Anesth Analg 2006;103:70-75
© 2006 International Anesthesia Research Society
doi: 10.1213/01.ane.0000221435.14002.4c
AMBULATORY ANESTHESIA
Section Editor:
Peter S. A. Glass
Recent Smoking Behavior and Postoperative Nausea and Vomiting
Francis Whalen, MD,
Juraj Sprung, MD,
Christopher M. Burkle, MD,
Darrell R. Schroeder, MS, and
David O. Warner, MD
From the Anesthesia Clinical Research Unit, Departments of Anesthesiology and Health Sciences Research, Mayo Clinic, Rochester, Minnesota.
Address correspondence and reprint requests to Francis Whalen, MD, 200 1st St SW, Rochester, MN 55905. Address e-mail to whalen.francis{at}mayo.edu.
 |
Abstract
|
|---|
The risk of postoperative nausea and vomiting (PONV) is reduced in cigarette smokers by unknown mechanisms. If protection is related to an acute effect of smoke constituents, smokers with the most recent exposure to cigarette smoke would be most protected. We tested the hypothesis that in cigarette smokers, postoperative nausea is correlated with recent exposure to cigarette smoke as quantified by exhaled carbon monoxide (CO) concentrations. In this observational study, exhaled CO levels were measured in 140 female smokers preoperatively. PONV was assessed over the first 24 h after surgery. There was no correlation (assessed with Spearman rank correlation) between preoperative CO and nausea scores at recovery room discharge. Significant correlations were found between nausea assessed over the first 24 h postoperatively and a history of PONV or motion sickness, the use of intraoperative antiemetic prophylaxis, duration of anesthesia, and use of opioids in the postanesthesia care unit. However, there was no correlation between preoperative CO and nausea over the first 24 h. These preliminary data suggest that the effect of smoking in reduced PONV is not directly related to preoperative exhaled CO levels.
|
Introduction
|
|---|
Smoking is an important public health problem. It is the most common preventable cause of death in the United States (1). Smoking status is a risk factor for several adverse perioperative outcomes, including cardiovascular, pulmonary, and wound-related complications (2–4). However, status as an active smoker actually protects against the development of postoperative nausea and vomiting (PONV) (5–12). The etiology of this action is not known. Two broad categories of mechanism are possible. First, chronic exposure to cigarette smoke may confer protection. For example, the initiation of smoking produces nausea, and subjects must develop tolerance to this effect if they continue smoking (13,14). Depending on the mechanism responsible for this tolerance, these patients may exhibit cross-tolerance to other emetic stimuli, such as those associated with anesthesia and surgery. Chronic exposure to smoke also produces changes in liver microsomal enzymes that metabolize nicotine and other constituents of cigarette smoke (15–17). This may affect the metabolism of drugs used in the perioperative period and the ability of these drugs to produce PONV. Second, it is also possible that acute exposure to substances in cigarette smoke may have an antiemetic effect (18). Candidate substances include carbon monoxide (CO), which may regulate intestinal motility and inhibit inflammation (19), or nicotine itself. At the same time, CO concentration can be used as a marker for recent smoking; this has been shown to correlate well with nicotine concentration (20).
If the protective effect of smoking is related to an acute effect of nicotine or other smoke constituents with relatively short half-lives (in the range of hours), we reasoned that those smokers who had most recent exposure to cigarette smoke preoperatively would be most protected. If, on the other hand, the protective effect is related to more chronic smoke exposure, such a relationship should not be present. The purpose of this observational study was to test the hypothesis that in cigarette smokers, the severity of postoperative nausea is correlated with recent exposure to cigarette smoke as quantified by exhaled CO levels.
|
METHODS
|
|---|
This study was approved by the Mayo Clinic IRB. Eligible subjects included females self-identified as smokers (defined as having smoked at least one cigarette per day in the week before surgery) undergoing elective noncardiac surgery requiring general anesthesia. Exclusion criteria included 1) age <18 yr; 2) ASA physical status 
IV; 3) emergency surgery, and 4) subjects not anticipated to require postanesthesia care unit (PACU) admission or not expected to be tracheally extubated in the PACU. After enrollment and giving informed consent, a brief smoking history questionnaire was administered, including the Fagerstrom test for nicotine dependence (21). Subjects were queried about risk factors for PONV, including history of motion sickness or PONV.
In the holding area immediately before the patient was taken to the operating room, exhaled CO levels were measured using a hand-held device (Micro Smokerlyzer®; Bedfont Scientific, Medford, NJ) (22). The patient was also asked to rate her current level of nausea using a nausea visual analog scale (NVAS) (23). The indicated position on this 100-mm linear scale was converted to a numerical value based on the distance from the leftmost point on the scale, ranging from 0 (no nausea) to 10 (worst nausea imaginable). We chose this scale as a sensitive measure that did not require verbalization in the immediate postoperative period.
