Midazolam reduces vomiting after tonsillectomy in children

William M. Splinter MDFRCPC 0 Helen B. MacNeill MDFRCPC 0 Eugene A. Menard MDFRCPC 0 Elliot J. Rhine MDFRCPC 0 David J. Roberts MDFRCPC 0 Marion H. Gould MDFRCPC 0 0 From the Department of Anaesthesia,Universityof Ottawa and Children's Hospital of Eastern Ontario , Ottawa , Ontario, Canada K1H 8L1. Department of Anaesthesia,Children's Hospital of Eastern Ontario , 401 Smyth Rd., Ottawa , Ontario, Canada , KIH 8LI. Presented in part at the 49th Annual Meeting of the Canadian Anaesthetist'sSociety, Toronto , June 1992. Accepted for publication 4th November , 1994 The purpose o f this study was to assess the effect o f midazolam on vomiting after tonsillectomy in children. We compared 215 children aged 1.5-14 )Jr undergoing tonsillectomy or adenotonsillectomy under general anaesthesia with nitrous oxide and halothane. In a double-blind fashion the subjects were administered either placebo or midazolam 75 lag" kg - t iv after induction o f anaesthesia. After the operation, the number o f emetic episodes and the length o f stay in hospital were recorded. The groups were similar with respect to age, weight, sex, mode o f induction, duration o f anaesthesia, surgicalprocedure, opioid administration and length o f stay in the PAR and the Day Care Surgical Unit. The 108 midazolam-treated children had a lower incidence (42% vs 57%) o f vomiting than the placebo group, P < 0.02. The placebo group had a higher incidence (9% vs 2%) o f unscheduled admissions to hospital due to nausea and vomiting, P < 0.05. It is concluded that midazolam administered intravenously to children intraoperatively reduces vomiting after tonsillectomy. Cette dtude vise d dvaluer chez les enfants l'effet du midazolam sur les vomissements post-amygdalectomie. Nous dtudions 215 enfants dont l'age se situe entre 1,5 et 14 ans soumis ~ une amygdalectomie ou Zt une addno-amygdalectomie sous anesthdsie gdndrale ~ l'halothane et au protoxyde d'azote. Les sujets refoivent en double aveugle soit un placebo soit du midazolam 75 lag" kg - t aprks l'induction de l'anesthdsie. Aprbs l'intervention, le hombre d~pisodes dmdtiques et la durde du sdjour ?t - l~@ital sont enregistrds. Les groupes sont comparables pour l'dge, le poids, le sexe, le mode d~nduction, la durde de I'anesth~sie, l'intervention chirurgicale, l'administration de morphinique et la dur~e du s~jour en salle d'op~ration et h l'unitd de chirurgie d'un jour. Les 108 enfants trait~s au midazolam vomissent moins (42% vs 57%) que le groupe placebo. L'incidence des admissions impr~vues d 17~tpital pour nausdes est vomissements est plus dlevde clans le groupe placebo (9070 vs 2%, P < 0,05). Nous concluons que l'administration iv de midazolam pendant l'intervention diminue l'incidence des naus~es et des vomissements aprbs l'amygdalectomie. Vomiting is an important, common and unpleasant sequel of paediatric general anaesthesia. Twenty per cent of children aged one to five years and 33% ages > six years vomit ~ after general anaesthesia. The reported incidence of vomiting after tonsillectomy in children is as high as 75%. 2 Persistent vomiting may result in expensive delays in discharge from hospital and unscheduled admissions and readmissions. 3,4 More important, prolonged postoperative vomiting (POV) may cause dehydration, metabolic derangements, seizures and if mismanaged, death. 5.6 The aetiology of postoperative vomiting is multifactorial. A variety of means has been employed to prevent and treat this problem. Established antiemetic drugs, such as droperidol have been shown to be effective.7 Benzodiazepines, such as diazepam, lorazepam and midazolam, are usually administered for their anxiolytic, amnestic and sedative effects. Recently, the administration of these drugs has been associated with a decrease in nausea and vomiting among ontology patients who have received chemotherapeutic agents.S-~~Khalii et al. ~1 studied the effect of lorazepam/v on POV after ophthalmic surgery in children. They observed a decrease in POV in patients treated with lorazepam. J2fis research has lead one to question whether other benzodiazepines may diminish POV. Recently, we observed that midazolam has antiemefic effects which are similar to droperidol among children undergoing strabismus surgery. 3 There have not been any studies of the perioperative antiemetic efficacy of midazolam Methods With parental consent and Hospital Ethics Committee approval we studied 215 ASA physical status I or II children aged 1.5-14 yr undergoing tonsillectomy or adenotonsillectomy. Subjects were excluded if they had an allergy to a study drug or a non-study anaesthetic was deemed more appropriate. Subjects did not receive a premedicant. After the placement of standard monitors, patients were sedated with N20, and anaesthesia was induced either by inhalation with halothane or with thiopentone 6 m g - k g -~ iv. After induction of anaesthesia, subjects were administered vecuronium 80 lag- kg -~, fentanyl 1.5 lag. kg -I, and a study drug (midazolam 75 lag. kg -I or placebo) in a random, blocked, double-blind fashion. Anaesthesia was maintained with 70% N20 and 0.75-1.5% halothane and titrated to maintain H R and SBP within normal limits. Upon completion of surgery, neuromuscular blockade was reversed with atropine 20 lag-kg -l and neostigmine 60 lag. kg -l. The tracheas were extubated after spontaneous ventilation had resumed and before the return of airway reflexes. After the operation, patients received fentanyl 0.5 lag. k g - ' iv q5min in the post-anaesthetic recovery room (PAR) if the modified C H E O P S pain score was >6.12 In the Day Care Surgical Unit (DCSU), acetaminophen elixir 10 lag" kg -~ a n d / o r codeine 1 mg. kg -I, po, pr, or im, was administered prn for pain. Excessive vomiting in the PAR or DCSU, that is vomiting >twice was treated with dimenhydrinate 1 mg. kg -t /v. The length of stay in hospital after the operation was recorded. (The minimum stay after tonsillectomy in our institution at the time of this study was 240 min.) Parents were contacted the day after surgery to ascertain the incidence of vomiting and other adverse events. Data were compared with paired and unpaired t tests, Mann-Whitney-U test, Chi-square analysis or Fisher Exact test, where appropriate. Acceptable alpha error was 0.05. For sample size calculation the accepted beta error was 0.2, the predicted incidence of vomiting in the placebo group and midazolam-treated group was 65% and 45%, respectively. TABLE I Demographic data TABLE !1 Anaesthesia and recovery times Number of episodes of emesis. The 108 midazolam-treated children had a lower incidence (42% vs 57%) of vomiting than the placebo group, P < 0.02, Chi-square analysis. The number of times a patient vomited was greater in the placebo group (Figure). The incidence of unscheduled admissions to hospital due to vomiting was higher in the placebo group than the midazolam-treated group, 9% vs 2%, P < 0.05, Fisher exact test. No patient r (...truncated)