Referral of Burn Patients in the Absence of Guidelines: A Rwandan Study

The management of severe burns and pediatric burns requires an organized system of care delivery, which includes referral guidelines. In Rwanda, the burn unit at the University Teaching Hospital of Kigali (CHUK) is the only dedicated burn unit in the country and admits patients of all ages referred from the other provinces. However, since there are no official referral guidelines, it is unknown whether patients with burns are appropriately referred. This study aims to analyze referral patterns among burn patients admitted at the CHUK burn unit and their adherence to the referral criteria listed by the American Burn Association (ABA), comparing patients transferred to the burn unit from facilities within Kigali and those referred from facilities outside Kigali.

This retrospective study included all patients with acute burns admitted to the CHUK burn unit with data available on the province of origin from 2005 to 2019. Patients with burns younger than 16 y were defined as pediatric burns, as per the CHUK routines. Characteristics of all burns referred from a facility within Kigali (Pat-K) and facilities outside Kigali (Pat-O) were compared using Fisher's exact test for categorical variables or Wilcoxon rank-sum test for continuous variables. The adherence to ABA referral criteria was assessed for variables with available data, which were total burnt surface area %, burn thickness, cause of burn, and age.

The study population consisted of 1093 patients, of which 1064 had data regarding if they were referred from other facilities to CHUK. Overall, the median age was 3 y (2-16 y), with Pat-O being older than Pat-K (P < 0.001). Scalds were the dominant cause of injury in both groups; flame was more common among Pat-O than among Pat-K (in 25.5% versus 10.6%, P < 0.01). Burns of larger size and depth were more common among Pat-O, as compared to the Pat-K group [median total burnt surface area % = 19% versus 15.5%, respectively (P < 0.001); presence of full-thickness burns = 55.6% versus 29.7 %, respectively (P < 0.001)]. Hospitalization was longer and in-hospital mortality higher in Pat-O than in Pat-K [LOS = 42 d (interquartile range 11-164) versus 28 d (interquartile range 9-132), P < 0.05; in-hospital mortality = 18.9% versus 10.1%, P < 0.001]. Among Pat-O, 85% had characteristics that mirrored one or more of the analyzed ABA criteria.

Although there are no clear guidelines, referral patterns indicate that patients are being appropriately referred for specialized burn care in Rwanda. Compared to patients from Kigali, rural patients had larger and more severe injuries, needed more surgical interventions, and had worse outcomes. Most of referred patients matched criteria listed among international referral guidelines. However, the statement of national referral criteria is essential to improve timely access to adequate care.