Recovering horses after general anaesthesia

Anaesthesia-related mortality has not improved in the last 50 years. In 1969, a mortality rate of 1.5% relating to equine anaesthesia was reported in a descriptive case series from the Royal (Dick) School of Veterinary Studies in Edinburgh (Mitchell, 1969). In 2002, the confidential enquiry into perioperative equine fatalities (CEPEF 1-2) reported an overall anaesthetic-related mortality of 1.9% (Johnston et al, 2002). In 2020, two large single-centre retrospective cross-sectional studies reported a general anaesthesia-related global mortality rate of 1.4% (Laurenza et al, 2020; Nicolaisen et al, 2020). Interestingly, further appraisal of these studies suggests that the recovery period has become the most crucial stage of anaesthesia as the incidence of fatalities during this period has increased, while fatalities during induction or maintenance of anaesthesia have decreased. In early reports, only 14.3% of anaesthetic-related mortalities occurred during recovery from anaesthesia (Mitchell, 1969), whereas later studies reported an incidence of 32.6–52.3% (Johnston et al, 2002; Bidwell et al, 2007). Furthermore, the most recent studies suggested that between 81.0% and 92.0% of anaesthesia-related complications occur during the recovery period (Laurenza et al, 2020; Nicolaisen et al, 2020). It is established that recovery is a crucial phase of anaesthesia but is also the most difficult phase to control (Hubbell, 2005).

The most common anaesthetic-related complications reported during the recovery period include:

• Fractures and dislocations (Johnston et al, 2002; Laurenza et al, 2020)
• Post-anaesthetic myopathy (Duke et al, 2006; Laurenza et al, 2020)
• Cardiopulmonary arrest or collapse (Johnston et al, 2002; Laurenza et al, 2020)
• Post-anaesthetic lameness (Richey et al, 1990)
• Upper respiratory airway obstruction and pulmonary oedema (Senior, 2005)
• Post-anaesthetic myelopathy (Ragle et al, 2011)
• Compartment syndrome (Louro et al, 2020).

While these events may occur or become apparent in the recovery box, contributing factors occurring well before the arrival of the horse in the recovery box may also have a significant impact on the outcome. Peri-anaesthetic management decisions made throughout the whole period, from pre-induction preparation right through to positioning of the horse in recovery, have the potential to influence the success of recovery.

In the perfect recovery scenario, the horse is placed in the recovery area, awakens from general anaesthesia and rises calmly to its feet without appreciable stress or ataxia (Hubbell, 2005). Achieving safe recovery from anaesthesia is likely influenced by a complex interaction of factors and events, some of which are outside of the veterinary surgeon's control, beginning in the very first stages of preparation of the horse for anaesthesia.

Pre-anaesthetic preparation

Considering the signalment of the horse is important in order to identify any aspects which may affect anaesthetic risk. Studies have identified the following to be associated with higher risks during recovery from anaesthesia:

• Increasing age (Dugdale et al, 2016)
• Breed (Louro et al, 2020)
• More invasive surgical procedures (Young and Taylor, 1993; Laurenza et al, 2020)
• Fracture repair (Johnston et al, 2002)
• An increased American Society of Anesthesiologists (ASA) physical status score (Johnston et al, 2002; Laurenza et al, 2020).

Increased anaesthetic durations also negatively influence recovery quality (Young and Taylor, 1993), so measures should be undertaken to avoid any unnecessary intraoperative delays. Preliminary surgical site preparation and verification that equipment is functioning correctly can be undertaken before induction of general anaesthesia to optimise preparatory efficiency once the horse is anaesthetised.

Implementation of surgical safety checklists in human medicine has long been established and is known to reduce perioperative mortality and morbidity (Haynes et al, 2009; Bergs et al, 2014). The Association of Veterinary Anaesthetists (AVA) designed a checklist with several important key objectives including outlining key anaesthetic procedures, optimising team work and communication and reinforcing recognised safe practices (Mair, 2020). While there is no doubt that using the checklist improves patient safety, it may also reduce intra-operative delays as it prompts the user to check that equipment is functioning and necessary images (such as radiographs) are accessible before the induction of general anaesthesia.

