The recent AOSpine North America prospective observational cohort study of functional and quality of life outcomes in geriatric odontoid fracture (GOF) patients with Type II dens fracture demonstrated significant benefits of surgical versus non-surgical treatment.1 These benefits included lower annual mortality rates, improved functional outcome based on the Neck Disability Index, and less pain based on the SF-36v2 Bodily Pain score. Notably, there were no significant differences in the overall rates of complications or in the rates of specific complications between the operative and nonoperative treatment groups, except that the surgical group had a significantly lower rate of nonunion. Although the study design was not randomized, due to IRB concerns and general lack of equipoise, the study does provide Level II evidence and is among the best data available to guide management of these patients.
The AOSpine North America GOF study has enabled multiple subanalyses that may provide additional insight into the care of GOF patients. One such analysis is the present study,2 which specifically focused on nonoperative treatment and the effect of fracture union versus nonunion on outcome. This subanalysis was not intended to determine whether an operative or nonoperative treatment is better, as the initial report1 has already documented the benefits of operative treatment through direct comparison of treatment groups. The primary objective of the present study was to assess the general outcomes among GOF patients who were treated nonoperatively and to specifically assess whether fracture nonunion had apparent impact on outcome over a one-year follow-up.
In the present study, we found that patients treated nonoperatively had high rates of nonunion and mortality. Unfortunately, due to IRB concerns we were unable to require routine CT imaging at follow-up to more definitely assess fusion status, and it is likely that many of the patients deemed to be fused only had fibrous unions that may not necessarily prove stable in the long term. This is evidenced by a substantial number of cases with delayed fracture displacement that developed within the first year, despite initially being classified as having achieved “union”. Although our data suggest that patients with nonunion did not have worse outcome compared with those who achieved union at one year, it is very important to recognize that the majority of patients with nonunion underwent surgical treatment and that the outcomes for these patients are reflective of surgical treatment.
Dr. Pearson’s Blog notes that the analysis in the present paper suggests that surgery can “likely be avoided in 78% of patients, and that there is no evidence that delayed surgery leads to worse outcomes than early surgery.” Indeed, the decision of whether to pursue operative or nonoperative treatment for GOF is personal and depends on many factors, including the health of the patient and the risk tolerance and personal preferences of the patient and family. As with any surgical procedure, it is important for the surgeon to discuss with patients the risks, benefits, and alternatives. With regard to the operative approach, a common and effective surgical treatment for GOF is posterior C1-C2 instrumented arthrodesis. This is a well-described procedure that in experienced hands can be achieved within ~2 hours with limited blood loss, a low complication rate, and very high fusion rates.3 Instrumented fusion provides immediate stabilization and facilitates early mobilization, which is particularly important in elderly patients in an effort to minimize deconditioning and medical complications that can occur with decreased mobility.
In contrast, with nonoperative treatment, there remain many unknowns, especially with regard to longer-term outcomes. The present study suggests that, if the nonoperatively treated patient survives, there is an approximately 20% chance that delayed surgery will be necessary within the first year of injury. For the remaining patients, many of whom may have only fibrous unions, the future is less certain. It is important to remember, when contemplating nonoperative care and counseling patients and families, that the present study only provides one-year of follow-up. In the nonoperatively treated patient, there is very likely at least some (albeit not well defined) increased risk of potentially catastrophic re-injury or fracture dislocation. Nevertheless, a subset of patients and families may choose the nonoperative approach, whether due to serious medical conditions or simply personal preference, and for these patients the present study provides a basis for the short-term outlook.
1. Vaccaro AR, Kepler CK, Kopjar B, et al. Functional and quality-of-life outcomes in geriatric patients with type-II dens fracture. The Journal of bone and joint surgery American volume 2013;95:729-35.
2. Smith JS, Kepler CK, Kopjar B, et al. The Effect of Type II Odontoid Fracture Nonunion on Outcome Among Elderly Patients Treated Without Surgery. Spine 2013.
3. Hamilton DK, Smith JS, Sansur CA, Dumont AS, Shaffrey CI. C-2 neurectomy during atlantoaxial instrumented fusion in the elderly: patient satisfaction and surgical outcome. Journal of neurosurgery Spine 2011;15:3-8.