C. A. Thiels1, S. Nigo3, M. Kasumba3, J. A. Brown3, S. M. Wren2 1Mayo Clinic,Rochester, MN, USA 2Stanford Medicine,Palo Alto, CA, USA 3Mbingo Baptist Hospital,Northwest Province, , Cameroon
Introduction: Trauma remains a leading cause of death in Africa. Mortality reviews aimed at identifying preventable deaths, or deaths which could be avoided if optimal care had been delivered, are underutilized but may provide information to guide improvement at trauma centers. We report our experience with implementing a trauma mortality review process in a rural teaching hospital in Central Africa.
Methods: A prospective trauma registry at Mbingo Baptist Hospital, Cameroon from 1/2014-3/2016 (n=1912) identified 36 deaths. Chart review was conducted using a standardized preventable death assessment form to identify demographics, cause of death, physician related factors (e.g. delayed diagnosis), system related factors (e.g. lack of medications), and patient related factors to identify themes for improvement. Preventable deaths were defined using American College of Surgeons criteria.
Results: Of the 36 trauma mortalities identified, 30 records were available and included. Median age was 29 years (IQR 19, 46) and 80% were male. Mechanism of injury included 16 road traffic related crashes, 8 thermal injuries, 3 falls, 2 blunt injuries, and 1 firearm injury. Traumatic brain/cord injuries accounted for half (n=15) of the injuries with the remainder being burns (n=8), extremity (n=4), abdominal (n=2), and one patient with tetanus. Fifty percent of patients presented in a delayed fashion (≥ 1-day delay) and 43% were transfers. Two patients died at initial resuscitation, 6 during non-operative management, and 22 after surgery, at median hospital day 2 (IQR 1,5). Causes of death included neurologic (47%), respiratory (33%), multi-organ failure (20%), infection (17%), and bleeding (17%) etiologies. Opportunities for improvement were identified in all cases including 16 preventable or possibly preventable deaths and 14 non-preventable deaths. Physician related issues were identified in 80% of cases with pre-operative (n=13, e.g. suspected under resuscitation in 5/8 burn patients) and post-operative (n=13) factors being the most commonly cited. Systems related issues were cited in 77% of cases including unavailable medications (n=12), lack of ventilator support (n=12) or cross-sectional imaging (n=9), and limited blood product availability (n=5).
Conclusion: Implementation of mortality review at a Central African rural hospital revealed that the majority of trauma deaths resulted from burns or neurologic injuries, with most occurring in the post-operative phase of care. Under resuscitation was noted as a recurring physician related area of improvement in burn patients and was exacerbated by the transfer status of many patients. Lack of cross-sectional imaging was noted as a contributing factor in many of the patients with neurologic trauma. These data may help facilitate quality improvement and allocation of resources while this method of structured review of trauma deaths may help improve the quality of trauma care at other trauma centers in Africa.