Abstract Background The objective of this study was to investigate the potential benefit of local haemostatic agents for the prevention of postoperative bleeding after thyroidectomy. Methods A systematic literature search was performed, and RCTs involving adult patients who underwent thyroid surgery using either active (AHA) or passive (PHA) haemostatic agents were included in the review. The main outcome was the rate of cervical haematoma that required reoperation. A Bayesian random-effects model was used for network meta-analysis with minimally informative prior distributions. Results Thirteen RCTs were included. The rate of cervical haematoma requiring reoperation ranged from 0 to 9·1 per cent, and was not reduced by haemostatic agents: AHA versus control (odds ratio (OR) 1·53, 95 per cent credibility interval 0·21 to 10·77); PHA versus control (OR 2·74, 0·41 to 16·62) and AHA versus PHA (OR 1·77, 0·12 to 25·06). No difference was observed in the time required for drain removal, duration of hospital stay, and the rate of postoperative hypocalcaemia or recurrent nerve palsy. AHA led to a significantly lower total postoperative blood loss and reduced operating time in comparison with both the control and PHA groups. Conclusion The general use of local haemostatic agents has not been shown to reduce the rate of clinically relevant bleeding.