Evaluation of neostigmine antagonism at different levels of vecuronium-induced neuromuscular blockade in isoflurane anesthetized dogs

Abstract

Residual neuromuscular block (NMB) during recovery from general anesthesia may be minimized by antagonizing NMB with neostigmine. We examined neostigmine for restoring neuromuscular function when administered at 2 levels of vecuronium-induced NMB in dogs. Eight healthy adult dogs received vecuronium 0.1 mg/kg body weight (BW), IV, during isoflurane anesthesia. Recovery from vecuronium occurred spontaneously (control group; C), or was enhanced with neostigmine, 0.04 mg/kg BW, IV, administered when 2 (N2) or 4 (N4) responses to train-of-four (TOF) stimulation were first observed. Duration of NMB was significantly shorter for N2 and N4 than for C. The period of complete NMB was equal for all groups; differences were observed during the recovery phase of NMB. Time of neostigmine-enhanced recovery was significantly shorter for N4 than N2, but overall duration of NMB was not reduced. Recovery from NMB was faster with neostigmine. There is no clinical advantage in delaying neostigmine administration once 2 responses to TOF are present.

Introduction

The use of neuromuscular blocking agents (NMBA) during general anesthesia carries an implicit risk: if neuromuscular function is not completely restored at the time of tracheal extubation, residual neuromuscular block ensues in the immediate postoperative period. In humans, even low levels of neuromuscular impairment contribute to well-recognized development of complications after termination of general anesthesia, such as upper airway collapse, tracheal aspiration, and postoperative hypoxia, which are associated with impaired clinical recovery from anesthesia (1–3).

The incidence of residual blockade, and its consequences, may be minimized by pharmacologically reversing neuromuscular blockade with the administration of an acetylcholinesterase inhibitor drug such as neostigmine. Neostigmine has been successfully used for the antagonism of vecuronium and atracurium-induced neuromuscular blockade in dogs (4,5). The ability of neostigmine to completely restore neuromuscular function depends not only on the dose administered, but also on the level of neuromuscular blockade at the time of antagonism; restoration of adequate neuromuscular function after neostigmine administration occurs faster as the extent of neuromuscular impairment at the time of antagonism becomes less (6–8). Although the expectation is that antagonism of neuromuscular block will require less time if neostigmine is administered during shallow neuromuscular blockade, such improvement might be offset by the time required before that level of blockade is spontaneously reached. In other words, delaying neostigmine administration until the level of neuromuscular blockade becomes more shallow may not result in a clinically useful reduction in the overall recovery time.

We evaluated the time required for neostigmine to reverse vecuronium-induced neuromuscular blockade in dogs, when administered at 2 submaximal levels of blockade. We also evaluated whether the overall recovery time is reduced by delaying administration of neostigmine until a shallow level of neuromuscular blockade has been reached. We hypothesized that i) the time between neostigmine administration and complete recovery from neuromuscular blockade is shorter when neostigmine is administered during shallow block than it is when administered at moderate block, and ii) that consequently the overall recovery time is reduced.

Material and methods

This project was approved by the local ethics and animal use committee. Eight healthy adult mixed breed dogs (6 males, 2 females), aged 2 to 5 y and with body weights of 10.4 to 29 kg were studied. Each dog was anesthetized 3 times in a crossover design, at least 7 d apart. The order of treatment allocation was randomized. After overnight fasting, dogs were sedated with dexmedetomidine (Dexdomitor; Pfizer, Buenos Aires, Argentina), 2.5 μg/kg body weight (BW), and midazolam (Midazolam, Richmond Vet Pharma; Buenos Aires, Argentina), 0.3 mg/kg BW, IM. General anesthesia was induced with propofol (Diprivan; Astra Zeneca, Buenos Aires, Argentina) administered to effect through a cephalic catheter, until tracheal intubation was possible. Anesthesia was maintained with isoflurane (Forane; Abbott, Buenos Aires, Argentina) in 100% oxygen (1.5 L/m) and dexmedetomidine, 1 μg/kg BW per hour. The isoflurane vaporizer was initially set at 1%, and then it was adjusted at the discretion of the attending veterinarian according to clinical signs of depth of anesthesia, mean arterial blood pressure, and to permit neuromuscular monitoring. The lungs were mechanically ventilated to keep the end-tidal CO₂ between 35 and 45 mmHg. Monitoring included electrocardiogram, pulse oximetry, oscillometric arterial blood pressure, capnography, and esophageal temperature (Monitor Multipar; Feas Electronica, Córdoba, Argentina).

