Postoperative nausea and vomiting

INTRODUCTION — Nausea, vomiting, and retching frequently complicate recovery from anesthesia. Postoperative nausea and vomiting (PONV) is a patient-important outcome; patients often rate PONV as worse than postoperative pain [1]. PONV usually resolves or is treated without sequelae, but may require unanticipated hospital admission and delay recovery room discharge [2,3]. In addition, vomiting or retching can result in wound dehiscence, esophageal rupture, aspiration, dehydration, increased intracranial pressure, and pneumothorax.

The term PONV is typically used to describe nausea and/or vomiting or retching in the post-anesthesia care unit (PACU) or in the immediate 24 postoperative hours. Postdischarge nausea and vomiting (PDNV) refers to symptoms that occur after discharge for outpatient procedures. Postoperative vomiting (POV) is usually measured and discussed in children, rather than PONV, because nausea may be difficult to assess in young children.

This topic will discuss the risk factors for PONV, preventive strategies, and treatment of PONV in adults and children.

PATHOPHYSIOLOGY — Nausea and vomiting may be induced through a variety of central and peripheral mechanisms. (See "Approach to the adult with nausea and vomiting", section on 'Pathophysiology'.)

Five principle neurotransmitter receptors mediate nausea and vomiting: muscarinic M1, dopamine D2, histamine H1, 5-hydroxytryptamine (HT)-3 serotonin, and neurokinin 1 (NK1) – substance P [4]. All of these receptors may be targets for prevention or treatment of PONV. The pathophysiologic pathways for PONV are shown in a figure (figure 1). (See 'Antiemetics' below.)

●Central mechanisms – Nausea and emesis can be triggered by higher cortical centers communicating with the central pattern generator (formerly called the vomiting center) in the medulla. In the perioperative period, fear, pain, anxiety, conditioned nausea related to environmental cues, and stimulation of the vestibular system are central stimuli that may cause nausea and vomiting. As an example, during tympanoplasty, surgical stimulation of the vestibular system, transmitted via the H₁ histamine and M₁ acetylcholine receptors, may result in profound PONV [5].

●Peripheral mechanisms – Direct gastric stimulation from gastric trauma, blood, or toxins induces release of substance P and serotonin from enterochromaffin cells, thereby activating the vagal and splanchnic nerve 5-HT₃ receptors [6]. Vagal and splanchnic nerve afferents terminate in the nucleus tractus solitarius in the brain stem, near or within the area postrema (also called the chemoreceptor trigger zone). Bowel surgery and blood in the gastrointestinal tract from oral or ear, nose, and throat surgery may cause nausea and vomiting via this pathway, though the mechanisms by which serotonergic stimuli cause nausea and vomiting are incompletely understood.

●Drugs and toxins – The molecular and neural mechanisms by which drugs and toxins, including anesthetics and opioids, cause nausea and vomiting are complex and incompletely understood [7]. Both opioids and inhalation anesthetics may cause nausea and vomiting by stimulation of the area postrema at the base of the fourth ventricle in the medulla. The area postrema then communicates with the central pattern generator via dopamine and serotonin to trigger the vomiting reflex [8,9].

RISK FACTORS — Without prophylaxis, PONV occurs in approximately 30 percent of children and adults after anesthesia [10-13]. The risk of PONV for an individual patient varies widely; the rate of PONV may be as high as 80 percent in high-risk patients [11].

The incidence of PONV varies with patient factors, anesthetic choices, and possibly the type of surgery.

Patient risk factors — Patient characteristics that increase the risk of PONV include the following, in decreasing order of risk:

●Preoperative nausea and vomiting – PONV may be the result of a condition that was causing nausea and vomiting prior to surgery (eg, renal colic).

●Female gender – Female gender is the most reliable patient-specific predictor for PONV [14,15]. A meta-analysis of 22 prospective studies including over 95,000 adults found that female gender was the strongest overall predictor for PONV (odds ratio [OR] 2.57) [14].

In children, prior to puberty, female gender does not increase the risk of PONV [16-18].

●History of PONV or motion sickness – Prior PONV and/or motion sickness increase the risk of PONV in adults (OR 2.09) [12,14,15,19].

In children, both a prior history of PONV or postoperative vomiting (POV) and a history of PONV or POV in a parent or sibling increase the risk of POV/PONV [13].

●Nonsmoking – Nonsmoking status is an independent risk factor for PONV (OR 1.82) [12,14,15,20].

Smoking status does not usually apply to children, and the risk of POV/PONV in children exposed to secondhand smoke has not been studied.

●Age – Most studies have reported a slight, progressive decrease in PONV in adults with increasing age [10,14,21]. A prospective study including approximately 2200 patients who underwent general anesthesia reported that age <50 was a risk factor for PONV in the post-anesthesia care unit (PACU) (OR 1.79) and postdischarge nausea and vomiting (PDNV) (OR 2.17) [21].

In children, young age appears to be protective. POV rarely occurs in children <3 years of age, and increases in frequency with age >3, decreasing again with puberty [16].

●Chemotherapy-induced nausea and vomiting – A history of chemotherapy-induced nausea and vomiting (CINV) may increase the risk of PONV (OR 3.15) [22]. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting".)

Anesthetic factors — A number of anesthetic factors have been associated with PONV, some of which may be modified to reduce risk.

●Anesthetic technique – General anesthesia is associated with a higher incidence of PONV compared with purely regional anesthesia [12]. Regional anesthesia, in both adults and children, may reduce PONV by reducing opioid administration for postoperative pain.

●Volatile anesthetics versus total intravenous anesthesia – The use of volatile anesthetics is an important risk factor for PONV [12,23-25]. In a study of 1180 surgical patients, those randomly assigned to receive potent inhalation anesthesia with isoflurane, enflurane, or sevoflurane had a higher risk of PONV compared with propofol, which has antiemetic properties (OR 2.4, 2.3, and 2.3, respectively) [23] (see 'Propofol' below). The degree of risk was similar for each of the volatile anesthetics, was dose-related (ie, based on the duration of anesthesia), and was limited to the early postoperative period.

