Various flap designs that consider postoperative complications or periodontitis of the mandibular second molars have been reported. Among them, TF and EF are the most frequently used and have been examined[10,14−18]. The discussion proceeded with comparison with these reports and the effectiveness of mEF was clarified.
To the best of our knowledge, the strongest evidence of TF vs. EF has been reported in three systematic reviews and meta-analyses[15,17,18]. Among them, Zhu et al. reported that EF is more effective than TF in reducing postoperative pain and swelling in the subgroup with low difficulty. This is because tissue trauma, including vertical incision in the buccal gum, may lead to the release of local inflammatory mediators in TF and sensitize peripheral pain receptors in low difficulty, resulting in more postoperative pain. However, as the invasive procedure of tooth extraction is greater in patients with high difficulty, the pain in the tissue trauma caused by vertical incision may not affect the postoperative pain . Our results suggest that mEF is more effective in reducing postoperative pain in low difficulty cases. The significant difference was observed in all cases, because it may be a predilection in the number of low and high difficulty cases (2:1). Two other studies have reported that there was no significant difference between the two flaps[17,18]. As a problem of previous RCTs, most reports have shown that pain score was statistically evaluated using the mean VAS variables. The VAS is a universal instrument used to measure pain. However, as the VAS is an ordinal scale, not a ratio scale, a statistical method that compares the mean VAS variables is incorrect. For an ordinal scale, variables should be categorized into two groups and evaluated using Fischer’s exact test or chi-square test. Therefore, most previous results related to postoperative pain should be interpreted with caution. To the best of our knowledge of the two studies with correct statistical analysis, Sandhu et al. reported that pain was significantly greater in EF compared with TF on postoperative day 1 and day 7, while Xie et al. reported that there was no significant difference between mEF and TF. However, both studies were single-center and had a small number of cases. Moreover, as background factors related to patients, such as smoking and drinking, or difficulty of tooth extraction were unclear, there may be some confounding factors in these studies. Our results suggest that postoperative pain was significantly lower in mEF than in TF, which is different from the results of the previous two studies. The management of our multicenter RCT has been made uniform to control for confounding factors, and the results of our study must be strong evidence. Taken together, tissue trauma caused by vertical incision led to enhanced postoperative pain and mEF is more effective in reducing postoperative pain in low difficulty cases.
On the other hand, Zhu et al. reported that TF was more effective than EF in lowering the incidence of dry socket. Two other studies have reported that there was no significant difference between the two flaps[17,18]. However, no previous reports have histologically discussed the reasons for the lower incidence of dry socket in TF than in EF. The risk factors of dry socket have been associated with tobacco use, the amount of anesthesia, menstrual cycles, older patient age, surgical difficulty, and some drugs. For this reason, Zhu et al. stated that their results should be interpreted carefully and analyzed accurately in a split-mouth controlled study. Elo et al.  reported that mEF was more effective than TF and EF in lowering the incidence of dry socket in split-mouth controlled study This study had a large number of cases and had one of the strongest evidence in the previous reports. Our study suggests that there was no significant difference between the two groups. However, our study was a parallel randomized controlled trial with a small number of dry socket. Therefore, our findings should be interpreted with caution.
There were no significant differences in postoperative hemorrhage, nerve paralysis, and infection between the two groups. Flap design was not a critical factor for the improvement of complications. Mobilio et al. reported that operation time was the most important factor for early postoperative outcomes in the third molar extraction. Although there was no significant difference in operation time, the use of mEF can save approximately one minute compared to TF. To enclose the mesial vertical incision, one stitch is needed, leading to a longer operation time. Although mEF has the disadvantage of a small surgical field, there was no difference in operation time. Trismus and swelling are among the most important postoperative complications of the third molar extraction. Xie et al. reported that mEF was more effective than TF in reducing postoperative trismus on postoperative day 3 and 7. In this study, postoperative pain, swelling and trismus were compared on days 3 and 7 after surgery. In all participating hospitals of our study, the postoperative examination was performed only on day 7 ± 1 after surgery. As detailed examinations on postoperative days 1, 3, and 7 could not be performed, evaluations of trismus and swelling were excluded.
This study has some potential limitations. The sample size was small, and the observation interval was short (7 ± 1days). Early postoperative pain was significantly lower in the mEF group, but the long-term outcome (two or three weeks later) was unclear. Moreover, it was unclear whether mEF is better than TF in terms of periodontitis of the mandibular second molar. Therefore, multicenter RCTs with more participants and long-term follow-up are required in the future.
In conclusion, the study shows that mEF resulted in less postoperative pain in low (ⅠA, ⅡA, ⅢA, ⅠB) difficulty classified by G.B. Winter classification. This is because tissue trauma caused by vertical incision led to enhanced early postoperative pain in low difficulty cases. Although mEF has the disadvantage of a small surgical field, there is no bad influence in operation time. However, the evidence at present is not sufficient to suggest the use of either flap design.