For horses, the safety checklist can be extended to facilitate ‘readiness’ in recovery for complications such as apnoea, respiratory obstruction or rapid emergence from general anaesthesia.

Pre-anaesthetic medication

Acepromazine is a phenothiazine tranquiliser widely used in equine anaesthesia. It produces mild sedation and has a calming effect in fractious horses, without producing significant ataxia (Knych et al, 2018; Morgan et al, 2020). Acepromazine has a delayed onset (20 minutes) and long duration of action (6 hours) when administered intravenously (IV) (Hashem and Keller, 1993). The authors use a dose of 0.02–0.03 mg/kg intramuscularly (IM) administered 1 hour before the induction of anaesthesia in healthy adult horses. Administration of acepromazine was associated with reduced anaesthetic-related mortality in a large retrospective study, although statistical significance was not quite reached in the multivariate model (Johnston et al, 2002). In this study, halothane was the most commonly used volatile agent and since halothane may sensitise the myocardium to catecholamines, the protective effects of acepromazine may be attributable to its anxiolytic action and reduction in arrhythmogenic circulating catecholamines (Dugdale and Taylor, 2016). While isoflurane and sevoflurane do not possess the same myocardial sensitising properties, acepromazine may continue to offer haemodynamic and oxygenation advantages when added to alpha 2 agonist and opioid premedication combinations (Marntell et al, 2005). Furthermore, the long-lasting sedative properties of acepromazine may contribute to a smoother transition into the recovery period (Driessen et al, 2011), although another study failed to identify a benefit in terms of recovery quality and duration when acepromazine was added to romifidine during pre-anaesthetic medication, before induction of anaesthesia with ketamine (Marntell and Nyman, 1996).

Analgesia

Invasive surgical procedures are known to adversely affect equine recovery quality, which is likely to be reflected with greater post-operative pain associated with some surgical procedures (Young and Taylor, 1993). Pre-emptive multimodal analgesia uses different classes of systemically administered analgesics, as well as locoregional techniques. Horses that are in pain during recovery are less likely to rest quietly in the recovery box and will attempt to stand earlier, making signs of ataxia more evident (Clark et al, 2008). Therefore, it is crucial to optimise analgesia throughout the perioperative period (including the recovery period), to encourage safe recovery. This statement is supported by a variety of studies suggesting that the systemic administration of morphine (0.1 to 0.2 mg/kg IV) improved recovery quality in horses (Mircica et al, 2003; Love et al, 2006; Clark et al, 2008). Furthermore, epidural administration of morphine positively influenced recovery quality in both experimental (Natalini et al, 2007) and clinical cases (Louro et al, 2020).

Temperament

During the preparation for general anaesthesia and administration of pre-anaesthetic sedation, an assessment of the horse's temperament can be made. A horse's temperament can influence recovery quality, with horses described as nervous, anxious, aggressive or unhandled having poorer recovery qualities than those described as calm in one prospective study by Leece et al (2008). During recovery from field anaesthesia, another study reported that all four of the horses which required intervention during recovery were considered to be excitable or dominant/aggressive and responded poorly to sedation. However, recovery conditions could not be controlled in this study on account of its clinical nature (Harðardóttir et al, 2019).

Intraoperative monitoring

Intraoperative hypotension

Hypotension may play a role in the development of myopathy and post-anaesthetic lameness (Richey et al, 1990). Post-anaesthetic myopathy, a form of compartment syndrome and ischaemic muscle damage, is associated with poor padding, poor positioning, long anaesthetic durations and hypotension (Dugdale and Taylor, 2016). Maintaining mean arterial pressure >70 mmHg may help reduce the severity of myopathy (Duke et al, 2006). Post-anaesthetic myopathy may hinder a horse's attempts to rise and may result in repeated unsuccessful attempts to stand, leading to trauma and exhaustion. It is also possible that sub-clinical myopathies go unrecognised, resulting in prolonged recumbency in recovery because of pain, paresis and lack of functioning muscle. This reinforces the importance of maintaining an appropriate mean arterial pressure during general anaesthesia in horses (Voulgaris and Hofmeister, 2009). It is also important to optimise padding and take care with positioning in order to avoid the development of pressure points or uneven distribution of load, which may result in nerve compression and neuropathy.

