The current WHO recommendations for snake bite management state that even patients with local reactions and a prolonged prothrombin time without any other features of systemic envenomation  are candidates for anti-snake venom. We in Pokhara, Nepal see a lot of snake-bites which are semi-poisonous and cause only local cellulites and markedly prolonged prothrombin time with no other features of systemic envenomation. None of them are ever treated with antivenom serum and all have recovered without any sequelae. We initially tried treating them with anti venom to tackle the prolonged prothrombin time ( according to the current recommendations) but found that it remained prolonged even after 20 to 30 vials ( in some of the patients who could afford that many, one of them also developed transverse myelitis as a hypersensitivity reaction ). These days we have switched to a policy of treating the local wound with appropriate measures and leave the prothrombin time alone, which almost invariably comes back to normal after 3 to 4 days. Is it an example of evidence base medicine being relevant only to local disease patterns?

 ( Good evidence may not be universally/globally applicable?)