Failure to Reverse Prolonged Vecuronium-Induced Neuromuscular Blockade with Edrophonium in an Anesthetized Dog

Introduction

Neuromuscular blocking agents (NMBAs) can be used as part of balanced anesthesia to permit profound muscle relaxation without the need for high doses of general anesthetics. Nondepolarizing NMBAs have an added advantage as their action can be terminated by specific pharmacologic antagonism. Unfortunately, in practice, complete antagonism of neuromuscular blockade is not always possible and may result in postoperative residual curarization (PORC). Among other things, PORC is responsible for skeletal muscle weakness, hypoventilation, laryngeal dysfunction, and a decline in respiratory function during recovery from anesthesia.^1,2

Routine use of peripheral nerve stimulators to assess neuromuscular function decreases the incidence of PORC.³ The use of objective monitors, such as acceleromyography (AMG), to quantify muscle activity evoked by peripheral nerve stimulation results in a further reduction in the incidence of PORC in humans and has been shown to improve monitoring of neuromuscular function compared with visual assessment of evoked twitches in horses.^4,5 Although AMG is relatively easy to use, care must be taken when positioning the accelerating-sensitive crystal on the distal limb if accurate values are to be obtained.

Surprisingly, monitoring using peripheral nerve stimulation is not routine in surgery rooms when humans are anesthetized.⁶ A similar scenario is likely in veterinary anesthesia. Presumably, anesthetists operating under these circumstances rely on either an accepted dose of an anticholinesterase producing complete antagonism of the neuromuscular blockade or the relatively predictable duration of action of modern NMBAs such as vecuronium or atracurium. This case report illustrates the risk associated with either of these two approaches.

Case Report

An 8 yr old male castrated Weimaraner dog weighing 40 kg presented to the emergency service at Cornell University Hospital for Animals with excessive incisional drainage and inappetance 4 days after abdominal surgery for foreign body removal. The patient was quiet and painful on physical examination; he was tachycardic (152 beats/minu), hyperthermic (40°C), and tachypneic (44 breaths/min). Pulses were bounding, and the mucous membranes looked hyperemic. A sample of peritoneal fluid was obtained and revealed the presence of degenerate neutrophils and bacteria. An intravenous catheter was placed and a blood sample obtained. Packed cell volume (PCV) was 43% (reference range, 41–60%), total solids were 6.8 g/dL (reference range, 5.9–7.8 mg/dL), glucose was 74 mg/dL (reference range, 60–120 mg/dL), and an AZO stick^a was 5–15 mg/dL (<30 is normal). Results from preoperative venous blood gas analysis are presented in Table 1. These values are considered normal at the authors' laboratory.

Table 1 Results of Venous and Arterial Blood Gas Analysis Presurgically and at the Time of Neuromuscular Blockade Reversal

BE, base excess; Ca, calcium; Cl, chloride; FiO₂, fraction of inspired oxygen; HCO₃⁻, bicarbonate; K, potassium; Mg, magnesium; NA, not available; pCO₂, partial pressure of carbon dioxide;. pO₂, partial pressure of oxygen.

View Fullscreen

The patient was transferred to the anesthesia and surgery services for exploratory laparotomy. After approximately 3 min of oxygen administration via face mask, general anesthesia was induced with fentanyl^b 10 μg/kg and midazolam^c 0.25 mg/kg IV. The trachea was intubated with a cuffed orotracheal tube (12 mm internal diameter) and anesthesia maintained with isoflurane^d in oxygen and an infusion of fentanyl (average rate 37 μg/kg/hr during a 3 hr period). The isoflurane vaporizer setting during surgery was between 0.4% and 0.7%. The lungs were mechanically ventilated with a tidal volume of 800 mL, and respiratory rate was between 10 and 12 breaths/min during the course of anesthesia. The peak inspiratory pressure oscillated between 10 and 12 cm H₂O. Monitoring consisted of electrocardiogram, invasive arterial blood pressure, pulse oximetry, esophageal temperature^e, and end-tidal carbon dioxide^f. Temperature was maintained between 36.2°C and 37.0°C using forced heated air during general anesthesia^g. Arterial blood samples were collected intermittently for blood gas analysis (Table 1). A second intravenous catheter was placed and a balanced crystalloid solution^h and a colloidⁱ were infused. During the 4 hr of general anesthesia, the patient received 12.5 mL/kg/hr of crystalloids, 3.12 mL/kg/hr colloids, and 120 mL of packed red blood cells^j; the latter occurred because the PCV decreased to 23%. By the end of the procedure, the PCV was 29%. In addition, dopamine^k, phenylephrine^l, and vasopressin^m were used to maintain mean arterial blood pressure above 60 mm Hg, and 0.3 mEq/kg magnesium sulfateⁿ was given after the induction of anesthesia as a 15 min IV infusion to treat multifocal ventricular tachycardia. Two hours after induction of anesthesia, a bolus of 0.1 mg/kg IV of vecuronium^o was used to enhance muscle relaxation and improve the conditions for abdominal exploration and mechanical ventilation. No further doses of NMBAs were administered, and neuromuscular function was not monitored during the procedure.