Intra-operative hypoxaemia

Anaesthetic agents generally depress respiratory function (Hubbell and Muir, 2015) and large differences in alveolar and arterial oxygen tension P(A–a)O₂ develop in dorsal or lateral recumbency (Hall, 1971). Hypoventilation, diffusion impairment, shunt and ventilation/perfusion inequality can all contribute to the development of large differences in P(A-a)O₂ during anaesthesia (West, 2008). Impairment of pulmonary function during equine anaesthesia may result in low arterial partial pressure of oxygen (PaO₂) (Auckburally and Nyman, 2017) and intraoperative hypoxaemia (PaO²< 60 mmHg) may subsequently have detrimental effects on recovery quality (Rüegg et al, 2016). Monitoring arterial oxygenation intra-operatively enables treatment to be implemented accordingly. Intra-operative measures which may improve PaO₂ include aerosolised salbutamol (Robertson and Bailey, 2002), controlled mechanical ventilation (Bardell et al, 2020) and other ventilatory strategies such as positive end expiratory pressure and recruitment manoeuvre (Hopster et al, 2011). Intraoperative hypoxaemia can persist into the recovery period (Trim and Wan, 1990) and Bardell et al (2020) demonstrated that PaO₂ did not return to pre-anaesthesia baseline values 1 hour after recovery from inhalational anaesthesia. Hypoxaemia in recovery can be difficult to treat once the horse starts to move. Therefore, avoiding the development of intraoperative hypoxaemia may help prevent this occurrence.

Electrolyte abnormalities

Electrolyte imbalances exist in the majority of horses presenting for colic surgery, with hypocalcaemia and hypokalaemia being the most common intraoperative imbalances (Adami et al, 2020). Muscle weakness can be associated with hypocalcaemia or hypokalaemia (Borer and Corley, 2006), which may hinder a horse's ability to rise and stand during recovery. Measurement and appropriate supplementation of electrolytes should be carried out to address preoperative deficits.

Maintenance of general anaesthesia

Inhalational agents

Sevoflurane has a lower blood:gas partition coefficient compared to isoflurane, which may translate to faster recovery times. However, in horses undergoing advanced imaging, Leece et al (2008) found no differences in recovery time or quality in horses recovering from inhalational anaesthesia with isoflurane or sevoflurane.

Intravenous anaesthetic agents

Total intravenous anaesthesia (TIVA) remains the easiest method of providing general anaesthesia in the field setting or for short (<60 minutes) hospital procedures (White, 2015). TIVA is generally associated with less cardiopulmonary depression compared to inhalational agent anaesthesia, although conclusions regarding recovery quality must be made with caution since anaesthetic duration tends to be considerably shorter for procedures for which TIVA is selected. In fact, over longer periods of infusion (1–1.5 hours), the cumulative effects of injectable agents used in TIVA methods can result in poor recoveries (White, 2015). In the UK, TIVA protocols usually consist of ketamine, an α-2 adrenoceptor agonist and guaifenesin. Recently, as a result of the lack of availability of guaifenesin in some countries, the use of midazolam in TIVA protocols has been investigated as a substitute. However, despite being licensed for co-induction with ketamine, midazolam remains unlicensed in the UK for this purpose. Good-to-excellent recovery qualities have been reported after ketamine-xylazine-midazolam IV maintenance combinations (Hopster et al, 2014), with horses in one study standing after an average of one attempt with little-to-moderate effort and only slight body shifting observed (Hubbell et al, 2012).

