Compared to the placebo group, patients who received the neostigmine-glycopyrrolate mixture had a significantly higher incidence of emesis and mean number of emetic episodes and required more rescue antiemetic therapy while in the PACU Table 2. There were no significant differences between the edrophonium-atropine and placebo groups in the incidence of emesis, mean number of emetic episodes, or need for antiemetic therapy while in the PACU. The mean number of emetic episodes and the need for rescue antiemetic therapy were significantly greater in the neostigmine-glycopyrrolate group compared to the edrophonium-atropine group Table 2. However, the differences between these two groups in the incidence of emesis in the PACU did not achieve statistical significance (34% vs 18%, P = 0.18). After discharge from the PACU, there were no significant differences in the incidence of emesis among the three groups Table 2.
When the comparisons of the incidence of emesis were repeated for a subset of patients who did not receive analgesic therapy with morphine in the PACU, similar results were noted. In this subpopulation, the incidence of emesis in the PACU remained significantly higher in the group that had received neostigmine-glycopyrrolate than in the group where spontaneous recovery from block had occurred (36% vs 10%, P = 0.04). Finally, the incidence of emesis after discharge from the PACU did not differ among the three subgroups.
It is unclear whether anticholinesterases should be used routinely to antagonize residual neuromuscular block from the shorter acting nondepolarizing drugs such as mivacurium or whether spontaneous recovery should be permitted [1,8]. As anticholinesterase drugs inhibit both acetylcholinesterase and plasma cholinesterase, they have multiple potential interactions with mivacurium, especially if the block is less intense . This study has demonstrated that avoidance of neostigmine may be of some benefit after moderate mivacurium-induced block, as its use may be associated with increased emesis in the PACU.
Neostigmine increases intestinal motility, and some surgeons have concerns regarding the integrity of their intestinal anastomoses if these drugs are used . However, the effects of neostigmine on postoperative nausea and vomiting are conflicting. Janhunen and Tammisto  reported that the use of neostigmine-atropine was associated with a decreased incidence, but not severity of emetic symptoms, in patients undergoing venous stripping or uncomplicated open cholecystectomy procedures. Boeke et al. , using a smaller dose of neostigmine (1.5 mg vs 2.5 mg) in a similar patient population, found no differences in the incidence of nausea or vomiting. However, this group reported a decreased requirement for antiemetic therapy in patients receiving neostigmine-atropine compared to those undergoing spontaneous recovery from block.
In contrast, King et al.  found that the combination of neostigmine (2.5 mg) with atropine (1.2 mg) was associated with increased emesis compared to spontaneous recovery in an older population undergoing knee or hip prosthesis operations. Recently, Ding and White  compared the influence of mivacurium (with or without reversal) and succinylcholine on the recovery profile in women undergoing laparoscopic surgery. In this study, patients who developed nausea and vomiting within 30 min of opioid administration in the PACU were excluded from the statistical analysis. The use of mivacurium without reversal was associated with decreased nausea (but not emesis) in the PACU compared to the use of succinylcholine. Furthermore, both nausea and vomiting were decreased after spontaneous recovery from mivacurium-induced block compared to antagonism with neostigmine-glycopyrrolate.
Our study confirmed that antagonism of mivacurium-induced block with a neostigmine-glycopyrrolate combination is associated with an increased incidence and severity of postoperative emesis in the PACU. Although significantly more patients in the neostigmine group required rescue antiemetic therapy compared to the group where block was permitted to resolve spontaneously, the times to discharge from the PACU did not differ between the three groups in our study (or in previous studies).
This study also demonstrated that the incidence and severity of emesis was not significantly increased by the use of edrophonium. Thus, if antagonism of mivacurium-induced block is indicated this drug should be considered, even though it is more expensive than neostigmine . However, the lack of differences in the times from end of surgery to tracheal extubation, arrival and discharge from the PACU, as well as the similar incidence of postdischarge emesis, suggest that the benefits of the use of edrophonium are small compared to the avoidance of anticholinesterase drugs.
In our study, the surgical procedures did not require the maintenance of an intense degree of neuromuscular block. The rapidity of spontaneous recovery from a moderate degree of mivacurium-induced block may explain the lack of significant differences in the times from the end of surgery to tracheal extubation and arrival in the PACU in patients who did or did not receive reversal drugs. With a more profound neuromuscular block, there may be differences in the time from end of surgery to tracheal extubation, as the time for achievement of a train-of-four ratio of 0.75 from a block in excess of 90% is greater with spontaneous recovery compared to recovery after the administration of neostigmine or edrophonium .
Our study was designed such that tracheal extubation was delayed until the patient demonstrated stimulus-evoked responses and clinical signs in keeping with satisfactory recovery from neuromuscular block. This may explain the lack of differences in the incidence of postoperative airway obstruction, hypoxemia, or duration of supplemental oxygen therapy in the PACU. The relatively small number of patients studied (n = 38) preclude us from concluding that it is safe to avoid the antagonism of residual neuromuscular blocking effects of mivacurium in all patients. The risks of emesis in the PACU may be judged to be of less concern than the risks of postoperative respiratory complications from residual neuromuscular block in the presence of factors such as preoperative airway obstruction, diminished respiratory reserve, atypical pseudocholinesterase, or after intense block [1,13]. These data would suggest that edrophonium may be used to advantage in these patients, as it provides for a more rapid recovery than neostigmine without increasing the incidence of postoperative emesis in the PACU [11,12].
