Editor

We noted several issues with the article by Christoffersen et al1. The primary endpoint considered is pain on 1st postoperative day, which is not clinically relevant. Neither the patients nor the clinicians are particularly bothered about this outcome when choosing a procedure. The authors should have considered more relevant outcomes like chronic pain and recurrence rate as the primary outcomes rather than the secondary outcomes. In addition, one of the exclusion criteria was a “small umbilical defect (0·5–2·0 cm)”. However, Table 1 shows a defect size 2·5 (1·5–4·0) which means there were some cases of defect size 1·5 cm in the study. Did this violate the protocol and does it affect the results?

It is mentioned that “Intention to treat principle” (ITT) was used for analysis. Consider data in Table 2. Day-30: Seroma 14 and 22. For ITT analysis, we have to use the proportions 14/40 & 22/40 for a Chi squared test (P = 0·072) but the article mentions P = 0·043. Similarly, Table 3 shows the cumulative recurrence rate of 5 and 12. We found 5/40 and 12/40 to have P = 0·055 and not P = 0·047 as mentioned in the article.

Recurrence rates (14% and 32%) are high against literature reported rates of 3-5%. This raises the question, “can we adopt the procedure”? The authors used Physiomesh™ (Ethicon) and the article mentions that the mesh is withdrawn from the market. So the answer is we cannot conclude recurrence and seroma rates are higher in the non-defect closure group. There are many ways to close abdominal wounds and defects as published recently, including guidelines2-6.