While envenomation by numerous snake species has been associated with respiratory failure in cats and dogs, only snakes in the Elapidae and Viperidae families have caused documented respiratory paralysis (11).

Elapid snakes, belonging to the Elapidae family, are characterized by a pair of rostrally positioned fangs (1, 12). Venom constituents and effects vary between species (12). Neurotoxin induced respiratory paralysis results in hypoventilation requiring mechanical ventilation in severe cases. Envenomation can pose a significant threat to life should treatment be delayed (13).

Elapids are distributed globally but the majority of species are located within North America, South-eastern Asia and Australia. Australian elapids have been reported as some of the most venomous species on earth (14). Those of significance include the Brown snakes (Pseudonaja spp.), Tiger snakes (Notechis spp., Austrelaps spp.), Black snakes (Pseudechis spp.), Taipans (Oxyuranus spp.), and the death adder (Acanthophis spp.) (15). Mechanical ventilation was most commonly required for hypoventilation in Australian dogs and cats envenomated by brown snake (19%, 14/59), tiger snake (11%, 5/46), and black snake (11%, 1/9) immunotypes in a retrospective multicenter study (7). Hypoventilation requiring MV has been documented in one case of Pseudechis prophyriacus envenomation (16). Hypoxemia secondary to pulmonary hemorrhage is a more common indication for MV in envenomation by this species (4, 7).

Kraits (Bungarus spp.), Cobras (Naja spp.), and the King Cobra (Ophiophagus hannah) are Elapidae that inhabit Asia. No studies in the veterinary literature describe the use of MV for envenomation by these species. One case report exists of two children suffering Bungarus caeruleus and Naja naja envenomation, requiring 10 and 11 days of invasive MV, respectively (17). Another human study reported 51% (17/33) of patients required MV for B. caeruleus envenomation, with all patients alive at the 9 month follow up after a median of 96 h of ventilation (18).

Eastern coral snakes (Micrurus fulvius) are Elapidae found in North America. Envenomation by Micrurus sp. has been documented to cause respiratory paralysis in cats and dogs necessitating mechanical ventilation, with a median duration of MV 58 h (25–84 h), median duration of hospitalization of 8.2 days (6–11 days) and survival rate of 87.5% (7/8) (19).

Within North America, envenomation is more commonly reported from members of the Viperidae family—Rattlesnakes (Crotalus spp.), and the Copperheads (Agkistrodon spp.) (20). Rattlesnakes account for most mortalities in this area but the severity of neurological signs varies between species (6). Positive pressure ventilation (PPV) was indicated or provided in between 1 and 11.8% (1/82 and 4/34) rattlesnake envenomation's in reviews of dogs with Crotalus spp. envenomation (6, 21). MV for Agkistrodon contortrix envenomation has not been documented in dogs or cats (22).

There are several recognized toxin types, which can be divided into neurotoxins, coagulants (both pro- and anti-coagulant), myotoxins, hemolysins, and cytotoxins (23, 24). As neurotoxins are the primary cause for respiratory paralysis, they shall be the focus of this article. Additional consideration should, however, be given to coagulants and myotoxins for their potential contributory role in the need for MV, as well as their influence in disease progression and outcome.

Two main types of neurotoxins inhibit synaptic transmission at the neuromuscular junction (NMJ), phospholipase A2 (PLA2) enzymes and three-finger alpha-neurotoxins. PLA2 enzymes irreversibly bind and hydrolyze the membrane phospholipids at the motor nerve terminal. This results in a calcium influx and subsequent release and depletion of acetylcholine, leading to degeneration of the nerve terminal (25). These include ammodytoxin A (Vipera ammodytes), β-bungarotoxin (Bungarus multicinctus), taipoxin (Oxyuranus scutellatus), and textilotoxin (Pseudonaja textilis) (26). Eastern coral snake venom contains PLA2 enzymes and a type of alpha-neurotoxin, a post-synaptic acetylcholine receptor antagonist, resulting in generalized muscle weakness and respiratory paralysis (2, 19). Alpha-neurotoxins, such as α-Bungarotoxin, bind and inhibit post-synaptic acetylcholine receptors at the NMJ (25, 27).

