Adequate recovery of muscle power is mandatory before discharging a patient to avoid serious complications after ambulatory anaesthesia. Incomplete neuromuscular recovery is a common problem in patients in postanaesthesia care units and the incidence of residual paralysis after administration of current intermediate-acting neuromuscular blocking agents may be up to 40–60% [1,2]. In addition, residual paralysis may lead to postoperative pulmonary complications, increasing patient morbidity and hospital stay and costs . Postoperative nausea and vomiting (PONV) is still one of the commonest complications after outpatient anaesthesia . Most patients consider PONV to be even more distressing than pain . In addition to reducing patient satisfaction, PONV may delay recovery, increase length of hospital stay, and lead to unanticipated hospitalization .
It has been suggested that antagonism of residual neuromuscular block at the end of surgery may increase the risk of PONV . As a consequence, omitting antagonism of the neuromuscular block would decrease the incidence of PONV. Omitting antagonism, however, introduces a non-negligent risk of residual neuromuscular paralysis, even with intermediate- or short-acting non-depolarizing neuromuscular blocking agents (NMBAs) [8,9]. Therefore the theoretical benefit of omitting pharmacological reversal of neuromuscular blockade at the end of surgery has to be balanced against the potential risk of residual paralysis . Thus, the question whether or not to antagonize is of particular clinical relevance in the context of ambulatory anaesthesia.
Influence of pharmacological reversal on PONV
Pharmacological reversal of residual neuromuscular blockade is usually achieved using an anticholinesterase drug (neostigmine or edrophonium), along with an anticholinergic agent (atropine or glycopyrrolate), the latter for prevention of muscarinic side-effects of the anticholinesterase drugs. This mixture may, at least theoretically, influence the incidence of PONV by the action of several mechanisms. The antiemetic properties of anticholinergic compounds have been utilized traditionally in premedication for several decades. Their antiemetic properties result from a central rather than a peripheral mode of action, as demonstrated by the finding that glycopyrrolate, when used as an alternative to atropine, was associated with no antiemetic properties . In addition, in a double-blind comparison of glycopyrrolate and atropine, it was found that PONV was twice as common after glycopyrrolate as after atropine . These results are consistent with the inability of the highly polarized glycopyrrolate molecule to cross the intact blood–brain barrier. However, no difference in the incidence of PONV was observed in another study comparing atropine, glycopyrrolate and placebo . In addition, the combination of atropine and neostigmine was shown to decrease the lower oesophageal sphincter pressure and thus increase the risk of PONV. Moreover, anticholinesterase drugs may have muscarinic effects on the gastrointestinal tract, which increase motility and stimulate secretion of gastric fluid. However, these effects are partly prevented by anticholinergic drugs. Thus, the definitive contribution of a mixture of anticholinesterase and anticholinergic drugs to PONV is still not clear.
Evidence from randomized controlled trials up to 1998 (for details see )
Evidence from randomized controlled trials concerning the influence of pharmacological reversal of neuromuscular blockade at the end of surgery on the incidence of PONV, the likelihood of harm when antagonism was omitted was recently published by Tramèr and Fuchs-Buder in a meta-analysis . For patients receiving placebo or no antagonism of neuromuscular block, and thus exposed to spontaneous recovery, the absolute risk (incidence) of early and late PONV were 10–40% and 10–70%, respectively. Corresponding values in patients receiving neostigmine suggested an increased risk of PONV with increasing doses of neostigmine. For both early and late outcomes, the highest incidence of PONV with the lowest dose of neostigmine (1.5 mg) used did not overlap with the lowest incidence of PONV with the highest dose of neostigmine (2.5 mg) used (Figure 1). In addition, the number needed to treat to prevent PONV by omitting neostigmine suggested dose-dependency and varied from negative to greater than 10. The 1.5 mg dose of neostigmine when compared with no treatment in one study had a negative needed-to-treat value, suggesting an antiemetic effect with neostigmine. In trials testing the highest dose of neostigmine (2.5 mg), the needed-to-treat values were positive for all outcomes, suggesting a clinically relevant dose-related emetogenic effect of neostigmine. In other words, there was an absence of emetic effect when this dose of neostigmine was not used. However, caution must be exercised in interpreting dose–response relations with such a relatively small number of patients. Omitting edrophonium had no beneficial effect on early or late emesis. (See  for full details of this meta-analysis.)