The use of total intravenous anesthesia (TIVA) rather than volatile anesthetics may also reduce hospital length of stay for patients or procedures that are at high risk for PONV or surgical complications related to vomiting. In a single institution retrospective cohort study of 448 patients who underwent laparoscopic Nissen fundoplication, the use of TIVA was associated with reduced hospital length of stay (OR for hospital length of stay ≤1 night 2.91 [95% CI 1.47-5.78]) and 7.8 percent reduction in cost of care [26].

Another meta-analysis of 57 trials in adults and 13 trials in children found that propofol anesthesia resulted in 3.5-fold reduction of PONV in adults and 5.7-fold reduction in children [27].

●Intravenous (IV) anesthetics – Etomidate does not independently increase PONV at doses commonly administered for induction of anesthesia [28]. Perioperative low-dose ketamine may reduce PONV [29] along with decreasing postoperative pain and opioid requirements [29-31]. There is lack of evidence to suggest an increase in PONV after an induction dose of ketamine (ie, 1 to 2 mg/kg IV) [32]. Use of ketamine for perioperative analgesia is discussed separately. (See "Nonopioid pharmacotherapy for acute pain in adults", section on 'Ketamine'.)

●Nitrous oxide (N₂O) – N₂O may modestly increase the risk of PONV, especially in high-risk adults and children, compared with N₂O-free anesthesia [25]. Studies of the emetogenic effect of N₂O have been conflicting, with some reporting an increase in PONV with N₂O [33-35], and others reporting no effect [36-38]. A meta-analysis of 30 studies including approximately 4600 patients reported a modestly-higher incidence of PONV with N₂O compared with N₂O-free anesthesia (33 versus 27 percent) [24]. The maximal risk reduction with avoidance of N₂O was in women (OR 0.7), and the difference between groups was eliminated by administration of a propofol infusion. In one subsequent study of 100 women who underwent laparoscopic-assisted hysterectomy, there was no difference in PONV in patients who were randomly assigned to have 70 percent N₂O versus air added during the last 30 minutes of isoflurane anesthesia [39].

●Duration of anesthesia – Longer duration of anesthesia with volatile anesthetics may increase the risk of PONV [14,15,40,41]. In addition to the increase in dose of anesthetics, longer procedures tend to be more invasive, and tend to require administration of larger doses of postoperative opioids than shorter procedures.

The emetogenic effect of N₂O may depend on the duration of exposure as well. A meta-analysis of 29 randomized trials involving 10,317 patients found an insignificant risk of PONV from administration of N₂O for less than one hour [42]. After 45 minutes of administration, N₂O increased the risk of PONV by 20 percent per hour. Conclusions from this study are limited by the possible effects of other anesthetics and/or the surgical procedure itself.

In children, surgery >30 minutes increases the risk of POV compared with shorter procedures [16].

●Opioid administration and reduction – Multiple studies have shown that administration of perioperative opioids increases the incidence of PONV [11,12,14] in a dose-related manner [43]. Multimodal perioperative pain control strategies that reduce or eliminate opioid administration reduce the incidence of PONV [44-48]. In children, administration of dexmedetomidine may be particularly beneficial for its opioid-sparing effects and to reduce the incidence of emergence delirium. (See "Emergence delirium and agitation in children", section on 'Prevention' and "Approach to the management of acute pain in adults", section on 'Use multimodal analgesia'.)

●Sugammadex versus neostigmine for reversal of neuromuscular blockade – We do not choose the reversal for neuromuscular blocking agents based on a possible effect on PONV. Contrary to popular belief, IV neostigmine for reversal of neuromuscular block does not appear to significantly increase the risk of PONV. A meta-analysis of 10 randomized trials including 933 patients reported that the combination of neostigmine with either atropine or glycopyrrolate did not significantly increase the incidence of overall nausea (relative risk [RR] 1.24) or vomiting (RR 0.91) [49].

There are limited data on the incidence of PONV after reversal of neuromuscular blocking agents with sugammadex. In one study of approximately 100 patients who were randomly assigned to reversal with neostigmine or sugammadex, the incidence of nausea and/or vomiting in the first postoperative hour was lower after sugammadex, but PONV over the first 24 hours was similar with the two drugs [50]. A systematic review and meta-analysis of six studies found that the PONV risk was decreased when sugammadex was used instead of neostigmine for the reversal of neuromuscular blockade, with the number needed to treat to prevent PONV cited at 16 [51]. However, conclusions from this study are limited by unclear baseline risk of PONV of study subjects and inclusion of open label studies.

Type of surgery — Studies of the effect of the type of surgery on the incidence of PONV have reported conflicting results. The best evidence suggests that cholecystectomy, gynecologic, and laparoscopic procedures are associated with modestly increased risk of PONV compared with other general surgical procedures [14].

In children, strabismus surgery is an independent, and possibly the most significant, predictor for POV [13,16,52]. In addition, POV occurs in as many as 70 percent of children without prophylaxis who undergo adenotonsillectomy [53], up to 60 percent of children who undergo otoplasty [54], and 40 percent of those who undergo inguinal scrotal or penile procedures [55].

Risk scores — A number of predictive models, or scoring systems, have been developed to stratify the risk of PONV, based on patient, anesthetic, and surgical risk factors. The score may be used to determine the appropriate degree of intervention for prophylaxis (figure 2).

Risk score for postoperative nausea and vomiting for adults — We use the simplified risk score for PONV that was created by Apfel, et al [11] to evaluate adults preoperatively, and base the preventive strategy on the resulting degree of predicted risk. The simplified risk score is easy to use and accurately predicts the risk for PONV in our adult patients. The components of the simplified risk score include the following four highly predictive risk factors:

●Female gender

●Nonsmoker

●History of motion sickness or previous PONV

●Expected administration of postoperative opioids

This risk score has been validated within and across institutions; the presence of 0, 1, 2, 3, and 4 of these risk factors corresponds to risk of PONV of 10, 20, 40, 60, and 80 percent, respectively [11].

Risk score for postoperative vomiting for children — The risk scores used for adult patients are not applicable for children [17]. We use a scoring system that includes preoperative, intraoperative, and postoperative risk factors, which appears in the 2020 Fourth Consensus Guidelines for the Management of Postoperative Nausea and Vomiting from the American Society of Enhanced Recovery and the Society for Ambulatory Anesthesia [56]. Those risk factors are shown in a table (table 1).

