History

A 4-year-old 125-kg (276-lb) Miniature Shetland pony gelding was examined because of a 5-week history of progressive hind limb ataxia. At the time of referral, the ataxia was described as severe, with the pony frequently stumbling and occasionally falling. Results of a complete physical examination, including neurologic examination, were normal apart from severe (grade 4/5) hind limb ataxia. The pony had a history of chronic, bilateral, upward fixation of the patellae and had been intermittently treated with suxibuzone^a (an NSAID licensed in the United Kingdom for treatment of musculoskeletal conditions in horses) and stall confinement.

General anesthesia for myelography, with spinal CT if required, was planned. The pony was premedicated with acepromazine (0.02 mg/kg [0.009 mg/lb], IM) 90 minutes prior to induction of anesthesia. A 16-gauge indwelling IV catheter was placed in the right jugular vein. The patient then received flunixin meglumine (1.1 mg/kg [0.5 mg/lb], IV) and dexamethasone (0.04 mg/kg [0.018 mg/lb], IV), as recommended by Grant and Paterson.¹ Immediately prior to induction of anesthesia, romifidine (0.06 mg/kg [0.027 mg/lb], IV) was administered, which resulted in profound sedation. Because of the severity of ataxia, once the pony was heavily sedated it adopted a so-called dogsitting posture. General anesthesia was immediately induced by administration of ketamine (2.2 mg/kg [1.0 mg/lb], IV) and midazolam (0.04 mg/kg, IV); anesthetic induction was considered smooth.

Endotracheal intubation was performed by means of a blind technique with a 14.0-mm cuffed endotracheal tube, and the patient was connected to a small animal circle breathing system. Anesthesia was maintained with isoflurane at a mean end-tidal concentration of 1.1%. The isoflurane was administered in a mixture of oxygen and medical air, with the aim of maintaining an inspired oxygen fraction (Fio₂) of 60% because a low Fio₂ may reduce intrapulmonary shunting and decrease the risk of hypoventilation.² The pony breathed spontaneously throughout the anesthetic period, and mechanical ventilation was not required. Hartmann solution was administered throughout at a rate of 5 mL/kg/h (2.3 mL/lb/h), IV. Monitoring included clinical assessment by the attending anesthesiologist of depth of anesthesia (palpebral reflex, presence or absence of nystagmus, tone of the neck muscles, and anal tone) and use of a multiparameter monitor.^b Three-lead base-apex ECG, pulse oximetry (for measurement of oxygen saturation), capnography, arterial blood gas analyses, and noninvasive blood pressure measurement (Doppler ultrasonic flow detector with a size 5 cuff on the right metatarsal area) were used.

The myelogram was nondiagnostic; therefore, the pony underwent spinal CT during the same anesthetic episode. The patient was repositioned from lateral to dorsal recumbency for CT, and once CT was completed, the pony was allowed to recover in a padded stall. Xylazine (0.3 mg/kg [0.14 mg/lb], IV) was administered during recovery (total dose, 37.5 mg), which was hand assisted with one person holding the halter and another holding the tail. Time from the end of anesthesia to standing was 27 minutes. Recovery was judged to be smooth, and the degree of ataxia was no more severe than before anesthesia. Total duration of anesthesia, from administration of induction agents to cessation of isoflurane administration, was 180 minutes; of this, 50 minutes was spent in dorsal recumbency for the CT scan.

Computed tomography revealed an intradural, extramedullary lesion at the level of T11, with neoplasia as the most likely differential diagnosis. Thus, the pony was scheduled to undergo a dorsal hemilaminectomy on the right side at the level of T11 to decompress the spinal cord³; the procedure was to be performed 24 hours later.

Question

What are the considerations for the anesthesia care plan for a patient such as this pony, scheduled for dorsal hemilaminectomy under general anesthesia?