The conduct of anesthesia was at the discretion of the attending anesthesiologist, including the decision of whether to provide prophylactic antiemetics. After surgery, the treatment of the patient in the PACU, including the use of opioids or antiemetics, was also entirely at the discretion of the attending anesthesiologist. Just before discharge from PACU, each patient was asked to rate the worst nausea experienced since emerging from anesthesia, again using the NVAS. The occurrence of any vomiting since emergence (defined as retching with or without oral expulsion of gastric contents) was also noted. Finally, another exhaled CO level was obtained.
After transfer to the outpatient area or the floor (for inpatients), the patients received standard nursing care. The occurrence of vomiting after discharge from the PACU (i.e., delayed postoperative vomiting) (24) was assessed the morning after surgery (approximately 24 h postoperatively) by personal interview (for inpatients) or via telephone (for outpatients). The assessment of nausea since PACU discharge was performed using a nausea verbal descriptive scale (NVSD) (23) (NVSD: 0 = none, 1 = mild, 2 = moderate, and 3 = severe nausea) because it was not possible to obtain the NVAS via a telephone interview. Additional information, including the duration of anesthesia and the use of antinausea medications or opioids, was obtained from the patients' medical records.
Unless otherwise specified, data are summarized using mean ± sd for continuous variables and percentages for categorical variables. Spearman rank correlation was used to assess the univariate association of expired CO concentration and selected other characteristics with the NVAS measured at discharge from the PACU and also with the NVSD measured on postoperative day 2 (POD2, approximately 24 h after surgery). Because multiple characteristics were found to be univariately associated with the NVSD measured on POD2, a multiple logistic regression analysis was performed to identify a subset of characteristics with evidence of an independent association with nausea as assessed using NVSD on POD2. For this analysis, the dependent variable was any nausea (mild, moderate, or severe). All characteristics were included in the first step, and a backward elimination procedure was used to eliminate variables that had no compelling evidence (P > 0.10) of an independent association with nausea. Other statistical methods include use of the two-sample Student's t-test, rank sum test, and Fisher's exact test as indicated. In all cases, two-tailed P values 
0.05 were considered statistically significant.
The sample size for this investigation was established for an analysis to test for a linear association between preoperative expired CO concentration and postoperative NVAS. For such an analysis, an effective sample size of n = 100 would provide statistical power of >80% to detect an association consistent with an R2 of 7.8% or more. Because the observed distribution of NVAS was highly skewed, nonparametric correlation analyses were performed and supplemental analyses were performed to compare the mean expired CO concentration of patients with NVAS = 0 versus those with NVAS>0. For the analysis comparing mean CO concentration between groups, the effective sample size provided statistical power of 80% to detect a difference in mean expired CO of 7.2 ppm.
|
RESULTS
|
|---|
There were 150 patients enrolled for the current investigation. Of these, 10 were excluded because their surgery was canceled or because the procedure was performed using regional rather than general anesthesia. Thus 140 patients are included in this report. Assessments at all evaluation times were obtained in 100 patients. PACU assessments were missing in 36 subjects and POD2 (the day after surgery) assessments were missing in 10 subjects. The majority of the missing data was at the time of PACU discharge where, because of logistical problems, we had difficulty obtaining assessments in a timely manner. Baseline patient and procedural characteristics are presented in Table 1. Approximately one third of patients reported a history of either PONV or motion sickness. Twenty patients (14%) reported a preoperative NVAS score of more than zero (Table 2).
The median time from the last cigarette to the preoperative assessment reported by the patients was 1.3 h (Table 2), a value that would be consistent with the CO values measured at this assessment. At the PACU assessment, the CO values were less than half of those measured preoperatively (Table 2).
The duration of anesthesia was 171 ± 98 min, and 6.5 ± 2.6 h elapsed from the preoperative assessment to the PACU assessment (at the time of PACU discharge). Most patients received intraoperative opioids (Table 3), and approximately two thirds received intraoperative antiemetic prophylaxis. Approximately two thirds of patients also required opioids in the PACU.
In the PACU, the NVAS was more than zero in 29% of the patients and 5% vomited (Table 2). Antiemetics were administered to 5% of the patients while in the PACU. At the assessment on POD2 (the day after surgery), 45% of patients reported at least mild nausea since the time of PACU discharge, and 15% reported vomiting (Table 2). Approximately two thirds of the patients maintained abstinence from smoking through the time of the POD2 assessment.