PLA2 enzymes also play a role in myotoxicity, causing either local swelling at the bite site, rhabdomyolysis, or both (28). Enzymes such as notexin, found in Notechis scutatus venom, cause extensive skeletal muscle damage, including muscles responsible for respiration (26). Regeneration of peripheral nerves begins 3–4 days following envenomation, with almost complete resolution of damage by 10–14 days (25). Rhabdomyolysis has been associated with the development of acute kidney injury (AKI) which may alter the patient's management and prognosis (25, 27, 29).

Coagulation disorders may cause respiratory failure by pulmonary thromboembolism or pulmonary hemorrhage (5). Elapids with procoagulant toxin resulting in consumption of clotting factors (known as venom-induced consumptive coagulopathy) include Pseudonaja spp., N. scutatus, and Oxyuranus spp. (30). Anticoagulant toxin present in Pseudechis spp., O. hannah, and Naja spp. inhibits factors X, II and platelet function (30–32).

Interference at the NMJ leads to progressive lower motor neuron signs. Clinical signs can range from generalized paresis to paralysis and death from respiratory failure (19). Localized swelling of the bite site can occur with certain snake species, and hemorrhage may be witnessed if the venom contains hemolytic toxins (16, 33). In Australian elapids, bite sites are often not found due to small sized fangs and minimal localized swelling in combination with thick hair coats of cats and dogs (34). Other common clinical signs include vomiting, hypersalivation, diarrhea, tachypnea, collapse, mydriasis, ataxia, and pigmenturia (4, 19, 33, 35).

Clinical signs typically are acute in onset (often within 1 h) but can be biphasic or delayed. Retrospective studies report the majority of envenomated patients are presented for veterinary assessment within 6 h (dogs) to 3 days (cats) (33, 36). The incidence of dogs and cats that are asymptomatic on presentation is unknown. Humans asymptomatic for known snakebites are monitored for 12–24 h in hospital, should testing be inconclusive (37, 38). One study reported development of clinical signs up to 14 h after Micrurus fulvius envenomation and a previous controlled experimental study showed development of paralysis within 24 h following subcutaneous administration of Pseudonaja sp. venom in cats (35, 39).

Antivenom is a crucial part of treatment for any patient with clinical evidence of snake envenomation. Antivenom is produced by injecting venom or components into a host (most commonly horses) and extracting the newly formed IgG antibodies from the plasma (40, 41). IgG antibodies bind to the active unbound venom antigen, forming masses of immune complexes that are readily phagocytized, inhibiting attachment to receptor sites (42). Commercially available antivenom products are summarized in Table 1.

Antivenom varies by geographical location. Correct identification of individual snake species can be difficult, yet fortunately is less crucial with the development of polyvalent antivenoms. In Australia, snake venom detection kits are utilized to aid in selection of the most appropriate antivenom therapy.¹ Specificity of this ELISA (Enzyme-linked Immunosorbent Assay) has been reported as 100%, however false negatives from acute envenomation, absence of circulating venom and the “hook effect,” whereby binding of antibodies in the presence of a high venom concentration inhibits sandwich formation, should be considered (33, 43, 44). This immunoassay can be performed patient side with a rapid result (5).

