Effect of Low-dose Droperidol on the QT Interval during and after General Anesthesia: A Placebo-controlled Study

After obtaining approval by the Institutional Review Board of the University of Texas Southwestern Medical Center at Dallas, 120 outpatients with American Society of Anesthesiologists physical status I–III who were scheduled to undergo otolaryngologic surgical procedures provided informed consent and were enrolled in this two-part clinical study. Patients with clinically significant cardiac disease, atrioventricular conduction delays or bundle branch blocks, a history of alcohol or drug abuse within the past 3 months, or morbid obesity (body mass index > 40 kg/m²) were excluded from participation in this study. Patients who had taken any antiemetic medication within 24 h before surgery or were pregnant or experiencing menstrual symptoms were also excluded. A detailed medical history and demographic information, including age, height, weight, and American Society of Anesthesiologists physical status, and history of previous PONV, motion sickness, or smoking were obtained from each patient.

A 12-lead electrocardiogram was obtained in the preoperative holding area before any preoperative medications were administered. At the time of arrival in the operating room, an automatic blood pressure cuff, five-lead electrocardiograph, capnograph, and pulse oximeter were applied. Mean arterial pressure, heart rate, end-tidal carbon dioxide, and oxygen saturation were measured at 1- to 5-min intervals throughout the intraoperative study period. General anesthesia was induced with 1.5–2 mg/kg propofol and 0.5–1 μg/kg intravenous remifentanil, and 0.6 mg/kg intravenous rocuronium was used to facilitate tracheal intubation. After intubation, anesthesia was maintained with 4% inspired desflurane (in an air–oxygen mixture) and a remifentanil infusion at 0.125 μg · kg⁻¹· min⁻¹. Approximately 15 min before the surgical incision, 60 patients received one of three study medications based on a computer-generated random number table. The study medication consisted of either saline (control) or 0.625 or 1.25 mg droperidol, prepared in identical-appearing 2-ml syringes by an individual not involved in the study. The patients were maintained under stable anesthetic conditions without surgical stimulation for a minimum 10-min interval after injection of the study medication.

The standard electrocardiographic lead II was continuously recorded at a paper speed of 50 mm/s and an amplification of 0.1 mV/mm for 2–3 min before and every minute for a minimum of 10 min after the injection of the study medication. In both the operating and recovery rooms, the cardiac rhythm was continuously monitored using a standard lead II. The electrocardiographic measurements were independently performed by two investigators who were unaware of which of the three treatment groups the patient had been assigned to. Heart rate was calculated from the three R-R intervals preceding the measured QT intervals. The QT intervals were measured manually from the onset of the QRS complexes to the end of the T wave (defined as the intersection of the isoelectric line and the tangent of the maximal downward limb of the T wave). Each registered measurement was a mean of two consecutive QT intervals. The QT interval was adjusted for the patient’s heart rate using the formula of Bazett17(QTc = QT/R-R^0.5). If the T-wave amplitude was too low (< 0.1 mV), the lead was excluded from the analysis.

In a follow-up study, 60 patients who had received either droperidol (1.25 mg intravenous) or saline (2 ml) after induction of anesthesia underwent repeated 12-lead electrocardiographic evaluations after arrival in the recovery room and at subsequent 1- and 2-h intervals. The QTc interval was automatically calculated by the Marquette® MAC® VU device (Milwaukee, WI) using the formula of Bazett. Finally, the incidences of nausea and vomiting/retching (requiring a “rescue” antiemetic drug), restlessness, dizziness, and headache were assessed at 4- and 24-h intervals after surgery in the initial study population.

