Surgical versus nonsurgical management for type II odontoid fractures in the elderly population: a systematic review

Background

Odontoid process fractures, of which type II constitute the majority, are an increasingly important cause of morbidity and mortality in the elderly population. The incidence of geriatric type II fractures is steadily increasing in line with the aging population. However, the decision between surgical and non-surgical intervention for type II fractures in the elderly remains controversial.

Purpose

The present study aims to synthesize the current published literature comparing outcomes following surgical and non-surgical interventions for type II odontoid fractures in the elderly population (≥65 years old).

Study design/setting

Systematic review and meta-analysis were performed.

Methods

A systematic search of MEDLINE, MEDLINE In-Progress & Other Non-Indexed Citations, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) was performed to identify available evidence in English language. Studies with extractable data for all type II odontoid fractures in participants aged 65 years or older and which compared surgical and non-surgical intervention were included. Methodological quality was assessed using the Downs & Black checklist. Primary outcomes were mortality at short-term follow-up (≤3 months), mortality at long-term follow-up (predetermined study endpoint or mean follow-up length), and radiological union rate. Funding was provided by The University of Edinburgh for travel expenses to present this paper at the Society of British Neurological Sciences 2016 Conference ($170).

Results

Twelve studies (n=1,098), all non-randomized, met eligibility criteria. Methodological quality was particularly poor in the confounding, bias, and power domains of assessment. Substantial methodological and statistical heterogeneity allowed only a narrative synthesis of the primary outcomes. Overall, data on mortality at short-term follow-up appeared to favor neither surgical nor non-surgical intervention. A small favorable outcome in surgically managed patients over non-surgically managed patients in terms of mortality at long-term follow-up was not proven conclusive because of considerable heterogeneity in study methodologies. Inadequate reporting of the time point of union assessment introduced the potential for significant intra- and interstudy heterogeneity and precluded assessment of union rates.

Conclusions

Evidence on this controversial topic is sparse, markedly heterogeneous, and of poor quality. Well-designed prospective trials adhering to guidance published by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) initiative are required to inform clinical practice on this contentious but growing issue. Future randomized controlled trials should include an assessment of frailty and medical comorbidities with suitable patients subsequently randomized to surgical or non-surgical treatment.