The ideal dose of antivenom is unknown. Research into Pseudonaja textilis envenomation identified neutralization of venom levels following administration of 4,000 units each of an experimental IgG tiger—brown snake antivenom (5). A previous study used between 3,000 and 6,000 units for N. scutatis envenomation in cats and dogs (13). A study of cats undergoing MV for elapid envenomation in Australia reported ~42% (5/12) of cats received one vial, 42% (5/12) received two vials and 16% (2/12) of cats received 3 vials of antivenom, containing 3,000 units of N. scutatis antivenom with 1,000 units of P. textilis antivenom or 3,000 units of N. scutatis with 4,000 units of P. textilis antivenom (45). This study did not evaluate outcomes between the two different antivenom treatment groups, however 11 of the 12 patients survived to discharge. Another study investigating elapid envenomation in Melbourne, Australia administered a median of 8,100 units of tiger/brown polyvalent antivenom to dogs and a median of 4,050 units to cats (9). Most cats and dogs with Micrurus sp. snakebite receive 5–10 mg of IgG equine-derived antivenom (2, 19). A mean of five vials of a polyvalent antivenom was reported in one canine study in Sri Lanka, with each vial capable of neutralizing 6 mg N. naja, 4.5 mg B. caeruleus, 6 mg D. russelii, and 4.5 mg E. carinatus venom (46). In humans, current guidelines support giving one vial (1,000 units) in both children and adults for most elapid species (37).

There is no evidence to suggest additional vials are of benefit to justify the increased risk associated with repeated antivenom administration (37). For all other elapid species, following human guidelines, one vial of appropriate antivenom should be provided with additional vials given if clinical deterioration is observed or if a repeated snake venom detection kit is positive (9, 37). Additional antivenom may be required in the event of multiple bites, however the number of bites may be irrelevant as dry bites can occur without release of venom (47). A maximum number of antivenom vials cannot be suggested based on available evidence, with further investigation into antivenom dosages required.

In a study of dogs and cats envenomated by P. porphyriacus, ventilated patients did not require more antivenom compared to non-ventilated patients (4). There was no significant difference between number of vials of antivenom and length of hospitalization or survival in another study of dogs and cats undergoing MV for rattlesnake envenomation (6). Antivenom is most effective when administered as soon as possible to reduce the need for mechanical ventilation given its inability to neutralize bound pre-synaptic toxin (48). Post-synaptic neurotoxins will be neutralized following antivenom administration, therefore a cat or dog presenting with snake envenomation in which mechanical ventilation is indicated will not be expected to immediately improve with antivenom administration and mechanical ventilation should proceed (45).

Antivenom should be diluted prior to administration (49). The rate of administration in humans is not associated with an increase in adverse effects (50). With any blood product, initial administration should be commenced slowly with the rate titrated, however in the face of lethal envenomation, antivenom has been reported to be given in as little as 5 min (10). Multiple studies report administration either rapidly (administered as a bolus) or over a 20 min period (5, 6, 10, 19). Rapid administration is anecdotally tolerated at the authors' institution, but no veterinary literature exists evaluating the incidence of transfusion reactions. There is no evidence to support premedication of patients with antihistamines or glucocorticoids prior to administration of antivenom (3, 51, 52).

There are currently four recognized indications for mechanical ventilation: (1) severe hypoxemia defined as partial pressure arterial oxygen (PaO₂) <60 mmHg despite oxygen therapy (fraction of inspired oxygen (FiO₂) >50%); (2) severe hypoventilation defined as partial pressure arterial carbon dioxide (PaCO₂) >60 mmHg; (3) unsustainable respiratory effort, and (4) hemodynamic compromise unresponsive to traditional therapy (53–55).

Hypoventilation and respiratory fatigue are the more common indications for MV in the envenomated patient. Hypoxemia is less frequently reported but can occur due to severe pulmonary hemorrhage or aspiration pneumonia (10, 45, 56–58).

Of the studies published describing MV in elapid envenomation, few outline the ventilator settings used. Compared to other patients with lower motor neuron diseases requiring MV, cats and dogs with elapid envenomation generally lack primary pulmonary disease and require short ventilation periods and minimally aggressive settings (45, 58).