This study can be criticized for not limiting the patient population to one operative procedure, and that the differences in the rate of recovery after the administration of the study drug may have unmasked the blinded nature of the study. However, the lack of significant differences in the type of operations performed suggests that this confounding factor did not affect the conclusions of our study. More importantly, the observer in the PACU was not present at the time of injection of the study drugs. Finally, the study can be criticized because of the use of different anticholinergic drugs (atropine and glycopyrrolate) to counter the muscarinic actions of the anticholinesterase drugs. Atropine, unlike the quaternary anticholinergic glycopyrrolate, can diffuse into the central nervous system, and is known to reduce emesis associated with morphine premedication . Thus, it is possible that the differing antiemetic effects are related to the anticholinergic drug rather than the anticholinesterase. Given the similarity in onset and duration of action of atropine and edrophonium, and of neostigmine and glycopyrrolate, these combinations have become the standard in clinical practice [1,15].
In conclusion, this pediatric study has demonstrated that the use of a neostigmine-glycopyrrolate combination to antagonize residual block from mivacurium is associated with an increase in vomiting in the PACU compared to spontaneous recovery or reversal with edrophonium. However, use of an edrophonium-atropine combination for reversal of residual mivacurium-induced block does not increase postoperative vomiting.
The authors are grateful to the Texas Scottish Rite Hospital for the support provided for this study. We thank Richard Browne, PhD, for his assistance in the randomization schedule and the nurses and pharmacists at the Texas Scottish Rite Hospital for Children for their patience during the study. We also gratefully acknowledge the help of C. T. Payne, BA, and A. E. Teufel, BA, for their assistance in data collection during the study, and Medfusion, Inc. for the loan of a pump for the study.
1. Bevan DR, Donati F, Kopman AF. Reversal of neuromuscular blockade. Anesthesiology 1992;77:785-805.
2. Ding YF, White PF. Use of mivacurium during laparoscopic surgery: effect of reversal drugs on postoperative recovery. Anesth Analg 1994;78:450-4.
3. King MJ, Milazkiewicz R, Carli F, Deacock AR. Influence of neostigmine on postoperative vomiting. Br J Anaesth 1988;61:403-6.
4. Boeke AJ, de Lange JJ, van Druenen B, Langemeijer JJM. Effect of antagonizing residual neuromuscular block by neostigmine and atropine on postoperative vomiting. Br J Anaesth 1994;72:654-6.
5. Grasela TH, McLesky CH, Walawander CA, et al. Postoperative nausea and vomiting and the effect of anticholinesterase agents [abstract]. Anesthesiology 1994;81(Suppl) A:1278.
6. Kao YJ, Mian T, McDaniel KE, et al. Neuromuscular blockade reversal agents induce postoperative nausea and vomiting [abstract]. Anesthesiology 1992;77(Suppl):A1120.
7. Hannallah RS, Broadman LM, Belman AB, et al. Comparison of caudal and ilioinguinal/iliohypogastric nerve blocks for the control of post-orchidopexy pain in pediatric ambulatory surgery. Anesthesiology 1987;66:832-4.
8. Fleming NW, Lewis BK. Cholinesterase inhibitors do not prolong neuromuscular blockade produced by mivacurium. Br J Anaesth 1994;73:241-3.
9. Bell CMA, Lewis CB. Effect of neostigmine on integrity of ileorectal anastomoses. BMJ 1968;1:587-8.
10. Janhunen L, Tammisto T. Postoperative vomiting after different modes of general anesthesia. Ann Chir Gynaecol 1972;61:152-9.
11. Naguib M, Abdulatif M, al-Gamdi A, et al. Dose-response relationships for edrophonium and neostigmine antagonism of mivacurium-induced neuromuscular block. Br J Anaesth 1993;71:709-14.
12. Goldhill DR, Whitehead JP, Emmott RS, et al. Neuromuscular and clinical effects of mivacurium chloride in healthy adult patients during nitrous oxide-enflurane anaesthesia. Br J Anaesth 1991;67:289-95.
13. Ostergaard D, Jensen FS, Jensen E, et al. Mivacurium-induced neuromuscular blockade in patients with atypical plasma pseudocholinesterase. Acta Anaesthesiol Scand 1993;37:314-8.
14. Dundee JW, Kirwan MK, Clarke RSJ. Anaesthesia and premedication as factors in postoperative vomiting. Acta Anaesthesiol Scand 1965;9:223-31.
© 1995 International Anesthesia Research Society
15. Cozanitis DA, Dundee JW, Merrett JD, et al. Evaluation of glycopyrrolate and atropine as adjuncts to reversal of nondepolarizing neuromuscular blocking agents in a true-to-life situation. Br J Anaesth 1980;52:85-9.