Preclinical Care, Clinical Management, and Outcomes of Dog Bite Injuries in High Rabies Burden Districts of Wakiso and Kampala, Uganda

Abstract

Uganda is endemic for rabies disease primarily transmitted to humans through dog bites. Although Uganda has established guidelines on appropriate wound care following dog bites, anecdotal reports of patients applying local regimens to treat dog bite wounds (DBWs) exist, posing the risk of clinical rabies and/or wound infections. The patterns of preclinical practices, their underlying causes, and their linkages to clinical management and outcomes are not well known. The main objective of the study was to assess the preclinical care practices, the clinical management for DBWs and their association with wound infection and other outcomes in the high rabies burden districts of Wakiso and Kampala in Uganda. The study was conducted in Mulago National and Entebbe General Referral Hospitals among patients presenting with dog bite injuries between March and October 2019. To assess the compliance with preclinical practices as recommended in the Uganda Clinical Guidelines (UCG), an explanatory sequential mixed methods study was conducted. A structured questionnaire was administered to patients with DBWs and thirteen in-depth interviews were conducted. To ascertain the determinants of compliance, prevalence ratios (PRs) were computed by the use of a generalized linear model (GLM) with Poisson family and a log link with robust standard errors. The qualitative data were analyzed following a deductive thematic approach. A total of 379 patients were recruited in the study. Nearly half (190/376, 50.5%) had a dog bite in Wakiso district, 201/376 (53.5%) were males, and 203/376 (54.0%) were aged 15 years and above. Seventy patients (18.6%) complied with the preclinical guidelines including washing the wounds with water and soap, and seeking medical care within 24 hours. Factors associated with a reduced likelihood of compliance include being aged 15 years or older; not being certain whether the same dog bit other people; and knowing the owner of the biter. Having a secondary or higher education, being employed, and believing the biting dog was sick was associated with increased likelihood of compliance. Secondly, to assess compliance to clinical guidelines on clinical management of dog bite injuries by healthcare workers, an exploratory qualitative study that used observation of healthcare worker-patient encounters; reviews of medical records; and in-depth interviews with healthcare workers was undertaken. A deductive thematic approach was used to analyse the data. The study found that verification and recording of history was not being done, DBWs were incorrectly classified, and ancillary laboratory tests such as culture and sensitivity were not being done. Antibiotics were being administered based on availability and affordability rather than UCG recommendations. Additionally, there was indiscriminate prescribing of the anti-rabies vaccine. Adherence to UCG was hampered by frequent anti-rabies vaccine stock outs, a lack of coordination/cooperation among post-exposure treatment (PET) centers, and a lack of knowledge and skills on DBWs and rabies management. Thirdly, to assess the antimicrobial resistance burden associated with DBWs, 199 patients with infected wounds had a swab sample from the wound collected and cultured under aerobic and anaerobic conditions. Antibiotic susceptibility testing was conducted using the disc diffusion method following the modified Kirby-Bauer method. Eighty four percent (168/199) of the swabs were culture-positive, yielding a total of 768 viS. intermedius, S. canis, and Corynebactrium spp, were resistant to three classes of antimicrobial agents, while S. aureus, S. pyogenes, E. feacalis, Lactobacillus spp and Lactococcus spp were resistant to 4 or more classes of antimicrobial drugs. Among the gram-negative isolates, P. vulgaris, C. werkmanii, E. asburiae, and Bacteriodes spp were resistant to antimicrobial agents in three classes, while P. mirabilis, K. pneumonae, K. oxytoca, M. wisconsensis, C. canimorsus, E. coli and B. zoohelcum were resistant to 4 or more classes of antimicrobial drugs. Fourthly, the predictors of wound infection were determined via a GLM with the Poisson family and a log link with robust standard errors. Time to detection of wound healing was assessed by using Kaplan- Meier survival curves and used log rank test to test differences in curves. Bivariate logistic regression was used to explore relationships between the selected variables and delayed wound healing. The rate of wound infection among the participants was 52.9% at PET initiation. By day 7, the infection rate had dropped by 40% and 56 new infections had been realized. Having complied with UCG preclinical recommendations and having received conventional treatment before reporting for PEP significantly reduced the chances of infections by approximately 40% and 23%, respectively. Conversely, Category III wounds were associated with a 20% more chances infection at initial presentation than category II. Taken together, the study showed that: 1) compliance with preclinical guidelines was low; 2) clinical management of DBWs did not fully follow the UCG; 3) bacteria from DBWs were highly resistant to metronidazole, and there is a high rate of multi-drug resistance (MDR) to antibiotics commonly used to treat DBWs; and 4) deviations from preclinical guidelines by patients and PEP protocols by clinicians resulted in poor outcomes. This highlights the need for targeted health education programs; regulation of the activities of herbalists with regard to DBWs; interventions that reduce human-dog interactions in public spaces; adoption of an integrated bite case management system; continuing medical education programs for healthcare workers; revisiting metronidazole as one of the antibiotics of choice in UCG for the management of DBWs; enlisting DBWs for surveillance during routine antimicrobial resistance programs and the need for risk assessments before prescribing antibiotics for DBWs in rabies-endemic settings, and creation of awareness on post-exposure prophylaxis and making it available and accessible to the wider communities to minimize risk of human cases of rabies.