TRANSACTIONS OF THE ROYAL SOCIETYOF TROPICAL MEDICINE AND HYGIENE(2003) 97, 312-317 Envenoming by Bothropsjararaca in Brazil: association between venom antigenaemia and severity at admission to hospital F. O. S. Fran~a 1'4, K. C. Barbaro 2, H. W. Fan t, J. L. C. Cardoso 1, I. S. Sano-Martins 3, S. C. Tomy 3, M. H. Lopes °, D. A. WarrelP ,6, R. D. G. Theakston 6 and the Butantan Institute Antivenom Study Group* 1Hospital Vital Brazil, 2Laboratory of Immunopathology, and ~Laboratory of Pathophysiology, Instituto Butan- tan, Sglo Paulo, SP, Brazil; 4Department of Infectious and Parasitic Diseases, Faculty of Medicine, University of S~o Paulo, SP, Brazil; 5Centrefor Tropical Medicine, Nuffield Department of Clinical Medicine, University of Oxford, John Radcliffe Hospital, Headington, Oxford 0 X 3 9DU, UK; 6Alistair Reid Venom Research Unit, Liverpool School of TropicalMedicine, Pembroke Place, LiverpoolL3 5QA, UK Abstract The association between the clinical severity of Bothrops jararaca envenoming at admission and serum venom and plasma fibrinogen concentrations before antivenom administration is reported in 137 patients admitted to Hospital Vital Brazil, Instituto Butantan, Silo Paulo, Brazil, between 1989 and 1990. Other variables such as age, gender, site of the bite, use of tourniquet and the time interval between the bite and start of antivenom therapy, spontaneous systemic bleeding, and the 20 minute whole blood clotting test (20WBCT) at admission showed no association with either severity or serum venom antigen concentra- tion (SVAC). Mean SVAC in patients with mild envenoming was significantly lower than in the group with moderate envenoming (P = 0.0007). Patients with plasma fibrinogen concentrations > 1.5 g/L had a lower mean SVAC than patients with plasma fibrinogen concentrations ~ 1.5 g/L (P = 0.02). Those admitted with a tourniquet in place had significantly higher plasma fibrinogen concentrations than those without a tourniquet (P = 0.002). A multiple logistic regression model showed independent risk factors for severity: bites at sites other than legs or forearms, SVACs/> 400 ng/mL, and the use of a tourniquet. Rapid quantification of SVAC before antivenom therapy might improve initial evaluation of severity in B. jararaca bites. Keywords- snakebite, Bothropsjararaca, serum venom antigen concentration, plasma fibrinogen concentration, enzyme immunoassay, Brazil Introduction Some 20 000 snakebites, caused mainly by lance- headed vipers (Bothrops spp.), are recorded annually in Brazil. These result in some morbidity but fewer than 100 deaths/year (Minist6rio da Safide, Brasil, 1998) thanks to the availability of an effective antivenom. Local envenoming by Bothrops spp. is characterized by bleeding, swelling, pain and sometimes blisters, and may be complicated by the development of local ab- scesses and necrosis. Occasionally, compartmental syn- drome may develop. Signs of systemic envenoming include gingival haemorrhage, microscopic haematuria, ecchymosis and consumption coagnlopathy and, more rarely, macroscopic haematuria, epistaxis, haemoptosis, menorrhagia and haematemesis (Cardoso et al., 1993a). Deaths are usually attributed to renal failure, shock, severe bleeding, and complicating septicaemia. Several factors have been suggested as being asso- ciated with the severity of envenoming. Of these, the quantity and composition of venom injected by the snake seem likely to be the most important. The selec- tion of other epidemiological, clinical, and laboratory variables in our study (e.g. time interval between the bite and start of antivenom therapy, bite site, patient's age, and use of tourniquet) was based on earlier reports which attempted to evaluate factors responsible for the severity of envenoming. Their influence on seventy was well known (Rosenfeld, 1971) before the development of techniques such as the enzyme immunoassay (EIA) which permitted estimation of the amount of venom actually circulating in the blood (Theakston et al., 1977; Ho et al., 1986a, 1986b; Barral-Netto et al., *Butantan Institute Antivenom Study Group: J. L. C. Cardoso, H. W. Fan, F. O. S. Franqa, M. T. Jorge, R. P. Leite, S. A. Nishioka, A. Avila, I. S. Sano-Martins, S. C. Tomy, M. L. Santoro, A. M. Chudzinski, S. C. B. Castro, A. S. Kamiguti, E. M. A. Kelen, M. H. Hirata, R. M. S. Mirandola, R. D. G. Theakston, and D. A. Warrell. Address for correspondence: Francisco Oscar de Siqueira Franqa, Hospital Vital Brazil, Instituto Bntantan, Av. Vital Brazil, 1500, Zip code 05503-900, Silo Paulo, SP, Brazil; phone +55 11 3726 7962, fax +55 11 3726 1505, e-mail fosfranca@butantan.gov.br 1991; Audebert et al., 1992, 1994; Theakston & Kami- guti, 1998). The aim of this study was to investigate the associa- tion of venom antigenaemia and clinical severity at hospital admission with some epidemiological, clinical, and laboratory variables in patients envenomed by lance-headed vipers (B. jararaca). Patients and Methods Patients This study was carried out at Hospital Vital Brazil (HVB), Instituto Butantan, S~o Paulo, Brazil, between 1989 and 1990. During this period, 233 patients were admitted either with signs and symptoms suggestive of B. jararaca envenoming or because they brought the B. jararaca snake to hospital with them. Of these, 96 patients were excluded either because they did not fulfill the inclusion criteria, because they refused to participate in the study, or because no venom was injected. The remaining 137 patients were classified clinically as 'mild' or 'moderate', in whom local swell- ing at the site of bite or incoagulable blood were present (Cardoso et al., 1993a). The severity of envenoming grades (mild, moderate, and severe) used were as defined by the Brazilian Ministry of Health (Minist6rio da Saflde, Brasil, 1998). 'Mild' envenoming, occurring in 87 (63.5%) patients, was characterized by swelling involving 1 or 2 segments of the bitten limb and/or coagulopathy (incoagulable blood). 'Moderate' enve- noming, occurring in 32 (23.4%) patients, was defined as swelling involving 3 or 4 segments of the bitten limb with or without coagulopathy. We have no information about the dose given in 18 (13.1%) patients. Patients exhibiting features of severe envenoming, such as renal failure, hypotension (supine systolic blood pressure < 70 mmHg on at least 2 occasions up to 10 min apart at admission), and swelling at admission which in- volved the entire bitten limb and spread to the trunk or in which the summed circumferences of the swollen segments of the bitten limb exceeded those of the same segments of the control limb by more than 30%, were excluded. Also excluded were patients aged < 7 and > 70 years, pregnant women, those admitted > 48 h after the bite, and those who had been treated with