Evidence from randomized, controlled trials after 1998 [14–18]
Since March 1998, four more studies have investigated the contribution of pharmacological reversal of residual paralysis to PONV. Data from 503 patients on different doses of neostigmine combined with different doses of atropine or glycopyrrolate in adults and children were analysed (Table 1). NMBAs used were mivacurium, rocuronium and atracurium. Four reports included a placebo group not receiving pharmacological reversal [14–16,18] and one study evaluated the association of neostigmine with atropine or glycopyrrolate on PONV in children . Joshi and his colleagues reported no difference in the incidence of nausea, vomiting and the need for antiemetics after pharmacological reversal with neostigmine 2.5 mg and glycopyrrolate 0.5 mg, i.v., compared with patients without reversal . McCourt and his colleagues found that the incidence of emetic symptoms was comparable between controls, and neostigmine/glycopyrrolate and neostigmine/atropine groups. They concluded that administration of reversal and the type of anticholinergic drug has no influence on the incidence of PONV . Nelskyla and colleagues further confirmed that compared with placebo, a mixture of neostigmine (2 mg, i.v.) and glycopyrrolate (0.4 mg, i.v.) did not increase the occurrence of PONV in women undergoing outpatient gynaecological laparoscopy . Chhibber and colleagues reported that reversal of neuromuscular blockade with atropine and neostigmine was associated with a lower incidence of PONV compared with glycopyrrolate and neostigmine . Interestingly, the need for antiemetics did not differ between the groups in this latter study. Recently, Lovstad and coworkers reported that antagonism of neuromuscular block with high dose of neostigmine (i.e. 50 mg kg−1) increases postoperative nausea and the use of antiemetic drugs .
Risk of residual paralysis when omitting reversal
In the meta-analysis reported by Tramèr and Fuchs-Buder, the number needed to harm to produce one patient with clinically relevant muscle weakness by omitting reversal compared with giving anticholinesterase drugs was 30 . Thus, there is evidence that one in 30 patients undergoing surgery and given a muscle relaxant but not receiving reversal with neostigmine or edrophonium at the end of surgery will show clinically overt muscle weakness (i.e. requiring rescue anticholinesterase drugs or re-intubation). As indicated by the authors, this result lacks statistical significance. However, residual paralysis occurred only in patients who did not receive anticholinesterase drugs and not in any patients with reversal of neuromuscular block at the end of surgery. Moreover, only clinically overt muscle weakness was considered, further supporting the clinical relevance of this result because even one case of residual paralysis leading to permanent adverse effect is unacceptable.
Pathophysiological consequences of residual paralysis
Residual paralysis and respiratory muscle function
Restoration of adequate neuromuscular function postoperatively is essential to assure that patients are able to sustain adequate ventilation and cough, and maintain their airway open. For several years a train-of-four (TOF) ratio of 0.7, measured at the adductor pollicis, was considered synonymous with adequate recovery from non-depolarizing neuromuscular blockade because this index was generally accepted to predict adequate ventilatory function postoperatively. This was based mainly on the assumption that sustained maximum inspiratory force and minute ventilation indicated safe recovery from neuromuscular block . However, new insights into the pathophysiological consequences of residual neuromuscular blockade required more rigorous criteria for determining the adequacy of neuromuscular recovery.