Of the variables evaluated, none were found to be significantly correlated with the NVAS reported at PACU discharge with the exception of the use of antiemetics in the PACU (Table 4). In particular, NVAS at PACU discharge was not associated with preoperative CO concentration (r = +0.07; P = 0.487), or with the CO concentration measured in the PACU at the time of the NVAS assessment (r = +0.05; P = 0.646). Furthermore, the mean preoperative expired CO concentration did not differ significantly between patients with zero versus non-zero NVAS (15.7 ± 10.7 versus 18.1 ± 13.4 ppm for patients with zero versus non-zero NVAS respectively; Student's t-test P = 0.341), with the mean preoperative CO concentration observed to be larger by +2.4 ppm (95% confidence interval, -2.6 ppm to +7.4 ppm) for patients having non-zero NVAS scores. There was also no significant correlation between NVAS and the average number of cigarettes smoked per day (r = +0.05; P = 0.581) or the duration of preoperative abstinence (r = +0.04; P = 0.670). When this analysis was repeated after excluding patients who received intraoperative antiemetic prophylaxis, similar results were obtained (Table 4).
Significant univariate associations were found between the NVDS reported on POD2 and the duration of preoperative abstinence, a history of PONV or motion sickness, the use of intraoperative antiemetic prophylaxis, duration of anesthesia, and the use of opioids in the PACU (Table 5). Again, there was no significant association between preoperative CO concentration (P = 0.180) or postoperative CO concentration (P = 0.467) and the NVDS on POD2. The mean preoperative expired CO concentration also did not differ significantly between patients reporting none versus any (mild, moderate, or severe) nausea at POD2 (17.1 ± 10.4 versus 15.6 ± 11.7 ppm for patients with none versus any nausea respectively; Student's t-test P = 0.467). A multiple logistic regression analysis was performed to identify characteristics independently associated with nausea as assessed using NVSD on POD2 (Table 6). From this analysis, use of opioids in the PACU (odds ratio = 3.0; P = 0.005) and a history of PONV or motion sickness (odds ratio = 2.2; P = 0.051) were found to be associated with an increased likelihood of nausea. In addition, there was some evidence suggesting that the likelihood of nausea was increased with longer duration of anesthesia (odds ratio = 1.1 for each 30-min increase, P = 0.071). When added to the multiple logistic regression model that included these covariates, there was no evidence to suggest that preoperative CO concentration (P = 0.672), PACU CO concentration (P = 0.351), or duration of preoperative abstinence (P = 0.293) was associated with PONV.
View this table:
[in this window]
[in a new window]
|
Table 6. Multiple Logistic Regression Assessing Characteristics Associated with Any Nausea from PACU to Postoperative Day 2
|
|
|
DISCUSSION
|
|---|
Several studies show that status as an active smoker is an independent factor protective of PONV in multivariate analysis (5,8,10–12), supporting other studies that used univariate analysis to compare PONV in smokers and nonsmokers (9). Categories of proposed mechanisms for this effect include acute exposure to smoke constituents and chronic changes caused by chronic smoke exposure.
Two plausible pharmacologic candidates for acute effects include nicotine and CO. Nicotine can inhibit the function of 5HT3 (serotonin) receptors (18), which may affect nausea and vomiting. Nicotine may also have an effect on gastrointestinal motility (25,26), which could influence symptoms of nausea. Inhaled CO in relatively small concentrations (250 ppm) attenuates postoperative ileus in mice apparently by inhibiting gut inflammation, an effect that may also reduce nausea (27); endogenous CO regulates gut motility (19). In addition, there are thousands of other smoke constituents that can have acute pharmacologic effects (15).
Chronic exposure to cigarette smoke may also confer protection against PONV. Such exposure causes extensive changes in the function of nicotinic receptors throughout the nervous system (28), changes which apparently include tolerance to the acute emetic effects of nicotine or other smoke constituents (13,14). This tolerance may extend to factors producing PONV. Cigarette smoke also has significant effects on drug metabolism. For example, smoke constituents induce some cytochrome P450 (CYP) pathways (15–17), including CYP1A2 and CYP2E1, which are responsible for the metabolism of several drugs used during anesthesia (including volatile anesthetics) (29). Such differences in metabolism could affect the pharmacokinetics of these drugs and thus, perhaps, their ability to induce PONV (9).
