Comment on "Absorbable versus non-absorbable suture in percutaneous laparoscopic inguinal hernia repair in children:...

Shi et al. report a pooled odds ratio of 0.03, 95% confidence interval 0.01 to 0.15, for suture knot reaction, which they read as a significantly lower risk in the absorbable suture group in percutaneous laparoscopic inguinal hernia repair in children. Recomputing that estimate from the counts printed in the review's own forest plots reproduces both the published odds ratio and the 67.2% weight carried by the study that contributes most of it. It also shows that this study enters Figs. 3 and 4 with its two groups exchanged. The source article ties each denominator to a specific group. One of the two assignments makes the absorbable column of Fig. 3 exceed the absorbable total the review states for itself, and the other does not. Applying the assignment used in Fig. 4 returns 2.06, 95% confidence interval 1.15 to 3.67; setting the study aside returns 0.06, 95% confidence interval 0.007 to 0.50. This Comment asks the authors to confirm the group assignment and, if it is transposed, to publish a corrected estimate.