Answer

Considerations for optimal anesthesia care for any equid undergoing major surgery performed under general anesthesia must include consideration of the patient's preexisting history and status. The plan should be tailored to provide appropriate management with minimal risk of perioperative complications, including hypotension, hypoventilation, hypoxemia, bradyarrhythmias, musculoskeletal injury, and upper airway obstruction.⁴ In addition to these usual considerations, a major consideration in the patient described here was the need to perform dorsal hemilaminectomy with the pony positioned in sternal recumbency. This is an unusual position for an equid undergoing general anesthesia. As such, an important additional consideration that was discussed during preanesthetic planning included limb positioning to minimize the risk that postanesthetic myopathy or neuropathy would develop.

The anesthetic regimen included premedication with acepromazine (0.02 mg/kg, IM) and morphine (0.1 mg/kg [0.045 mg/lb], IM). A 16-gauge indwelling IV catheter was placed in the right jugular vein, and penicillin G procaine (22,000 U/kg [10,000 U/lb], IM), gentamicin (6.6 mg/kg [3.0 mg/lb], IM), and flunxin meglumine (1.1 mg/kg, IV) were administered. Once the pony was sedated, immediately prior to induction of general anesthesia, romifidine (0.06 mg/kg, IV) was administered. Ketamine (2.2 mg/kg, IV) and midazolam (0.04 mg/kg [0.018 mg/lb], IV) were used for anesthetic induction. After induction of anesthesia, the trachea was intubated with a 14.0-mm cuffed endotracheal tube.

After induction, the patient was positioned in sternal recumbency on a padded surgical table^c with an inflatable air mattress. The forelimbs were flexed at the carpi with the hind limbs extended caudally. With the mattress inflated, the head and neck were maintained in a natural position.

Once positioned in sternal recumbency, the patient was connected to a small animal circle breathing system, and isoflurane in an oxygen-air mixture was administered with a goal of maintaining an Fio₂ of 60%. Mean end-tidal isoflurane concentration throughout was 1.2%. The pony also received constant rate infusions of morphine⁵ (0.1 mg/kg/h, IV) and ketamine (30 μg/kg/min [13.6 μg/lb/min], IV), with the ketamine infusion rate decreased to 15 μg/kg/min (6.8 μg/lb/min) after 150 minutes of general anesthesia in an attempt to avoid excessive accumulation of the drug. The patient was allowed to breathe spontaneously throughout with a plan to provide positive-pressure ventilation should the fractional concentration of CO₂ in expired gas exceed 70 mm Hg. Because this did not occur, positive-pressure ventilation was not required at any time during anesthesia.

The patient was monitored throughout, including assessment of the previously reported reflexes. A multiparameter monitor^b was used for 3-lead base-apex ECG, pulse oximetry, and capnography. Blood pressure was measured directly via a 21-gauge butterfly cannula placed in the right auricular artery and connected to a strain gauge pressure transducer. Dobutamine (1 μg/kg/min [0.45 μg/lb/min], IV, titrated to effect) was administered when mean arterial pressure was 80 mm Hg. The total dobutamine dose administered between minutes 55 and 70 of anesthesia was 15 μg/kg. Arterial blood gas analyses were performed 60 and 220 minutes after anesthetic induction to assess ventilation and oxygenation status (Table 1). Hartmann solution was administered at a rate of 5 mL/kg/h, IV, throughout the anesthetic period.

One hundred sixty minutes after induction of anesthesia (90 minutes after the start of surgery), acepromazine (5 μg/kg, IV) was administered in an attempt to optimize muscle perfusion, by inducing mild peripheral vasodilation while maintaining cardiac output and adequate mean arterial blood pressure,⁶ and reduce the risk of postanesthetic myopathy. Prior to administration of acepromazine, the mean arterial pressure was 85 mm Hg, and the heart rate was 35 beats/min. No changes in measured cardiovascular parameters were noted after administration of acepromazine.

Ketamine and morphine infusions were ceased 215 minutes after anesthetic induction, prior to the cessation of inhalation anesthesia, to reduce their potential negative influence on recovery quality. After 240 minutes of anesthesia, surgery was completed, administration of isoflurane and the oxygen–medical air mixture was discontinued, and the patient was moved to a padded recovery stall. Recovery was hand assisted and was judged to be of good quality. Time from cessation of inhalant anesthetic administration to standing was 30 minutes. No further medication was administered in the recovery period.

