Comment on ‘Do TEP and TAPP have similar effect on postoperative pain? Author’s reply’

Abstract

We are glad to receive a reply on our letter regarding recently published trial sequential analysis pertaining to total extraperitoneal (TEP) versus laparoscopic transabdominal preperitoneal (TAPP) comparison in the setting of uncomplicated inguinal hernia in Hernia [1]. We, now, realize the reason of excluding randomized controlled trial by Bansal et al. [2] in their synthesis through the explanation in their reply [1]. In addition to the explanation regarding eligibility criteria, Aiolfi and Bonitta also mentioned some important issues, including (1) unfit statistical methodology (time-dependent problem), (2) heterogeneity, and (3) small effect size. Indeed, we agree with that these three issues have to be taken into consideration in a meta-analysis. Further discussions would shed light on the synthesized evidence regarding TEP and TAPP for inguinal hernia. Firstly, regarding unfit statistical methodology, we know that pain score after laparoscopic herniorrhaphy is timedependent data due to overtime repeated measurement, and to pool the data across time is inappropriate. However, what we did is like what Aiolfi et al. had performed in their metaanalysis. Data on postoperative pain in the meta-analysis by Aiolfi et al. were appropriately pooled based on measurement time points. Similarly, our supplementary analysis of postoperative pain also pooled data for each time point [3, 4]. We completely agree with that a problematic analysis with data from time-dependent patient reported outcomes would lead to misleading conclusion. Secondly, with regard to heterogeneity, it could be overstated in our previous letter since we know that the between-study variance was not reflected from a secondary combination. Consequently, our previous letter did not emphasize heterogeneity. In fact, because we have noticed high heterogeneity in the pooled mean difference of postoperative pain in our previous wok [5], we really understand the concerns regarding heterogeneity that Aiolfi and Bonitta mentioned in their reply. The highly heterogeneous pooled estimates would limit the confidence of findings. The final point is about small effect size. We cannot be certain if statistically significant reduced postoperative pain for the TEP approach quantified in a reduction of −0.36 and −0.31 on the VAS scale at 1 and 3 months, respectively, really make a clinical difference; wherefore our previous letter did not make conclusion for pain prevention. In the previous letter, we just raised a question ‘do TEP and TAPP have similar effect on postoperative pain?’ particularly when a relevant trial was not included, and called further meticulous studies to evaluate postoperative pain since it would be concerned by patients. Collectively, postoperative pain between TEP and TAPP herniorrhaphy may need further evidence with much rigorous design. Although these correspondences have no definite conclusion, the relevant discussions on methodology may inspire studies to analyze data meticulously.