Background: Thyroid storm is a feared complication in patients with hyperthyroidism undergoing surgery. We assessed the risk of thyroid storm for different preoperative treatment options for patients with primary hyperthyroidism undergoing surgery. Methods: Pubmed, EMBASE, and The Cochrane Library were searched systematically for all studies reporting on adult hyperthyroid patients undergoing elective surgery under general anaesthesia. Selected studies were categorised based on preoperative treatment: no treatment, antithyroid medication (thionamides), iodine, b-blocking medication, or a combination thereof. Treatment effect, that is restoring euthyroidism, was extracted from the publications if available. Risk of bias was assessed using the Risk of Bias in Non-randomised Studies of Interventions (ROBINS-I) or the Cochrane Risk of Bias tool for randomised studies. Results: The search yielded 7009 articles, of which 26 studies published between 1975 and 2020 were selected for critical appraisal. All studies had moderate to critical risk of bias, mainly attributable to risk of confounding, classification of intervention status, and definition of the outcome. All studies reported on thyroidectomy patients. We found no randomised studies comparing the risk of thyroid storm between treated and untreated patients. Cases of thyroid storm were reported in all treatment groups with incidences described ranging from 0% to 14%. Conclusion: Evidence assessing the risk of perioperative thyroid storm is of insufficient quality. Given the seriousness of this complication and the impossibility of identifying patients at increased risk, preoperative treatment of these patients remains warranted.