PREVENTION

Our strategy — The management strategy for PONV should include risk assessment (including patient and surgical factors), an emphasis on multimodal preventive measures based on the risk assessment, and evidenced-based interventions when PONV occurs. Our approach is generally consistent with the 2020 Fourth Consensus Guidelines for the Management of Postoperative Nausea and Vomiting from the American Society of Enhanced Recovery and the Society for Ambulatory Anesthesia [56].

Preventive measures for all patients — We employ a multimodal, opioid-sparing or opioid-free strategy for effective perioperative pain control; the reduction of pain has been correlated with a reduction in PONV [57]. Additional preventive measures may include modification of anesthetic technique and medications used, administration of antiemetics, and use of nonpharmacologic measures for prophylaxis. (See "Approach to the management of acute perioperative pain in infants and children", section on 'Opioid avoidance' and "Approach to the management of acute pain in adults", section on 'Minimize opioids'.)

Risk stratification — The decision to administer prophylaxis for PONV should be made with the following factors in mind:

●The patient-specific risk of PONV (ie, the number of risk factors for PONV)

●The consequences of nausea and vomiting related to the surgical procedure (eg, wound disruption, increase in intracranial pressure, clot disruption after angiographic procedures)

●Patient and clinician preferences

●Implications for a change in routine anesthetic technique (eg, using total intravenous anesthesia [TIVA] in pediatric patients, who would otherwise receive sevoflurane anesthesia)

Prophylaxis based on risk — We recommend prophylaxis for PONV for all patients who have risk factors for PONV, rather than no prophylaxis. We suggest multimodal prophylaxis, with antiemetics from different drug classes, and possibly modification of the anesthetic.

Adults — Our approach to prophylaxis for PONV in adults is shown in an algorithm (algorithm 1).

Adults with ≥3 risk factors — For patients with three or four risk factors, we use three or more interventions, including at least three antiemetics from different drug classes, and may also use propofol anesthesia, and/or acupuncture [56,58]. (See 'Risk score for postoperative nausea and vomiting for adults' above.)

In our hospital-based practice of both ambulatory and inpatient surgeries where we have a limited hospital-based drug formulary, our strategy for PONV prophylaxis for high-risk adults is described here. The efficacy of specific interventions is detailed below. (See 'Options for prophylaxis in adults and children' below.)

●Anesthetic technique – We offer regional anesthesia rather than general anesthesia if appropriate for the surgery. If general anesthesia is required, we prefer TIVA with propofol and opioid-sparing anesthetic techniques.

●Antiemetics – We usually administer the following antiemetics:

•Scopolamine patch – We have the patient apply a scopolamine patch at least two hours prior to the induction of anesthesia, with instructions to remove the patch within 24 hours after surgery. If necessary, we apply the patch preoperatively. (See 'Anticholinergics' below.)

•Dexamethasone – 4 to 8 mg intravenously (IV) after induction. (See 'Glucocorticoids' below.)

•Ondansetron – 4 mg IV at the end of surgery. (See 'Serotonin (5-hydroxytryptamine) receptor antagonists' below.)

•For patients who do not receive a scopolamine patch (eg, patients with angle closure glaucoma), we administer dimenhydrinate 25 to 50 mg IV, prochlorperazine 5 to 10 mg IV, or haloperidol 1 mg IV, as a third antiemetic. (See 'Antihistamines' below and 'Phenothiazines' below and 'Antidopaminergics' below.)

●Rescue antiemetics – If nausea and vomiting occur in the post-anesthesia care unit (PACU), we administer an antiemetic from a different class of drug from those used for prophylaxis. Our usual options include:

•Prochlorperazine 5 to 10 mg IV (see 'Phenothiazines' below), or

•Droperidol 0.625 mg IV (see 'Antidopaminergics' below)

Adults with one or two risk factors — For patients with one or two risk factors for PONV (including patients with any risk factor for PONV and patients who receive inhalation anesthesia or TIVA that includes opioids), we use two or more interventions. We administer two antiemetics of different drug classes, and may also choose to either modify the anesthetic technique (ie, regional anesthesia or TIVA with propofol) or perform acupuncture. (See 'Reduction of baseline risk' below and 'Antiemetics' below and 'Nonpharmacologic techniques' below.)

Adults without risk factors

●For adults without preoperative risk factors for PONV who receive inhalation anesthesia or TIVA that includes opioids, we usually administer dexamethasone 4 to 8 mg IV after induction of anesthesia, and also ondansetron 4 mg IV at the end of surgery.

●For adults without preoperative risk factors who receive regional anesthesia or TIVA with propofol without intraoperative or postoperative opioid, we do not usually administer PONV prophylaxis, though others do. (See 'Antiemetics' below.)

Children — Ours is a hospital-based practice of both ambulatory and inpatient surgeries where we have a limited hospital-based drug formulary. Our approach to PONV in children is shown in an algorithm (algorithm 2).

Children with ≥3 risk factors — For high-risk children (ie, with ≥3 risk factors), our strategy is as follows (see 'Risk score for postoperative vomiting for children' above):

●Anesthetic technique – We offer regional anesthesia with sedation to older children. If general anesthesia is required, as it is in most children, we prefer TIVA with propofol for very-high-risk patients.

●Antiemetics – We administer the following antiemetics:

•Dexamethasone 0.25 mg/kg IV, maximum 4 mg (see 'Glucocorticoids' below)

•Ondansetron 0.1 mg/kg IV, maximum 4 mg (see 'Serotonin (5-hydroxytryptamine) receptor antagonists' below)

●Rescue antiemetics – For POV rescue, our usual choices include:

•Dimenhydrinate or diphenhydramine 0.5 mg/kg IV, maximum 25 mg (see 'Antihistamines' below)

•Ondansetron, repeat dose 0.1 mg/kg IV, maximum 4 mg, total perioperative maximum 8 mg (see 'Serotonin (5-hydroxytryptamine) receptor antagonists' below)

Children with one or two risk factors — We manage moderate-risk children (ie, those with one or two risk factors) as we would high-risk children, as described above, without using TIVA. (See 'Risk score for postoperative vomiting for children' above and 'Children with ≥3 risk factors' above.)

