Snakebites as cause of deaths in the Western Brazilian Amazon: Why and who dies? Deaths from snakebites in the Amazon.

Snake envenoming represents a major burden for public health worldwide. In the Amazon, the official number of cases and deaths detected is probably underestimated because of the difficulty riverine and indigenous populations have reaching health centers in order to receive medical assistance. Thus, integrated analysis of health information systems must be used in order to improve adequate health policies. The aim of this work is to describe a series of deaths and identify risk factors for lethality from snakebites in the state of Amazonas, Brazil. All deaths from snakebites reported to the Brazilian Notifiable Diseases Surveillance System (SINAN) and to the Mortality Information System (SIM; ICD10-10th revision, X.29), from 2007 to 2015, were included. Variables were assessed by blocks with distal (ecological variables), intermediate (demographics) and proximal (clinical variables) components to identify predictors of case fatality. A total of 127 deaths from snakebites were recorded, with 58 pairs found through linkage of the SINAN and SIM databases (45.7%), 37 (29.1%) deaths found only in SINAN and 32 (25.2%) found only in the SIM. Deaths occurred mostly in males (95 cases; 74.8%) living in rural areas (78.6%). The most affected age group was the ≥61 years old (36 cases; 28.4%). Snakebites were presumably due to Bothrops snakes in 68.5% of the cases and Lachesis in 29.5% based on clinico-epidemiological diagnosis. A proportion of 26.2% of the cases received treatment over 24 h after the bite ocurred. On admission, cases were mostly classified as severe (65.6%). Overall, 28 patients (22.0%). Deceased without any medical assistance Antivenom was given to 53.5%. In the multivariate analysis, a distance from Manaus >300 km [OR = 3.40 (95%CI = 1.99-5.79); (p < 0.001)]; age ≥61 years [OR = 4.31 (95%CI = 1.22-15.21); (p = 0.023)] and Indigenous status [OR = 5.47 (95%CI = 2.37-12.66); (p < 0.001)] were independently associated with case fatality from snakebites. Severe snakebites [OR = 16.24 (95%CI = 4.37-60.39); (p < 0.001)] and a lack of antivenom administration [OR = 4.21 (95%CI = 1.30-13.19); (p = 0.014)] were also independently associated with case fatality. Respiratory failure/dyspnea, systemic bleeding, sepsis and shock were recorded only among fatal cases. In conclusion, i) death from snakebites was underreported in the mortality surveillance system; ii) older age groups living in remote municipalities and indigenous peoples were the population groups most prone to death; iii) lack or underdosage of antivenom resulted in higher case fatality and iv) systemic bleeding, circulatory shock, sepsis and acute respiratory failure were strongly associated to fatal outcome.