Abstract
STUDY DESIGN: Retrospective study of data abstracted from the diagnosis procedure combination (DPC) database. OBJECITIVES: The etiology of cervical fracture dislocation (CFD) remains unknown because of the limited clinical investigations that involve approximately 100 individuals. This study investigated the association of CFD-impaired levels with patient outcomes. METHODS: This study included 4653 adult patients with a definitive CFD diagnosis from the DPC database. The database consisted of nationwide inpatient data collected from >1000 acute care hospitals in Japan. This DPC database contains information regarding the hospitalization course, such as diagnosis, treatment, medical history, complications, and hospitalization outcomes. We included 866 CFD patients whose impaired levels were specified. We compared the outcomes between the patients whose CFD occurred in the upper cervical levels ("high" group) and the intermediate-low levels ("low" group). This study determined 121 pairs of patients after one-to-one propensity score matching (PSM). The main outcomes included inhospital death, 30-day mortality, and major complications after admission. The secondary outcomes were the length of hospital stay, discharged home rate, and improvement in Barthel index. RESULTS: No differences in inhospital death and 30-day mortality were found between the high and low groups, even after adjusting for PSM (P > 0.05, respectively). Moreover, this analysis revealed that the high group exhibited a higher incidence of respiratory complications than the low group (37.2% vs. 24.8%; P = 0.0256). CONCLUSIONS: This study revealed no difference in mortality but a significantly higher incidence of respiratory complications with high-level CFD compared to low-level CFD.