Comparison of Bothropoides jararaca bites with and without envenoming treated at the Vital Brazil Hospital of the Butantan Institute , State of São Paulo , Brazil

Introduction: This study analyses the cases of all bites (including dry bites) caused by Bothropoides jararaca attended at the Vital Brazil Hospital of the Butantan Institute, State of São Paulo, Brazil. Methods: A retrospective study was conducted of patients bitten by Bothropoides jararaca (n=792) from January 1990 to December 2004. The characteristics of the snake specimen, data related to the accident and clinical manifestations on admission were obtained from patient medical records. Results: The majority of the cases in this study were caused by female and juvenile snakes. No stomach contents were found in 93.4% of the snake specimens after dissection. No statistical difference was observed between the occurrence of dry bites and the maturity or sex of the snake. The median SVL of snakes in mild and moderate cases was 40.5cm and in severe cases, SVL increased to 99cm. Necrosis was more common in the digits of the feet and hands (4.8%) compared to the other body regions (1.8%). A significant difference was verified between severity and a time interval greater than six hours from the bite to hospital admission. A significant association was verified between gingival bleeding and abnormal blood coagulability. In accidents caused by adult snakes, necrosis was more frequent (7.2%) compared to accidents caused by juvenile snakes (1%). Conclusions: In this work, the association between certain epidemiological data and the evolution of biological parameters in the clinical course of Bothrops sensu latu accidents were highlighted, contributing to the improvement of snake bite assistance. Key-words: Bothrops. Bothropoides. Snake bites. Envenoming. Brazil. Clinical aspects. RESUMO Introdução: Neste estudo, analisou-se todos os casos de picadas (incluindo picadas secas) causadas por Bothropoides jararaca atendidos no Hospital Vital Brazil do Instituto Butantan, São Paulo, Brasil. Métodos: Estudo retrospectivo em que foram incluídos pacientes atendidos no Hospital Vital Brazil do Instituto Butantan, picados por serpentes da espécie Bothropoides jararaca (no=792) entre 1990 a 2004. Os dados foram obtidos através de prontuários médicos. Resultados: No presente estudo, a maioria dos acidentes foi causada por serpentes fêmeas e filhotes. Não havia presença de conteúdo estomacal em 93,4% das serpentes dissecadas. Não houve diferença estatística entre a ocorrência de picada seca e o sexo da serpente. O comprimento rostro cloacal das serpentes nos casos leves e moderados foram 40,5cm e nos casos severos 99cm. Necrose foi mais comum nos dedos dos pés e das mãos (4,8%) em comparação com outras regiões do corpo (1,8%). Houve diferença estatística entre a gravidade e o intervalo de tempo entre a picada e a admissão hospitalar superior a seis horas. Encontramos uma associação significativa entre gengivorragia e incoagulabilidade sanguínea. Nos acidentes causados por serpentes adultas, a necrose foi mais frequente (7,2%) quando comparado aos acidentes causados por serpentes filhotes (1%). Conclusões: Neste estudo, destaca-se a associação entre os dados epidemiológicos e biológicos em relação à evolução do quadro clínico nos acidentes botrópicos, contribuindo para a melhoria da assistência nos acidentes ofídicos. Palavras-chaves: Bothrops. Bothropoides. Acidentes ofídicos. Envenenamento. Brasil. Aspectos clínicos. In Brazil, nine genera of venomous snakes are recognized: Bothrops, Bothropoides, Bothriopsis, Bothrocophias, Rhinocerophis, Crotalus, Lachesis, Leptomicrurus and Micrurus1. Due to similar manifestations and treatment with the same antivenom, the first five genera are grouped here as Bothrops sensu lato. These snakes are responsible for 86.9% (Bothrops sensu lato), 8.7% (Crotalus), 3.6% (Lachesis) and 0.8% (Leptomicrurus and Micrurus) of accidents involving venomous snakes2. In 2009, 21,446 cases of snakebite were reported, with a fatality rate of 0.4%3. The highest incidences were recorded in the southeast (28%), north (27%) and northeast (24%) of Brazil4. There is a predominance of cases in the hot and rainy months (October to April), affecting mostly rural, male workers, aged