Partial intravenous anaesthesia

Partial intravenous anaesthesia (PIVA) aims to reduce inhalational agent requirements, providing additional analgesia while having minimal cardiorespiratory impact (White, 2015). An international survey of anaesthetists showed that practise was fairly evenly split between those that routinely used continuous rate infusions (CRIs) during inhalational anaesthesia and those that did not routinely use them or only used them for selected cases (Wohlfender et al, 2015).

Injectable agents most commonly used in PIVA protocols include the α-2 adrenoceptor agonists, lidocaine and ketamine. Studies have demonstrated that intraoperative CRIs of xylazine (Pöppel et al, 2015), romifidine (Kuhn et al, 2004), medetomidine (Neges et al, 2003), ketamine (Pöppel et al, 2015) or lidocaine (Dzikiti et al, 2003) have an isoflurane-sparing effect. Some studies have suggested that horses with a lower end-tidal isoflurane concentration at the end of anaesthesia have a faster time to standing (Voulgaris and Hofmeister, 2009). This may be the result of a shorter period of brain anaesthetic levels within the ‘ataxic’ range allowing horses to stand successfully earlier (Young and Taylor, 1993). However, while PIVA may reduce inhalational agent requirements, total isoflurane administered throughout the anaes-thetic and overall anaesthetic duration remains relevant in terms of recovery time (Eger and Johnson, 1987). Furthermore, pharmacodynamic effects on inhalational agent elimination and individual agent properties may influence the effect of PIVA on recovery.

Lidocaine

Intravenous lidocaine is known to exert an isoflurane sparing effect (Dzikiti et al, 2003), as well as having analgesic (Murrell et al, 2005) and anti-inflammatory effects (Cook et al, 2009). However, its intraoperative use has been associated with ataxia in recovery so it is important to cease lidocaine at least 30 minutes before recovery (Valverde et al, 2005). Intraoperative CRI of a combination of both lidocaine and ketamine had a significant isoflurane sparing effect (Enderle et al 2008; Villalba et al, 2011) and reduced dobutamine requirements (Enderle et al, 2008). Villalba et al (2011) commented that recoveries were of worse quality, with more attempts to stand in horses receiving lidocaine and ketamine compared to the isoflurane alone. However, differences in recovery scores did not reach statistical significance in either study.

Ketamine

Ketamine provides effective analgesia but when ketamine was infused IV intraoperatively for several hours, muscle rigidity and involuntary limb movements were observed during the recovery period (Muir and Sams, 1992). Recoveries were faster but of poorer quality in horses receiving S-ketamine intraoperative infusions, compared to medetomidine in isoflurane-anaesthetised horses (Larenza Menzies et al, 2016). Emergence from ketamine anaes-thesia may depend mainly on rapid redistribution of the drug from the central compartment, rather than elimination, which may explain the abrupt emergence observed in horses (Waterman et al, 1987). Furthermore, additional intraoperative ‘top up’ boluses of ketamine exceeding 2 mg/kg or thiopentone exceeding 4 mg/kg administered in the event of a sudden decrease in anaes-thetic depth had a negative impact on recovery quality (Louro et al, 2020). During field anaesthesia for castration, horses receiving ketamine induction doses of 2.2 mg/kg IV had subjectively better recoveries compared to those receiving a higher dose (5 mg/kg ketamine IV) (Harðardóttir et al, 2019).

Alpha 2 adrenoreceptor agonists

α-2 adrenoreceptor agonists are a very popular choice for PIVA protocols. α-2 agonists offer analgesia and sedation which may positively influence recovery quality.

