs included were in the last 10 years. A rationale for the study was the fact that similar ones had only used a small number of trials on which to base their findings. The value of this study was to be an analysis of more trials than previously evaluated, therefore it would have strengthened this meta-analysis to have identified as many trials as possible. It is not clear if this was attempted or not. The study was also limiting in its inclusion criteria by selecting trials reporting on ‘any’ one of the primary or secondary outcome criteria. This limited the scope of the results reported on as not all seven trials reviewed gave outcome data for all six outcomes reported on.
The method of study selection and data extraction was more rigorous and the statistical analysis thorough. The authors used the Cochrane tool to assess risk of bias in the RCTs but state that this quality control was based on the authors’ judgment, therefore admitting that the assessment was subjective. Little discussion is devoted to the risk of bias in the trials even though there is a high degree reported with regard to 4/7 elements. Sample size of the trials is also questioned but not validated.
Although the statistical analysis of the data was well conducted, with results being supported by Confidence Intervals and Relative Risk, there is not enough detail about the individual studies to support these results. The authors state that there is no significant heterogeneity between trials yet there are factors which should be reported on, for example age. There is a twenty year mean age difference between some trials. Similarly, rectal surgery treatment is known to be such that length of stay is longer for these patients than for colon surgical patients. These aspects could be confounding factors. They also report in Table 1 ‘cancer type’ but not all trial patients are operated on for cancer only.
One big limitation of the study is in the lack of analysis of fast track in the context it is applied by the RCTs. No definition is given; seven elements are selected as ones implemented by the trials, but obviously not exclusively nor necessarily with the same degree of compliance. Bowel preparation for example is undertaken by a number of trials, some of which do not include rectal surgical patients; in the others, it is not clear if this is for all patients or only rectal surgical, as recommended by ERAS. Any further elements included in the fast track program of each trial are merely reported on as a total figure and catagorised as ‘other’. Indeed the authors do not categorically state that ERAS is the official fast track program used. All these factors have important implications for verification of the results and therefore their clinical value. The findings mirror those of a similar analysis, but with fewer RCTs, thus adding some weight to the overall body of evidence. The scarcity of detail however in some qualitative aspects of the study questions the accuracy of the results.
Finally, it is worth noting that in terms of applicability, this is an analysis of a small sample of the population and is not specific to one geographical area or demography, and, although results are apparently similar despite this, they are not necessarily representative enough to inform local practice, especially since the authors state that their aim is to look at the efficacy and not the efficiency of laparoscopic surgery and fast track for colorectal surgical patients.