Children without risk factors — Prophylaxis for POV for children without risk factors should be based on clinician and patient preferences, formulary choices, and cost, and may include zero or one prophylactic intervention. For children with no risk factors for POV, we administer ondansetron 0.1 mg/kg IV, maximum 4 mg. For rescue antiemetic, we administer dimenhydrinate or diphenhydramine 0.5 mg/kg IV, maximum 25 mg, or ondansetron, repeat dose 0.1 mg/kg IV, maximum 4 mg, total perioperative maximum 8 mg.

Options for prophylaxis in adults and children — All patients should receive a multimodal, opioid-sparing approach for effective postoperative pain control (algorithm 1 and algorithm 2). (See 'Reduction of baseline risk' below and 'Antiemetics' below and 'Nonpharmacologic techniques' below.)

Reduction of baseline risk — The baseline risk of PONV and POV may be reduced as follows:

●Modification of anesthetic technique (see 'Anesthetic factors' above)

•Use of regional anesthesia rather than general anesthesia

•Use of TIVA with propofol, rather than volatile anesthetics

●Adequate hydration – In clinical practice, decisions regarding IV fluid administration in adults are usually based on factors other than the risk of PONV. The literature on the effect of the type and amount of IV fluid on PONV in adults is conflicting. A meta-analysis of 15 randomized trials including approximately 1600 patients found that liberal intraoperative crystalloid administration (15 to 30 mL/kg) reduced the risk of late and overall PONV but not early PONV compared with restrictive IV fluids (0 to 2 mL/kg) [59].

A 2019 meta-analysis of 41 randomized trials including approximately 4200 healthy patients who underwent general anesthesia for ambulatory or short stay surgery found that administration of at least 10 mL/kg IV crystalloid was associated with reduced risk of PONV and reduced requirement for treatment for PONV, compared with lower volume crystalloid [60]. However, the quality of evidence was moderate, with heterogeneous methods and volumes of IV fluid administration, and unclear risk of bias. Risk of serious adverse events and unanticipated postoperative hospital admission related to vigorous IV crystalloid administration were not reported.

Liberal fluid administration may be beneficial in children for prevention of PONV. In a meta-analysis of seven small randomized trials of children who underwent surgery with volatile anesthetics without PONV prophylaxis, administration of supplemental IV fluid (20 or 30 mL/kg/hour) reduced PONV compared with more conservative IV fluid administration at 10 mL/kg/hour (25 versus 50 percent) [61]. In our practice, we consider administration of liberal amounts of IV fluid for children who have had POV despite multiple preventive strategies with prior anesthetics, unless the child is undergoing bowel or bariatric surgery. Similar to adults, restrictive or goal-directed fluid management is increasingly used for children who undergo these procedures [62]. Intraoperative fluid management is discussed separately. (See "Intraoperative fluid management".)

●Intravenous dextrose solution – Studies of the effect of perioperative administration of IV dextrose solutions on PONV have reported mixed results. A 2019 meta-analysis of 10 randomized trials including approximately 1000 patients who underwent a variety of surgical procedures round that use of an IV dextrose infusion was not associated with a reduction in nausea and vomiting in the postanesthesia care unit (PACU) or in the first 24 hours after surgery [63]. In a 2020 meta-analysis of 11 trials of almost exclusively female patients, perioperative dextrose administration slightly reduced the incidence of early and postoperative nausea, without an effect on vomiting [64]. Overall quality of evidence was rated low.

Intraoperative administration of IV dextrose solutions can cause hyperglycemia, even in nondiabetic patients [65]. In one study of dextrose administration for PONV prevention in nondiabetic patients who underwent laparoscopic cholecystectomy, mean blood glucose after the infusion of 5% dextrose was 282 mg/dL, versus 92 mg/dL after infusion of Ringer's lactate [66].

●Multimodal postoperative pain control – A multimodal strategy for perioperative pain control including regional anesthesia techniques, nonopioid analgesics, other nonopioid adjuvants, and nonpharmacologic therapies, may reduce the need for postoperative opioids and the risk of PONV and POV. (See 'Anesthetic factors' above and "Approach to the management of acute pain in adults", section on 'Use multimodal analgesia'.)

Antiemetics — A variety of antiemetics acting via different mechanisms are used for prevention and treatment of PONV. In general, clinicians choose among these agents based upon side effect profile, personal experience, cost, and formulary considerations (table 2 and table 3). Side effects of many of these agents when used for PONV prophylaxis have not been well studied [67]; where appropriate they are mentioned in descriptions of drugs below.

The mechanisms of action of antiemetics are discussed separately. (See "Characteristics of antiemetic drugs".)

General principles — The following principles apply to the use of antiemetics in this setting:

●Commonly used antiemetics reduce the risk of PONV by approximately 25 percent (table 2) [68].

●The absolute benefit of an antiemetic depends on the degree of baseline risk, with higher-risk patients benefiting more than low-risk patients [68]. As examples, for a patient with a baseline risk of PONV of 80 percent (eg, a female nonsmoker with a history of motion sickness who requires postoperative opioids), ondansetron would reduce the risk of PONV by approximately 20 percent to a risk of 60 percent. By contrast, for a patient with baseline risk of 10 percent (eg, a male smoker without motion sickness who does not require postoperative opioids), ondansetron would reduce the risk of PONV by only 2.5 percent, to a risk of 7.5 percent.

●The effects of drugs that act on different receptors are additive rather than synergistic. Because each drug reduces PONV by a fraction of the existing risk, each added drug results in less antiemetic benefit [68].

●The antiemetic benefit from TIVA with propofol is approximately the same as administration of one antiemetic with volatile inhalation anesthesia [68]. There is further risk reduction of PONV when TIVA with propofol is used with another TIVA agent (ie, dexmedetomidine for opioid sparing effect) [69].

●All antiemetics have side effects, some more serious than others (eg, headache, sedation, electrocardiogram [ECG] QT interval prolongation).

●Rescue treatment should include an antiemetic from a different class than the drugs that were used for prophylaxis unless the effect of the previous drug has likely worn off.

Commonly used antiemetics are described in the following sections (table 2).

Serotonin (5-hydroxytryptamine) receptor antagonists — The 5-hydroxytryptamine type 3 (5-HT₃) receptor antagonists are commonly used for prevention and treatment of PONV. Their lack of sedative effects is an advantage in the postoperative period; some clinicians avoid the 5-HT₃ antagonists intraoperatively to save these drugs for possible rescue in the PACU.