Carboxyhemoglobin has a half-life of approximately 4 hours (30), and exhaled CO is widely used to objectively quantify recent smoking behavior (31,32). There was a wide range of preoperative exhaled CO values observed among study patients, consistent with a wide range of times for preoperative abstinence although, consistent with prior work (33,34), the majority of patients smoked within 2 hours of their preoperative assessment. Despite this wide range of exhaled CO values, we did not find an association between this index of smoke exposure and any measure of PONV. In univariate analysis, there was a significant association between the duration of preoperative abstinence and nausea assessed at POD2; however, this association was not significant in multivariate analysis. These findings suggest that the acute pharmacologic effects of cigarette smoke constituents do not confer protection against PONV.
This study has several limitations. First, there were a number of missing assessments, especially in the PACU. Reasons included logistical problems in contacting study patients before PACU dismissal and patient refusal to provide assessments. The missing assessments reduced the power to detect correlations, although we note that the number of assessments completed still provided a statistical power of >80% to detect an association consistent with an R2 of 7.8% or more. Comparing those patients with complete and incomplete assessments, no differences were noted in demographic characteristics (Table 1) or preoperative CO (data not shown). Nonetheless, we cannot exclude bias arising from these missing assessments. Second, we only examined the correlation of CO levels and postoperative nausea. We did not directly measure other potentially relevant factors, such as nicotine levels. Third, a major limitation was that because this was an observational study of clinical practice, we did not control for many factors that can influence PONV. For example, the majority of patients received intraoperative PONV prophylaxis. This occurred even though such prophylaxis was not indicated for most patients according to departmental guidelines and may have accounted for the relatively low nausea scores in the PACU. Although similar results were observed when patients receiving prophylaxis were excluded from analysis, it is possible that a small effect of acute smoke exposure could have been obscured by such confounding uncontrolled factors and, thus, that our conclusion of a lack of correlation between exhaled CO and PONV is not valid. However, our analysis of PONV over approximately the first 24 hours after surgery (assessed on POD2) was able to confirm other risk factors previously identified, such as a history of PONV, anesthesia duration, and the use of opioids postoperatively, suggesting that our methodology had reasonable sensitivity to detect risk factors. Nonetheless, this observational study does not constitute proof that chronic (versus acute) smoking provides protection against PONV. Rather, it provides information that could guide future controlled studies, although we note that given the difficulty many patients have in abstaining from smoking for even a short time, such studies could be challenging.
In summary, the concentration of exhaled CO, an index of recent exposure to cigarette smoke, was not correlated with the intensity of PONV in female subjects. This suggests that the protective aspect of smoking on PONV is not directly related to preoperative exhaled CO levels.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank Paul Decker (Department of Biostatistics, Mayo Clinic Rochester) for statistical support, Jonathan Warner, Lavonne Liedl and Anita Baumgartner (all from the Dept. of Anesthesiology, Mayo Clinic Rochester) for assistance with data collection, and Dr. Gianrico Farrugia (Division of Gastroenterology and Hepatology, Mayo Clinic Rochester) for stimulating discussions regarding carbon monoxide.
|
Footnotes
|
|---|
Accepted for publication March 14, 2006.
Supported, in part, by the Mayo Foundation.
|
REFERENCES
|
|---|
- Cigarette smoking among adults: United States, 2000. MMWR Mortal Wkly Rep 2002;51:642–3.
- Warner DO. Helping surgical patients quit smoking: why, when, and how. Anesth Analg 2005;101:481–7.[Abstract/Free Full Text]
- Warner DO. Perioperative abstinence from cigarettes: physiological and clinical consequences. Anesthesiology 2006;104:356–67.[ISI][Medline]
- Warner DO. Preoperative smoking cessation: the role of the primary care provider. Mayo Clin Proc 2005;80:252–8.[ISI][Medline]
- Apfel CC, Greim CA, Haubitz I, et al. A risk score to predict the probability of postoperative vomiting in adults. Acta Anaesthesiol Scand 1998;42:495–501.[ISI][Medline]