Neuromuscular blockade and monitoring

After induction of anesthesia the dogs were positioned in left lateral recumbency. The dependent limb was supported parallel to the surgical table, so that the tarsus could flex freely and unopposed during nerve stimulation. Neuromuscular function was assessed in the nondependent pelvic limb with acceleromyography (AMG) using a TOF-Watch monitor (TOF-Watch; Organon Ltd, Swords, Dublin, Ireland). Two needle electrodes were placed subcutaneously over the peroneal nerve and an acceleration-sensitive crystal was taped over the dorsal aspect of the paw to measure the peak acceleration of the evoked tarsal flexion. Train-of-four (TOF) stimulation was delivered every 15 s (60 mA, 2Hz, pulse duration 0.2 ms). Following each TOF, the AMG monitor automatically calculates and displays the TOF ratio, that is, the magnitude of the fourth twitch of the train (T4) as a percentage of the first twitch (T4/T1) × 100. A period of at least 30 min of anesthesia and TOF stimulation was allowed before data collection began. The AMG monitor was calibrated immediately prior to the start of data collection. During calibration, a number of single twitches are elicited automatically and the response measured; the device then automatically sets the response to 100%, and that reference value is stored in the monitor’s memory for the duration of that procedure. After calibration was completed and baseline values obtained, neuromuscular block was achieved by the administration of vecuronium (Galaren, Fada Pharma, Buenos Aires, Argentina), 0.1 mg/kg BW, IV.

Treatment allocation and data collection

The vaporizer setting was recorded every 5 min throughout the duration of the procedure. The acceleromyographic TOF ratio was measured continually every 15 s but values were recorded manually every minute after vecuronium administration. Dogs were assigned randomly to 1 of 3 treatment groups; the control treatment group (C) received no neostigmine and recovery from vecuronium-induced block was spontaneous. The second treatment group received neostigmine (Neostigmina; Laboratorio Drawer, Buenos Aires, Argentina), 0.04 mg/kg BW, IV, when 2 responses to TOF stimulation (T2) were first detected visually (moderate neuromuscular block; N2) during offset of neuromuscular block. The third group received the same dose of IV neostigmine when all 4 responses to TOF stimulation (T4) were first detected visually (shallow neuromuscular block; N4) during offset of neuromuscular block. Neostigmine injections were immediately preceded by atropine (Klonatropina, Klonal Laboratorios, Buenos Aires, Argentina), 0.02 mg/kg BW, IV.

The average of 3 consecutive values of TOF ratio measured after calibration of the AMG monitor was used as the baseline TOF ratio for each dog. Vecuronium was then injected over 5 s through a free-flowing infusion of a balanced crystalloid solution. Recovery from neuromuscular block was considered adequate when the TOF ratio first reached ≥ 90% with no subsequent decreases (2,9). Duration of neuromuscular block was defined as the time from vecuronium administration to a TOF ratio ≥ 90%, and was recorded for all groups. Data were collected from the AMG monitor by an investigator who was aware of the treatment allocation.

The duration of neuromuscular block was further divided into 2 periods: the period of surgical block (from vecuronium administration to return of T2), and the recovery period (from the return of T2 to TOF ≥ 90%). In those groups receiving neostigmine (N2 and N4), the time between neostigmine administration and a TOF ratio ≥ 90% was also recorded.

Once recovery from neuromuscular block was complete, the vaporizer was turned off and the dogs were allowed to recover from general anesthesia.

Statistical analysis

Distribution of all data was evaluated with the Shapiro-Wilk test (Statistix 9.0; Analytical Software, Tallahassee, Florida, USA). Log-transformation was imposed to 1 variable which was not normally distributed. The significance of differences between all groups for baseline TOF ratio, duration of neuromuscular block, period of surgical block, and recovery period for the 3 groups was evaluated with a mixed effect model, accounting for the crossover design. Tukey’s post-hoc analysis was also used (JMP, Version 10; SAS Institute, Cary, North Carolina, USA). Results are summarized as mean [± standard error (SE)] [min-max]. The same analysis was used when comparing the dose of propofol and the isoflurane vaporizer setting between the 3 groups. The vaporizer setting recorded every 5 min was averaged for each individual for the duration of the procedure; the average vaporizer setting for each animal was used for the comparisons between groups. Propofol dose and vaporizer setting are summarized as mean (± SE). Differences are considered significant when P < 0.05.

