John V. Donlon

Metoclopramide reduces the incidence of vomiting after tonsillectomy in children.

The efficacy of intravenous metoclopramide in controlling vomiting in children after tonsillectomy was determined in a prospective randomized, double-blind investigation. One hundred two unpremedicated, ASA physical status I or II children between the ages of 1 and 15 yr who were undergoing surgical removal of the tonsils, with or without adenoidectomy, were studied. Anesthesia was induced either with halothane, nitrous oxide, and oxygen by mask or by intravenous thiopental and was maintained with halothane, nitrous oxide, oxygen, and intravenous morphine (0.1 mg/kg). Each child randomly received either 0.15 mg/kg of metoclopramide or saline solution placebo intravenously after transfer to the postanesthesia care unit. All episodes of vomiting were recorded for 24 h after completion of surgery. The incidence of vomiting in the saline solution group was 70%, compared with 47% in the metoclopramide group (P = 0.026). The authors conclude that the administration of intravenous metoclopramide in a dose of 0.15 mg/kg on arrival in the postanesthesia care unit significantly decreases the incidence of vomiting in children after tonsillectomy.

Human dose-response curves for neuromuscular blocking drugs: a comparison of two methods of construction and analysis.

This study was done to demonstrate the validity of the log-probit method of analyzing dose–response data and to test the accuracy of the incremental dose method for generating dose–response data for pancuronium and d-tubocurarine. During balanced general anesthesia, cumulative dose–response data were obtained from ten patients each for pancuronium and d-tubocurarine. Dose–response data were also obtained using a single bolus injection method in 46 patients given pancuronium and 27 patients given d-tubocurarine. Evoked thumb adduction response was measured using an FT-10 force-displacement transducer and recorded on a Grass®-7 polygraph. Dose–response data were plotted on log-probit paper and analyzed by the Litchfield-Wilcoxon approximation method to determine mean ED50 and ED95 for each muscle relaxant. The dose–response data were also plotted on arithmetic scales and analyzed by the linear regression method to determine ED50 and ED95. These results were compared with those obtained by use of the log-probit method. The validity of the incremental dose method of obtaining dose–response curves was determined by comparison with the dose–response curves obtained using the conventional single bolus injection method. For each neuromuscular relaxant studied, the mean dose–response curve obtained by the cumulative dose method was not statistically significantly different from the bolus method dose–response curve. Comparing methods of analysis, the log-probit method yields results that are statistically not different from those obtained by use of conventional linear regression analysis. The log-probit plot is a simple, accurate, appropriate approach for analyzing neuromuscular relaxant dose–response data. The log-probit curve yields results that are not statistically significantly different from those of the linear regression method, and is a better fit to the data at the clinically important extremes (85–99 per cent) of the response scale. For pancuronium and d-tubocurarine, the incremental dose method is efficient and accurate for generating dose–response curves.

A new approach to the study of four nondepolarizing relaxants in man.

&NA; Relative potency and recovery time of four commonly used nondepolarizing agents, pancuronium, gallamine, d‐tubocurarine, and dimethyltubocurarine, were studied in 36 class‐I adult patients, using incremental dose‐response curves plotted on log‐probit paper. The authors propose this method of comparison for consideration in establishing relative potency and recovery time for such drugs.

Single-Dose Ondansetron Prevents Postoperative Vomiting in Pediatric Outpatients

This randomized, double-blind, parallel-group, multicenter study evaluated the safety and efficacy of ondansetron (0.1 mg/kg to 4 mg intravenously) compared with placebo in the prevention of postoperative vomiting in 429 ASA status I-III children 1-12 yr old undergoing outpatient surgery under nitrous oxide- and halothane-based general anesthesia. The results show that during both the 2-h and the 24-h evaluation periods after discontinuation of nitrous oxide, a significantly greater percentage of ondansetron-treated patients (2 h 89%, 24 h 68%) compared with placebo-treated patients (2 h 71%, 24 h 40%) experienced complete response (i.e., no emetic episodes, not rescued, and not withdrawn; P < 0.001 at both time points). Ondansetron-treated patients reached criteria for home readiness one-half hour sooner than placebo-treated patients (P < 0.05). The age of the child, use of intraoperative opioids, type of surgery, and requirement to tolerate fluids before discharge may also have affected the incidence of postoperative emesis during the 0- to 24-h observation period. Use of postoperative opioids did not have any effect on complete response rates in this patient population. We conclude that the prophylactic use of ondansetron reduces postoperative emesis in pediatric patients, regardless of the operant influential factors. Implications: Postoperative nausea and vomiting often occur after surgery and general anesthesia in children and are the major reason for unexpected hospital admission after ambulatory surgery. Our study demonstrates that the prophylactic use of a small dose of ondansetron reduces postoperative vomiting in pediatric patients. (Anesth Analg 1997;85:538-45)

Interaction of Intraocular Air and Sulfur Hexafluoride withNitrous Oxide

The diffusion dynamics of intravitreal gas bubbles injected during retinal reattachment procedures were studied using a mathematical model. This model predicts the effect of 70 per cent nitrous oxide anesthesia on the volume of the intravitreal bubble. The calculations indicate that when 70 per cent nitrous oxide administration is continued following intravitreal gas injection, there is a rapid, almost threefold increase in the volume of the injected bubble. When nitrous oxide is discontinued at the time of intravitreal injection, a maximum initial bubble expansion of only 35 per cent occurs. If nitrous oxide is discontinued 15 minutes prior to intravitreal injection, the mathematical model indicates that, at most, there will be a 15 per cent expansion of the bubble volume. The model calculations indicate that there is little difference in intravitreal bubble volume whether air of 100 per cent oxygen is used during the anesthetic. The two major factors that influence intravitreal bubble volume are the mixture of air or SF6 injected and the pattern of nitrous oxide use during anesthesia. These factors can be controlled. The importance of bubble volume changes on intraocular pressure and retinal blood flow also depend on other factors such as scleral rigidity, blood pressure, the presence of glaucoma, and the size of the injected gas bubble relative to the total vitreal volume.

