OTA 2013 Posters

Scientific Poster #30 Geriatric OTA 2013

The Utility of Injury Severity Indices in the High-Energy Geriatric Trauma Population with High-Mortality Orthopaedic Injuries

Sanjit R. Konda MD; William D. Lack, MD; Matthew Wilson, MD; Rachel Seymour, PhD; Madhav A. Karunakar, MD;
Carolinas Medical Center, Charlotte, North Carolina, USA

Background/Purpose: Injury severity indices (injury severity score [ISS], new injury severity score [NISS], revised trauma score [RTS], and trauma score–injury severity score [TRISS]) have been devised to better characterize injury severity and predict outcomes in the general trauma population. Their usefulness in the geriatric population is less clear as preexisting conditions (PECs) have been shown to significantly affect mortality and these indices do not take these into consideration. We sought to evaluate the predictive ability of the above injury indices in high-energy geriatric trauma patients (HE-GTPs) with high-mortality orthopaedic injuries to better understand if anatomic and/or physiologic parameters should be taken into consideration when devising new injury severity indices in this elder population.

Methods: We conducted a retrospective review of prospectively collected data entered into the North Carolina Trauma Registry for all GTPs (age ≥55 years) who presented to our Level I trauma center from 2008 to 2011. High-energy mechanisms of injury were defined as falls from height, motor vehicle and motorcycle crashes, and pedestrians struck by motor vehicles. We identified 1605 HE-GTPs for which the overall mortality rate was 7.0%. We identified extremity, shoulder girdle, and pelvic/sacral fractures that had a mortality rate greater than 7.0% and labeled these as high-mortality orthopaedic fractures. We found 556 HE-GTPs who met the criteria for high-mortality orthopaedic fractures and used this cohort to compare the ability of the ISS, NISS, RTS, and TRISS to predict inpatient mortality using receiver operating characteristic (ROC) curves.

Results: The mean age of our cohort was 66.5 ± 9.7 years and had a mortality rate of 10.6%. High-mortality orthopaedic fractures included (mortality %): clavicle (13.6%), scapula (14.5%), humerus (14.5%), proximal ulna/radius (11.8%), distal radius (8.4%), femur (13.7%), patella (12.2%), tibial shaft (11.5%), pelvis (12.3%), and sacrum (14.5%). The ability of ISS (area under curve [AUC]: 0.769), NISS (AUC: 0.776), and RTS (AUC: 0.752) to predict mortality was graded as fair (AUC range, 0.70-0.80) and there was no difference among these indices. The ability of TRISS (AUC: 0.860) to predict mortality was graded as good (AUC range, 0.81-0.9) and was significantly better than the ISS (P <0.01), NISS (P = 0.01), and RTS (P <0.01). TRISS predicted 86% of deaths at its most predictive value of 0.87. In contrast, ISS, NISS, and RTS only predicted 67.8%, 74.6%, and 62.7% of deaths at their most predictive values of 27, 27, and 7.55, respectively.

Conclusion: In the high-energy geriatric trauma population with high-mortality orthopaedic fractures, TRISS outperforms ISS, NISS, and RTS in the ability to predict inpatient mortality. TRISS is a combination of an anatomic injury index (ISS) and a physiologic index (RTS), thus both anatomic and physiologic profiles should be included in future studies that aim to develop geriatric trauma specific scoring indices. Since TRISS does not factor in PECs, we feel that new scoring indices that also include PECs are likely to improve the ability to predict inpatient mortality and guide triage to appropriate levels of care.

• The FDA has not cleared this drug and/or medical device for the use described in this presentation   (i.e., the drug or medical device is being discussed for an “off label” use).  ◆FDA information not available at time of printing. Δ OTA Grant.