A best evidence topic in thoracic surgery was written according to a structured protocol. The question addressed was ‘in patients undergoing thyroidectomy for retrosternal goitre, what factors predict sternotomy?’ Altogether 165 papers were found as a result of the reported search, of which only 3 prospective studies, 1 review paper and 2 retrospective studies represented the best evidence to answer the clinical question. The authors, journals, date and country of publication, patient group studied, study type, relevant outcomes and results of the papers are tabulated. We conclude that a combination of preoperative clinical and radiological risk factors, alongside informed patient choice can be used to predict the need for sternotomy in thyroidectomy for goitre with retrosternal extension. Clinically, a history of goitre with retrosternal extension beyond 160 months is a risk factor for sternotomy. Thyroid tissue density, posterior mediastinal location and subcarinal extension, as measured using computed tomography (CT) imaging, are independent preoperatively obtained risk factors for sternotomy, which are supported by both prospective and retrospective studies. Thyroid tissue density is the strongest factor and increases the risk of sternotomy 47-fold. Minimal upper sternotomy (sternal-split) can be used instead of median sternotomy where there is evidence of retrosternal extension to the aortic root. CT evidence of an ectopic nodule, a dumbbell-shaped goitre, a conical-shaped goitre constricted by an isthmic thoracic inlet or a thoracic goitre component wider than the thoracic inlet can also predict the need to undergo sternotomy. Finally, informed consent should include a discussion that patients with bilateral multinodular goitre and evidence of intrathoracic extension, who are undergoing total thyroidectomy via cervicotomy, have an independently increased risk of complications, specifically recurrent laryngeal nerve injury. After explanation of these risks, a patient may be unwilling to accept the increased risks of cervicotomy per se versus those of combined cervicotomy and sternotomy.