Reversal of Neuromuscular Blockade

In Response: We thank White and Watcha for their careful review of our paper [1] and for pointing out an error in our Table 2. We apologize for the mistake. In Table 2, the percentage of patients in Group A (neuromuscular blockade not reversed) for vomiting in the postanesthesia care unit (PACU) 1-2 h after anesthesia should be 3% (2 of 80 patients), not 13%. Accordingly, 3% vs 0% (or 2 vs 0 patients) is the correct comparison, and there is no statistical difference between the groups. To detect early differences in our study postoperative nausea and vomiting (PONV) was recorded at 1-h intervals (0-1, 1-2, and 2-3 h after anesthesia) during the PACU stay. It seems that PONV has been recorded and reported in a slightly different way in different studies. In some studies [2], patients who vomited are also listed as having nausea, which makes it difficult to determine the total incidence of PONV (nausea and vomiting). In our studies [1,3,4] and in a well cited benchmark study of Bellville et al. [5], nausea and vomiting are listed separately as "none," "nausea," or "vomiting" (i.e., if patients had both nausea and vomiting, they were listed as having had vomiting). When listed separately, the total incidence of PONV is the sum of nausea and vomiting (e.g., in [1]). If the primary end point is "none" (i.e., neither nausea nor vomiting), then the percentage of patients having PONV is 100 - the percentage having "none." When one patient is registered in only one category, nonparametric statistics (e.g., chi squared test or Fisher's exact test) can be used for comparisons between groups. Because methods to register PONV are not standardized, one must carefully describe (and readers must understand) how PONV was analyzed to prevent misinterpretation of results [6]. To keep the title as brief as possible, details of patient material (e.g., that surgery was abdominal hysterectomy) and anesthesia were given in the abstract and in Methods. We have just finished another controlled, randomized, double-blind study of 100 patients undergoing outpatient gynecological laparoscopy. In this new, out-patient study, the incidence of PONV in the PACU (0-90 min after anesthesia) was 14% and 17%, and overall PONV was 48% and 50% in patients with mivacurium neuromuscular blockade reversed and in patients with no reversal, respectively. According to data obtained in controlled, randomized, double-blind studies, the omission of reversal of mivacurium neuromuscular block with neostigmine does not decrease the incidence or severity of PONV in patients undergoing inpatient or outpatient gynecological operations. A recent review of literature [7] indicates that evidence is growing to support the use of combinations of antiemetics to prevent and treat PONV. For patients with intractable vomiting (e.g., young women with a history of motion sickness and PONV), the authors' approach is to use propofol for both the induction and maintenance and a small dose of droperidol (0.5-1.0 mg) and ondansetron (4 mg) given toward the end of the surgery to prevent PONV. Jaromir Hovorka, MD, PhD Kari Korttila, MD, PhD Kaisa Nelskyla, MD Department of Anaesthesia; Helsinki University Central Hospital; Hyks 00029, Finland