Results

All dogs completed the investigation and recovered from general anesthesia without complications. Esophageal temperature was maintained between 35.8°C and 38°C in all dogs. The doses of propofol (mg/kg BW) for treatments C, N2, and N4 were 1.5 (0.19), 1.8 (0.19), and 1.7 (0.19), respectively (P = 0.55). The average isoflurane vaporizer setting (%) was 1.1 (0.04) for group C, 0.9 (0.04) for N2, and 1.0 (0.04) for N4 (P = 0.045); N2 being lower than the other 2 groups.

Baseline TOF ratios are shown in Table 1 and were similar between treatment groups (P = 0.31). Duration of neuromuscular block (from vecuronium injection to TOF ≥ 90%) was shorter for N2 and N4 than for C (P = 0.01); with no difference between N2 and N4 (Table 1, Figure 1). The period of surgical block, from vecuronium administration to the return of T2, was not different between any of the 3 treatment groups (P = 0.52) (Table 1, Figure 1). The recovery period (from return of T2 to TOF ≥ 90%) was significantly shorter for N2 and N4 than for the control group (P = 0.004); with no difference between N2 and N4 (Table 1, Figure 2). The time from administration of neostigmine to TOF ≥ 90% was shorter for N4 than for N2 (P = 0.008) (Table 1, Figure 2).

Table 1.

Mean (± SE) [min-max] baseline TOF ratio and variables describing the time-course of recovery from vecuronium-induced neuromuscular block in dogs

	Baseline TOF ratio (%)	Duration of neuromuscular block (min)	Period of surgical block (min)	Recovery period (min)	Time from neostigmine to TOF 90% (min)
Control	99 (0.7) [95–100]	42.6 (2.31) [36–60]	25.6 (1.42) [21–30]	17 (1.89) [9–35]	NA
N2	97 (0.7) [94–100]	33.6 (2.31) [24–40]a	27.4 (1.42) [18–34]	6.3 (1.89) [3–12]a	6.3 (0.83) [3–12]
N4	97 (0.7) [93–99]	31.7 (2.31) [27–38]a	24.8 (1.42) [21–30]	6.8 (1.89) [5–9]a	3.6 (0.83) [2–6]b

NA — not applicable; SE — standard error.

^aSignificantly different from Control.

^bSignificantly different from N2.

Figure 1.

Dot plots of the duration of neuromuscular block (from vecuronium administration to a TOF ratio ≥ 90%) for all groups (A), and the period of surgical block [from vecuronium administration to return of T2 (B)]. Each dot represents an individual dog and the horizontal line represents the mean value.

*Significantly different from C.

Figure 2.

Dot plots of the recovery period (from return to T2 to a TOF ratio ≥ 90%) in all groups (A). B — time between neostigmine administration to a TOF ratio ≥ 90% in groups N2 and N4 only. Each dot represents an individual dog and the horizontal line represents the mean value.

*Significantly different from C.

^Significantly different from N2.

The TOF ratio measured with AMG at the time of neostigmine administration in N4 was 22% (0, 37).

Discussion

Although the data show that neostigmine reverses vecuronium-induced neuromuscular block in a shorter period if administered when 4 responses to TOF have returned, rather than when only 2 responses are present, the total duration of neuromuscular block and the time defined as recovery period were not different between the 2 groups receiving pharmacological antagonism. This means that while delaying administration of neostigmine indeed accelerates pharmacological antagonism of the NMBA, this improvement is offset by the period required before neostigmine is administered. Hence, although the data support our first hypothesis, overall recovery time was not improved by delaying neostigmine administration. When dogs were given neostigmine at a TOF count of 4, rather than 2, the variability in the time to reach adequate recovery was smaller and all dogs reached a TOF ≥ 90% in 6 min or less. When administered at a count of 2, 1 dog required 12 min before adequate recovery was reached. Although the difference in the times between neostigmine administration and a TOF ratio ≥ 90% was significantly shorter, delaying the time of neostigmine administration until 4 responses to TOF return offers only a small clinical advantage.

Neostigmine antagonizes nondepolarizing neuromuscular blockade indirectly by inhibiting the actions of the acetylcholinesterase enzyme, allowing the local concentration of acetylcholine (ACh) to increase. Ultimately, it is ACh which competes with the NMBA and reverses neuromuscular block. The maximum concentration of ACh that can be reached in the neuromuscular junction is determined by the rate of release of ACh from the nerve cell and the rate of ACh hydrolysis; hence, once the activity of actylcholinesterase has been fully inhibited, further doses of neostigmine will fail to increase the local concentration of ACh. This mechanism provides the theoretical basis for the observation that shallow neuromuscular block is antagonized faster than a deeper one, and that antagonism of nondepolarizing block might be completely ineffective during intense levels of blockade (9). When neostigmine was administered during intense vecuronium-induced neuromuscular block in humans (no response to TOF stimulation), up to 60 min were required before reversal was complete (10). Since it has been shown in humans that even low degrees of residual neuromuscular block can result in a decreased response to hypoxia and an increased risk for tracheal aspiration, oxygen supplementation and monitoring should not be interrupted until recovery from neuromuscular blockade is complete (1–3).

