Background Large head and neck reconstructions have often been accompanied with elective tracheostomies to bypass the pharyngolaryngeal oedema and ensure unobstructed ventilation perioperatively. Recently, high volume centres have trialled delayed extubation (DE) which theoretically aims to provide a faster recovery to oral intake and speech. There remains a paucity of synthesised evidence comparing tracheostomies and delayed extubation in head and neck reconstruction surgeries. Methods This study represents the largest systematic review till date on this topic from databases including PubMed, Embase and Scopus which found 4423 studies and included 9 after screening and appraisal. A meta-analysis was conducted using Mantel-Haenszel statistical method with risk ratios and 95% confidence intervals calculated. Results Data from 1450 patients was analysed. For tracheostomy patients, weighted mean hospital stay and intensive care stay was 13.3 and 1.83 days respectively. For DE, the results were 12.25 days and 1.47 days respectively. Meta-analysis for airway complications was statistically significant with a risk ratio of 3.07 (95% CI, 1.91-4.91) which favoured the DE group. Risk ratio for flap complications was also significant at 8.19 (95% CI, 3.41-19.65) favouring the DE group. Studies reporting overall survival and disease-free survival at 88.2% and 68.3% for tracheostomy group respectively, compared to 95.7% and 89.1% in the DE cohort. Conclusion DE may significantly decrease complications in the perioperative setting but this has not been controlled for a variety of confounding factors including stages of cancer and types of resections. DE also does not decrease hospital lengths of stay. Regarding functional outcomes, collection of speech questionnaires and oral intake data should be mandated to understand the true difference.