Results are presented as mean (± SD), percentage, and number. A power analysis (α= 0.05, β= 80%) was performed before the initiation of the study using the software nQuery Advisor® version 1.0 (Statistical Solutions, Saugus, MA). This analysis suggested that 20 patients in each group should be adequate to detect a 15% or greater prolongation of the QTc interval in the droperidol groups versus  the saline group assuming a mean QTc of 400 ms with an SD of 40 ms in the control group.16These electrocardiographic data were analyzed using analysis of variance, with a Bonferroni correction applied for multiple comparisons. The changes in the QTc intervals recorded on the serial electrocardiographic tracings were analyzed using repeated-measures analysis of variance. All QTc data were subjected to normality testing using the D’Agostino skewness, kurtosis, and omnibus tests after performing a log transformation of these data. The incidences of postoperative side effects were analyzed using chi-square test with Fisher exact test where appropriate. A P  value of less than 0.05 was considered statistically significant.

The three study groups were similar with respect to their demographic characteristics, durations of anesthesia, and total dosages of anesthetic and analgesic drugs, as well as intravenous fluid administration during the operation (table 1). The recovery times from the end of anesthesia to awakening (i.e. , eye opening on verbal command), orientation (i.e. , correctly stating name, date of birth, and location), and recovery room discharge were similar among the three groups. Although the 0.625- and 1.25-mg droperidol groups had lower overall incidences of nausea and vomiting compared with the saline group (15% and 20% vs.  45%, respectively), this difference only achieved statistical significance when the two droperidol groups were combined (vs.  saline), because of the small group sizes.

The mean maximum QTc prolongation values in the 0.625- and 1.25-mg droperidol groups were 15 ± 40 and 22 ± 41 ms, respectively, at 3–6 min after administration of the study medication compared with 12 ± 35 ms in the saline group. The average QTc intervals were nonsignificantly prolonged in all three groups at 10 min after intravenous injection of the study medication (table 2). There were no differences among the three study groups with respect to the incidence of QTc prolongation greater than 10% of the baseline value. However, two patients in each of the droperidol groups had QTc prolongation lasting more than 60 ms during the 10-min observation period. The maximum prolongation of the QTc interval was 133 ms in a patient receiving 1.25 mg intravenous droperidol. There were no significant differences among the three groups with respect to hemodynamic changes during the 10-min observation period (table 3). In addition, postoperative side effects were also not significantly different among the three treatment groups (table 3). No electrocardiographic rhythm disturbances were observed during the continuous electrocardiographic monitoring in the operating room or in the postanesthesia care unit before discharge.

In the follow-up study involving preoperative and postoperative 12-lead electrocardiographic monitoring, the mean QTc intervals and changes in QTc (from baseline values) were similar in patients who did or did not receive 1.25 mg intravenous droperidol (table 4). However, the QTc was significantly prolonged (14–16 s) in both groups at the time of arrival in the recovery room compared with the baseline values (P < 0.01). No arrhythmias were observed on the 12-lead electrocardiographic tracings at any of the testing intervals.

Use of small doses of droperidol (0.625–1.25 mg intravenous) for antiemetic prophylaxis was not associated with clinically significant prolongation of the QT interval during or immediately after surgery. Although these small antiemetic doses of droperidol were able to produce significant (> 10%) prolongation of the QT interval in 10–15% of the study patients, this effect was short lasting, and there was no evidence of any adverse effect on cardiac rhythm. The residual effect of general anesthesia was also associated with a small but statistically significant prolongation in the QTc at the time of arrival in the recovery room, both in those who did and in those who did not receive intraoperative droperidol. Given the lack of any direct evidence that use of droperidol in the therapeutic dosage range (0.625–1.25 mg intravenous) was associated with any clinically significant effects on the electrocardiogram or adverse cardiovascular outcomes, these data support the position that the current FDA recommendations for additional perioperative electrocardiographic monitoring are unnecessary and, in fact, wasteful of valuable healthcare resources.

