To the Editor:—
We congratulate Dr. Arnberger et al. 1on their recent work using nerve stimulation at the P6 acupuncture site to reduce the incidence of postoperative nausea and vomiting (PONV). This is obviously an important clinical work that could influence the management and multimodal approach to PONV. However, we are concerned about the validity of the conclusions secondary to the disproportionately higher number of smokers in the treatment group. The reasons for this concern are as follows.
According to the recently published guidelines from the Society for Ambulatory Anesthesia, the risk factors for PONV fall into four criteria. These include patient risk factors, anesthesia risk factors, surgical risk factors, and type of surgery.2The three most important patient risk factors are female gender, prior history of PONV or motion sickness, and nonsmoking status. In Arnberger et al. ,1they controlled very well for the gender (all patients female, as mentioned in the editorial) and the history of motion sickness risk factors (P = 0.37). Of particular concern is the higher percentage of smokers in the treatment P6 group. Any descriptive statistics with a P value less than 0.15 should be accounted for as a potential confounder, especially if the variable is known to influence the dependent variable. It has been demonstrated in multiple randomized controlled studies that smoking decreases the incidence of PONV and having a higher percentage in the treatment group has a strong possibility of affecting the conclusion. A stratified random sample procedure would have been a much better option to achieve equal numbers of smokers in each group. Given that this was not the case, to account for this potential confounder, a logistic regression analysis should have been used to ensure that the higher percentage of smokers didn’t affect the clinical outcome in this study.
Several studies have shown that smoking is an important risk factor in the development of PONV. A Canadian study retrospectively reviewed the charts for 16,000 patients postoperatively for over 2 yr in four teaching hospitals using a multiple logistic regression analysis to show that nonsmokers were 1.79 times more likely to develop PONV.3Stadler et al. 4had a sample size three times higher than Arnberger et al. ,1and showed that nonsmoking status was highly significant for both nausea and vomiting (P = 0.0070 and 0.0074). In a Korean model derived from more than 5,000 patients, a multiple regression analysis showed that smoking status reduced PONV with an odds ratio of 2.0.5
The exact mechanism of smoking reducing the incidence of PONV is not fully explained; however, there are several potential explanations. Chronic exposure to one of the chemicals in tobacco may desensitize the patient to anesthetic gas or may a have direct antiemetic effect.6Another explanation is that the cytochrome p450 may be up-regulated in chronic smokers, which may increase metabolism of anesthetic agents and result in less PONV.7
In conclusion, we believe that the results of this study may be hampered by the uneven distribution of smokers in the treatment group. Our intent is to encourage further studies in the area and/or further statistical analysis performed to truly elucidate if this amazing effect is caused by this seemingly simple maneuver.
*The Ohio State University, Columbus, Ohio. hamdy.elsayed-awad@osumc.edu