The pony's neurologic status following recovery from anesthesia was similar to its status before surgery, but clinical status slightly worsened over the following 2 days before gradual improvement was seen. This pattern was considered consistent with postoperative inflammation. Postoperative analgesia consisted of flunixin (1.1 mg/kg, IV, q 24 h), morphine (0.1 mg/kg, IM, q 4 h), and dexamethasone (0.1 mg/kg, IV, q 24 h). Surgical and postanesthetic management have been reported elsewhere.³ Briefly, a diagnosis of lymphoma was made and chemotherapy was instigated. The pony was discharged 8 days after surgery. At 16 months after surgery, no adverse reactions to the chemotherapy were reported, and the pony was noted to have only mild ataxia.

Discussion

A 2002 study⁷ reported that general anesthesia in horses may have a high rate of complications, compared with the complication rate for general anesthesia in small animals, with a mortality rate of 0.9% for nonabdominal emergency surgical cases. That study found approximately a third of deaths occurred as a result of fracture or myopathy, with fractures most likely occurring during the recovery period.⁸ As for any equine patient with severe ataxia, we considered the risk of injury or death as a complication of general anesthesia and recovery to be higher for our patient than for patients undergoing routine elective surgery.

The dorsal hemilaminectomy procedure required that the pony be positioned in sternal recumbency for optimal surgical site access. A literature search did not reveal any other reports of such positioning in horses. Spinal surgery in horses including cervical dorsal laminectomy for treatment of cervical vertebral malformation has been reported, but with the patient positioned in lateral recumbency.⁸

One of our biggest concerns was how best to position the limbs so as to minimize the risk of postanesthetic myopathy. We cannot completely exclude the possibility that postanesthetic myopathy occurred, as muscle enzyme activities were not evaluated and the ataxia initially worsened postoperatively. However, this was considered to be consistent with postoperative inflammation. No other evidence of myopathy was present; the muscles were soft with no signs of pain on palpation, and there was no hematuria or evidence of patient distress.

It is well accepted that poor perfusion pressure to the muscles⁹ and poor positioning¹⁰ resulting in high intracompartmental muscle pressure can contribute to the development of postanesthetic myopathy. It is for this reason that we chose a higher cutoff (80 mm Hg) than typical (70 mm Hg) for the lowest acceptable mean arterial pressure that would prompt administration of dobutamine. Current evidence indicates that maintenance of cardiac output, rather than blood pressure, ensures adequate tissue perfusion. However, routine cardiac output monitoring is not available at our institution; therefore, the surrogate measure of blood pressure was used. Nevertheless, a previous study⁹ has shown that increasing blood pressure by increasing systemic vascular resistance adversely affects muscle blood flow and cardiac function. For this reason, we hoped that the administration of acepromazine, with its vasodilatory properties, partway through the anesthetic period would help to optimize muscle perfusion. An additional factor that probably contributed to the lack of overt complications, including evidence of myopathy, was the small size of the pony (body weight, 125 kg [276 lb]) versus a typical 500-kg (1,100-lb) adult horse.

When planning for the surgical procedure, we thought that the standard positioning used for a dog undergoing a similar surgical procedure was not appropriate. This would have included maintenance of sternal recumbency with flexing of all hind limb joints, with the hind limbs positioned under the body and the forelimbs extended and positioned on each side of the head.¹¹ Such hind limb positioning increases the curvature of the spine and allows for easier surgical access. Although it may have increased compartmental pressure on the triceps brachii and biceps brachii muscles, we thought that positioning the pony with the forelimbs flexed at the carpi was the most natural position. With regard to the hind limbs, we were concerned that adopting a position similar to that used for dogs would have compressed much of the larger bulk of the upper limb musculature, thus potentially reducing muscle perfusion. Pulling the hind limbs cranially, to lay on either side of the abdomen, was thought to put strain on the muscles on the caudal aspect of the limb, such as the gluteal, biceps femoris, and gastrocnemius muscles. It was considered that in horses positioned in dorsal recumbency, extension of the limbs can stretch the femoral nerve and create pressure in the gluteal muscles.¹⁰ However, with the patient in sternal recumbency, we thought that this position would likely provide the least pressure on the largest muscle masses of the hind limb, although the risk of femoral nerve damage still existed. We had also been advised that similar patient positioning had been used for laminectomy in a foal previously^d with no apparent complications. Although not ideal, this was the patient position chosen.