●First-generation serotonin antagonists – The first-generation serotonin antagonists (ie, ondansetron, granisetron, dolasetron, and, outside the United States, tropisetron) are equally efficacious for PONV at equipotent doses (relative risk [RR] 0.76 versus placebo) [70]. Ondansetron may be the most commonly used antiemetic in this class, given its low cost and benign side effect profile. All of these drugs have potential to prolong the electrocardiogram intervals, particularly the QT interval, and should be avoided for patients at risk for QTc prolongation (ie, patients with congenital long QT syndrome, congestive heart failure, bradyarrhythmias, or patients who are already taking medications that prolong the QT interval) [71]. In 2021, the US Food and Drug Administration (FDA) mandated a label change for ondansetron to include a warning that there have been reported cases of myocardial ischemia after administration of the drug. The incidence is likely extremely rare, since there have been very few reported cases and ondansetron is very commonly used. A causal link has not been definitively established. Coronary spasm has been proposed as a possible etiology.

These agents are administered as a single dose IV at the end of surgery. Ondansetron is available as an orally disintegrating film (ODF) and as an orally disintegrating tablet (ODT), both of which are administered at twice the IV dose. Oral preparations are as effective as IV ondansetron, administered 30 to 60 minutes prior to surgery [72,73]. The ODT may be useful for postdischarge administration and may be administered to children over the age of five [74].

•Ondansetron – Adults 4 mg IV or 8 mg orally prior to surgery [68,75-77], children 0.1 mg/kg IV, maximum 4 mg [78].

•Granisetron – Adults 0.35 to 3 mg IV [79,80], children 40 mcg/kg IV, maximum 0.6 mg [81].

•Dolasetron – Adults 12.5 mg IV [82], children 0.35 mg/kg IV, maximum 12.5 mg. Dolasetron is contraindicated at the high doses previously used for prevention of chemotherapy-induced nausea and vomiting (CINV), due to the risk of QTc prolongation and fatal torsade de pointes at those doses [83]. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Dolasetron'.)

•Tropisetron – Adults 2 mg IV at the end surgery [84].

●Second-generation serotonin antagonist

•Palonosetron – Palonosetron is a second-generation serotonin antagonist. Compared with the first-generation drugs in this class, palonosetron has unique and higher receptor binding affinity [85], and a longer half-life of 40 hours [86], with similar efficacy for prevention of PONV [87,88]. Because of its prolonged action, palonosetron may be particularly effective for prevention of late or postdischarge nausea and vomiting (PDNV). (See 'Postdischarge nausea and vomiting' below.)

Palonosetron does not appear to affect the QT interval. Palonosetron 0.075 mg is more efficacious at preventing PONV compared with ondansetron 4 and 8 mg, granisetron 1 mg, dexamethasone 5 and 8 mg, tropisetron 2 mg, and ramosetron 0.3 mg [89-91].

-Palonosetron dose – Adults 0.075 mg IV [87] at induction of anesthesia, children 0.5 to 1.5 mcg/kg IV [92,93] (limited data)

•Ramosetron – Ramosetron is a second generation 5 hydroxytryptamine type 3 (5HT-3; serotonin) receptor antagonist, not licensed for use in the United States. Elsewhere, ramosetron is approved for the treatment of chemotherapy induced nausea and vomiting. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'First-generation agents'.)

-Ramosetron dose – Adults 0.3 to 0.6 mg IV at the end of surgery [94,95].

Glucocorticoids — Dexamethasone is as effective as ondansetron and droperidol for prevention of PONV (RR 0.76 versus placebo) [68,96,97], and is the most commonly used and studied corticosteroid for POV prophylaxis in children [98].

Dexamethasone may be beneficial because of a direct antiemetic effect and by reducing postoperative pain and the need for postoperative opioids. Lower doses of dexamethasone may be required for PONV prophylaxis than for pain relief. A meta-analysis of 60 randomized trials with 6700 patients found that a dose of 4 to 5 mg IV dexamethasone was as effective as 8 to 10 mg IV for reduction of PONV [99]. In contrast, two meta-analyses of studies comparing lower-dose (<0.1 mg/kg IV) dexamethasone with higher doses (>0.1 mg/kg) found that higher doses were required for reduced opioid requirement [100,101]. A subsequent multicenter randomized trial including 1350 patients who underwent bowel surgery reported that a single postinduction dose of dexamethasone 8 mg IV reduced the incidence of PONV at 24 hours and the need for rescue antiemetic for up to 72 hours [102]. In the randomized international trial described below (PADDI), dexamethasone 8 mg IV reduced PONV in the first 24 hours after surgery (42.2 versus 53.9 percent, risk ratio 0.78; 95% CI 0.75–0.82) [103]. In a substudy of 1466 patients in the PADDI trial, dexamethasone reduced the incidence and severity of PONV on days one and two postoperatively [104]. Conclusions from this study are limited by the small number of PONV events.

●Safety of dexamethasone – Data on the safety of prophylactic dexamethasone are inconclusive, and the use of dexamethasone should be individualized. (See "Anesthesia for tonsillectomy with or without adenoidectomy in children", section on 'Dexamethasone'.)

•Surgical site infection – Most studies have shown no increase in wound infection with one dose of dexamethasone [105-109]. In a randomized multicenter international noninferiority trial (PADDI) including approximately 8700 patients who had general anesthesia for noncardiac surgery, the incidence of surgical site infection within 30 days of surgery was similar in patients who received dexamethasone 8 mg IV during surgery versus those who received placebo (8.1 versus 9.1 percent) [103]. The results were similar in patients with and without diabetes, and across types of infection (ie, superficial, deep, organ space).

•Increase in blood glucose – Dexamethasone at the doses used for PONV prophylaxis (ie, up to 8 mg IV) can cause a modest glucose elevation in patients with or without diabetes, over and above the typical increase in glucose that occurs as a result of surgery [109-112]. The increase in glucose after dexamethasone administration may be greater in patients without diabetes [113], and in patients with diabetes the increase may depend on the degree of control of diabetes [114].

-In a meta-analysis of 56 randomized trials that evaluated glucocorticoid administration in patients who had elective surgery, the mean increase in perioperative glucose concentration was 20 mg/dL (95% CI 11.4–28.6) [108].