- Hough M, Sweeney B. The influence of smoking on postoperative nausea and vomiting. Anaesthesia 1998;53:932–3.[ISI][Medline]
- Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997;52:443–9.[ISI][Medline]
- Cohen MM, Duncan PG, DeBoer DP, Tweed WA. The postoperative interview: assessing risk factors for nausea and vomiting. Anesth Analg 1994;78:7–16.[Abstract/Free Full Text]
- Chimbira W, Sweeney BP. The effect of smoking on postoperative nausea and vomiting. Anaesthesia 2000;55:540–4.[ISI][Medline]
- Stadler M, Bardiau F, Seidel L, et al. Difference in risk factors for postoperative nausea and vomiting. Anesthesiology 2003;98:46–52.[ISI][Medline]
- Apfel CC, Laara E, Koivuranta M, et al. A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers. Anesthesiology 1999;91:693–700.[ISI][Medline]
- Sinclair DR, Chung F, Mezei G. Can postoperative nausea and vomiting be predicted? Anesthesiology 1999;91:109–18.[ISI][Medline]
- Srivastava ED, Russell MA, Feyerabend C, et al. Sensitivity and tolerance to nicotine in smokers and nonsmokers. Psychopharmacology (Berl) 1991;105:63–8.[Medline]
- Pomerleau CS, Pomerleau OF, Namenek RJ, Marks JL. Initial exposure to nicotine in college-age women smokers and never-smokers: a replication and extension. J Addict Dis 1999;18:13–9.[ISI][Medline]
- Sweeney BP. Why does smoking protect against PONV? Br J Anaesth 2002;89:810–3.[Free Full Text]
- Zevin S, Benowitz NL. Drug interactions with tobacco smoking: an update. Clin Pharmacokinet 1999;36:425–38.[ISI][Medline]
- Pelkonen O, Maenpaa J, Taavitsainen P, et al. Inhibition and induction of human cytochrome P450 (CYP) enzymes. Xenobiotica 1998;28:1203–53.[ISI][Medline]
- Breitinger HG, Geetha N, Hess GP. Inhibition of the serotonin 5-HT3 receptor by nicotine, cocaine, and fluoxetine investigated by rapid chemical kinetic techniques. Biochemistry 2001;40:8419–29.[Medline]
- Gibbons SJ, Farrugia G. The role of carbon monoxide in the gastrointestinal tract. J Physiol 2004;556:325–36.[Abstract/Free Full Text]
- Jarvik ME, Madsen DC, Olmstead RE, et al. Nicotine blood levels and subjective craving for cigarettes. Pharmacol Biochem Behav 2000;66:553–8.[ISI][Medline]
- Heatherton TF, Kozlowski LT, Frecker RC, Fagerstrom KO. The Fagerstrom test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br J Addict 1991;86:1119–27.[ISI][Medline]
- Irving MJ, Clark EC, Crombie IK, Smith WC. Evaluation of a portable measure of expired-air carbon monoxide. Prev Med 1988;17:109–15.[ISI][Medline]
- Boogaerts JG, Vanacker E, Seidel L, et al. Assessment of postoperative nausea using a visual analogue scale. Acta Anaesthesiol Scand 2000;44:470–4.[ISI][Medline]
- Apfel CC, Roewer N, Korttila K. How to study postoperative nausea and vomiting. Acta Anaesthesiol Scand 2002;46:921–8.[ISI][Medline]
- Bertolino M, Kellar KJ, Vicini S, Gillis RA. Nicotinic receptor mediates spontaneous GABA release in the rat dorsal motor nucleus of the vagus. Neuroscience 1997;79:671–81.[ISI][Medline]
- Rausch T, Beglinger C, Alam N, et al. Effect of transdermal application of nicotine on colonic transit in healthy nonsmoking volunteers. Neurogastroenterol Motil 1998;10:263–70.[ISI][Medline]
- Moore BA, Otterbein LE, Turler A, et al. Inhaled carbon monoxide suppresses the development of postoperative ileus in the murine small intestine. Gastroenterology 2003;124:377–91.[ISI]
- Perkins KA, Grobe JE, Fonte C, et al. Chronic and acute tolerance to subjective, behavioral and cardiovascular effects of nicotine in humans. J Pharmacol Exp Ther 1994;270:628–38.[Abstract/Free Full Text]
- Kharasch ED, Thummel KE. Identification of cytochrome P450 2E1 as the predominant enzyme catalyzing human liver microsomal defluorination of sevoflurane, isoflurane, and methoxyflurane. Anesthesiology 1993;79:795–807.[ISI][Medline]
- Kambam JR, Chen LH, Hyman SA. Effect of short-term smoking halt on carboxyhemoglobin levels and P50 values. Anesth Analg 1986;65:1186–8.[Abstract/Free Full Text]
- Patrick DL, Cheadle A, Thompson DC, et al. The validity of self-reported smoking: a review and meta-analysis. Am J Public Health 1994;84:1086–93.[Abstract/Free Full Text]
- Gilbert DD. Chemical analyses as validators in smoking cessation programs. J Behav Med 1993;16:295–308.[ISI][Medline]
- Warner DO, Patten CA, Ames SC, et al. Smoking behavior and perceived stress in cigarette smokers undergoing elective surgery. Anesthesiology 2004;100:1125–37.[ISI][Medline]
- Warner DO, Patten CA, Ames SC, et al. Effect of nicotine replacement therapy on stress and smoking behavior in surgical patients. Anesthesiology 2005;102:1138–46.[ISI][Medline]
Top
Abstract
Introduction