One retrospective study of eight cats and dogs with M. fulvius envenomation performed volume-controlled positive pressure ventilation using synchronized intermittent mandatory ventilation (SIMV), with initial pressure support of 8 cm H₂O and positive end expiratory pressure (PEEP) 3 cm H₂O. This study reported a median duration of reliance on mandatory breaths of 15 h (8–24 h), with a median duration of ventilation of 58 h (25–84 h) and 7/8 patients were successfully weaned (19). In a study of 12 cats, the authors followed a pre-established protocol using volume-controlled positive pressure ventilation with a respiratory rate of 20 breaths/min, tidal volume of 10 ml/kg, PEEP of 3–5 cm H₂O and a spontaneous breathing trial performed every 12 h (45). Eleven of the 12 cats were successfully weaned, with a median duration on MV of 19.5 h (7–37 h). In a case report, one dog with P. porphyriacus envenomation was managed using pressure-controlled ventilation and was weaned after 18 h, however the breath pattern was not reported (16).

Invasive ventilation facilitates airway protection so is preferred to non-invasive ventilation in cats and dogs with generalized neuromuscular disease. Dynamic positive airway pressure systems such as high flow nasal oxygen therapy are currently not considered appropriate if the patient's primary cause of respiratory failure is hypoventilation (59).

No veterinary literature compares pressure vs. volume controlled mandatory modes of ventilation. It has been suggested that pressure-controlled ventilation may reduce the work of breathing and therefore improve patient comfort and patient-ventilator synchrony (60). Mode of MV is uncommonly documented in the human literature with specific reference to snake envenomation. Successful ventilation with survival to discharge is documented in case reports using SIMV and mandatory pressure-controlled ventilation for Bungarus sp. envenomation (61, 62). Volume-controlled mandatory ventilation was compared to adaptive support ventilation in a prospective randomized trial of 48 adults with neuroparalytic snake envenomation; mode did not appear to significantly affect outcome with respect to time to weaning, incidence of VAP or length of hospital stay (63). Median duration of MV was 17 h in a prospective study of adult envenomated humans (64). Of these, 60% (8/14) patients required MV for <24 h and 13/14 (93%) survived to discharge (64). Mode of MV was not specified in this study. Further research is required in both human and veterinary medicine. At the authors' institution, invasive ventilation is most commonly performed in mandatory pressure-controlled mode with anecdotal success.

Patients should be initially commenced on FiO₂ 100% at the start of MV. Oxygen concentration is titrated down incrementally if adequate arterial oxygenation can be maintained, as assessed by pulse oximetry or preferably arterial blood gas analysis (65). To reduce the risk of oxygen toxicity and minimize absorption atelectasis, this process should commence as soon as the patient is stable on the ventilator. Targeting the lowest possible FiO₂ concentration to maintain an SpO₂ or SaO₂ at >95% (66).

Lung protective ventilation involving reduced inspiratory pressures (<30 cmH₂O), lower tidal volumes (6–8 ml/kg), higher PEEP and presence of permissive hypercapnia have been associated with reduced mortality in people with acute lung injury and acute respiratory distress syndrome (ARDS) (67, 68). This approach should be considered for envenomated cats and dogs with pulmonary disease such as hemorrhage or pneumonia (69). For cats and dogs without pulmonary disease, initial recommended setting includes tidal volume 8–12 ml/kg, respiratory rate 10–20 breaths per minute, inspiratory pressures of 10–20 cm H₂O and inspiratory to expiratory ratio 1:2 s. Low levels of PEEP (3–5 cm H₂O) are described to avoid atelectasis (70). In all cats and dogs receiving MV, ventilator settings should be adjusted to the least aggressive settings required to achieve normotension and normoxemia (69).

Provision of a balanced diet at an appropriate proportion of RER based on individual nutritional assessment is beneficial for patients undergoing MV and should be considered for snake envenomated cats and dogs after 72 h of hospitalization (90).