-In the large international trial described above, the median difference between preoperative and peak blood glucose in patients who received dexamethasone was 65 mg/dL compared with a median increase of 43 mg/dL in patients who received placebo [103].

-In one trial, 302 elective noncardiac surgical patients were randomly assigned to dexamethasone 4 or 8 mg or placebo, during surgery [114]. Blood glucose increased in all groups. The increase was similar after 4 or 8 mg dexamethasone in patients without diabetes and patients with well controlled diabetes. In patients with high preoperative hemoglobin A1C, the increase was greater with an 8 mg dose, compared with a 4 mg dose of dexamethasone.

-In another trial, 185 patients (49 with diabetes) who had major noncardiac surgery were randomly assigned to receive 8 mg dexamethasone IV during surgery versus placebo [113]. Mean blood glucose increased in both groups; increase in blood glucose was greater in patients without diabetes than in those with diabetes (72 ± 45 versus 63 ± 69 mg/dL). In patients with diabetes, the increase in blood glucose was similar with dexamethasone versus placebo. In patients without diabetes, dexamethasone increased blood glucose by a mean of 29 mg/dL compared with placebo.

•Concerns for patients with cancer – In pediatric oncology patients who undergo anesthesia for bone marrow biopsy or lumbar puncture, glucocorticoids can interfere with cell counts and, ultimately, chemotherapeutic management or, rarely, can cause tumor lysis syndrome [58]. The patient's oncologist should be consulted prior to administration of dexamethasone. (See "Tumor lysis syndrome: Pathogenesis, clinical manifestations, definition, etiology and risk factors".)

●Dexamethasone dose — Because of its slow onset, dexamethasone is more effective if administered after induction rather than at the end of surgery [96].

•Adults 4 mg to 8 mg IV [115,116] after induction

•Children 0.25 mg/kg, maximum 4 mg [117]

Anticholinergics — Scopolamine is an anticholinergic medication applied as a 1.5 mg transdermal preparation for PONV prophylaxis. Transdermal scopolamine (TDS) is a sustained-release, long-acting patch that is applied at least several hours prior to anesthesia, and is as effective as ondansetron and droperidol for PONV prophylaxis [118]. A meta-analysis of 25 randomized trials with 3300 patients found that TDS was associated with a reduced risk of PONV (RR 0.73) compared with placebo during the first 24 hours after anesthesia [119]. TDS was as effective if applied the evening prior to surgery as when applied two to four hours prior to anesthesia. The side effects of scopolamine are mild, and include dry mouth and blurry vision [119,120]. Confusion or agitation may occur, especially in older adults with baseline cognitive impairment [121,122]. Acute angle closure glaucoma is a possible side effect of scopolamine [123]. Scopolamine patch is contraindicated in patients with angle closure glaucoma [124].

Outside the United States, scopolamine is available as an IV preparation as well.

Antidopaminergics — The antidopaminergic class of antiemetics includes droperidol, haloperidol, and amisulpride. Low doses of droperidol and haloperidol are as effective as ondansetron for PONV prophylaxis (droperidol 1.25 mg IV; RR 0.74 versus placebo) [125,126], though they are associated with an increase in sedation [127]. The US Food and Drug Administration (FDA) issued a "black box" warning regarding the potential for sudden cardiac death with droperidol in 2001. Since then, droperidol has rarely been used in the United States, though at the doses recommended for PONV prophylaxis, the risk of adverse cardiac events is extremely low. The butyrophenones are associated with an increase in the QT interval to a degree that is similar to ondansetron [128]. Administration of ondansetron with droperidol or with haloperidol is more effective for PONV than either drug alone, without additional QT prolongation [129-131]. Similar to ondansetron, butyrophenones should be avoided for patients at increased risk of QT prolongation (table 4).

Both haloperidol and droperidol are typically administered as a single IV dose at the end of surgery. Haloperidol may also be administered orally or intramuscularly (IM). These drugs are not administered for POV prophylaxis for children.

For patients who receive droperidol, we follow the FDA recommendation to monitor ECG for arrhythmias for two to three hours after administration [128].

Amisulpride is a dopamine receptor antagonist that was approved by the FDA in February 2020 for use in prevention and treatment of PONV [132]. Amisulpride 5 mg is more efficacious than placebo in the prevention of PONV [133,134]. Amisulpride 10 mg IV, not 5 mg, is more effective compared with placebo as rescue treatment of PONV in patients who have already received a PONV prophylaxis antiemetic that is not in the antidopaminergic class [135]. Amisulpride can cause a mild elevation in serum prolactin, hypokalemia, chills, hypotension during injection, and pain at the injection site [132]. Amisulpride should be avoided in patients who receive droperidol and in patients with congenital long QT syndrome. ECG monitoring should be performed in patients with cardiac conduction disorders and in patients on medications (ie, ondansetron) which cause QT prolongation [136].

●Droperidol – 0.625 to 1.25 mg IV.

●Haloperidol – 1 mg IV, or 1 mg oral, or 1 mg IM.

●Amisulpride – 5 mg IV injected over one to two minutes at induction of anesthesia for prevention of PONV [137]; 10 mg IV injected over one to two minutes as a rescue antiemetic after surgery if an antidopaminergic agent has not already been administered previously [135].

Neurokinin 1 receptor antagonists — The neurokinin 1 (NK1) receptor antagonists comprise a relatively new class of long-acting antiemetics that may be especially effective for prevention of POV and PDNV. In a network meta-analysis of 585 randomized trials (97,516 patients) involving 44 single drugs and 51 drug combinations, the NK1 receptor antagonists were the most effective class of antiemetics, and as effective when administered alone as other drug combinations [67]. Fosaprepitant was the single most effective antiemetic, though this conclusion was based on a very small number of studies that compared it with only placebo or ondansetron, and the evidence was of moderate quality. The only NK1 receptor antagonist that is FDA approved for PONV is oral aprepitant.

Aprepitant and fosaprepitant may reduce the efficacy of hormonal contraceptives [138]. Patients should be instructed to use a non-hormonal method of contraception while using and following the last dose of these drugs [139,140].

