Study design: Systematic methodological review.
Objective: Identify predictors for cervical spine injury (CSI) mortality in elderly patients by reviewing the available literature.
Summary of background data: The proportion of active elderly individuals in society is increasing. This population is at high risk for CSI mortality. The results of studies identifying predictors for CSI mortality in the elderly population are often inconclusive or even conflicting. Currently, there is no set of predictors that can adequately identify and describe CSI mortality risk for the elderly. Thus, we performed a systematic review to identify the predictors for mortality in elderly patients with CSI.
Methods: We performed searches in the MEDLINE, EMBASE, ScienceDirect, and OVID databases (articles published prior to May 2012) for noninterventional studies that evaluated predictors for CSI mortality in the elderly. Only those observational studies with eligible data were included. Study quality was assessed using a modified quality assessment tool that was designed previously for an observational study. Study outcomes were combined with study quality scores using a best-evidence synthesis model.
Results: Twenty-three observational studies involving 2325 patients were included. These studies were published between 1993 and 2011. According to the quality assessment criteria, 8 studies were of high quality, 11 studies were of moderate quality, and 4 studies were of low quality. We identified 3 strong evidence predictors for CSI mortality, including pre-existing comorbidities, spinal cord injury, and age. We also identified 3 moderate evidence predictors, 7 limited evidence predictors and 1 conflicting evidence predictor.
Conclusion: Although there is no conclusive evidence regarding the mortality of elderly patients with CSI, these data provide information that can help us to make recommendations and to counsel patients and their families. Special attention should be paid to the 3 strong predictors. Further studies will be required to validate these predictors.