Methodological systematic review: mortality in elderly patients with cervical spine injury: a critical appraisal of the reporting of baseline characteristics, follow-up, cause of death, and analysis of risk factors.
Review
Overview
abstract
STUDY DESIGN: Methodologic systematic review. OBJECTIVE: To determine the validity of reported risk factors for mortality in elderly patients with cervical spine injury. SUMMARY OF BACKGROUND DATA: In elderly patients with cervical spine injury, mortality has frequently been associated with the type of treatment. To date, however, no review evaluating the validity of reported risk factors for mortality in elderly patients with cervical spine injury has been published. METHODS: Studies evaluating the treatment of cervical spine injuries in elderly (>/=60 years of age) patients were searched through the Medline and EMBASE databases. In addition to standard methodologic details, reporting of putative confounding baseline characteristics and analysis of risk factors for mortality were appraised critically. For this purpose, patient data presented in included studies were pooled. Exploratory descriptive statistics were used for data analysis. RESULTS: Twenty-six eligible studies were identified, including a total of 1550 pooled elderly subjects. Except for 2, all studies reported presence or absence of spinal cord injury. Details concerning the severity and/or extent of the injury were reported in 12 (46%) studies. Pre-existing comorbidities were reported in 9 studies (35%). In the pooled subjects, the cause of death was not reported in 155 of 335 deceased patients (42%). Based on own results, 18 (69%) studies reported on risk factors for mortality. Of these studies, 6 (23%) performed statistical analyses of risk factors for mortality outcomes. Only 1 study statistically adjusted potential risk factors for mortality for confounding. CONCLUSION: Overall, pre-existing comorbidities, concomitant injuries, follow-up and cause of death have been underreported in studies investigating the treatment of cervical spine injuries in elderly patients. To strengthen the validity of risk factors for mortality in future clinical trials, adjustments for appropriately reported putative confounders by regression analysis are mandatory.