There are no published guidelines on re-alimentation in mechanically ventilated veterinary patients (91–96), and the frequency of nutrition provision in mechanically ventilated cats and dogs remains poor (90). Veterinary patients without appropriate nutritional support are at risk of starvation. Starvation may prolong mechanical ventilation in the envenomated patient due to impaired respiratory muscle function from respiratory muscle catabolism, reduced surfactant production, decreased immune response and hypoalbuminemia that may exacerbate pulmonary edema (55, 91). Excessive caloric intake increases carbon dioxide production which may affect MV settings (91).

Enteral nutrition (EN) is preferred as it is associated with decreased mortality and duration of hospitalization and more ventilator-free days in humans (92, 97). Physiological benefits include maintenance of gastrointestinal integrity as well as modulation of stress and the immune response (92–95, 98). Nasogastric/nasoesophageal feeding tubes are appropriate as envenomated patients generally have a short disease course without ongoing gastrointestinal dysfunction.

Trophic feeding is preferred to total caloric provision, with one study in humans reporting a reduced incidence of gastrointestinal intolerance to EN but no overall difference in ventilator free days or mortality (92, 96). Patients who are likely to tolerate EN should be given a balanced, digestible commercially available veterinary diet (99). One study in hospitalized dogs requiring nasogastric or nasoesophageal feeding described the use of veterinary (Clinicare² and Clinicare RF³) and human (Ensure⁴) liquid enteral diets with no significant difference in outcome (90, 92, 100, 101). Other diets have been described in two dogs receiving EN at 25% RER (Royal Canin Convalescence Support Instant Diet⁵ and Enteral Care KC⁶) (90). At the authors' institution, patients undergoing MV are initially fed at 25% RER and increased in 25% increments according to the patients' tolerance to feeding.

Parenteral nutrition (PN) can be provided as adjunctive nutrition for patients that may not tolerate EN (such as those with severe ileus) or require supplementation to an EN feeding plan (100). Administration requires strict aseptic technique, intensive monitoring, and a specific catheter to reduce the risk of complications (100, 101). PN has not been evaluated in mechanically ventilated veterinary patients but based on human literature, it should not be used before or in preference to EN (90, 92, 93).

Gastrointestinal dysmotility is a reported side effect of snake envenomation, which may be worsened by shock or mechanical ventilation (102). Megaesophagus has been documented in a case report of four dogs with tiger snake (Notechis scutatus) envenomation (56).

MV can cause relaxation of the lower esophageal sphincter and the development of gastroesophageal reflux as a consequence of intrathoracic pressure changes and anesthetic agents used (103). Prokinetic therapy is indicated if gastric ileus is identified ultrasonographically or if gastric residual volumes exceed 10 ml/kg (93, 103). Metoclopramide 2 mg/kg/d CRI is an appropriate first line prokinetic in dogs (103). Proton pump inhibitors are indicated only when there is evidence of gastrointestinal ulceration or esophagitis to reduce the risk of VAP development, given gastrointestinal signs are generally brief and resolve with antivenom and anti-emetics (33, 104).

Mechanically ventilated cats and dogs require continuous care and are at risk of complications including ocular and oral ulceration, decubital ulcers and aspiration of gastric fluid (71). Suboptimal patient care is associated with increased mortality in humans (105, 106).

Envenomated cats and dogs may be at greater risk of ocular ulceration due to tear desiccation, the inability to blink and reduced tear production in the critically ill (107). One study describing MV in Australia reported corneal ulceration in 58% (7/12) of envenomated cats, and a study of M. fulvius envenomation in dogs and a cat reported corneal ulceration in 25% (2/8) (19, 45). Other studies in dogs and cats describing MV for mixed causes report the incidence of corneal ulceration between 5 and 10% (69, 108). Regular sterile saline lavage, daily fluorescein stains, and administration of lubricating ointment every 2 h is recommended to reduce and prevent the risk of exposure keratitis (71). Polyethylene eye covers are utilized in human ICU settings, with one veterinary study reporting use of lenses in envenomated cats and dogs undergoing MV (5, 109). Further investigation of protective ocular equipment and their use in veterinary patients is required (110, 111). Development of blepharospasm, conjunctival hyperemia, mucoid discharge and episcleral congestion should prompt clinicians to investigate and escalate ocular care. Superficial corneal ulcers require treatment with a broad-spectrum antimicrobial ointment with close monitoring for progression.

