Mantovani 2014.

Study characteristics
Methods	Study design: RCT, split‐mouth design Conducted in: Italy
Participants	Inclusion criteria: the third molar in question had to be Class A or B and in position 1, 2, or 3 according to the radiographic classification of Pell and Gregory, 18 based on the spatial relations of the tooth to the ascending ramus of the mandible and to the occlusal plane; the bilateral molars had to be in the same angulation (horizontal, mesioangular, or vertical); the indication for surgery was based on a diagnosis of pericoronitis Exclusion criteria: a clinically significant medical history (e.g. systemic infective disease, cardiovascular disease, liver disease, haematologic disease, bleeding tendency, diabetes, or neoplastic disease), recent anti‐inflammatory treatment, regular use of medications with possible anti‐inflammatory activity (e.g. antihistamines, non‐steroidal anti‐inflammatory drugs, corticosteroids, and antidepressants), women who were pregnant or breastfeeding, current heavy tobacco smokers (> 10 cigarettes daily), patients undergoing orthodontic therapy, and patients unwilling to undergo the data collection procedures Age: mean 24.02 years Number randomised: 125 Number evaluated: 100 (bilateral extraction)
Interventions	Piezoelectric (ultrasound) device versus traditional surgery using burs Group A (n = 125): included all operations carried out with the bur Group B (n = 125): surgeries carried out with the piezoelectric technique
Outcomes	The primary outcomes reported were postoperative pain, objective orofacial swelling, and surgical duration; secondary outcomes were gender, age, radiologic position, and possible adverse events (e.g. paraesthesia or infection). Participants were given a questionnaire about their subjective experience of the 2 different surgeries regarding the presence of vibrations and noise, which intervention was more comfortable, if further dental surgery was necessary, and which one they would prefer. Participants also were asked to describe their pain in detail; the symptoms score was obtained using a VAS. The VAS consisted of a 100‐millimetre horizontal line marked from 0 (no pain) to 100 (most severe pain ever experienced). Participants were asked to mark the scale, late in the evening, daily for 6 days after surgery. Clinical assessments were performed at 2, 7, 14, and 28 days after the surgery. Facial measurements were collected at baseline preoperatively and on day 7 after suture removal to evaluate any swelling.
Notes	Sample size calculation: reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote; "randomization was performed with a table of random numbers" Comment: random number table used
Allocation concealment (selection bias)	Low risk	Quote: "by a researcher not involved in the study and who was blinded to the type of procedure" Comment: concealment methods were put in place. Randomisation was done by an independent researcher who was blinded to the type of procedure.
Blinding (performance bias and detection bias) patient	Unclear risk	Comment: not mentioned, but it is possible the participant was aware of the technique used on each side based on noise levels
Blinding (performance bias and detection bias) assessor	Low risk	Quote: "Postsurgical clinical assessments were performed by a single blinded examiner" Comment: the assessor was blinded to the allocation of sides
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "The study sample was derived from 140 patients. Fifteen of these patients did not show periodontal parameters lower than 20% and thus were excluded; 14 patients underwent only 1 intervention and 11 patients did not attend all follow‐up visits. Thus, the final study sample consisted of 100 patients." Comment: the dropouts from the originally recruited 140 participants have been described and justified
Selective reporting (reporting bias)	Low risk	Planned outcomes were reported.
Other bias	Low risk	No other sources of bias were identified.