E. DeSouza1, T. J. Zens1, G. Leverson1, H. Jung1, S. Agarwal1 1University Of WI School Of Medicine And Public Health,General Surgery,Madison, WI, USA
Introduction: The number of rib fractures an individual suffers has long been considered an independent predictor of morbidity and mortality; however, in previous studies all rib fractures were considered to be the same. We hypothesized that not only total number of ribs fractured, but also location of fracture is a strong predictor of patient outcomes in terms of mortality, length of stay, and discharge disposition.
Methods: An IRB-approved retrospective chart review was performed at an academic, level one trauma center. Patients who suffered traumatic rib fractures between January 2013 and April 2015 were identified by CPT codes for possible inclusion in the study. Individual computer tomography scans of the chest were reviewed and validated by staff radiology reads. The location of the rib fractures were characterized in terms of anterior, posterior, lateral, upper, middle, lower, and right vs. left. SAS statistical software, logistic regression curves and ANOVA data analysis examined the data for relationships between rib fracture location and patient outcomes in terms of length of stay, ICU length of stay, discharge disposition and overall mortality.
Results: 929 patients were initially reviewed for possible inclusion in the study and 248 excluded. A total of 3,864 fractures were identified in the patient population. Statistically significant positive correlation coefficients were identified between length of stay and number of rib fractures in all locations with the strongest relationships seen in the upper (0.191, p=<0.0001), middle (0.185, p=<0.0001) and lateral (0.189, p=<0.0001) locations. Similarly, statistically significant correlations were seen between all rib fracture locations and ICU days with the strongest relationships seen in patients with upper (0.184, p=<0.0001), middle (0.189, p=<0.0001) and lateral (0.205, p=<0.0001) fractures. In addition, there was a statistically significant association between in hospital mortality and lateral rib fractures, with the non-survivors having on average 2 more lateral rib fractures than survivors (p=.0192), and left-sided rib fractures, with the non-survivors also having on average 2 more left-sided rib fractures than survivors (p=.0287). A statistically significant association was seen between discharge disposition and left-sided fractures (p=0.0160) as well as lateral rib fractures (p=0.0002). There was a linear relationship between type of discharge, in terms of increasing need for support, and number of lateral rib fractures.
Conclusion: Rib fracture location, particularly left sidedness and lateral location, help predict outcomes in terms of mortality and discharge disposition. This may assist in discussing care and planning resources for traumatically injured patients. A larger validation trial is needed to confirm these results.