RibScore: A Novel Radiographic Score Based on Fracture Pattern that Predicts Pneumonia, Respiratory Failure, and Tracheostomy
Author(s):
Brandon Chapman, University of Colorado Denver; Benoit Herbert, Denver Health Medical Center; Jennifer Salotto, Denver Health Medical Center; Maridi Rodil, Denver Health Medical Center; Robert Stovall, Denver Health Medical Center; Walter Biffl, Denver Health Medical Center; Jeffrey Johnson, Denver Health Medical Center; Clay Burlew, Denver Health Medical Center; Carlton Barnett, Denver Health Medical Center; Charles Fox, Denver Health Medical Center; Ernest Moore, Denver Health Medical Center; Gregory Jurkovich, Denver Health Medical Center; Frederic Pieracci, Denver Health Medical Center
Background: There is currently no scoring system for rib fractures that relates detailed anatomic variables to patient outcomes. The purpose of this study was to develop and validate a radiographic rib fracture scoring system based on initial CT chest findings.
Hypothesis: A radiographic rib fracture score will predict adverse pulmonary outcomes.
Methods: We reviewed our trauma registry from Sept 2012- April 2014 for all blunt trauma patients with ≥ 1 rib fracture visualized on CT chest. We identified six candidate radiographic variables and tested their individual associations with pneumonia, defined as clinical suspicion plus quantitative microbiology: 1) ≥ 6 rib fractures, 2) bilateral fractures, 3) flail chest, 4) ≥ 3 severely (bi-cortical) displaced fractures, 5) first rib fracture, and 6) at least one fracture in all three anatomic areas (anterior, lateral, and posterior). Using these variables, we developed the “RibScore,” which ranges from 0-6, and assigns one point for each of the aforementioned variables. The RibScore was then validated among the sample. Our primary outcome measure was pneumonia; secondary outcomes were acute respiratory failure and tracheostomy. Statistics: chi-squared, alpha=0.05.
Results: A total of 385 patients with ≥ 1 rib fracture were identified; 274 (71.2%) males, mean age 48.4 years, and mean Injury Severity Score 19.5. Of these patients, 156 (40.5%) had ≥ 6 rib fractures, 120 (31.2%) had bilateral fractures, 46 (11.9%) had flail chest, 32 (8.3%) had ≥ 3 severely displaced fractures, 91 (23.6%) had a first rib fracture, and 58 (15.1%) had fractures in all three anatomic areas. Each of the RibScore component variables was associated individually with pneumonia by univariate analysis (p<0.05 in each case). The median RibScore was 1 (range 0-6). Only 2 patients (0.5%) had a RibScore of 6 and were excluded from analysis. The distribution of the rib score was: zero (39.2%); one (23.6%); two (16.9%); three (10.1%); four (8.3%); and five (1.3%). RibScore was associated both linearly and significantly with pneumonia (p<0.01), acute respiratory failure (p<0.01), and tracheostomy (p<0.01) (Figure).
Conclusions: The RibScore successfully predicted adverse pulmonary outcomes, including pneumonia. The RibScore represents a standardized assessment of fracture severity that may be used for communication and prognostication of the severely injured trauma patient.