The Use of Midazolam for Persistent Postoperative Nausea and Vomiting T. DI FLORIO* Department of Anaesthesia, King Edward Hospital, Perth, Western Australia Key Words: ANAESTHESIA; COMPLICATIONS: nausea, vomiting, midazolam The management of postoperative nausea and vomiting in some patients remains a difficult problem. Many precautions can be taken perioperatively to decrease the likelihood of vomiting but once it occurs our pharmacological intervention relies mainly on antidopaminergic drugs. The successful use of midazolam in conjunction with antidopaminergic drugs to control postoperative vomiting is reported here. Case J A 38-year-old, 85 kg woman was scheduled as a day-case for laparoscopy for the investigation of infertility. Her previous general anaesthetics for tubal ligation and tubal reanastomosis were both complicated by postoperative nausea and vomiting. She was a smoker of 20 cigarettes per day. Her preoperative assessment was otherwise unremarkable. Temazepam 20 mg and metoclopramide 10 mg were given orally one hour before operation. Droperidol 2 mg and fentanyl 100 mcg were given intravenously immediately prior to induction of anaesthesia with propofol 100 mg. She was easy to ventilate by mask and intubation was facilitated by vecuronium 6 mg. General anaesthesia was maintained with nitrous oxide:oxygen (70:30) and enflurane 0.6%. During the thirty-minute anaesthetic she received Hartmann's solution 1000 ml and remained haemodynamically stable. An indomethacin suppository 100 mg was inserted at the end of the operation. She was reversed with neostigmine 2.5 mg and atropine 1.2 mg, extubated on her side and taken to the recovery room. In the recovery room she vomited once and complained of abdominal pain, for which she received pethidine 25 mg intravenously. Over the next thirty minutes she experienced no further pain or vomiting and remained haemodynamically stable and so was returned to the ward. *B.Sc., M.B.B.S., Anaesthetic Registrar. Address for Reprints: Dr. T. Di Florio, Department of Anaesthesia. Sir Charles Gairdner Hospital, Perth. W.A. 6109. Accepted for publication April 22, 1992 Anaesthesia and Intensive Care. Vo!. 20. No. 3, August, 1992
Approximately one hour after returning to the ward she became nauseated and began to vomit with a frequency of five to ten times per hour. She was given metoclopramide 10 mg intramuscularly. There was no improvement in the next 90 minutes, so droperidol 1.5 mg was given intramuscularly and intravenous fluids were commenced. She was still vomiting with the same frequency two hours later when she received prochlorperazine 12.5 mg intravenously. She was quite distressed at this stage and was also given lorazepam 2.5 mg orally which she subsequently vomited. Over the next 90 minutes there was no improvement and, because she was so distressed, she was given midazolam 3 mg intravenously followed by an infusion of 1 mglhour. Her nausea and vomiting ceased following the commencement of midazolam. She was monitored with a pulse oximeter in the High Dependency Unit while on the infusion of midazolam and she remained awake with Sp02 values greater than 96% on air. She was free from nausea except for two episodes of vomiting. The infusion was ceased after eight hours. Forty-five minutes after the cessation of midazolam, vomiting recommenced with a frequency of three to five times per hour. The midazolam infusion was recommenced and again her vomiting decreased to less than one episode per hour. The infusion was continued for a further seven hours and then ceased. Two hours later she complained of mild nausea for which she was given droperidol 1.5 mg intravenously. Following this she remained free of nausea and vomiting for the rest of her admission. Case 2 A 72-year-old 65 kg woman was booked for a left total hip replacement. She had hypertension controlled with metoprolol and had a previous anaesthetic for cholecystectomy without complications. One hour prior to operation she received temazepam 20 mg and metociopramide 10 mg orally. Papaveretum 10 mg was given intravenously prior to induction of anaesthesia with thiopentone 200 mg. Intubation was
facilitated by vecuronium 8 mg. Anaesthesia was maintained with nitrous oxide:oxygen (70:30) and enflurane 1.0%. Papaveretum 10 mg was given prior to incision and vecuronium 2 mg was given as required. Blood loss during the two-hour procedure was 600 ml and she received Hartmann's solution 2000 ml and Haemaccel 500 m!. She was extubated awake following reversal with neostigmine 2.5 mg and glycopyrrolate 0.4 mg. Approximately fifteen minutes after arriving in the recovery room she became nauseated and was given metoclopramide 10 mg intravenously. Over the next thirty minutes she vomited four times and so was given droperidol 1.5 mg intravenously. She was also given papaveretum 10 mg intramuscularly for pain. During the following hour she vomited six times. She received prochlorperazine 12.5 mg intramuscularly and was transferred to the high dependency unit where her nausea and vomiting persisted. In the High Dependency Unit she was given midazolam I mg intravenously followed by an infusion of 0.5 mg/hour. She vomited once over the next hour and then her nausea and vomiting ceased. Her Sp02 values were greater than 95% on air and she did not become sedated while on the midazolam infusion. She received papaveretum 15 mg intramuscularly as required without associated nausea. The infusion was ceased after six hours. She complained of nausea 35 minutes after stopping the infusion and vomited twice over the following hour. The infusion was recommenced and her nausea ceased after 45 minutes. Six hours later the infusion was ceased again with no further problems.
