Background: Surgery is an important part of the management of oral cavity cancer with regard to both the removal of the primary tumour and removal of lymph nodes in the neck. Surgery is less frequently used in oropharyngeal cancer. Surgery alone may be treatment for early-stage disease or surgery may be used in combination with radiotherapy, chemotherapy and immunotherapy/biotherapy. There is variation in the recommended timing and extent of surgery in the overall treatment regimens of people with these cancers. This is an update of a review originally published in 2007 and first updated in 2011. Objectives: To determine which surgical treatment modalities for oral and oropharyngeal cancers result in increased overall survival, disease-free survival and locoregional control and reduced recurrence. To determine the implication of treatment modalities in terms of morbidity, quality of life, costs, hospital days of treatment, complications and harms. Search methods: Cochrane Oral Health's Information Specialist searched the following databases: Cochrane Oral Health's Trials Register (to 20 December 2017), the Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 11), MEDLINE Ovid (1946 to 20 December 2017) and Embase Ovid (1980 to 20 December 2017). We searched the US National Institutes of Health Trials Registry (ClinicalTrials.gov) and the World Health Organization International Clinical Trials Registry Platform for ongoing trials. There were no restrictions on the language or date of publication. Selection criteria: Randomised controlled trials where more than 50% of participants had primary tumours of the oral cavity or oropharynx, or where separate data could be extracted for these participants, and that compared two or more surgical treatment modalities, or surgery versus other treatment modalities. Data collection and analysis: Two or more review authors independently extracted data and assessed risk of bias. We contacted study authors for additional information as required. We collected adverse events data from included studies. Main results: We identified five new trials in this update, bringing the total number of included trials to 12 (2300 participants; 2148 with cancers of the oral cavity). We assessed four trials at high risk of bias, and eight at unclear. None of the included trials compared different surgical approaches for the excision of the primary tumour. We grouped the trials into seven main comparisons. Future research may change the findings as there is only very low-certainty evidence available for all results. Five trials compared elective neck dissection (ND) with therapeutic (delayed) ND in participants with oral cavity cancer and clinically negative neck nodes, but differences in type of surgery and duration of follow-up made meta-analysis inappropriate in most cases. Four of these trials reported overall and disease-free survival. The meta-analyses of two trials found no evidence of either intervention leading to greater overall survival (hazard ratio (HR) 0.84, 95% confidence interval (CI) 0.41 to 1.72; 571 participants), or disease-free survival (HR 0.73, 95% CI 0.25 to 2.11; 571 participants), but one trial found a benefit for elective supraomohyoid ND compared to therapeutic ND in overall survival (RR 0.40, 95% CI 0.19 to 0.84; 67 participants) and disease-free survival (HR 0.32, 95% CI 0.12 to 0.84; 67 participants). Four individual trials assessed locoregional recurrence, but could not be meta-analysed; one trial favoured elective ND over therapeutic delayed ND, while the others were inconclusive. Two trials compared elective radical ND with elective selective ND, but we were unable to pool the data for two outcomes. Neither study found evidence of a difference in overall survival or disease-free survival. A single trial found no evidence of a difference in recurrence. One trial compared surgery plus radiotherapy with radiotherapy alone, but data were unreliable because the trial stopped early and there were multiple protocol violations. One trial comparing positron-emission tomography-computed tomography (PET-CT) following chemoradiotherapy (with ND only if no or incomplete response) versus planned ND (either before or after chemoradiotherapy), showed no evidence of a difference in mortality (HR 0.92, 95% CI 0.65 to 1.31; 564 participants). The trial did not provide usable data for the other outcomes. Three single trials compared: surgery plus adjunctive radiotherapy versus chemoradiotherapy; supraomohyoid ND versus modified radical ND; and super selective ND versus selective ND. There were no useable data from these trials. The reporting of adverse events was poor. Four trials measured adverse events. Only one of the trials reported quality of life as an outcome. Authors' conclusions: Twelve randomised controlled trials evaluated ND surgery in people with oral cavity cancers; however, the evidence available for all comparisons and outcomes is very low certainty, therefore we cannot rely on the findings. The evidence is insufficient to draw conclusions about elective ND of clinically negative neck nodes at the time of removal of the primary tumour compared to therapeutic (delayed) ND. Two trials combined in meta-analysis suggested there is no difference between these interventions, while one trial (which evaluated elective supraomohyoid ND) found that it may be associated with increased overall and disease-free survival. One trial found elective ND reduced locoregional recurrence, while three were inconclusive. There is no evidence that radical ND increases overall or disease-free survival compared to more conservative ND surgery, or that there is a difference in mortality between PET-CT surveillance following chemoradiotherapy versus planned ND (before or after chemoradiotherapy). Reporting of adverse events in all trials was poor and it was not possible to compare the quality of life of people undergoing different surgical treatments.