Musculoskeletal injuries in trauma patients: a Swedish nationwide register study including 37,266 patients.

BACKGROUND AND PURPOSE: Trauma causes over 4 million annual deaths globally and accounts for over 10% of the global burden of disease. Trauma patients often sustain multiple injuries in multiple organ systems. We aimed to investigate the proportion and distribution of musculoskeletal injuries in adult trauma patients. PATIENTS AND METHODS: This is a register-based study using data from the national Swedish trauma register (SweTrau) collected in 2015-2019. By categorizing Abbreviated Injury Scale (AIS) codes into different injury types, we provide a detailed description of the types of musculoskeletal injuries that occurred in trauma patients. RESULTS: 51,335 cases were identified in the register. After exclusion of 7,696 cases that did not have any trauma diagnosis (AIS codes) registered from the trauma and 6,373 patients aged < 18, a total of 37,266 patients were included in the study. 15,246 (41%) had sustained musculoskeletal injury. Of the patients with musculoskeletal injuries, 7,733 (51%) had more than 1 such injury. Spine injuries were the most common injury location (n = 7,083 patients, 19%) followed by lower extremity injuries (n = 5,943 patients, 16%) and upper extremity injuries (n = 6,273 patients, 17%). Fractures were the dominating injury type with 30,755 (87%) of injuries being a fracture. CONCLUSION: 41% of the trauma patients had at least 1 musculoskeletal injury. A spine injury was the most common injury location. Fractures was the dominating injury type constituting 87% of all injuries. We also found that half the patients (51%) with spine or extremity injuries had ≥ 2 such injuries.

Prediction of mortality among severely injured trauma patients A comparison between TRISS and machine learning-based predictive models.

BACKGROUND: Given the huge impact of trauma on hospital systems around the world, several attempts have been made to develop predictive models for the outcomes of trauma victims. The most used, and in many studies most accurate predictive model, is the "Trauma Score and Injury Severity Score" (TRISS). Although it has proven to be fairly accurate and is widely used, it has faced criticism for its inability to classify more complex cases. In this study, we aimed to develop machine learning models that better than TRISS could predict mortality among severely injured trauma patients, something that has not been studied using data from a nationwide register before. METHODS: Patient data was collected from the national trauma register in Sweden, SweTrau. The studied period was from the 1st of January 2015 to 31st of December 2019. After feature selection and multiple imputation of missing data three machine learning (ML) methods (Random Forest, eXtreme Gradient Boosting, and a Generalized Linear Model) were used to create predictive models. The ML models and TRISS were then tested on predictive ability for 30-day mortality. RESULTS: The ML models were well-calibrated and outperformed TRISS in all the tested measurements. Among the ML models, the eXtreme Gradient Boosting model performed best with an AUC of 0.91 (0.88-0.93). CONCLUSION: This study showed that all the developed ML-based prediction models were superior to TRISS for the prediction of trauma mortality.

The pelvic fracture - Indicator of injury severity or lethal fracture?

BACKGROUND: Presence of pelvic fractures in trauma patients has previously been related to high mortality. However, there are controversies on whether pelvic fractures are the underlying cause of death or if it is rather an indicator of injury severity. We aimed to assess whether the presence of pelvic fracture increased mortality among a cohort of trauma patients or if it was simply an indicator of severe injury. MATERIAL AND METHODS: Karolinska University Hospital is the largest trauma centre in Sweden. The hospital is linked to the Swedish National Trauma Registry, "SweTrau". Registry data was collected for the period January 2013 until December 2015 with a one year further follow-up regarding mortality. Patients in the pelvic fracture group were compared to the non-pelvic fracture group and regression analysis was performed adjusting for factors that could possibly affect mortality. RESULTS: Univariable analysis showed that pelvic fracture was associated with an increased mortality, OR 2.4 (CI 1.3-3.4). Multivariable analysis showed that the presence of a pelvic fracture was not associated with an increased 30-day mortality (OR 0.5, CI 0.2-0.9), while factors as Shock (OR 7.1, CI 4.6-10.9), GCS < 9 (OR 6.2, CI 3.9-9.8), ISS > 15 (OR 12.4, CI 8.1-18.9), Age >60 (OR 3.2, CI 2.1-4,9) and ASA 3-4 (OR 4.7, CI 3.1-7.3) were associated with an increased 30-day mortality. Factors affecting 1-year mortality was analysed in the same way and the results were similar. CONCLUSION: Presence of pelvic fractures in trauma patients is not correlated to increased mortality when adjusted for Age, ISS, ASA, GCS and Shock.

The Spanish flu in Uppsala, clinical and epidemiological impact of the influenza pandemic 1918-1919 on a Swedish county.

INTRODUCTION AND AIM: The Spanish flu reached Sweden in June 1918, and at least one-third of the population (then 5.8 million) became infected. Some 34,500 persons (5.9 per 1,000 people) died from influenza during the first year of the pandemic (when acute pneumonia is included, the number of deaths rose to 7.1 per 1,000 people). In this historical look back at the pandemic, our aim was to review the epidemiological impact on the Swedish county of Uppsala, the clinical outcomes and the economic impact on the regional hospital; a relevant backgound to consider the impact of a future virulent pandemic. We also focused on how the pandemic was perceived by the medical community and by health care authorities. METHODS: Health care reports, statistics, daily newspapers, medical journals, and records of patients treated for influenza at the Uppsala Academic Hospital from July 1918 to June 1919 were included in our review. RESULTS: An influenza related mortality rate of 693 persons (5.1 per 1,000 people) was reported in the Uppsala region from 1918-1919; from July 1918 to June 1919, 384 patients were treated for influenza at the Uppsala Academic Hospital. The first wave peaked in November 1918 with case fatality rates up to 30%; a second wave peaked in April 1919 with a lower rate of mortality. Of the patients treated, a total of 66 died. Of these, 60% were 20-29 years of age, and 85% were less than 40 years old. Autopsy reports revealed pneumonia in 89% of the cases; among these, 47% were hemorrhagic, 18% were bilateral, and 45% had additional extrapulmonary organ involvement. Signs of severe viral disease were documented, but secondary bacterial disease was the primary cause of death in the majority of cases. CONCLUSION: The epidemiologic and pathologic results were in accordance with other publications of this time period. The costs of running the hospital doubled from 1917 to 1920 and then reversed by 45%. Today, an influenza pandemic of the same virulence would paralyze health care systems and result in extremely high financial costs and rates of mortality.