●Aprepitant – Aprepitant is a neurokinin 1 (NK1) receptor antagonist available in oral and intravenous formulations. Aprepitant has a half-life of 40 hours [58]. Compared with ondansetron alone [141,142], and compared with ondansetron when both drugs have been administered with dexamethasone [143], aprepitant has been found to be equally effective for prevention of postoperative nausea and more effective for preventing vomiting at 24 and 48 hours.

•Aprepitant dose – Adults, 40 mg orally preoperatively [144-146]; or 32 mg IV over 30 seconds preoperatively

●Fosaprepitant – Fosaprepitant is a prodrug of aprepitant and is available only in IV form.

•Fosaprepitant dose – 150 mg IV preoperatively [147]

●Rolapitant – Rolapitant is a long-acting NK1 antagonist that is available in the United States in oral-form 90 mg tabs, and as an injectable emulsion for prevention of CINV (166.5 mg IV injected over 30 minutes). Rolapitant may be particularly effective for PDNV because of its very long half-life of 180 hours. A prospective dose ranging study of 600 high-risk patients reported that doses of ≥70 mg orally reduced emesis at 72 and 120 hours compared with placebo and with ondansetron [148]. The intravenous preparation has not been studied for PONV prophylaxis.

•Rolapitant dose – Adults, 90 mg oral, preoperatively [148]

Antihistamines — Dimenhydrinate and diphenhydramine are antihistamines with antiemetic efficacy similar to reported efficacy for dexamethasone, droperidol, and the 5-HT₃ antagonists at a dose of 1 mg/kg IV for adults, and 0.5 mg/kg, maximum 25 mg in children [149], though direct comparisons have not been done. Dimenhydrinate can also be administered IM, orally, and rectally. Dose responses and optimal timing of administration have not been established.

Common side effects of dimenhydrinate include sedation, dry mouth, dizziness, and urinary retention. In children, concerns for sedation and the possibility of emergence delirium related to urinary retention have limited the routine use of this drug.

Phenothiazines — The phenothiazines as a group are most effective at treating opioid-induced nausea and vomiting. Their use for PONV is limited by sedation at high doses and extrapyramidal effects. Data are limited regarding optimal doses and timing of administration. These drugs are not usually administered for children for POV prophylaxis.

●Promethazine – 6.25 to 12.5 mg IV at induction of anesthesia [150,151]

Note: Extravasation of promethazine into subcutaneous tissues is likely to cause damage. The IV should be tested for patency and visible during administration of the drug. Promethazine should be diluted with saline to a concentrations ≤1 mg/mL and infused over at least 20 minutes.

●Prochlorperazine – 5 to 10 mg IV at the end of surgery [152]

Propofol — Propofol is a sedative hypnotic that is used for induction and maintenance of anesthesia and for sedation. Propofol is an antiemetic when administered at doses required for induction and maintenance for TIVA [68,153-156] and when administered at subhypnotic doses in adults [157] and children [158]. In addition to a direct antiemetic effect, propofol may reduce PONV by reducing the required concentration of volatile anesthetic. (See 'Anesthetic factors' above.)

In one study, 80 children who underwent tonsillectomy with dexamethasone PONV prophylaxis were randomly assigned to receive a propofol bolus (1 mg/kg IV) followed by low-dose infusion (20 mcg/kg/minute IV) until the end of surgery, or no propofol [158]. A higher percentage of those assigned to propofol were free of vomiting or retching for the initial 24 hours after surgery (75 versus 38 percent).

Metoclopramide — Metoclopramide is a relatively weak antiemetic that is rarely administered as PONV prophylaxis in adults or children.

The literature on the efficacy of metoclopramide as an antiemetic for PONV is controversial. A number of published studies of antiemetics from a single author have been retracted from publication, including studies on metoclopramide [159]. Meta-analyses including some of those studies had reported that metoclopramide was ineffective for prevention of PONV. However, a subsequent meta-analysis that excluded the retracted studies found that the usual adult dose of metoclopramide of 10 mg IV reduced the incidence of 24-hour PONV (odds ratio [OR] 0.58) [160]. However, metoclopramide in usual doses is not as effective as other commonly used antiemetics. In a meta-analysis of trials (without the retracted studies) comparing antiemetics, ondansetron was found to be 57 percent more effective than metoclopramide 10 mg IV for PONV prevention [125].

In addition, studies in both adults [41] and children [52] report that metoclopramide is effective only at doses higher than usually used (25 to 50 mg IV in adults, 0.15 to 0.25 mg/kg in children). We typically do not administer these high doses because of increasing risk of adverse effects including hypotension, tachycardia, and rarely extrapyramidal symptoms.

Extrapyramidal symptoms may be as much as 20-fold more common in children compared with adults [161]. In infants, prolonged clearance may increase serum concentrations of metoclopramide; administration should be avoided for children <1 year of age. In older children, metoclopramide should only be administered when other antiemetics cannot be used, at a dose of 0.1 mg/kg IV, maximum 10 mg. (See "Metoclopramide: Pediatric drug information".)

Midazolam — The literature on the efficacy of midazolam for PONV prophylaxis is inconclusive. Meta-analyses of trials of perioperative midazolam have reported 38 to 55 percent reduction in overall PONV [162,163]. However, there was high risk of bias in the included studies, and a clinically relevant increase in postoperative sedation could not be ruled out.

We do not administer midazolam solely for PONV prophylaxis because of the potential for postoperative sedation and delirium in some patient populations (eg, older adults, mentally ill).

Cannabinoids — Cannabinoids may be effective for treatment of CINV. However, a randomized trial involving 340 patients with a high risk of PONV who underwent general anesthesia reported that a single preoperative dose of nabilone 0.5 mg orally, in addition to routine PONV prophylaxis, had no effect on the incidence of PONV, pain scores, or opioid consumption [164].

Olanzapine — Olanzapine is a second-generation antipsychotic that blocks both serotonin type two 5-hydroxytryptamine (5-HT2) receptors and dopamine D2 receptors, and is used for antiemetic therapy for highly emetogenic chemotherapy regimens. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Olanzapine'.)