Sublingual Hematoma following Difficult Laryngoscopy

A 71-year-old man weighing 75 kg was scheduled for elective repair of a detached retina under general anesthesia. Pertinent past medical history included cervical arthritis and arthritic limitation of the temporomandibular joint. He had received no medication, including aspirin, for his arthritis for more than 1 year. Premedication consisted of meperidine, 50 mg IM, promethazine, 25 mg IM, and atropine, 0.4 mg IM. After induction of anesthesia with thiamylal, 350 mg IV, he was given succinylcholine, 100 mg IV. Although a good airway was maintained, four independent attempts to visualize even the posterior commissure were unsuccessful. These four attempts were made during a time interval of less than 4 minutes and the patient therefore required only the original dose of succinylcholine. Both Miller #3 and MacIntosh #3 laryngoscope blades were used. The main difficulties were an anterior placement of the larynx and limited mobility of the mandible and cervical spine. The patient was then allowed to breathe spontaneously and awaken. The right naris was anesthetized with a 4% cocaine solution and blind nasotracheal intubation was accomplished on the first attempt. The case then proceeded uneventfully. At the termination of the 3-hour procedure, a massive right sublingual hematoma was noted. The visible extent of this hematoma was such as to make the tongue protrude from the mouth and touch the hard palate on the right. Our first concern in this waking patient was that the hematoma might cause airway obstruction. The endotracheal tube was therefore left in place for the next 3 hours until the patient was fully awake and the size of the hema-

Cumulative Dose-response Curves for Gall amine: Effect of Altered Resting Thumb Tension and Mode of Stimulation

Neuromuscular blockade studies often use the thumb adductor twitch response to ulna nerve stimulation. Two factors that may influence the results are the resting muscle tension (initial fiber length) and the pattern of the stimulus waveform. This study was undertaken to improve understanding of the effect of these factors and lead to better-controlled study conditions and more consistent data. During nitrous oxide barbiturate-narcotic anesthesia in 10 normal adult patients, as the resting thumb tension was increased from 50 to 200 g, the evoked thumb adductor twitch response (Grass stimulator, 0.25 Hz) was augmented by 28%. There was only a 2.5% increase in the evoked (developed) tension when the resting tension was further increased from 200 to 300 g. Developed tension at 50 g was significantly (p < 0.001) less than at the other resting tensions. The developed tension at 100 g was also significantly (p < 0.05) less than at resting tensions of 200 or 300 g.Cumulative dose-response curves for gall amine in nine patients were not significantly altered by increasing resting tension from 50 to 200 g. Biphasic (Block-Aide monitor) or single square wave (Grass stimulator) stimuli wave forms in nine normal adult patients yielded gall amine dose-response curves that were not statistically different. The muscle response to biphasic stimulation during a non-depolarizing blockade was not affected by the average muscle refractory period.Because of the significantly lower developed tension at resting tension settings of 50 to 100 g, a practical consideration during neuromuscular function studies would be to have the resting thumb tension adjusted and rechecked for each patient and kept within the 200–300-g range to ensure maximum uniform developed tension. The type of stimulus wave form selected will not affect results as long as it is used consistently throughout the study.

Acquired QT interval changes and neck dissections

STUDY OBJECTIVE To determine if acquired long QT syndrome following right or left, radical or modified, neck dissections result in malignant arrhythmias or deaths. DESIGN Prospective study. SETTING Inpatient head and neck service of the Massachusetts Eye and Ear Infirmary. PATIENTS 69 patients who underwent extensive neck surgery, without congenital long QT syndrome, medications known to prolong the QT interval, preoperative ventricular arrhythmias, or electrolyte abnormalities. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Preoperative and postoperative electrolytes were evaluated. Preoperative and postoperative electrocardiograms and QT intervals were evaluated. Continuous intraoperative and 10- to 12-hour postoperative monitoring of lead II or V5 were evaluated. Twenty-six patients (Group 1) underwent either right radical neck dissection or modified right radical neck dissection, 25 patients (Group 2) underwent either left radical neck dissection or modified left neck dissection, and 18 patients (Group 3) underwent extensive neck surgery without radical or modified neck dissection. Postoperatively, 38 patients (19 Group 1, 11 Group 2, and 8 Group 3 patients) developed a QT interval corrected for heart rate (QTc) of greater than 440 milliseconds. Repeated measures analysis of variance, comparing preoperative and postoperative QTc showed a statistically significant preoperative to postoperative change, but no significant difference among the three groups. No malignant arrhythmias or deaths were recorded in any of the three groups. CONCLUSIONS Acquired long QT syndrome following radical neck dissection, without congenital, metabolic, or pharmacologic disturbance, is unlikely to trigger malignant arrhythmias, as previously reported for right radical neck dissection.