Our data reinforce the point that when neuromuscular block is allowed to recover spontaneously, a wide variation might be seen between individuals. The duration of surgical block, that is, the period until 2 responses to TOF returned, was the same between groups. In our investigation, vecuronium produced a predictable duration of effect (~25 minutes in each group). Differences and variability were observed during the recovery period. The recovery period (time between T2 and TOF ≥ 90%) ranged between 9 and 35 min in those animals not receiving neostigmine. While adequate recovery was reached quickly once responses to TOF reappeared in some animals, it took several times longer in others. Such variation is not surprising; in horses receiving rocuronium, duration of neuromuscular block (to a TOF ≥ 90%) ranged between ~20 and 85 min (11). In dogs, duration of rocuronium blockade ranged between ~18 and 47 min (12). The time in which all 4 responses to TOF are present, but the TOF ratio is < 90% represents a period when patients may be at risk because residual block might go undetected. Fade during TOF stimulation is difficult to assess visually or by palpation. In humans, fade can only be detected visually or by palpation, when the TOF ratio is < 40% to 50%. In other words, different degrees of moderate residual neuromuscular block (TOF ratio > 50%) cannot be identified by subjective methods (13,14). The same is true in horses (15). Furthermore, when horses were evaluated visually and with AMG, a period of 42 min was identified in 1 individual in which no fade could be detected visually, yet the acceleromyographic TOF ratio was < 90% (15). This constitutes a potential period of undetected residual block if quantitative monitoring had not been used.

Other tests to assess recovery of neuromuscular function can also provide misleading information. Return of spontaneous ventilation, and measurements of tidal volume, end-tidal CO₂, or inspiratory flow have historically been used to assess recovery from neuromuscular block. However, it has been shown recently that substantial neuromuscular impairment was still present when these measures of spontaneous ventilation had returned to baseline values (16). Therefore, the longer the period in which fade during TOF is present, the more opportunities to extubate a patient in which residual neuromuscular block might still be present but undetected. As a result, routine antagonism of neuromuscular block should be considered when responses to TOF stimulation can be seen, but the TOF ratio is not being quantified (17).

There are some limitations in our study. First, we arbitrarily defined the period of surgical paralysis until the return of the second response to TOF. We did this because our usual practice is to re-dose NMBA when 2 responses to TOF return. Second, return of the responses to TOF was detected visually rather than with AMG. This was done by design because objective monitoring of neuromuscular block (e.g., with AMG) is not frequently used in veterinary anesthesia. We think that obtaining results from tests that reflect routine practice would provide useful information in terms of speed of recovery after neostigmine administration. Third, we arbitrarily defined moderate and shallow block as TOF counts of 2 and 4, respectively. There are no universally accepted definitions of what constitutes shallow or moderate neuromuscular block. Shallow neuromuscular block could have been defined, for example, as a TOF ratio of 50% rather than simply a count of 4 (18). We chose to use visual detection of 4 twitches rather an acceleromyographic TOF ratio for the reasons explained. Lastly, the same dose of neostigmine was used for groups N2 and N4, despite the fact that the depth of neuromuscular block was different at the time of reversal. It is possible that a lower dose of neostigmine might be as effective for the antagonism of shallow neuromuscular block, but such determination exceeded the objectives of this investigation.

When neostigmine was administered to dogs in N4, the TOF ratio measured with AMG at that time was 22% (0, 37). In 1/8 dogs the AMG monitor did not display a TOF ratio, but instead reported a count of 4. TOF-Watch monitors display counts of 4 (and not TOF ratio) when 4 twitches of small magnitude are detected. Only when the magnitude of T1 exceeds 20% of baseline is a TOF ratio calculated (19). It can therefore be inferred that T1 was ≥ 20% at the time of neostigmine administration in 7/8 dogs, even when its magnitude of T1 was not being recorded.

In conclusion, neostigmine (0.04 mg/kg BW) antagonizes vecuronium-induced block faster if administered when 4 responses to TOF are present, rather than when only 2 responses are present. However, this does not result in a reduction in overall recovery time and hence, it is not necessary to delay neostigmine administration once 2 twitches during TOF are visually detectable. The variability encountered in recovery times underscores the importance of neuromuscular function monitoring. CVJ