Residual paralysis, and laryngeal and pharyngeal muscle function
There is increasing evidence that despite adequate recovery of central respiratory muscles, such as the diaphragm, considerable weakness of muscles involved in the maintenance of the airway may persist. Pavlin and his colleagues reported that adequate ventilation of the lungs in human volunteers recovering from d-tubocurarine neuromuscular blockade did not necessarily indicate ability to maintain a functionally intact airway . According to them a maximum inspiratory pressure of −25 cmH2O did not assure muscle strength sufficient for airway protection. An inspiratory pressure of greater than −50 cmH2O is required to perform all manoeuvres, including swallowing, which are necessary for protection of the airway, at least in volunteers. Finally, Eriksson and his colleagues assessed the pharyngeal function at rest and during swallowing in partially paralysed human volunteers . They reported that the upper oesophageal sphincter function was more sensitive to NMBAs than the pharyngeal constrictor muscle and concluded that normal pharyngeal function is not restored until an adductor pollicis TOF ratio of >0.9 is reached and that partial paralysis corresponding to a TOF ratio of <0.9 is associated with pharyngeal dysfunction, as well as with an increased risk of pulmonary aspiration.
Residual paralysis and hypoxic ventilatory control
The increase in ventilation during hypoxia is governed mainly by afferent neuronal input from peripheral chemoreceptors of the carotid bodies. There is evidence that non-depolarizing neuromuscular blocking agents interfere with this hypoxic ventilatory control (for current review, see ). The hypoxic ventilatory response was reduced by ≈ 30% in awake volunteers at the TOF ratio of 0.7 at the adductor pollicis, formerly accepted as indicating sufficient neuromuscular recovery. The magnitude of reduction was the same whether an aminosteroidal or a benzylisoquinolinium muscle relaxant was used, but for all drugs a considerable interindividual variation of the extent of depression of human hypoxic ventilatory response was observed. The mechanism behind this interaction seems to be a reversible depression of the carotid body chemoreceptor activity during hypoxia, although the exact mechanism of this interaction remains to be clarified as does whether anticholinesterase drugs can reverse this effect. Obviously, patients in the early postoperative period with possible residual effects of opioids, hypnotics and/or inhaled anaesthetics may be at increased risk of ventilatory depression in the presence of a partial neuromuscular block. An adductor pollicis TOF ratio of 0.9 or greater should therefore be aimed at prior to tracheal extubation and spontaneous breathing to eliminate any significant contribution of residual paralysis on hypoxic ventilatory control .
Residual paralysis and patient comfort
Kopman and his colleagues correlated the subjective feelings of residual paralysis with TOF ratios of 0.7 and 0.9 in healthy, awake volunteers . Their results were as surprising as they were significant, all volunteers felt uncomfortable at a TOF ratio <0.75, and all reported considerable visual disturbance at TOF ratios of <0.9. The authors concluded that any definition of ‘satisfactory’ recovery of neuromuscular block should be context-sensitive. In a subject recovering from intra-abdominal surgery and receiving opioids in amounts sufficient to control postoperative pain, diplopia or diminished grip strength is probably unlikely to be of major concern. But these may be disquieting to the patient who has otherwise fully recovered from the effects of anaesthesia and is attempting to ambulate. Thus, ‘adequate’ recovery of neuromuscular function in the outpatient setting requires return of the TOF ratio to a value = 0.9 .
Residual paralysis and safety margin of neuromuscular transmission
According to the concept of the margin of safety of neuromuscular transmission, the TOF response may be normal with ≈70% of acetylcholine (ACh) receptors at the neuromuscular junction still blocked with a non-depolarizing NMBA. However, in this situation, any decrease in the ACh concentration at the neuromuscular junction may enhance neuromuscular block with subsequent clinical consequences, such as recurarization. Thus, patients in the early postoperative period, with a large percentage of ACh receptors still occupied with the neuromuscular blocking agent, are particularly vulnerable to drug interaction at the neuromuscular junction . It may be particularly disquieting in ambulatory patients leaving the day care unit for home with a certain percentage of ACh receptors still occupied by the neuromuscular blocking agent. In this context, neuromuscular monitoring should give further information about the safety margin of neuromuscular transmission. Actually, only the tetanic stimulation pattern is able to do so. However, without this information, complete reversal of neuromuscular blockade should be the best way for preventing recurarization.
In view of these results, a TOF ratio of 0.7 can no longer be accepted as a index of sufficient recovery of neuromuscular blockade; it is now generally accepted that a TOF ratio of at least 0.9 is required. However, even these more rigorous criteria of neuromuscular recovery give no information about the safety margin.