between 10 and 40 years of age5. In the northeast, most accidents occur between April and May, while in the north, cases occur in the first months of the year4. The Bothropoides jararaca, formerly Bothrops jararaca5, species has high adaptive capacity, thriving in urban and wild environments6, and it is the most common species in the southeast7. The venom has local acute inflammatory, coagulative and hemorrhagic effects8. The most common signs and symptoms include pain, swelling and bleeding, occurring within the first six hours following envenomation. Six hours after the accident, blistering, bruising, necrosis, secondary infection sites and, in severe cases, acute renal failure occur. Frequently the blood clotting time (CT) is increased9. Bothrops sensu lato envenomation severity is classified as mild, moderate and severe. A dry bite is defined here as a bite without clinical signs or systemic envenoming10. The number of dry bites which do not cause envenoming is not known. When considering all Bothrops sensu lato bites, most studies in Brazil do not include dry bites. According to the World Health Organization (WHO), 25% to 35% of all cases of

In 2009, 21,446 cases of snakebite were reported, with a fatality rate of 0.4% 3 .The highest incidences were recorded in the southeast (28%), north (27%) and northeast (24%) of Brazil 4 .There is a predominance of cases in the hot and rainy months (October to April), affecting mostly rural, male workers, aged between 10 and 40 years of age 5 .In the northeast, most accidents occur between April and May, while in the north, cases occur in the first months of the year 4 .
The Bothropoides jararaca, formerly Bothrops jararaca 5 , species has high adaptive capacity, thriving in urban and wild environments 6 , and it is the most common species in the southeast 7 .
The venom has local acute inflammatory, coagulative and hemorrhagic effects 8 .The most common signs and symptoms include pain, swelling and bleeding, occurring within the first six hours following envenomation.Six hours after the accident, blistering, bruising, necrosis, secondary infection sites and, in severe cases, acute renal failure occur.Frequently the blood clotting time (CT) is increased 9 .
Bothrops sensu lato envenomation severity is classified as mild, moderate and severe.A dry bite is defined here as a bite without clinical signs or systemic envenoming 10 .The number of dry bites which do not cause envenoming is not known.When considering all Bothrops sensu lato bites, most studies in Brazil do not include dry bites.According to the World Health Organization (WHO), 25% to 35% of all cases of

RESULTS
Nicoleti AF et al -Comparison of Bothropoides jararaca bites treated at the Vital Brazil Hospital snakebites worldwide are dry bites 11 .Silveira and Nishioka 12 state that in dry bites, the amount of venom delivered to the victim is negligible.
This study analyzed the cases of all bites (including dry bites) caused by Bothropoides jararaca attended at the Vital Brazil Hospital (HVB) of the Butantan Institute, São Paulo, Brazil, only when the snake was brought by the victim.The study analyzed the epidemiological, clinical and laboratorial data in these cases.
Retrospective study of data obtained from medical records for all patients attended at the HVB with snake bite caused by the snake species Bothropoides jararaca.The snake specimens brought by the patients were identified and stomach contents were analyzed by the Herpetology laboratory of the Butantan Institute.Data collection and analysis covered a period of 15 years, between 1990 and 2004.
The information collected included epidemiological data from the patient (age, sex and geographic area), characteristics of the snake specimen (sex, maturity, length and stomach contents), data related to the accident (date and time of the accident, the anatomical site, presence of incisions, tourniquets, substances applied to the site), clinical manifestations on admission: presence of signs and symptoms including pain, local hemorrhaging, edema, erythema, ecchymosis, gingival bleeding, blistering, infection, local wound discharge, necrosis, systemic hemorrhaging, sepsis and acute renal failure.Blood CT was measured on admission and the quantity in vials of serotherapy was documented.
