Non-venomous snakebites in the Western Brazilian Amazon

Introduction: In this study, we examined the clinical manifestations, laboratory evidence, and the circumstances of snakebites caused by non-venomous snakes, which were treated at the Regional Hospital of Juruá in Cruzeiro do Sul. Methods: Data were collected through patient interviews, identification of the species that were taken to the hospital, and the clinical manifestations. Results: Eight confirmed and four probable cases of non-venomous snakebites were recorded. Conclusions: The symptoms produced by the snakes Helicops angulatus and Philodryas viridissima, combined with their coloration can be confused with venomous snakes (Bothrops atrox and Bothrops bilineatus), thus resulting in incorrect bothropic snakebite diagnosis.

Snakes from the families Colubridae and Dipsadidae are traditionally classified as non-poisonous, despite having the Duvernoy's gland and the capacity for producing toxic secretions, which eventually cause envenomations 1 . The Duvernoy's gland, located in the supralabial region, posterior to the eyes, is considered equivalent and precursory to the venom glands in elapids and viperids 1 . In these two families, two types of fangs can occur: aglyphous, indicating huge fangs without grooves or inoculating channels and opisthoglyphous, indicating a pair of grooved fangs located at the rear of the mouth 1 . These types of fangs and the muscles associated with them are not efficient in the inoculation of venom when compared to elapids and viperids; however, there have been cases of fatalities from bites from some opisthoglyphous and some aglyphous species 1 .
In Brazil, 27,000 snakebite incidents are registered annually, with a lethality rate of 0.43% and, in the Amazon, the highest incidence of cases is recorded (56.1 per 100,000 inhabitants) 2 . Of these, bites by non-venomous snakes are also computed (Boidae, Colubridae, and Dipsadidae) which, depending on the region, such as in Cuiabá (MT) for example, can represent up to 56% of cases 3 . At the Vital Brasil Hospital, in the Instituto Butantan (SP), approximately 32% of the snakes that caused the snakebites (brought in by the patients or their companions) corresponded to non-venomous species 4 . In a hospital in Uberlândia (MG), 19% of the patients showed no signs of envenoming or were bitten by non-venomous snakes 5 . Also, it is possible that some snakebites caused by viperids (Bothrops and Crotalus) are considered "dry bites" when the victims do not present any clinical or laboratory signs of envenomation 5,6 ; therefore, it is not necessary to use serum therapy-only patient observation. This demonstrates the importance of health professionals evaluating the need for serum therapy in cases of snakebites in which patients do not show signs of envenoming, thus avoiding possible allergic reactions and also the waste of antivenom.
Acre is one of the states with the highest incidence of snakebites in Brazil, 61.10 per 100,000 inhabitants 2 and the Upper Juruá in the extreme west of the Brazilian Amazon is also characterized by a high incidence of snake bites 7 . We present here cases of bites caused by the snakes Colubridae and Dipsadidae which were attended to at the Regional Hospital of Juruá in Cruzeiro do Sul (AC).
The study was conducted between June 15, 2017 and June 15, 2018 at the Juruá Regional Hospital located in the municipality of Cruzeiro do Sul (Acre) (07° 62' S; 72° 67' W), an area which has approximately 87,673 inhabitants 8 and is located in the western Brazilian Amazon.
The clinical data regarding the manifestations from the envenoming of the victims and laboratory parameters were obtained from the observation of patients and medical records. When possible, the 20-minute blood coagulation test 9 was performed with a blood sample collected from the patients.
Patients were interviewed to find out the circumstances of the incident and whether they recognized the snake that caused the poisoning by showing them boards containing color photographs of the most common snake species in the region. The snakes were identified with the aid of a regional book on snakes 10  During the study period, 133 victims of snakebites were treated at the Regional Hospital of Juruá. Twelve (9%) of them (eight confirmed and four probable) were caused by snakes that are considered non-venomous ( Table 1). The snakes confirmed by their capturer or a photograph taken by the victims were four Helicops angulatus, two Philodryas viridissima, one Hydrops triangularis, and one Xenodon severus, all belonging to the Dipsadidae family (Table 1, Figure 1). In seven cases (58%), at the site of the bite the marks were abrasive lesions and five cases showed signs of puncture (Table 1; Figure 2). In two cases, three ampoules (for H. angulatus) and four ampoules (for O. fulgidus or suspected P. viridissima) of Bothrops antivenom were administered due to prolonged blood coagulation time.
The Helicops angulatus species represented 36% of the snakes that were taken to the hospital, a result similar to the study carried out in Rio Branco (Acre) 11 , where this snake also represented a significant percentage of the snakes brought to the hospital (58% of the specimens). All those who were affected were bitten in the evening between 19:00 and 19:40, since this snake is commonly active at night and at dusk 10 . Of those who were bitten by this snake, two had no symptoms and two cases reported mild symptoms with local pain. However, snakes of the genus Helicops present the Duvernoy's gland that can produce toxic substances, and are considered moderately venomous snakes, whose venom has proteolytic and neurotoxic components 12 . Two patients bitten by the H. angulatus snake showed changes in coagulation time, and coagulopathy was observed in cases involving other species of the genus 4 . In one of these cases, besides the change in clotting time, there was also a change in bleeding time and irreversible clot formation. Due to these characteristics of altering blood coagulation time, signs and symptoms of envenoming and the coloration pattern (similar to Bothrops atrox), cases of bites from these snakes may be confused with Bothrops envenoming 4,5,11 .  In this study, confirmed cases of Philodryas viridissima bites had a proportion of 19%, where victims were bitten during the day (in the morning), a time at which this snake is active 10 . The bite signs observed showed only one puncture, and also signs of various punctures, such as an abrasive lesion, which portrays characteristics of the type of opisthoglyphous dentition 1 . Patients who were bitten by this species had more evident envenoming symptoms and experienced vomiting, pain, and headache. This is a snake that, like Helicops spp., also has the Duvernoy's gland and may cause symptoms of envenoming 4 . The signs and symptoms presented confer with those described for the genus 13,14,15 , where the victims reported pain, bleeding, edema, erythema, and ecchymosis, but not headache and vomiting. There are few reports of snakebites by P. viridissima 14 ; the majority of bites in the genus is caused by P. olfersii and P. paragoniensis 13,15 . The headache and vomiting observed in all cases of P. viridissima bites in this study may be associated with specific venom characteristics of this species. In all confirmed cases, there was no change in clotting time, which is somewhat expected for species of this genus, as is also reported in other studies 13,15 .
In this study, four cases (33%) were considered probably caused by Oxybelis fulgidus or Philodryas viridissima, where the victims reported having been bitten by green-colored snakes whose size ranged from 70 to 100 cm, all of them being bitten during the day, in lower limbs and with abrasive signs or in circumstances and environments characteristic of the activity of these species. The time of the snakebites' occurrence coincides with the greater activity of these snakes, since they are diurnal 10 . In addition, the punctate and arcuate lesions left by the snake are characteristic of these species 13 . The symptoms that arose in the victims-pain, headache, and blurred vision-may be present in incidents caused by Colubridae and Dipsadidae snakes. In two of these cases, there was incoagulability in the WBTC 20 test, which is unexpected for Philodryas spp. 13,14,15 .
In the snakebite caused by Hydrops triangularis, the patient had mild local pain, headache, and nausea. The popular name given to the "surucucu-de-brejo" or "surucucu-traíra" snake may be related to the aquatic environments where it is found 10 . In this case, there was no change in coagulation time. The patient received no antivenom, though this had been prescribed before the researchers could confirm the causative species.

