Surgical vs. non-surgical management of displaced type-2 odontoid fractures in patients aged 75 years and older: study protocol for a randomised controlled trial

Background: Displaced odontoid fractures in the elderly are treated non-surgically with a cervical collar or surgically with C1-C2 fusion. Due to the paucity of evidence, the treatment decision is often left to the discretion of the expert surgeon. Methods: The Uppsala Study on Odontoid Fracture Treatment (USOFT) is a multicentre, open-label, randomised controlled superiority trial evaluating the clinical superiority of the surgical treatment of type-2 odontoid fractures, with a 1-year Neck Disability Index (NDI) as the primary endpoint. Fifty consecutive patients aged >= 75 years, with displaced type-2 odontoid fracture, are randomised to non-surgical or surgical treatment. Excluded are patients with an American Society of Anaesthesiologists (ASA) score >= 4, dementia nursing care or anatomical cervical anomalies. The minimal clinically important difference of the NDI is 3.5 points. A minimum of 16 patients are needed in each group to test the superiority with 80% power. By considering a 1-year mortality forecast of 29%, up to 25 participants are recruited in each group. The non-surgical group is fitted with a rigid cervical collar for 12 weeks. The surgical group is treated with a posterior C1-C2 fusion. All participants are monitored with regard to the NDI, EuroQol score (EQ-5D), socio-demographics and computed tomography (CT) at the time of injury, at 6 weeks, 3 months and 12 months. At 12 months, a dynamic radiographical investigation of upper cervical stability is performed. The secondary endpoints are: EQ-5D score, activities of daily living (ADL), bony union, upper cervical stability and mortality. Discussion: USOFT is the first randomised controlled trial comparing non-surgical and surgical management of type-2 odontoid fractures in the elderly. Using the NDI and EQ-5D as endpoints, future value-based decisions may consider quality-adjusted life years gained. Major limitations are (1) the allocation bias of the open-label study design, (2) that only higher training levels of all core specialties of spine surgery are included in the surgical treatment arm and (3) that only one type of surgical stabilisation is investigated (posterior C1-C2 fusion), while other methods are not included in this study.