Intra-operative romifidine CRIs resulted in more horses standing after the first attempt with no ataxia, compared to those receiving saline CRI (Devisscher et al, 2010). In a prospective clinical trial, recovery score was not influenced by the use of intraoperative detomidine CRI compared to saline in isoflurane-anaesthetised horses. However, all horses received detomidine in recovery which may have made subtle differences harder to detect (Schauvliege et al, 2011). Medetomidine and the active enantiomer dexmedetomidine are not licensed for use in horses in the UK but as a result of their high α-2 selectivity (Virtanen et al, 1988) and pharmacodynamic suitability for CRI, their use in PIVA has been widely investigated. Intra-operative medetomidine CRIs have been associated with significantly better recovery scores than S-ketamine CRIs (Larenza Menzies et al, 2016) or lidocaine CRIs (Ringer et al, 2007) in isoflurane-anaesthetised horses. Dexmedetomidine intraoperative CRI resulted in significantly better recovery quality with minimal cardiovascular impact when compared to saline (Marcilla et al, 2012) or medetomidine CRI (Sacks et al, 2017) in isoflurane-anaesthetised horses.

The recovery period

Recovery from general anaesthesia describes the period of time between discontinuation of the maintenance agent, either a volatile anaesthetic agent or a combination of agents administered IV, and the time point by which the horse has recovered full control over posture and can safely be moved back to the stable (Clark-Price, 2013). The recovery period may also be extended, if desired, up to when the patient is eating and drinking or even up to 7 days post-anaesthesia.

Generally speaking, six phases can be observed during recovery from anaesthesia (Clark-Price, 2013):

• An initial period of transition from the operating table to recovery
• Time of first movement
• Attaining sternal recumbency
• First attempt to stand
• Initial stance
• Firm stance.

The anaesthetist should monitor recovery constantly in order to respond promptly to any complications that may occur during this period. Rapid recognition of such complications and early intervention is key to a successful outcome. It is absolutely crucial to ensure personal safety at all times and intervention under unsafe conditions is discouraged by the authors.

Recovery environment

In horses recovering from field anaesthesia, Harðardóttir et al (2019) reported that horses recovering with a companion present had improved attitude to recovery, which may reflect the horse's herd instinct.

In the hospital setting, the design of a recovery box varies immensely from hospital to hospital. The size and headroom are often dependent on the type of horses that present to the hospital. For example, a hospital where the majority of horses presented for surgery are Warmbloods of considerable size may require a more spacious recovery box with higher headroom than a hospital that serves a flat racing population. The height of suspended horses can be estimated from measurements taken from the standing animal, which can help when planning the construction of equine surgical facilities (Clutton et al, 2010). The recovery box can be shaped as a square, rectangle, octagon or be round, and should be designed to prevent a horse from accelerating and increasing its momentum, preventing violent impacts with the walls. Adequate floor padding and slip-resistant soft flooring is essential to provide a secure grip and minimise injuries during recovery from general anaesthesia. The doors of a recovery box should be wide to allow a conscious and anaesthetised patient to pass through without difficulty and should also be easily accessible to enable the removal of dead or euthanised patients. These doors should be sturdy and secured with door bars and bolts to prevent opening when a horse clashes against them. Small windows, platforms or video camera systems are generally installed for remote viewing of the inside of the recovery box. Personal safety should always be the main concern when recovering horses. The authors strongly discourage entering a recovery box in any situation where personal safety cannot be guaranteed. New assisted methods of recovery (such as head and tail rope recovery) have been developed, but these systems have not been tested in terms of safety to personnel and horses and should be used with caution.

Administration of sedatives to improve recovery quality

Volatile anaesthetic agent washout occurs when the agent molecules present within active sites in the central nervous system are either eliminated or redistributed to other tissues (Yasui et al, 2007). At low partial pressures of a volatile anaesthetic agent, immobility may not be achieved, but the remaining agent present in the locus coeruleus may still be sufficient to produce emergence agitation (Yasui et al, 2007). During the recovery period, horses regain awareness and responsiveness to their surroundings, but this is not accompanied by efficient neurologic and motor function because of the remaining volatile anaesthetic agent present in the central nervous system (Brosnan et al, 2012).

