Background: Pathology relating to mandibular wisdom teeth is a frequent presentation to oral and maxillofacial surgeons, and surgical removal of mandibular wisdom teeth is a common operation. The indications for surgical removal of these teeth are alleviation of local pain, swelling and trismus, and also the prevention of spread of infection that may occasionally threaten life. Surgery is commonly associated with short-term postoperative pain, swelling and trismus. Less frequently, infection, dry socket (alveolar osteitis) and trigeminal nerve injuries may occur. This review focuses on the optimal methods in order to improve patient experience and minimise postoperative morbidity. Objectives: To compare the relative benefits and risks of different techniques for surgical removal of mandibular wisdom teeth. Search methods: Cochrane Oral Health’s Information Specialist searched the following databases: Cochrane Oral Health Trials Register (to 8 July 2019), the Cochrane Central Register of Controlled Trials (CENTRAL) (the Cochrane Library; 2019, Issue 6), MEDLINE Ovid (1946 to 8 July 2019), and Embase Ovid (1980 to 8 July 2019). We searched ClinicalTrials.gov and the World Health Organization International Clinical Trials Registry Platform for ongoing trials. We placed no restrictions on the language or date of publication. Selection criteria: Randomised controlled trials comparing different surgical techniques for the removal of mandibular wisdom teeth. Data collection and analysis: Three review authors were involved in assessing the relevance of identified studies, evaluated the risk of bias in included studies and extracted data. We used risk ratios (RRs) for dichotomous data in parallel-group trials (or Peto odds ratios if the event rate was low), odds ratios (ORs) for dichotomous data in cross-over or split-mouth studies, and mean differences (MDs) for continuous data. We took into account the pairing of the split-mouth studies in our analyses, and combined parallel-group and split-mouth studies using the generic inverse-variance method. We used the fixed-effect model for three studies or fewer, and random-effects model for more than three studies. Main results: We included 62 trials with 4643 participants. Several of the trials excluded individuals who were not in excellent health. We assessed 33 of the studies (53%) as being at high risk of bias and 29 as unclear. We report results for our primary outcomes below. Comparisons of different suturing techniques and of drain versus no drain did not report any of our primary outcomes. No studies provided useable data for any of our primary outcomes in relation to coronectomy. There is insufficient evidence to determine whether envelope or triangular flap designs led to more alveolar osteitis (OR 0.33, 95% confidence interval (CI) 0.09 to 1.23; 5 studies; low-certainty evidence), wound infection (OR 0.29, 95% CI 0.04 to 2.06; 2 studies; low-certainty evidence), or permanent altered tongue sensation (Peto OR 4.48, 95% CI 0.07 to 286.49; 1 study; very low-certainty evidence). In terms of other adverse effects, two studies reported wound dehiscence at up to 30 days after surgery, but found no difference in risk between interventions. There is insufficient evidence to determine whether the use of a lingual retractor affected the risk of permanent altered sensation compared to not using one (Peto OR 0.14, 95% CI 0.00 to 6.82; 1 study; very low-certainty evidence). None of our other primary outcomes were reported by studies included in this comparison. There is insufficient evidence to determine whether lingual split with chisel is better than a surgical hand-piece for bone removal in terms of wound infection (OR 1.00, 95% CI 0.31 to 3.21; 1 study; very low-certainty evidence). Alveolar osteitis, permanent altered sensation, and other adverse effects were not reported. There is insufficient evidence to determine whether there is any difference in alveolar osteitis according to irrigation method (mechanical versus manual: RR 0.33, 95% CI 0.01 to 8.09; 1 study) or irrigation volume (high versus low; RR 0.52, 95% CI 0.27 to 1.02; 1 study), or whether there is any difference in postoperative infection according to irrigation method (mechanical versus manual: RR 0.50, 95% CI 0.05 to 5.43; 1 study) or irrigation volume (low versus high; RR 0.17, 95% CI 0.02 to 1.37; 1 study) (all very low-certainty evidence). These studies did not report permanent altered sensation and adverse effects. There is insufficient evidence to determine whether primary or secondary wound closure led to more alveolar osteitis (RR 0.99, 95% CI 0.41 to 2.40; 3 studies; low-certainty evidence), wound infection (RR 4.77, 95% CI 0.24 to 96.34; 1 study; very low-certainty evidence), or adverse effects (bleeding) (RR 0.41, 95% CI 0.11 to 1.47; 1 study; very low-certainty evidence). These studies did not report permanent sensation changes. Placing platelet rich plasma (PRP) or platelet rich fibrin (PRF) in sockets may reduce the incidence of alveolar osteitis (OR 0.39, 95% CI 0.22 to 0.67; 2 studies), but the evidence is of low certainty. Our other primary outcomes were not reported. Authors' conclusions: In this 2020 update, we added 27 new studies to the original 35 in the 2014 review. Unfortunately, even with the addition of these studies, we have been unable to draw many meaningful conclusions. The small number of trials evaluating each comparison and reporting our primary outcomes, along with methodological biases in the included trials, means that the body of evidence for each of the nine comparisons evaluated is of low or very low certainty. Participant populations in the trials may not be representative of the general population, or even the population undergoing third molar surgery. Many trials excluded individuals who were not in good health, and several excluded those with active infection or who had deep impactions of their third molars. Consequently, we are unable to make firm recommendations to surgeons to inform their techniques for removal of mandibular third molars. The evidence is uncertain, though we note that there is some limited evidence that placing PRP or PRF in sockets may reduce the incidence of dry socket. The evidence provided in this review may be used as a guide for surgeons when selecting and refining their surgical techniques. Ongoing studies may allow us to provide more definitive conclusions in the future.