The pulmonary ventilation-perfusion ratio may vary from 0, when there is perfusion but no ventilation, to infinity, when there is ventilation but no perfusion. An ideal ventilation-perfusion ratio is 0.8 throughout the lungs. It has long been known that in horses undergoing general anesthesia, positioning can have a large effect on the ventilation-perfusion ratio. As early as 1974, it was demonstrated that horses in dorsal recumbency had a lower Pao₂ than did those in lateral recumbency.¹² In dorsally recumbent horses, it has been demonstrated that the functional residual capacity of the lungs is less than the closing capacity.¹³ This results in alveoli that are continually closing and reopening with expiration and inspiration, respectively, which can predispose to greater areas of atelectasis and therefore shunt formation. This effect is magnified with the administration of isoflurane.¹⁴ Additionally, in recumbent horses, there are large areas of atelectasis,^15,16 and the use of multiple inert gas elimination techniques has shown that this atelectasis results in extensive shunt formation, resulting in large areas of the lungs that are perfused but not ventilated.¹⁷ The phenomenon is worse in patients positioned in dorsal versus lateral recumbency.

Previous studies^13,18 suggest that ventilation distribution may be more uniform with decreased shunt formation (as evidenced by a higher Pao₂) when the horse is in sternal, versus lateral, recumbency. Additionally, it has been demonstrated in anesthetized horses that although provision of a low Fio₂ results in decreased Pao₂, there is a reduced intrapulmonary shunting, compared with the degree of shunting when the Fio₂ is > 95%.² This is presumable because provision of oxygen-rich gas causes collapse of intermittently closed alveoli owing to rapid oxygen absorption (absorption atelectasis). Therefore, we hypothesized that in this pony, administration of an oxygen-air mixture would improve the distribution of ventilation, resulting in less chance of shunt formation and poor oxygenation. At both times arterial blood gas analyses were performed during surgery, the Pao₂ was > 500 mm Hg with an Fio₂ of 77% and evidence of mild hypoventilation. Use of the alveolar gas equation, with atmospheric pressure at sea level assumed to be 760 mm Hg, partial pressure of water vapor assumed to be 47 mm Hg, and the respiratory quotient assumed to be 0.9,¹⁶ would suggest that the alveolar partial pressure of oxygen was between 471 and 483 mm Hg. Although it is unusual for Pao₂ to be greater than alveolar partial pressure of oxygen, this discrepancy may be explained by the normal working limitations for accuracy of the blood gas analysis machine and the device used to measure Fio₂. Alternatively, Henry's law states that the partial pressure of a gas is proportional to its concentration at a given temperature and pressure. As the temperature decreases, the solubility of oxygen and carbon dioxide in the blood increases.¹⁹ The pony was mildly hypothermic (35.8°C [96.4°F]) at the time samples were collected for arterial blood gas analyses, and given that results of arterial blood gas analyses were not corrected for patient body temperature, it is possible that the measured Pao₂ was slightly greater than the actual value.¹⁹

The present report described a successful outcome for a pony undergoing general anesthesia and thoracolumbar dorsal laminectomy while positioned in sternal recumbency. We have highlighted some interesting and important considerations regarding anesthetic case management. Results of arterial blood gas analyses for this patient supported experimental data suggesting that horses in sternal recumbency have improved oxygenation, presumably because of reduced alveolar shunt formation. Although we saw no evidence of myopathy in this patient, we suggest that such positioning of any equine patient for surgery is still likely to be associated with a high risk of postoperative myopathy, although the risk was probably mitigated by the small size of this patient.