Eighty minutes after the vecuronium was given, abdominal closure began and infusion of fentanyl was reduced to 3 μg/kg/hr. A peripheral nerve stimulator^p was connected using alligator clips to subcutaneous needles over the ulnar nerve and a train-of-four (TOF) stimulation was applied (delivered current 50 mA). TOF stimulation is a process whereby four electrical impulses are delivered to the patient, and the muscular response (twitches) is evaluated. This technique is used to assess the depth of neuromuscular blockade. Four small but visible muscle twitches were elicited. Because the evoked muscle contraction seemed weak, it was difficult to ascertain whether there was fade (i.e., the magnitude of the evoked twitches decreases during stimulation). In this case, the anesthetist could not distinguish whether all four twitches of the TOF were of similar magnitude or whether fade was present. Therefore, a double burst stimulation (DBS) pattern was used. This is a pattern in which two impulses of bigger magnitude are delivered, generating more vigorous muscle twitches. With DBS, fade was evident; the second muscle twitch was visually weaker than the first one.

To exclude residual paralysis, 0.25 mg/kg IV of edrophonium^q was given slowly. After 5 min, there was no visible improvement in the fade during DBS testing and an additional dose of 0.25 mg/kg IV of edrophonium was given (cumulative dose 0.5 mg/kg). Another 5 min were allowed and fade during DBS was still evident; therefore, neostigmine^r 40 μg/kg and atropine^s 20 μg/kg IV were administered because neostigmine is a potent cholinesterase inhibitor and its administration, if not given with atropine, can result in life-threatening bradyarrhythmias. A transient increase in heart rate from 120 beats/min to approximately 160 beats/min was observed. In less than 5 min, the heart rate returned to approximately 120 beats/min. Three minutes later fade during DBS was no longer visible. Several TOFs were then evaluated, separated by 10 sec intervals. All four twitches were of bigger magnitude, and fade could not be seen. Spontaneous ventilation was reestablished during the course of the subsequent minutes (respiratory rate was 12 breaths/min), although hypoventilation was still evident on the capnograph (partial pressure of carbon dioxide was approximately 55 mm Hg). Isoflurane administration was then discontinued, and the patient's orotracheal tube was removed. Swallowing reflexes and head lifting appeared normal after extubation. The dog was transferred to the intensive care unit and recovered uneventfully.

Discussion

Vecuronium is a nondepolarizing NMBA with an intermediate duration of action. In healthy dogs, after a dose of 0.38 mg/kg, its effects are spontaneously, completely reversed after 60 min.⁷ Although approximately one quarter of that dose was given to this patient, muscle relaxation was still evident after 80 min. At that time, a conventional dose of edrophonium (0.5 mg/kg IV) did not completely reverse neuromuscular block. Hence, this patient was at increased risk for PORC.

The action of a nondepolarizing NMBA can either be allowed to terminate spontaneously as the drug is metabolized and excreted by the patient, or it can be antagonized pharmacologically with an acetylcholinesterase inhibitor. In either case, there is a risk of PORC. The likelihood of PORC can be reduced by using electrical stimulation of a peripheral motor nerve and observing the evoked muscle contraction; the latter can be assessed either subjectively (e.g., visually) or objectively (e.g., with AMG). The incidence of PORC was 64% with vecuronium in humans when no antagonist was used. Although introduction of pharmacologic antagonism on its own did not decrease the incidence of PORC, the use of subjective monitoring of peripheral nerve stimulation decreased, but did not eliminate, the incidence of PORC.⁸ Other work confirmed that visual or tactile inspection of evoked twitches did not guarantee exclusion of PORC in either people or horses.^5,9,10 The use of AMG to objectively measure evoked muscle twitches increased the sensitivity for detection of PORC beyond that possible with visual detection.⁵

Had this dog been extubated based solely on the time since administration of vecuronium, it would have been at risk from PORC-related complications. Moreover, even after a conventional dose of edrophonium was administered, the dog still had partial neuromuscular blockade. In nonanesthetized people, tidal volume and respiratory rate are adequate when the TOF ratio is as low as 0.6, suggesting that these respiratory measurements cannot be used to exclude PORC.¹¹ Other tests such as hand-grasp, eye-opening, tongue-protrusion, and head-lift reportedly correlated better with TOF ratios close to 75% and were superior clinical indicators of adequate recovery from neuromuscular blockade; however, these tests require a level of patient compliance that is impossible in veterinary patients.¹² Therefore, monitoring neuromuscular transmission with a peripheral nerve stimulator should be even more important in veterinary patients than it is in people.

The most commonly used pattern of electrical nerve stimulation for monitoring neuromuscular function in the authors' practice is TOF evaluated either visually or by AMG. The main advantage of TOF is that it obviates the need for measuring baseline values of neuromuscular transmission before blockade that would be used for comparison during the recovery phase of neuromuscular blockade. Each TOF acts as a control for itself: the magnitude of the fourth twitch is expressed as a fraction of the magnitude of the first one (T4/T1). In this case, the nerve stimulator was placed after neuromuscular blockade was instituted, which means that baseline measurements were not available. TOF was only used for evaluating recovery. In this patient, because the four elicited twitches during the TOF were small in magnitude and difficult to assess, the authors decided to switch to DBS consisting of two short tetanic stimuli or bursts (three impulses at a frequency of 50 Hz) separated by a 750 msec interval.¹³ DBS evoked bigger twitches than those produced by TOF. Bigger muscle twitches might be easier to assess either visually of by palpation; therefore, DBS might be more sensitive than TOF for subjective evaluation of fade.