A number of research articles have focused on the impact of the administration of sedatives, before recovery from anaesthesia, on recovery quality (Matthews et al, 1998; Santos et al, 2003; Woodhouse et al, 2013; Guedes et al, 2020; Hector et al, 2020). In an international survey regarding anaesthetic practice, 55.8% of anaes-thetists administered an α-2 adrenoreceptor agonist bolus before recovery from anaesthesia while 24.5% of responders did so only in certain cases (Wohlfender et al, 2015). The popularity of using α-2 adrenoreceptor agonists close to recovery may be in response to studies showing an association with improved quality of recovery (Matthews et al, 1998; Santos et al, 2003). These agents produce reliable sedation and provide analgesia (England and Clarke, 1996) and are frequently administered as a bolus before recovery from general anaesthesia, with the objective of improving recovery quality (Gozalo-Marcilla et al, 2015). The main objective is to provide adequate sedation to allow time for further elimination of volatile anaesthetic agent. If more volatile anaesthetic agent is eliminated during this period, the adverse effects on the central nervous system may be reduced, thereby reducing muscle weakness, incoordination, shivering and excitement; promoting an improved recovery quality. Selection of one agent over another may be based on individual clinician preference and experience. Currently, xylazine, detomidine and romifidine are licensed for use in horses in the UK. Medetomidine and dexmedetomidine are not currently licensed for use in horses, although a number of studies investigating their use in horses are available (Bryant et al, 1991; Bettschart-Wolfensberger et al, 1999; Guedes et al, 2020; Hector et al, 2020). Romifidine (20 μg/kg IV) produced better results in terms of recovery quality when compared to xylazine (200 μg/kg IV) (Woodhouse et al, 2013) or detomidine (2.5 μg/kg IV) (Alonso et al, 2020). More recently, dexmedetomidine (0.875–1 μg/kg IV) was found to result in equivalent recovery time and quality scores compared to romifidine (20 μg/kg IV) (Hector et al, 2020) and xylazine (200 μg/kg) (Guedes et al, 2020), when administered before recovery from anaesthesia. However, in large scale epidemiological studies, no statistically significant association was detected between recovery quality and attempts to keep horses recumbent for longer using α-2 adrenoceptor agonists administered close to recovery from anaesthesia (Young and Taylor, 1993; Dugdale et al, 2016).

Aside from the aforementioned desirable effects, the administration of α-2 adrenoreceptor agonists can produce significant cardiovascular and respiratory side effects such as dose-dependent bradycardia, decreases in cardiac output and increases in systemic vascular resistance, arrhythmias, respiratory depression, transient decreases in PaO2 and ataxia (Yamashita et al, 2000; Santos et al, 2003). Administration of α-2 adrenoreceptor agonists increases urine output in anaesthetised horses as a consequence of hyperglycaemia caused by hypoinsulinaemia (Steffey and Pascoe, 2002) and a reduction in arginine vasopressin secretion (Alexander and Irvine, 2000). When α-2 adrenoreceptor agonists are administered in anaesthetised patients, bladder catheterisation is advisable and the bladder should be emptied before recovery. In the authors' experience, patients with distended bladders may feel uncomfortable and are more likely to try to stand too soon in an attempt to urinate.

Oxygen supplementation

Hypoxaemia in recovery may not be recognised and it is hard to prevent and challenging to treat (Auckburally and Nyman, 2017). In recovery, measures which may be effective in improving arterial oxygen tension include oxygen supplementation via a demand valve (Mason et al, 1987) and nasotracheal or nasopharyngeal insufflation (10-15 l/minute) of oxygen (McMurphy and Cribb, 1989; Bardell et al, 2020). A patient-activated demand valve can be positioned at the proximal end of the endotracheal tube, which delivers oxygen during spontaneous inspiration. In addition, the anaesthetist may choose to deliver a controlled breath in apnoeic patients using the same device. Once a regular breathing pattern is established, the trachea can be extubated and oxygen supplementation can be administered using nasal insufflation via a narrow bore tubing attached to an oxygen flowmeter. The authors recommend an oxygen flow rate of 10-15 l/minute for an average 500 kg horse. Some anaesthetists prefer to leave an endotracheal tube in place during recovery to guard against airway obstruction (Clark-Price, 2013). While oxygen supplementation cannot necessarily prevent arterial hypoxaemia, there is evidence that it can significantly reduce the incidence (Bardell et al, 2020). Additionally, oxygen requirements increase dramatically during stormy recoveries and horses may become hypoxic, exhausted and unable to stand as a result. In humans, hypoxic events can result in loss of coordination, blurred vision, weakness and dizziness, possibly as a result of to mild brain injury (Turner et al, 2015). Therefore, altered cognition in horses recovering from general anaesthesia can have a detrimental impact on recovery quality (Auckburally and Nyman, 2017).

