Sir
An important aspect to the ethics approval for this trial was that the authors monitored all significant clinical incidents. Such incidents included postoperative deaths and clinically significant anastomotic leaks. It was as a result of this monitoring system that the number of anastomotic leaks requiring reoperation was reviewed and an independent decision was made to halt the trial. Monitoring adverse events is an essential element to any clinical study design.
Laparoscopic procedures were excluded from this trial because the patients were involved in a separate clinical trial. The ethics committee within our institution rightly considers that patients can only be involved in one clinical trial at a time. The authors' preference in patients undergoing laparoscopic bowel resection is to prepare the bowel with PEG since this facilitates bowel handling intra-operatively and assists in disease location.
It is important to highlight that this study did not demonstrate that bowel preparation prevents anastomotic leaks. As commented, there was no significant difference in the leak rate between the two groups. It did, however, demonstrate that the consequences of anastomotic leaks were more severe in patients who did not have full bowel preparation. The Chi-square test was used to demonstrate a significant difference in anastomotic leaks requiring re-operation between the two groups. With Fischer's exact test, the results still demonstrated a significant difference between the two groups.
We would also like to highlight the differences in the type of procedures performed in this study compared with previous clinical trials. Nearly 30 per cent of patients had a high-risk anastomosis within 6 cm of the anal verge. Therefore the results are not really comparable to previous studies, a point that should be considered in any further meta-analysis.