EPIDEMIOLOGY AND SURVIVAL AFTER SURGICAL STABILIZATION OF SPINAL METASTASES DURING 28 YEARS

Abstract

Introduction: The skeleton is the most common location for metastases, with life-time prevalences of 15% and 70% during autopsies. The Vienna Bone and Soft Tissue Tumor Registry has been established in 1965 and is collecting data on primary and secondary malignancies of the musculoskeletal apparatus ever since. The objective of this study was to review the epidemiology and survival of patients undergoing spinal stabilization, including an analysis of trends over time.

Methods: Data on patients operated on between 1980 and 2007 were available from the Tumor registry. Information on location of metastases, number of metastases within and outside the spine, primary tumor, as well as complications and recurrences after treatment were extracted. Survival after surgery was assessed using the Kaplan-Meier method, adjusting for patient age by Lexis expansion. Furthermore, the dataset was expanded on calendar time to test changes in epidemiology and survival during the observed 28 years. The effect of the abovementioned variables on survival was assessed in a Cox regression model using patient age and calendar time as time frame.

Results: Data on 254 patients could be ed. The most common primary diagnoses were hypernephroma (26.4%), breast cancer (19.75), and lung cancer (12.2%), mostly metastasizing to thoracic and lumbosacral spine (40% both). 104 patients (47.9%) had multiple spinal and 41 (16.1%) other osseous non-spinal metastases. 14.6% had complications in the immediate postoperative follow-up, 4.3% suffered from recurrences postoperatively. Average survival after surgery was 0.9 years (95%CI 0.7 to 1.0) with an average patient age of 60.4 years (95%CI 58.8 to 62.0). In the regression model location of metastases (p=0.008), primary malignancy (p< 0.001), and recurrences (p=0.008) were associated with decreases in survival. There was no association between survival time and the decade during which patients were treated (p=0.157). However, there were significantly less complications in patients treated in later decades, demonstrating the patient’s benefit of being referred to a specialized centre (p=0.015).

Discussion: For our study, we are able to draw from experience of 28 years with stabilization due to spinal metastases. Analyzing the data from 254 patients we observed that spinal metastases stem from a similar pattern of primary malignancies. The average survival in our cohort was less than a year, with a very narrow confidence interval. While survival was not associated with later periods of follow-up, number of complications was, supporting the recommendation to refer cancer patients to specialized centers with appropriate experience. The external validity of our findings, however, is confined by the source of our patients, which comprises mostly central and eastern European patients, and few cases referred from more remote areas.