Complications, revision fusions, readmissions, and utilization over a 1-year period after bone morphogenetic protein use during primary cervical spine fusions

Abstract Background context Nationwide estimates examining bone morphogenetic protein (BMP) use with cervical spine fusions have been limited to perioperative outcomes. Purpose To determine the 1-year risk of complications, cervical revision fusions, hospital readmissions, and health care services utilization. Study design A retrospective cohort study from 2002 to 2009 using a nationwide claims database. Patient sample There were 61,937 primary cervical spine fusions of which 1,677 received BMP. Outcome measures Complications, revision fusions, 30-day hospital readmission, and health care utilization. Methods Data for these analyses come from the Thomson Reuters MarketScan Commercial Claims and Encounters Database 2010. Patients were aged 18 to 64 years, receiving and not receiving BMP with a primary (C2–C7) cervical spine fusion. All outcomes were defined by International Classification of Diseases, 9th edition Clinical Modification and Current Procedural and Terminology, 4th edition codes. Complications were analyzed as any complication and stratified by nervous system, wound, and dysphagia or hoarseness. Cervical revision fusions were determined in the 1-year follow-up. Hospital readmission discharge records defined 30-day hospital readmission and reason for the readmission. The utilization of at least one health care service of cervical spine imaging, epidural usage or rehabilitation service was examined. Poisson regression models were used to estimate the relative risk and 95% confidence interval (CI). Linear regression was used to determine the time to hospital readmission. Results were stratified by anterior or posterior and circumferential approaches. Results Patients receiving BMP were 29% more likely to have a complication (adjusted relative risk [aRR]=1.29 [95% CI, 1.14–1.46]) and a nervous system complication (aRR=1.42 [95% CI, 1.10–1.83]). Cervical revision fusions were more likely among patients receiving BMP (aRR=1.69 [95% CI, 1.35–2.13]). The risk of 30-day readmission was greater with BMP use (aRR=1.37 [95% CI, 1.07–1.73]) and readmission occurred 27.4% sooner on an average. Patients receiving BMP were more likely to receive computed tomography scans (aRR=1.34 [95% CI, 1.06–1.70]) and epidurals with anterior surgical approaches (aRR=1.29 [95% CI, 1.00–1.65]). Conclusions These findings question both the safety and effectiveness of off-label BMP use in primary cervical spine fusions.