This prospective, randomised controlled study included patients with vertical or mesioangular partially impacted 3Ms between 01/2017-09/2017 and the last data collected to 12/2017. Patients were excluded if they have medical illness or medication that could influence the course of postoperative wound healing. This clinical study was registered by Australian New Zealand Clinical Trials Registry with the trial number of ACTRN12618001551280. Patients between the ages of 18 and 30 were included. Patients were excluded from randomisation if they had acute pericoronitis, a pre-existing abscess or cellulitis, any restorations or carious lesions on the distal surfaces of the 2Ms, or pathological conditions associated with the 3Ms.Oral contraceptive users and smokers were also excluded.
Patients were randomised to three groups by the envelope method: the secondary closure (SC) group, includes partial closure of the extraction site to provide secondary healing; the primary closure (PC) group, includes sliding the flap and suturing primarily to total closure of the extraction site; and the membrane-based primary closure (MBPC) group, includes placing a collagen membrane and sliding the flap to total closure of the extraction site.
Surgical operations were carried out by a single surgeon with each patient under local anesthesia, which was achieved with up to4 mL of Articaine-HCl and a 1:100,000 ratio of epinephrine (Ultracaine D-S Forte, Aventis). An incision was made from the anterior border of the mandibular ramus, extending to the distal surface and the buccal gingivodental sulcus of the 2M. The incision was continuous, with vertical incision, oblique into the mandibular vestibular fornix, aligned with the mesiobuccal cusp of the 2M. A full-thickness mucoperiosteal triangular flap was elevated. If necessary, osteotomy was performed under constant irrigation.
The wound closure was performed with atraumatic silk sutures depending on the groups.In the SC group, the flap was positioned to its former position by a single suture distal to the 2M, with leaving a gap. If necessary, a wedge of mucosa was removed. In the PC group, the sliding flap was sutured. In the MBPC group, the extraction socket was closed with a resorbable collagen membrane (Evolution®, Osteobiol-Tecnoss, Italy), and the sliding flap was sutured primarily.
The patients were given standard postoperative medication; antibiotics (amoxicillin + clavulanic acid), nonsteroidal anti-inflammatory drugs (dexketoprofen trometamol) and mouthwash (with 0.12% chlorhexidine). The sutures were removed after 7 days.
Parameters determined in this study included gender, duration of surgery, surgical difficulty, and the position of the 3Ms. Surgical difficulty was recorded immediately after the procedure and evaluated using a four-class difficulty scale, as follows: 1,extraction requiring forceps only; 2, extraction requiring osteotomy; 3, extraction requiring osteotomy and coronal sectioning; and 4, complex extraction(root sectioning).The level of impaction of the 3Mswas classified as being either on the level of the occlusal plane or between the occlusal plane and the cervical line of the 2M (below the occlusal plane), according to Pell and Gregory’s classification. 3M angulation was classified on a two-class scale according to Winter’s classification, being either 1, mesioangularor2, vertical. Digital panoramic radiographswas obtained to ensure the similarity of the impaction types according to angulation and relationship to the occlusal plane.
Another primary outcome variable was the periodontal status of the 2Ms. Periodontal measurements were performed preoperatively (T1) and three months postoperatively (T2). The pocket depths and plaque index scores of the distobuccal (DB), distolingual (DL), midbuccal (MidB) and midlingual (MidL) surfaces of the 2Ms were recorded. All periodontal pocket depth (PPD)measurements were performed as millimeterusing William’s periodontal probe (Aesculap AG&Co., Tuttlingen, Germany).To assess the periodontal plaque index (PPI) scores, the Silness&Löe plaque index was registered on the distal aspects of the 2Ms,and the mean values were used.
Digital panoramic radiographs were obtained preoperatively andthree months postoperatively. Cemento-enamel junction (CEJs) was used as reference points, and the distances between the CEJ and the alveolar bone crest on the distal aspect of the adjacent 2Mweremeasured as millimetre using calibrated radiograph measurement software (MedData Medical Software, Ankara, Turkey). All radiological measurements were performed twice and independently, and mean values were recorded preoperatively (T1) and three months postoperatively (T2). The differences (positive or negative) between the pre-and postoperative radiographic measurements constituted one of the primary outcomes of this study.
The study data were subjected to power analysis to a posteriori check the acceptability of the sample size. The sample size was 30 participants per group (total=90) were analysed with an estimated error (α) of 5 and 80% power, and the sample size was found to be appropriate(0,69). Statistical analysis was performed using the SPSS statistical software package, version 20.0 (IBM, Chicago, IL, USA). The normalities of the distributions were tested using the Kolmogorov-Smirnov test. The Wilcoxon signed rank test was used for intragroup comparisons. For intergroup comparisons, the Kruskal-Wallis orchi-square tests was used, and for post hoc tests, the Bonferroni correction was used.