A conventional dose of edrophonium (0.5 mg/kg IV, given in two aliquots) did not reverse neuromuscular blockade completely in this patient. If given when the first twitch of the TOF is visible, this dose of edrophonium completely reverses blockade induced by 0.05 mg/kg vecuronium in healthy dogs.¹⁴ The authors used a higher dose of vecuronium in this patient (0.1 mg/kg), consistent with doses described previously, and antagonism was not attempted until all four twitches of the TOF were visible.¹⁵

Neostigmine, administered subsequent to the edrophonium, restored neuromuscular function as evaluated visually by lack of fade during DBS and TOF. Edrophonium was initially chosen over neostigmine in this case because of its lower incidence of bradycardia and the reduced need for concomitant administration of anticholinergic to a patient that was already dysrhythmic. There were several studies comparing edrophonium with neostigmine for reversing vecuronium in anesthetized people. Seven of 20 human patients given vecuronium had a TOF ratio of <0.7 after edrophonium was given, but all 20 individuals reached this value after neostigmine was used.¹⁶ In a group of people given vecuronium, edrophonium (0.75 mg/kg IV) did not fully reverse paralysis in 4 of 13 patients, whereas 0.05 mg/kg IV neostigmine reversed all 10 patients treated with neostigmine.¹⁷ Interestingly, two doses of 0.75 mg/kg of edrophonium succeeded by a dose of neostigmine failed to produce complete reversal in two of the patients. Dernovoi showed that increasing the dose of edrophonium from 0.5 to 1 mg/kg did not improve antagonism of vecuronium in people.¹⁸ This case report, along with the aforementioned studies, suggested that when vecuronium was used in dogs, neostigmine should be preferred to edrophonium for reversal.

The patient was in shock and was given vasoactive drugs (phenylephrine and vasopressin) to maintain acceptable systemic perfusion pressure. These factors could have impaired redistribution, hepatic metabolism, and renal excretion of vecuronium, and hence, contributed to prolonged duration of action of the vecuronium. Reduced perfusion of skeletal muscle could also have impaired distribution of edrophonium to neuromuscular junctions and contributed to impaired antagonism of the NMBA.

Presumably, as a consequence of reduced perfusion, the patient was considerably acidemic at the time of reversal. Acidemia prolongs the time–response curve of muscle relaxants. Respiratory and metabolic acidosis not only potentiated NMBAs but also interfered with pharmacologic reversal.^19,20 Reversal of blockade in this animal might have been aided by restoring the pH toward normal with increased alveolar ventilation to induce respiratory alkalosis or by administering a buffering agent.

Hypothermia is also a cause of prolonged blockade.²¹ In this patient, esophageal temperature was between 36°C and 37°C, which is probably close to enough to normothermia to exclude this possible cause.

Plasma electrolyte concentrations in this patient were close to reference values before, during, and immediately after surgery. Ionized calcium and potassium plasma levels were low at the time of reversal and could have contributed to the lack of response to edrophonium.²² The patient was given magnesium sulfate to treat cardiac dysrythmias. The concomitant use of a magnesium sulfate and a nondepolarizing NMBA could produce prolonged paralysis in people; the mechanism for this was likely interference by magnesium with release of acetylcholine at the motor end plate.^23,24 A case of recurarization after magnesium administration in a patient that had been paralyzed and then reversed was reported,²⁵ and Fuchs-Buders et al. showed that pretreatment with magnesium sulfate not only prolonged the action of vecuronium but also decreased the efficacy of neostigmine as a reversal agent.²⁴ The authors could not exclude hypermagnesemia as a cause of the prolonged action of vecuronium and decreased ability to antagonize it, because plasma magnesium concentration was not measured after the magnesium was given; however, plasma magnesium was at the low end of the normal range before induction of anesthesia before the magnesium was given. Because vecuronium was given 2 hr after magnesium and the first dose of edrophonium was given almost an hour and a half after that, it was unlikely that hypermagnesemia contributed to the prolonged block and inability to antagonize the NMBA in this patient.

The degree of blockade at the time of administration of an antagonist also affected the efficacy of reversal. Specifically, antagonism might be unsuccessful if attempted during profound neuromuscular blockade.²⁶ In this patient, all four twitches of the TOF were visible before reversal was first attempted, and even under these circumstances, the edrophonium failed to reverse the NMBA.

Conclusion

In summary, edrophonium might not produce complete pharmacologic reversal of vecuronium-induced neuromuscular blockade and, for this reason, neostigmine might be preferred. Whether recovery from neuromuscular block was spontaneous or produced by pharmacologic antagonism, the use of peripheral nerve stimulation, preferably with objective measurement of neuromuscular transmission such as AMG was necessary to exclude PORC.