Snakebite is mostly a disease of rural farmers with the highest incidence recorded in Southeast Asia and Sub-Saharan Africa [1,21] and tends to afflict the economically active age group [22–25]. The nature of their occupation brings farmers, herdsmen and hunters [7,21] close to the natural habitat of these snakes. Most of the snakebites occur on the lower limb when the snake is accidentally trod upon by farmers without adequate protective footwear during farming activities [1,7]. We found that most of our snakebite victims were below 30 years of age and were bitten whilst engaged in farming activities. Late presentation to hospital characterizes snakebite cases in Africa [5,6]. Initial reliance on ineffective traditional remedies accounts partly for this late presentation. We found that at least a third of snakebite victims resorted to the used of traditional herbal remedies prior to presentation to hospital.

Identification of the snake species responsible for the bite allows likely course of the envenoming and potential complications to be anticipated and treated. This is usually difficult except when the dead snake is brought along to hospital. Envenomation by the carpet viper or saw-scaled viper (Echis ocellatus) results in a painful swelling of the bitten site which may result in tissue damage and necrosis [14]. The venom causes a consumptive coagulopathy as well as causing direct damage to the vascular endothelial basement membrane which may lead to a life-threatening hemorrhage [14]. Almost all (95.3%) our snakebite victims presented with local swelling and/or coagulopathy which is consistent with the envenomation pattern of the carpet viper although a viper was identified in only 11.5% of the cases.

The World Health Organization’s (WHO) guidelines on snakebite and its management in Africa [1] are probably the most comprehensive guidelines available on the subject in this Region and forms the basis for most local and institutional guidelines. Currently, a large number of individual approaches are employed by doctors in the case management of snakebites in TTH and other facilities in Ghana [19]. A study in India and Pakistan, two countries with high snake bite mortalities, showed doctors were ill-prepared to handle snakebite cases [26]. Non-reliance on local protocols results in a wide variation in the management principles employed in areas such as the assessment of the severity of envenomation, dosage of antivenin used and the use of ancillary medications. The implementation and adherence to a written protocol together with staff training and patient sensitization have resulted in a significant reduction in snakebite mortality in a rural hospital in Ghana [19]. These interventions resulted in improved health-seeking behavior by shortening the average delay before presentation and ensuring an adequate initial antivenom dose whilst limiting its excessive use [18,19]. Non-beneficial or potentially harmful interventions are avoided by relying on evidenced-based guidelines whilst allowing a more rational assessment of envenomation with clear endpoints to antivenom therapy.

The most important decision to be made concerning any patient presenting with a snakebite is whether or not to give antivenom, the only specific antidote for snakebite [1]. Antivenoms may be monovalent (against a single snake venom) or polyvalent (against venoms of several species of snakes). The antivenom used during the study period is a polyvalent lyophilized equine antivenom manufactured in India and supplied to the hospital courtesy the Ministry of Health of Ghana. It is encouraging that antivenom was available to most of the patients who needed them. The rate of use of antivenom and the number of vials administered in this study was higher than that reported by another study in Ghana [27]. This disparity may be related to the relative efficacy of the antivenoms administered or the fact that antivenom was freely available to snakebite victims in this study.

The use of antivenoms is not without risks. Reactions ranging from mild to serious/life threatening can result from their use and are particularly common with the polyvalent preparations. Studies on premedication and adverse drug reactions have produced mixed results [28,29]. Interventions that have shown some efficacy in reducing the frequency and severity of anaphylactic reactions include premedication with intravenous hydrocortisone, antihistamines, subcutaneous adrenaline and administration of a diluted antivenom by slow IV infusion over 60 minutes [28]. In the absence of compelling evidence of benefit, prophylaxis against allergic reactions may only be considered in high risk individuals, that is, those with previous severe reactions or have severe atopy [1,7]. Rigors and fever shortly after the infusion/injection was the main adverse reaction to antivenom reported in this study. Almost two-thirds (62%) of patients received steroids (mainly as intravenous hydrocortisone). It is not clear whether this was administered as prophylaxis against allergic reactions to antivenom or as part of the treatment. The widespread administration of steroids observed in this study is consistent with that found in an earlier study in two district hospitals in Ghana [27]. Steroids per se are of no value in the management of snakebites and may interfere with the venom-antivenom reaction [30].

Tranexamic acid was used in some of our patients (22.9%) to control coagulopathy. There is little data on the use of antifibrinolytics in snakebite management. However, as in consumptive coagulopathies due to other causes, the use of antifibrinolytics (without heparin) in envenomated patients with coagulopathies may worsen outcome through their prothrombotic effects by causing microvascular thrombosis. Hence, the use of antifibrinolytics and heparin is generally discouraged in envenomated patients [1,7].

The use of prophylactic antibiotics in cases of snakebite is not appropriate [31–33] unless there are definite signs of infection or the wound has been grossly interfered with (such as incised with an unsterile instrument) or is necrotic. Most local effects of snakebite are attributable directly to cytolytic activities of the venom itself (rather than bacterial infection-mediated) so antibiotic treatment can be delayed in many cases. In one study, there was no difference in the length of hospitalization for patients with mild to moderate swelling (without necrosis or abscess) between those who were given antibiotics and those who were not [32]. All but one of the snakebite victims in this study (99.5%) received antibiotics. More than half of them (51%) received at least two different antibiotics whilst 11.5% received at least three different antibiotics. Studies in other parts of Africa have demonstrated a similarly high rate of routine prophylactic antibiotic usage [34].

A hospital-based study of snake bite always runs the risk of leaving out a significantly large group of patients who do not present to hospital for various reasons. A community-based approach would have helped overcome this limitation. Being a retrospective study, we were limited by the information documented in the patient folders, and this was compounded by the fact that no formal or written protocols were followed. For example, it was not clear on many occasions why steroids were administered, whether in anticipation of a drug reaction or as part of the management of the inflammation. A bigger study to explore the factors associated with adverse outcomes following snakebite in this environment may form the bases for an improved protocol for snakebite management in the country.