Positioning in recovery

For horses positioned on the surgical table in lateral recumbency, the patient should be positioned in recovery in the same recumbency and not turned onto the contralateral side. In ponies in lateral recumbency, the lung volume (McDonell et al, 1979) and blood flow (Stolk, 1982) is lower in the dependent lung. Therefore, changing recumbency in anaesthetised horses is likely to result in a decrease in PaO2 and widened alveolar–arterial oxygen gradient caused by atelectasis and ventilation–perfusion mismatch (Niyom et al, 2018). If the horse is positioned in dorsal recumbency on the surgery table, the authors often position the horse with the catheter site upper-most which normally corresponds to the side of the surgical site (for example, in orthopaedic procedures the authors prefer to position the operated limb uppermost to improve recovery).

Upper airway obstruction

In horses, accumulation of secretions, surgically-induced haemor-rhage and swelling, recumbency-related nasal mucosal oedema, perioperative trauma to the larynx, left or bilateral laryngeal paralysis and dorsal displacement of the soft palate have been reported as potential causes of upper airway obstruction in recovery from general anaesthesia (Senior, 2005). In order to prevent such complications, the authors often elevate the poll to facilitate drainage of secretions and place a nasopharyngeal tube with the aim of maintaining airway patency. Administration of intranasal phenylephrine reduces mucosal oedema (Lukasik et al, 1997) and the use of nasotracheal or orotracheal tubes during recovery has also been suggested. In higher risk cases, or when complications arise, tracheostomy can be performed to alleviate or prevent upper airway obstruction (Senior, 2005; Ronaldson et al, 2020).

Assisted recovery methods

Horses can recover freely or using an assisted method. This decision is generally based on several factors, such as temperament and health status of the patient, comorbidities, size of the patient, type and duration of surgery, clinician preference, staff experience, facilities, equipment and resources available.

Various assisted recovery methods have been described and include hand assistance, darkened recovery boxes, sling systems, inflated air pillow, tilt table, recovery pool, and head and tail rope assistance. Most of these articles suggested that assisted methods contribute to safer recoveries (Sullivan et al, 2002; Tidwell et al, 2002; Taylor et al, 2005; Elmas et al, 2007; Clark-Price et al, 2008; Clark-Price, 2013; Niimura Del Barrio et al, 2018). However, in the authors' opinion, this is extremely difficult to gauge because of the absence of control groups in the majority of these studies, as well as the large diversity in animal populations, surgical procedures, anaesthetic management protocols, recovery score systems and recovery methods.

Recovering horses in a darkened box did not improve quality of recovery and time to standing was similar to those recovering in an illuminated box (Clark-Price et al, 2008). A single centre, randomised, prospective, clinical trial comparing air-pillow assisted and free recovery methods showed no statistically significant difference in terms of occurrence of recovery-associated complications or recovery quality (Ray-Miller et al, 2006).

It is the authors' opinion that minimising noise and avoiding hypothermia and shivering in recovery by maintaining an adequate room temperature, removing excess sweat and drying the patients could contribute to good recovery quality from general anaesthesia.

Rope-assisted recovery

The use of assisted recovery methods has been increasing in popularity among anaesthetists. In an international survey of equine anaesthetic practice, 40% of respondents stated that they routinely use assisted recovery techniques on a daily basis, while 31.0% used assistance in selected cases and 28.5% did not use an assisted recovery teachnique (Wohlfender et al, 2015).

