Rib fractures in the elderly: physiology trumps anatomy. Issue 1 (12th March 2019)

• Record Type: Journal Article
• Authors: Schmoekel, Nathan; Berguson, Jon; Stassinopoulos, Jerry; Karamanos, Efstathios; Patton, Joe; Johnson, Jeffrey L
• Is Part Of: Trauma surgery & acute care open. Volume 4:Issue 1(2019)
• Publication Date: 2019-03-12
• Subjects: rib fractures -- frailty -- elderly -- respiratory failure; Traumatology -- Periodicals; Critical care medicine -- Periodicals; Wounds and injuries -- Periodicals; 617.1
• DOI: 10.1136/tsaco-2018-000257
• Languages: English
• ISSNs: 2397-5776

Abstract

Abstract : Introduction: Rib fractures in elderly patients are associated with increased morbidity and mortality. Predicting which patients are at risk for complications is an area of debate. Current models use anatomic, physiologic or laboratory parameters in isolation to answer this question. The 'RibScore' is an anatomic model that assesses fracture severity. Given that frailty is a major driver of adverse outcomes in the elderly, we hypothesize that the combined analysis of fracture severity, physiologic reserve and current pulmonary function are better predictors of respiratory compromise in this population. Methods: This is a retrospective chart review of 263 trauma patients age ≥55 from January 2014 to June 2017. Criteria included blunt mechanism and ≥ 1 rib fracture identified by CT. Variables indicating adverse pulmonary outcomes were defined by: pneumonia, respiratory failure and tracheostomy. Three models were assessed: (1) RibScore, (2) Modified Frailty Index (mFI) and (3) initial partial pressure of carbondioxide (PaCO2 ). Results: A total of 263 patients met inclusion criteria. 13% developed pulmonary complications. Increased RibScore, mFI and PaCO2 were each statistically associated with risk of complications. Receiver operating characteristics area under the curve analysis of individual models predicted complications with the following concordance statistic (CS): anatomic (RibScore) yielded a CS of 0.79 (95% CI 0.69 to 0.89); physiologic (mFI) yielded a CS of 0.83 (95% CI 0.75 to 0.91) and laboratory (PaCO2 ) yielded a CS of 0.88 (95% CI 0.80 to 0.95). The PaCO2 had the highest discriminative ability of the three individual models. Combining all three models yielded the best performance with a CS of 0.90 (95% CI 0.81 to 0.97). Discussion: The RibScore maintains discriminative ability in the elderly. However, models based on mFI and PaCO2 individually outperform the RibScore. A combination of all three models yields the highest discriminative ability. This combined approach is best for assessing the severity of rib fractures and prediction of complications in the elderly. Level of evidence: Prognostic Study, Level III.