Incidence and clinical consequences of residual paralysis
It has been shown previously that residual paralysis, defined as a TOF ratio of <0.7, occurs in 30–40% of patients after pancuronium neuromuscular block and in 5–10% of patients when intermediate-acting compounds, such as vecuronium, rocuronium or atracurium, have been used. Of interest in this context is the finding that without neuromuscular monitoring and without pharmacological reversal, the incidence of postoperative residual paralysis after vecuronium-neuromuscular blockade may be as high as 42% . In addition, after mivacurium — the NMBA with the shortest clinical duration — postoperative residual paralysis has been reported in adults when pharmacological reversal has been omitted .
The frequency of residual paralysis may be markedly higher when the stronger criterion of a TOF ratio of 0.9 is applied. Dilly and his colleagues assessed the incidence of residual paralysis (TOF ratio <0.9) after a single bolus of an intermediate-acting NMBA in relation to the duration of anaesthesia. For relatively short-acting procedures (60–90 min duration) 60% of patients had a TOF ratio of <0.9 at the end of anaesthesia. Even for longer procedures with a duration of anaesthesia of more than 2 h, up to 30% of patients did not fulfil the current criterion of sufficient neuromuscular recovery, i.e. a TOF ratio = 0.9 . Thus, without pharmacological reversal, insufficient neuromuscular recovery at the end of anaesthesia seems to be a frequent adverse event. In an important study, residual paralysis has recently been identified as an independent risk factor for postoperative pulmonary complications, at least when long-acting NMBAs such as pancuronium were used .
PONV is distressing and interferes with patient comfort, but it never becomes chronic and is almost never life-threatening. Residual paralysis, on the other hand, can lead to hypoventilation, attenuate the ventilatory response to hypoxia, impair coughing, compromise laryngeal and pharyngeal function and increase the incidence of pulmonary complications. Thus, unlike PONV, residual paralysis in the postoperative period is not a minor adverse event and should be regarded as potentially very harmful. Despite the widespread belief that anticholinesterase drugs may be emetic, there is no convincing evidence, based on clinical studies, that omitting pharmacological reversal at the end of anaesthesia decreases the risk of PONV . Omitting antagonism, however, introduces a non-negligent risk of residual paralysis, even with short-acting neuromuscular blocking agents . Thus, in ambulatory patients, neuromuscular recovery should be monitored routinely and residual paralysis antagonized, unless there is strong evidence that it is not required.
To further improve management of residual paralysis in ambulatory patients and thus contributing to patient safety and comfort, the following issues should be considered:
- • Trials testing the hypothesis of a dose-dependency of the effects of neostigmine are needed.
- • Neuromuscular monitoring mainly addresses whether a patient's neuromuscular recovery is sufficient or not. However, a link between the degree of residual paralysis and the dose of neostigmine needed to reverse it effectively would allow the further implementation of neuromuscular monitoring in this decision-making process. Information about the dose requirements of neostigmine to antagonize moderate residual neuromuscular block, i.e. TOF ratio >0.5, is lacking.
- • Finally, regarding the large and potentially harmful spectrum of neostigmine-associated side-effects, the need for new, safer anticholinesterase compounds becomes obvious.
1 Baillard C, Gehan G, Reboul-Marty J et al.
Residual curarization in the recovery room after vecuronium. Br J Anaesth
2000; 84: 394–395.
2 Dilly MP, Plaud B, Thery V, Debaene B. Incidence of residual neuromuscular block in the PACU after a single intubating dose of intermediate duration muscle relaxant. Br J Anaesth
1999; 82 (Suppl. 1): A30.
3 Berg H, Viby-Mogensen J, Roed J et al.
Residual neuromuscular block is a risk factor for postoperative pulmonary complications. Acta Anaesthesiol Scand
1997; 41: 1095–1103.
4 Chung F, Mezei G. Factors contributing to a prolonged stay after ambulatory surgery. Anesth Analg
1999; 89: 1352–1359.
5 Van-Wijk MGF, Smalhout B. A postoperative analysis of the patient's view of anaesthesia
in a Netherlands teaching hospital. Anaesthesia
1990; 45: 679–682.