The severity classification was established according to Brazilian Ministry of Health guidelines 13 .These guidelines define dry bite as a case in which patients show no signs or symptoms consistent with envenoming and present normal blood CT.Mild cases are defined as the presence of mild local signs, such as edema.Moderate cases include regional edema, and severe cases are defined as the presence of swelling around the full length of the affected limb and/or acute renal failure and/or shock, and/or severe hemorrhaging.These definitions do not take into account normal or abnormal blood CT.
All the patients were divided into three groups according to age: children (12 years), adults (13-59 years) and elderly individuals (above 60 years).
The maturity of the snake specimens were classified according to snout-vent length (SVL): adults above 65cm in males and above 75cm in females 6 .The stomach contents were classified according to the presence or absence of food.
The data were analyzed with SPSS (Statistical Package for Social Sciences) version 15.0.The Chi-square χ 2 , Fisher Exact and Kolmogorov-Smirnov tests were used to establish associations between the variables.

Ethical
The work was approved by the Ethics in Research Committee of the Faculty of Medicine of the University of São Paulo (CAPPesq no.0387/08).
A total of 792 patient medical records were included in the study, 628 (79.3%) were male, age ranged from 4 to 80 years-old, with a mean age of 33.23 ± 17.67 years-old and median of 31 years-old.
Most accidents occurred between the months of October and April (n = 638; 80.6%), with 711 (90.2%) between 7am and 8pm.The cities directly surrounding central São Paulo carried the highest burden of cases (n = 646; 81.9%).Analysis verified that 56.8% (n = 450) of the patients included were working when bitten.
Furthermore, in pre-hospital observations, 221 (27.9%) patients had used a tourniquet, 17 (2.1%) had made an incision at the site of the bite and 75 (9.5%)used other substances, for example garlic, tobacco or kerosene, on the site of the bite.
Five hundred and thirty-four (68.5%) patients were admitted to HVB less than three hours after the bite and according to severity, 117 (14.8%) patients showed no symptoms of envenoming.Five hundred and thirteen (68.4%) cases were classified as mild, 142 (17.9%) as moderate and 20 (2.5%) as severe.
Comparison of the epidemiological variables in the absence and presence of envenoming revealed no statistical difference in relation to the sex of the patient (p = 0.296), patient age when divided into child, adult and elderly groups (p = 0.615), the sex of the snake (p = 0.386), its maturity (p = 0.381) and the presence of stomach contents (p = 0.236).The median SVL of snakes in mild and moderate cases was 40.5cm, while in the severe cases, the SVL increased to 99cm.The majority (71.7%) of cases occurred during the spring and summer (p < 0.001) (Table 1).
Only 19 (2.4%) of the patients who used tourniquet had dry bites.Signs of envenoming were less frequent in patients bitten on the arms compared to patients bitten on the legs (p < 0.05).The use of a tourniquet was significantly more frequent in patients bitten on the legs (n = 146; 18.7%) compared to patients bitten on the arms (n = 75; 9.6%).Cases involving the distal limbs (feet, hands and digits) were caused more frequently by juvenile snakes (n = 413; 60.1) (Table 2).
A significant difference was verified in envenomation severity of patients attended at hospital less than or greater than 6 hours after the accident (p < 0.05) (Table 3).A significant association between the presence of gingival bleeding and blood incoagulability was detected (p < 0.001) (Table 4).
A statistical association was observed in relation to the anatomical site affected and the presence of necrosis, in that necrosis more commonly affected the digits of the hands or feet (4.8%) compared to the other body regions (1.8%).Observation verified that in cases caused by adult snakes, necrosis occurred more frequently (7.2%) compared to cases caused by juvenile snakes (1%) (Table 5).