Frequent oral care and airway management can reduce the likelihood of oral ulceration, lingual oedema, aspiration of gastric contents and bacterial colonization to lower risks of ventilator associated pneumonia (VAP). Complete oral care protocols are described elsewhere (71), but should be performed every 4 h, incorporating appropriate hand hygiene and use of sterile gloves, cleaning of the oral cavity and attachments with a 0.05% chlorhexidine solution and regular suctioning. Frequent deflation and repositioning of the endotracheal tube cuff has been recommended after suctioning to reduce the risk of pressure necrosis in cats and dogs, however is associated with increased risk of VAP in humans (112, 113). Ensuring cuff pressure is maintained at 25 cm H₂O through the use of a commercial endotracheal tube cuff inflation device is recommended to reduce the development of VAP and pressure necrosis (71, 112, 114). If a monitoring device is unavailable, the cuff should be deflated and repositioned to reduce the risk of tracheal necrosis (114).

Indwelling urinary catheterization allows close monitoring of urine production and increases patient comfort, reducing the potential for patient-ventilator asynchrony. Aseptic technique and a closed collection system facilitates gross assessment of urine, sampling for urinalysis and accurate urine output measurement. Appropriate catheter care is required to minimize urinary tract infection risk. Acute kidney injury secondary to envenomation is documented in several studies, thought to be a consequence of venom-induced hemolysis or rhabdomyolysis leading to pigmenturia (8, 19, 115, 116).

Prolonged patient recumbency can result in decubital ulcers, generalized muscle weakness and edema development, with ulcer severity associated with increased mortality in humans (114, 117). Patients undergoing long term ventilation (>48 h) are at a higher risk. Passive range of motion exercises and regular recumbency changes are recommended every 4 h, with padded bedding provided to maximize patient comfort.

Veterinary literature reports the use of isotonic crystalloids as fluid therapy for snake envenomated cats and dogs (8, 16, 19, 118), although many did not specify the type of fluid and the optimal rate is unclear. Intravenous fluid therapy should aim for neutral fluid balance, providing maintenance fluid requirements in combination with correcting dehydration, shock or ongoing losses where required. Human literature suggests a positive fluid balance significantly increases the risk of ventilator-associated adverse events (119).

Red blood cell containing products such as packed red blood cells (pRBC) or fresh whole blood (FWB) may be required in cases of severe hemolysis or hemorrhage. Packed red blood cell therapy was administered in 14/88 patients with P. porphyriacus envenomation that developed severe anemia with 57% survival (8/14) (4).

Fresh frozen plasma (FFP) therapy is controversial in snake envenomation. Snake envenomation causes hemorrhage due to venom induced consumptive coagulopathy (VICC) (120). Procoagulant toxins causing VICC can be neutralized by snake antivenom in in-vitro models (5, 10). However, VICC still occurs despite prompt antivenom treatment due to the rapid effects of these toxins. Human literature recommends prioritization of antivenom administration, with administration of FFP only in instances of ongoing bleeding (121). A small randomized clinical trial of humans administered FFP within 4 h of antivenom treatment failed to demonstrate improvement in clinical outcomes or time to resolution of clinical coagulopathy, although clotting factor dysfunction resolved more rapidly (121).

Veterinary literature has documented the administration of FFP for severe hemorrhage following antivenom administration. In 10 dogs with severe pulmonary hemorrhage secondary to P. textilis envenomation, 70% (7/10) were administered FFP. Of these, 43% (3/7) survived to discharge (10).