4/4
One case of snakebite involving Xenodon severus was recorded (12.5% of the total); this is an aglyphous snake with enlarged and well-adapted fangs for preying on Bufonidae toads 10 . The lesion of the bite presented four punctures, which is associated with the type of dentition of the animal. There was no change in coagulation time or other alterations in the laboratory data for this case.
In four cases of non-venomous snake incidents in this study, there was an unnecessary use of antivenom. The snakes Helicops angulatus (similar to Bothrops atrox) and Philodryas viridissima (green coloration) may be confused with venomous species of the region causing misdiagnosis. Thus, it is recommended that when the snake is not taken to the hospital, the medical team must take into account the lesions left by the snake 6 , together with the clinical and laboratory evaluation of the cases, monitoring evolution taking into account the particularities of the venom of non-venomous species (for example, change in coagulation time). Antivenom can reverse fractions of envenoming by snakes of the genus Philodryas, but its efficacy is highly discussed, and symptomatic treatment is recommended 15 .
Through this study regarding non-venomous snakes implicated in snakebites reported in the Upper Juruá region (AC), the circumstances of the snakebites and the clinical manifestations produced were described. In this regard, it was found that the frequency of bites caused by snakes which are considered to be non-venomous during the study period was 9%. The main non-venomous snake involved in snakebite incidents was Helicops angulatus, with a frequency of 50% of cases confirmed for this group. The possibility of an alteration of the coagulation time in patients bitten by H. angulatus, besides the pain and edema that may arise, may contribute to the misdiagnosis of the snakebite. In the cases of Philodryas viridissima bites, the symptoms were more intense when compared to the cases of H. angulatus bites, although no case presented alterations in coagulation time.
In some cases, the symptoms and description of green-colored snakes, were confused with Bothrops bilineatus smaragdinus or with Bothrops atrox, as occurred in cases of bites involving Helicops angulatus, may have influenced the diagnosis of a Bothrops snakebite, leading to the decision to use antivenom. For this reason, the dissemination of information through training, the need for a more accurate look at the diagnosis of the circumstances of the snakebite, the evidence of the injuries, and the clinical and laboratory manifestations that arise is recommended.
The act of bringing the dead snake that caused the snakebite, or even photographing it, is important in order to facilitate the diagnosis, and in the case of a non-venomous snake, it is also fundamental in view of the fact that it will avoid the unnecessary use of antivenom, as well as other medicines, but it does not exclude the need for observation.