6 Gold BS, Kitz DS, Lecky JH et al.
Unanticipated admission to the hospital following ambulatory surgery. JAMA
1989; 262: 3008–3010.
7 Rabey PG, Smith G. Anaesthetic factors contributing to postoperative nausea and vomiting
. Br J Anaesth
1992; 69 (Suppl. 1): 40S–45S.
8 Boeke AJ, Lange JJ, Van Druenen B, Langemeijer JJM. Effect of antagonizing residual neuromuscular blockade
and atropine on postoperative nausea and vomiting
. Br J Anaesth
1994; 72: 654–656.
9 Bevan DR, Kahwaji R, Ansermino JM et al. Residual block
after mivacurium with or without edrophonium
reversal in adults and children. Anesthesiology
1996; 84: 362–367.
10 Tramèr MR, Fuchs-Buder T. Omitting antagonism
of neuromuscular block: effect on postoperative nausea and vomiting
and risk of residual paralysis. A systematic review. Br J Anaesth
1999; 82: 379–386.
11 Mirakhur RK, Dundee JW. Lack of antiemetic action of glycopyrrolate. Anaesthesia
1981; 36: 819–820.
12 Salmenpera M, Kuoppamaki R, Salmenpera A. Do anticholinergic agents affect the occurrence of postanaesthetic nausea? Acta Anaesthesiol Scand
1992; 36: 445–448.
13 Chisakuta AM, Mirakhur RK. Anticholinergic prophylaxis does not prevent emesis following strabismus surgery in children. Paediatr Anaesth
1995; 5: 97–100.
14 Joshi GP, Garg SA, Hailey A, Yu YS. The effects of antagonizing neuromuscular blockade
and glycopyrrolate on nausea and vomiting after ambulatory surgery. Anesth Analg
1999; 89: 628–631.
15 McCourt KC, Mirakhur RK, Kerr CM. Dosage of neostigmine
for reversal of rocuronium block from two levels of spontaneous recovery. Anaesthesia
1999; 54: 651–655.
16 Nelskyla K, Yli-Hankala A, Soikkeli A, Korttila K. Neostigmine
with glycopyrrolate does not increase the incidence or severity of postoperative nausea and vomiting
in outpatients undergoing gynaecological laparoscopy. Br J Anaesth
1998; 81: 757–760.
17 Chhibber AK, Lustik SJ, Thakur R, Francisco DR, Fickling KB. Effects of anticholinergics on postoperative vomiting, recovery and hospital stay in children undergoing tonsillectomy with or without adenoidectomy. Anesthesiology
1999; 90: 697–700.
18 Lovstad RZ, Thagaard KS, Berner NS, Raeder JC. Neostigmine
50 µg/kg with gylcopyrrolate increases postoperative nausea in women after laparoscopic gynaecological surgery. Acta Anaesthesiol Scand
2001; 45: 495–500.
19 Ali HH, Wilson RS, Savarese JJ, Kitz RJ. The effect of tubocurarine on indirectly elicted train of four muscle response and respiratory measurements in humans. Br J Anaesth
1975; 47: 570–575.
20 Pavlin EG, Holle RH, Schoene R. Recovery of airway protection compared with ventilation in humans after paralysis with curare. Anesthesiology
1989; 70: 381–385.
21 Eriksson LI, Sundman E, Olsson R et al.
of the pharynx at rest and during swallowing in partially paralyzed humans. Anesthesiology
1997; 87: 1035–1043.
22 Eriksson LI. The effects of residual neuromuscular blockade
and volatile anaesthetics on the control of ventilation. Anesth Analg
1999; 89: 243–251.
23 Kopman AF, Yee PS, Neuman GG. Relationship of the train-of-four fade ratio to clinical signs and symptoms of residual paralysis in awake volunteers. Anesthesiology
1997; 86: 765–771.
24 Fuchs-Buder T. Magnesium sulphate enhances residual neuromuscular block induced by vecuronium. Br J Anaesth
1996; 76: 565–566.