Case 3 A 74-year-old, 67 kg man was scheduled for a left total hip replacement. He had not had previous general anaesthetics and apart from hypertension, controlled by captopril, his preoperative assessment was unremarkable. Temazepam 20 mg and metoclopramide 10 mg were given orally one hour prior to the operation. Following preloading with Hartmann's solution 1000 ml an epidural catheter was placed at the L3-4 interspace via a 16-gauge Tuohy needle. A total of 20 ml of bupivacaine 0.5% plain solution was injected, in small aliquots, via the epidural catheter. This resulted in a dense block of the operative site. General anaesthesia was induced with propofol 200 mg and a size 4 laryngeal mask was inserted. Anaesthesia was maintained with nitrous oxide:oxygen (70:30) and enflurane 0.8%. During the 95 minute procedure a 400 ml blood loss was recorded. Haemaccel 500 ml and
Hartmann's solution 1000 ml were given during the case. He was taken to the recovery room awake and comfortable at the completion of the case. An epidural infusion of pethidine 15 mg/hour was commenced. He remained haemodynamically stable and pain-free over the next 45 minutes and was then transferred to the High Dependency Unit for management of the epidural infusion. Soon after arriving in the High Dependency Unit he became nauseated and was given prochlorperazine 12.5 mg intramuscularly. The nausea persisted and he was given droperidol 1.5 mg intravenously and the pethidine infusion was ceased. He remained nauseated and vomited five times over the following two hours. He was then given midazolam I mg intravenously and commenced on an infusion of 0.5 mg/hour and his nausea and vomiting ceased. The epidural pethidine infusion was recommenced and the midazolam infusion was continued for nine hours. Vomiting did not return following the cessation of midazolam. He did not become sedated and his Sp 02 values remained above 95% on air during the midazolam infusion. DISCUSSION
Benzodiazepines have been used by oncologists against anticipatory nausea and vomiting associated with chemotherapyl and to provide amnesia of the experience. Midazolam has been used in combination and as a sole agent in patients having chemotherapy and was found to be an effective antiemetic. 2 There are no reports of benzodiazepine use in postoperative nausea and vomiting. These cases suggest a place for benzodiazepines, in conjunction with antidopaminergics, in patients with refractory postoperative vomiting. In all three cases nausea and vomiting subsided with the introduction of midazolam. More importantly nausea was ablated by recommencing the midazolam infusion in the two cases in which nausea and vomiting returned after ceasing midazolam. It is also interesting that only low doses of midazolam (0.5-1.0 mg/hour) are required to produce an anti emetic effect. Sedation and respiratory depression were not observed at this dose. On the strength of these cases it would seem reasonable to consider a trial of midazolam in patients with persistent postoperative vomiting that is unresponsive to conventional antiemetics. Close monitoring of these patients, in a high dependency unit, during the infusion ofmidazolam is recommended even though the doses needed are low. Midazolam and lorazepam have been shown to AnaeSlhes/Q and Intensive Care.