Comparing the three five-year periods studied (1990 to 1994; 1995 to 1999; 2000 to 2004), no statistical difference was determined in relation to the use of pre-hospital procedures (tourniquet, incision and substances applied to the site of the bite).However, a difference was observed in the timing between bite and hospital admission for the different periods.Between 1990 and 1994, 36 (53.7%) patients were admitted after 6 hours, whereas between 1995 and 1999, 19 ( Analysis of the data collected showed that most envenoming occurred in male patients with an average age of 33 years-old, while at work.This is in agreement with data from other studies published in Brazil [14][15][16][17][18][19] .The data also showed that snakes are more active in the hot rainy periods, whereas in the cold dry periods, snakes show little movement.This may be related to low humidity and the lower availability of prey 20 .
Regarding the time of the accident, although venomous snakes in Brazil follow nocturnal habits and rural workers are engaged in field activities during the day, it is during this period that thermoregulation of the snake occurs, when they search for sites with suitable temperatures, not exposed to direct solar radiation.This may increase human-snake encounters 20 .On the other hand, asynchrony between snake activity period and time of the accidents does not imply that inactive hidden snakes cannot bite defensively.In fact, it is noted in the Ministry of Health Manual that 52.3% of bites occur during normal daylight hours 13 .
The majority of the cases in this study were caused by adult female and juvenile snakes.This could be explained by the fact that female Bothropoides jararaca grow to longer lengths than males 20 .Gravidity is another factor, since females have higher food demands and forage more actively, demanding greater thermal control, consequently, they are more susceptible to human encounters.It is likely that juvenile snakes also cause more bites because visualization of smaller snakes is more difficult.In this case, a juvenile snake can hide coiled under a small leaf, making their presence hard to detect by humans.
The mean SVL was 52.2cm, which implies a large proportion of juvenile snakes.No stomach contents were found in 93.4% of the snake specimens, though in regard to this fact, there is little literature available for comparison.The relationship between the presence of food and the severity of bite in Daboia russelii siamensis did not found correlation 21 .Most (81.9%) cases occurred in metropolitan areas and following decentralization of the distribution of antivenom, the HVB began to serve residents of towns closest to the city of São Paulo.
Regarding the use of tourniquets, incision and substance use on the bite site, such measures are harmful because they increase the risk of infection and tissue necrosis.
Regarding the anatomical regions affected, the use of boots could prevent 44% of bites to the feet and toes.The use of appropriately thick gloves would prevent accidents in 36% of bites affecting the hands and fingers.One study determined that the use of boots and leggings prevent bites to the legs (16%) 13 .
The majority of cases caused by Bothropoides jararaca treated at the HVB were admitted less than three hours after the bite.These results are similar to other studies published in Brazil [22][23][24] .
Concerning severity, approximately 80% of the cases showed no evidence of envenoming or were classified as mild.This fact explains the low mean number of vials used per case (3.8 vials).The most common signs and symptoms in these cases were: pain (82.2%) and edema (81.6%).According to Silveira and Nishioka 12 ,in Bothrops sp cases, edema is an early sign of envenoming.Many studies report that pain and edema are, respectively, the dominant symptom and sign in Bothrops sensu lato cases, present in more than 90% of patients 22,25,26 .
In relation to blood CT, 41% of patients had normal CT, 18.2% had prolonged CT and incoagulable blood was verified in 40.8%.A study was conducted for the HVB by Ribeiro and Jorge 22 on 3,139 patients diagnosed as Bothrops sensu lato cases, admitted between 1981 and 1990.In comparison, the authors found 2,990 of CT results, which verified that 42.1% were normal, 11.2% were prolonged and 46.7% were incoagulable.
The length of admission for patients at the HVB, admitted between 1990 and 2004, was 2 days (median).This is the same as that obtained by Ribeiro and Jorge 22 in the same hospital between 1981 and 1990.
No statistical difference was observed between the distribution of dry bites and the sex of the patient.Similarly, no difference occurred in severity in relation to the sex of the patient.These results are consistent with the literature 13,17 .