A previous study involving higher doses of droperidol (0.1–0.25 mg/kg) suggested that this butyrophenone could produce dose-related prolongation of the QT interval.15Interestingly, despite the use of these high-doses of droperidol as part of a neurolept anesthetic technique for more than 30 yr,18there has not been a single report of a serious cardiac arrhythmia during or after anesthesia in the peer-reviewed literature. In carefully examining the specific case reports that lead the FDA to place a black box warning on the use of droperidol, investigators have concluded that “there is no evidence of a cause-and-effect relationship,”11and in most of the cases, there was “a more plausible explanation for the etiology of the described event.”19 

Food and Drug Administration officials have argued that the Food, Drug and Cosmetics Act allows that “reasonable evidence of association between drug and serious adverse event is evidence enough to place a warning in a drug label.”5Curiously, despite ample evidence that the popular 5-hydroxytryptamine type 3 antagonists dolasetron and ondansetron prolong the QTc interval in a dose-related fashion for up to 4 h,20these drugs have not been issued black box warnings. In defense of the FDA decision, Shafer6argued that the mere fact that there is a mechanistic basis for droperidol to prolong QT21,22“paints a convincing story of genuine risk from droperidol for a very small number of patients.” Because similar interactions have been reported between the 5-hydroxytryptamine type 3 antagonists and human cardiac ion channels,23it is unclear why this same reasoning would not also be applied to this popular class of antiemetic drugs.

Using a mathematical model to estimate the average prolongation of the median QTc interval by small doses of droperidol, Zhang et al.  16reported that 0.625–1.25 mg intravenous droperidol would be expected to prolong the median QTc interval by 9–18 ms beyond the saline “control” values. The current clinical findings confirm these previous predications. However, even in the two patients displaying more than 100 ms prolongation of the QTc interval, there was no evidence of any ectopic electrocardiographic activity or changes in their cardiac rhythm. Of note, the maximum effect of droperidol on the QTc interval during general anesthesia occurred at 3–6 min after the intravenous bolus injection. Therefore, the FDA recommendation for prolonged electrocardiographic monitoring after administration of droperidol seems illogical.

This two-part study can be criticized because the group sizes were too small to detect rare cardiovascular events. However, we were able to demonstrate that even small, antiemetic doses of droperidol produce the predicted effect on the QTc interval.16Another potential criticism of this study relates to the method used to evaluate droperidol-induced QT prolongation. A recent analysis by Sadanaga et al.  24found that the overall accuracy of predicting QT prolongation was significantly increased (98.1%) when using the formula of Bazett compared with using any fixed QTc cutoff value. In fact, the formula of Bazett compared favorably to the formula of Friderica (98.7%) in improving the accuracy of evaluating QT prolongation.24Other investigators have similarly concluded that the formula of Bazett is at least as good as other proposed heart rate corrections for the QT interval.25,26Finally, these findings examined the effect of low-dose droperidol on the QTc interval of anesthetized patients and may not apply to situations where the drug is used for the treatment  of PONV.

In light of these findings, we urge the FDA and the drug manufacturer (Akorn, Inc., Buffalo Grove, IL) to modify the black box warning to reflect a more reasonable position with respect to the clinical use of low-dose droperidol (< 2.5 mg intravenous) for the treatment and prevention of PONV. As stated in the consensus guidelines for managing PONV sponsored by the manufacturer of a competing 5-hydroxytryptamine type 3 antagonist,27droperidol would have been the panel’s “overwhelming first choice for the prevention of PONV” were it not for the black box warning. From a societal cost–benefit perspective,28,29low-dose droperidol should be made available to practitioners for routine clinical use without the medical legal concerns engendered by the current black box warning in the droperidol package insert. As pointed out by Roden30in his recent review article on drug-induced prolongation of the QT interval, rare poorly understood side effects occur with many highly effective drugs, and “restrictions in the use of these drugs may harm more patients than it would help.”

We conclude that 0.625–1.25 mg intravenous droperidol produces clinically insignificant prolongation of the QT interval and that the peak effect occurs less than 6 min after the intravenous bolus injection. Therefore, the current FDA recommendations for additional preoperative and postoperative electrocardiographic monitoring after administration of low-dose droperidol are unwarranted.