Head and tail rope assistance has become the most commonly used method of assisted recovery in horses. This technique aims to provide stability and direction to a horse during attempts to stand. One rope attaches to the halter and aims to control the direction of the horse's attempts to rise while the second rope, attached to the tail, aims to assist and stabilise the horse during rising (Castillo and Matthews, 2005; Wilderjans, 2008; Arndt et al, 2020). Failures of equipment and/or technique have been reported for head and tail rope-assisted recovery, including loose head collar, tail hair breakage, tail knot slippage, facial paralysis and even a fracture of the first coccygeal vertebra, with associated neurological dysfunction of the rectum, anus, tail and surrounding skin (Niimura Del Barrio et al, 2018; Bird et al, 2019).

The majority of studies present conflicting evidence regarding the value of head and tail rope assisted recovery in reducing fatal complications and improving recovery quality from general anaesthesia (Rüegg et al, 2016; Arndt et al, 2020; Laurenza et al, 2020; Nicolaisen et al, 2020). Nicolaisen et al (2020) demonstrated a reduction in fatal complications during recovery of 289 horses undergoing emergency exploratory laparotomies, when head and tail rope-assisted recovery was used. However, this study did not establish the same effect for horses recovering from elective surgery. The remaining studies failed to show a statistically significant reduction of recovery-associated fatalities when head and tail rope recovery was used (Rüegg et al, 2016; Arndt et al, 2020; Laurenza et al, 2020). The infrequent occurrence of recovery-associated fatalities means that larger studies may be required to detect an advantage of assisted vs unassisted recoveries.

In terms of minor complications and technical difficulties, mixed evidence is given by the different studies. Two studies could not detect a statistically significant effect of the recovery method used on the incidence of recovery-associated non-fatal complications observed (Laurenza et al, 2020; Nicolaisen et al, 2020). Others, despite showing a similar incidence of minor complications, reported a higher frequency of technical problems relating to the equipment resulting from head–tail rope-assisted recovery (Rüegg et al, 2016). Finally, a single centre, randomised, prospective, clinical trial, reported a higher incidence of skin abrasions and post-anaesthetic lameness in horses recovering unassisted, but similar incidence of technical problems in both recovery groups (Arndt et al, 2020).

In terms of recovery quality, no differences were found between head and tail rope assisted recovery and unassisted recovery methods (Rüegg et al, 2016). In contrast, other authors suggested that horses recovered using a head and tail rope method had fewer attempts to stand, shorter recovery times and better recovery quality scores, compared to horses recovering unassisted (Arndt et al, 2020).

The authors conclude that scientific evidence for the unequivocal benefit of head and tail rope-assisted recovery is currently lacking. In the authors' opinion, head and tail rope assistance may improve recovery quality in horses after general anaesthesia, but experienced personnel and correctly functioning equipment are vital components of the system.

Conclusions

The interplay of many factors during the preoperative, intraoperative and recovery period is likely to influence recovery quality and ultimately success. The complexity of surgical procedures, the equine population and the choice of anaesthetic and analgesic techniques are continually evolving, as is our understanding of their risk factors. Preservation of human safety must be paramount in endeavours to continue to investigate and improve methods of safe recovery of horses after general anaesthesia.

KEY POINTS

• The recovery period is a crucial stage of equine anaesthesia and is the least controllable phase.
• Complications which may occur in recovery include fracture and dislocation, post-anaesthetic myopathy, cardiopulmonary arrest or collapse, post-anaesthetic lameness, upper respiratory airway obstruction and pulmonary oedema, post-anaesthetic myelopathy and compartment syndrome.
• Provision of multimodal analgesia and balanced anaesthetic technique can improve recovery quality in horses.
• Rope-assisted recovery techniques are increasingly used in equine anaesthesia and are an area of active research.
• Awareness of factors which influence recovery quality in horses can help tailor anaesthetic technique and optimise patient safety.