BACKGROUND: Mesh cages have commonly been used for reconstruction after corpectomy. Recently, expandable cages have become a popular alternative. Regardless of cage type, subsidence is a concern following cage placement.
OBJECTIVE: To assess whether subsidence rates differ between static and expandable cages, and identify independent risk factors for subsidence and extent of subsidence when present.
METHODS: A consecutive population of patients who underwent corpectomy between 2006 and 2009 was identified. Subsidence was assessed via x-ray at 1-month and 1-year follow-ups. In addition to cage type, demographic, medical, and cage-related covariates were recorded. Multivariate models were used to assess independent associations with rate, odds, and extent of subsidence.
RESULTS: Of 91 patients, 44.0% had expandable cages and 56.0% had static cages. One-month subsidence rate was 36.3%, and the 1-year subsidence rate was 51.6%. Expandable cages were independently associated with higher rates and odds of subsidence in comparison with static cages. Infection, trauma, and footplate-to-vertebral body endplate ratio of less than 0.5 were independent risk factors for subsidence. The presence of prongs on cages and posterior fusion 2 or more levels above and below corpectomy level had lower rates and odds of subsidence. Infection and cage placement in the thoracic or lumbar region had greater extent of subsidence when subsidence was present.
CONCLUSION: Expandable cages had higher rates and risk of subsidence in comparison with static cages. When subsidence was present, expandable cages had greater magnitudes of subsidence. Other factors including footplate-to-vertebral body endplate ratio, prongs, extent of supplemental posterior fusion, spinal region, and diagnosis also impacted subsidence.
ABBREVIATIONS: ANOVA, analysis of variance
CI, confidence interval
PACS, picture archiving and communications system
‡University of Michigan Medical School, Ann Arbor, Michigan
§Department of Neurosurgery, University of Michigan, Ann Arbor, Michigan
Correspondence: Paul Park, MD, Department of Neurosurgery, University of Michigan, 1500 E Medical Center Dr, Room 3552 TC, Ann Arbor, MI 48109-5338. E-mail: firstname.lastname@example.org
* These authors have contributed equally to this article.
Received July 23, 2012
Accepted November 27, 2012