No statistical difference was determined between the severity of cases and the age of patients when divided into children, adults and elderly groups.Previous studies reported no association between the bite severity and the age of the patient in relation to Bothrops sensu lato cases 22,27 .
Regarding the sex of the snake, there is scant literature regarding the comparison of venoms from male and female specimens of the same species.One study reported that females of Bothropoides jararaca produce larger amounts of venom (220mg) with higher protein content and higher lethal activity compared to males (40mg) 28 .The venom activity of females demonstrates increased hemorrhagic action, while male venom shows greater procoagulant action 28 .However, the present study verified no difference in the occurrence of dry bites in relation to the sex of the snake.
No statistical difference was observed between the occurrence of dry bites and the maturity of the snake.However, when dry bites are excluded from the analysis, observation showed that the median SVL of snakes in mild and moderate cases was 40.5cm, while in severe cases it was 99cm.In agreement with these results, other studies with Bothropoides jararaca and Bothrops moojeni showed that accidents caused by juvenile snakes rarely cause severe cases of envenomation 29,30 .Furthermore no statistical difference was verified between the presence of stomach contents and the occurrence of dry bites.This relation was also not observed in Russell´s Viper (Daboia russelli siamensis) studies 21 .
Analysis revealed that dry bites occurred at the same frequency over each year studied.However, such cases were more frequent during the spring and summer months.
The use of a tourniquet by patients did not correlate with the presence of a dry bite.The use of a tourniquet can compromise venous return and lymphatic flow of an affected limb and can cause edema formation.In turn, the presence of this type of edema can be misinterpreted by the physician as being caused by the presence and action of Bothrops sensu lato venom.According to França 5 , the use of a tourniquet can contribute to the severity of the bite site, especially in envenoming with local inflammatory activity on the inflicted limb.
Regarding the anatomical area of the cases, the legs were more frequently affected, in 60% of cases, compared with the arms (39%).Only 1% of cases involved the head or trunk.This finding is in agreement with other national studies [14][15][16] .
Dry bites are relatively more common (7.5:1) in the arms than in the legs (p < 0.001).It appears that cases involving the arms, between juvenile:adult snakes, occur in a 3:1 ratio and in the legs, in a 2:1 ratio.However, when evaluating all cases of dry bite involving the use of a tourniquet, it appears that they are used more frequently (1:2 ratio) on the legs than on the arms (1:3 ratio) and are responsible for edema in such cases.
A significant difference was verified between severity and a time interval greater than 6 hours between the bite and hospital admission.Patients arriving later tend to present more severe reactions (p = 0.009).These results are consistent with official data 1 in which the lethality of Bothrops sensu lato accidents treated in less than six hours was 0.5%, while in those treated after 12 hours lethality was 1.5% 3 .
A significant association was determined between gingival bleeding and abnormal blood coagulability.A study conducted in the HVB between 1989 and 1991, involving 100 patients bitten by Bothropoides jararaca, observed that patients with systemic bleeding upon hospital admission presented thrombocytopenia 31 .In contrast, patients with local bleeding had identical platelet counts compared to patients without bleeding.The authors concluded that thrombocytopenia at admission is associated with the development of systemic bleeding and the severity of the accident 31 .
Necrosis was more common in the digits of the feet and hands (4.8%) compared to other body regions (1.8%).According to Ribeiro 26 , necrosis is associated with the bite site, occurring more frequently in the fingers due to a high concentration of venom in regions with minor circulation.We believe that the likelihood of developing compartment syndrome of the legs is higher than in other regions of the body.
In accidents caused by adult snakes, necrosis was more frequent (7.2%) compared to accidents caused by juvenile snakes (1%), since larger snakes can inoculate larger amounts of venom 8 .It may also be related to the difference in dietary habits discussed previously 6 .
The identification of 35 cases where the presence of pre-hospital adjuncts at the site of the bite were verified (including tourniquet, incision and substance use), suggests that the degree of knowledge