20. ,Vo. 3. August. 1992
be effective anti emetics in chemotherapy associated nausea and vomiting. 2 ,3 The effectiveness of midazolam in these postoperative cases raises the possibility that lorazepam may also be useful in postoperative emesis. In some patients postoperative vomiting is a recurrent problem following each anaesthetic. These patients may benefit from premedication with a benzodiazepine, such as lorazepam, which lasts well into the postoperative period. The mechanism by which benzodiazepines have their beneficial effect in nausea and vomiting is not yet known. Benzodiazepines have sedative/ hypnotic, amnestic, anxiolytic, anticonvulsant and muscle relaxant effects. It may be that they are effective in anticipatory nausea and vomiting by decreasing anxiety and therefore the psychic input to the vomiting centre. The postulated mechanisms of action of benzodiazepines have included glycine mimetic inhibitory effects in the spinal cord and brainstem and the enhancement of the inhibitory effects of gamma amino butyric acid (GABA) in the brain. 4 It has now become clear, however, that the enhancement of GABA does not adequately explain all the actions ofbenzodiazepines. Recently it has been suggested that some of the effects of benzodiazepines may involve adenosine, 5 which is a neuromodulator causing central nervous system depression via adenosine type I (AI) receptors. 6 Benzodiazepines block the re-uptake of adenosine 7 leading to enhanced adenosinergic effect. This could have some relevance in explaining the possible anti emetic properties of benzodiazepines, since adenosine antagonists such as aminophylline are capable of causing vomiting. Aminophylline has been shown to partially reverse lorazepam sedation and it is postulated that it does this by antagonising adenosine. 8 The presence of adenosine uptake sites in areas of established importance in the production of nausea and vomiting would support the hypothesis that benzodiazepines have anti emetic actions by enhancing adenosinergic effects. Adenosine uptake sites and Al receptors have been shown to be present in the area postrema in the medulla. 9 Thus, adenosine may be a neuromodulator of dopaminergic input in the chemoreceptor trigger zone which is located in the area postrema. Evidence exists for the role of adenosine as a modulator of dopamine effects centrally. In the anterior pituitary, for example, dopamine (prolactin inhibiting hormone) and adenosine antagonists decrease prolactin release whereas adenosine agonists increase prolactin release. 10 A recent study has revealed the presence of both Al and adenosine type 2 (A 2) receptors on neurons Anaesthesia and Intensive Care, Vo!. 20, No. 3, August, 1992
in the neostriatum postsynaptic to the dopaminergic input. 11 Also nigrostriatal dopamine synthesis and release are decreased by adenosine via Al receptors. 12 So the neuromodulation of dopamine effects by adenosine may involve both postsynaptic inhibition of dopamine action as well as decreased synthesis and release of dopamine. The evidence presented suggests that benzodiazepines not only reduce the psychic input to the vomiting centre but by blocking the re-uptake of adenosine may lead to an adenosinemediated inhibition of dopamine synthesis, release and action in the chemoreceptor trigger zone. Benzodiazepines would therefore be expected to be synergistic with antidopaminergic drugs in the control of nausea and vomiting. The combination of midazolam and antidopaminergic drugs proved effective in each of these cases. Conventional anti emetics are often inadequate in the management of persistent postoperative nausea and vomiting. On the basis of these reports it seems reasonable to consider a trial ofmidazolam in patients with postoperative vomiting that is not responsive to conventional antiemetics. It is hoped that these cases will stimulate further research in this area. ACKNOWLEDGEMENT
I would like to thank Professor John Utting for his encouragement. REFERENCES
I. Grunberg SM. Advances in the management of nausea and vomiting induced by non-cisplatin containing chemotherapeutic regimens. Blood Reviews 1989; Dec 3(4):216-221. 2. Olynyk JK, Cullen SR, Leahy MF. Midazolam: an effective antiemetic for cytotoxic chemotherapy. Med J Aust 1989; 150:466. 3. Maher J. Intravenous lorazepam to prevent nausea and vomiting associated with cancer chemotherapy. Lancet 1981; i:91-92. 4. Richter JJ. Current theories about the mechanisms of benzodiazepines and neuroleptic drugs. Anesthesiology 1981; 54:66-72. 5. Phillis JW, O'Regan MH. Benzodiazepine interaction with adenosine systems explains some anomalies in GABA hypothesis. Trends in Pharmacological Science 1988 May; 9(5): 153-154. 6. Dunwiddie TV. The physiological role of adenosine in the nervous system. In: International Review of Neurobiology. Smythies JR, Bradley RJ, ed. Academic, New York 1985; 27:63-139. 7. Phillis JW, Siemens RK, Wu PH. Effects of diazepam on adenosine and acetylcholine release from rat cerebral cortex: further evidence for a purinergic mechanism of action of diazepam. Br J Pharmacol 1980; 70:341-348. 8. Wangler MA, Kilpatrick DS. Aminophylline is an antagonist of lorazepam. Anesth Analg 1985; 64:834-836.
9. Deckert J, Bisserbe JC, Klien E, Maranges PJ. Adenosine uptake sites in brain: Regional distribution of putative subtypes in relationship to adenosine Al receptors. J Neurosci 1988 July; 8(7): 2338-2349. 10. OndoJG, Walker MW, Wheeler DD. Central actions of adenosine on pituitary secretion of prolactin, luteinizing hormone and thyrotropin. Neuroendocrinology 1989; 49:654-658.
11. Alexander SP, Reddington M. The cellular localization of adenosine receptors in rat neostriatum. Neuroscience 1989; 28(3): 645-651. 12. Wood PL, Kim HS, Boyer WC, Hutchison A. Inhibition of nigrostriatal release of dopamine in the rat by adenosine receptor agonists: Al receptor mediation. Neuropharmacology 1989; 28( 1): 21-25.
Anaesthesia and Intensive Care. Vo!. 20. No. 3. August. 1992