Olanzapine has a peak effect at six hours after oral administration and half-life of 30 hours, making it potentially useful for prevention of PDNV. In one single institution randomized trial, administration of olanzapine 10 mg orally preoperatively in addition to ondansetron and dexamethasone reduced the incidence of nausea and vomiting in the first 24 postoperative hours by approximately 60 percent, compared with the other drugs alone [165]. Similarly, in another single institution randomized trial including 100 patients with multiple risk factors for PONV plus a history of chemotherapy induced nausea and vomiting, addition of preoperative olanzapine to multidrug prophylaxis with dexamethasone, ondansetron, and the use of propofol anesthesia reduced the incidence of PONV in the first 24 hours by 60 percent (26 versus 63 percent) [166].

The role of olanzapine for PONV prophylaxis in adults and children, and optimal doses for this purpose have not been established. Olanzapine is associated with sedation, and doses should be reduced for older patients.

Combination therapy — We agree with guidelines that recommend a combination of at least two antiemetics from different drug classes for PONV prophylaxis for any patient with risk factors for PONV [56,57]. Combination therapy using different classes of drugs is usually more effective than single-drug therapy for PONV prophylaxis in both children and adults [68,129,167-174]. This concept was demonstrated by a large randomized trial of six interventions for PONV prophylaxis, including three antiemetics (dexamethasone, droperidol, and ondansetron) alone and in combination [68]. Each antiemetic reduced PONV to a similar extent (RR 26 percent), and each drug acted independently, such that the risk reduction from combinations could be estimated by multiplying individual risk reductions.

One single institution retrospective cohort study evaluated the effect of a combination of 5 prophylactic antiemetics on the incidence of PONV in 536 high risk surgical patients and case matched controls, all in a Veterans Administration hospital [175]. Study patients were at high risk of PONV due to either receiving intrathecal morphine or having bariatric surgery, or both. The 5-drug regimen included preoperative perphenazine 4 to 8 mg orally and aprepitant 40 mg orally, diphenhydramine 12.5mg IV. dexamethasone 4 mg IV, and either Palonosetron 75 mcg IV or ondansetron 8mg IV. The incidence of PONV requiring rescue on postoperative days 0 to 1 was 12 percent in patients who received the 5-drug regimen, compared with 22 percent in patients who received 1 to 4 drugs, and 56 percent for patients who received 0 to 1 drug. Further study is required before broadly advocating 5 drug PONV prophylaxis, even to patients who are at high risk for PONV.

Existing data support the strategy to administer antiemetics from different drug classes [176]. However, there is insufficient data to recommend one specific combination rather than others. The chosen combinations of prophylactic antiemetics must be individualized based on patient and surgical factors, and patients should be monitored for potential side effects when combining medications.

Examples of studies of combinations of antiemetics include the following:

●Dexamethasone and ondansetron or palonosetron decreased the risk of PONV compared with only 5-HT3 receptor antagonist therapy [174].

●Aprepitant with ondansetron is more effective than only monotherapy with ondansetron [177,178].

●Aprepitant with dexamethasone is more effective for preventing postoperative vomiting than ondansetron with dexamethasone [143].

●Amisulpride and ondansetron or dexamethasone was more efficacious than lone therapy with ondansetron or dexamethasone in the reduction of PONV and in the necessity of a rescue antiemetic [137].

●Triple multimodal antiemetic prophylaxis with aprepitant 80 mg, dexamethasone 4 to 8 mg, and ondansetron 4 mg was more effective in the reduction of PONV compared with the combination of dexamethasone and ondansetron in patients having elective laparoscopic surgery under neuraxial blockade and TIVA [179].

●In children undergoing surgery for strabismus, the addition of ondansetron to dexamethasone reduces POV from 23 to 5 percent [180].

Nonpharmacologic techniques — A number of nonpharmacologic methods have been suggested and studied for PONV prophylaxis. Of these, acupuncture seems to be the most promising.

Acupuncture — Stimulation of the pericardium-6 (P-6) acupuncture point may be effective for PONV prophylaxis for clinicians familiar with this technique (picture 1). A meta-analysis of 59 studies including 7700 adults and children reported that acupoint stimulation with a variety of techniques (eg, acupuncture needles, acupressure wristbands) was as effective as the commonly used antiemetics for prevention of PONV [181]. A 2020 meta-anaylsis of 16 randomized trials in children found that acupuncture reduced the risk of PONV and antiemetic use, particularly during the first four hours after the procedure [182]. The quality of the evidence was rated as low.

The side effects of acupuncture and acupressure are low, including skin redness or irritation and, very rarely, nerve damage, bleeding, or skin infection [183].

Isopropyl alcohol — Inhalation of isopropyl alcohol (eg, having the patient inhale alcohol from a skin prep pack or alcohol-soaked sponge) may be a modestly effective short-term treatment for PONV. A meta-analysis including six small studies found that alcohol inhalation reduced the need for rescue antiemetics compared with placebo, but that it was less effective than standard antiemetics [184].

Isopropyl alcohol is a readily available, low-risk treatment that may be most useful for immediate administration while other antiemetics are prepared for administration.

RESCUE THERAPY — When postoperative nausea and vomiting (PONV) occurs in the post-anesthesia care unit (PACU) or operating room, rescue treatment should include a drug from a different class than those that have already been administered unless the effect of the first drug has worn off or a potentially inadequate dose has been administered [185].

The serotonin receptor antagonists are particularly beneficial as rescue drugs in the PACU, especially for same-day surgery patients, because they are nonsedating. (See 'Serotonin (5-hydroxytryptamine) receptor antagonists' above.)

Whereas dexamethasone is an effective prophylactic antiemetic, it may be less effective as a rescue medication. In a multicenter trial of 280 patients who had PONV after general anesthesia without PONV prophylaxis, dexamethasone (at doses of 3, 6, or 12 mg IV) was ineffective for treatment of PONV within the first 24 hours after surgery, compared with placebo [186]. Conclusions are limited by the small study size, and the fact that the study was stopped early for futility.

For opioid-induced PONV or postoperative vomiting (POV), low-dose naloxone infusion (0.25 mcg/kg/hour IV) can reduce opioid-induced side effects including nausea and vomiting, without affecting analgesia [187,188].

POSTDISCHARGE NAUSEA AND VOMITING — Postdischarge nausea and vomiting (PDNV) (ie, nausea and vomiting within 48 hours of discharge) occurs in 37 percent of adult patients who undergo general anesthesia for same-day surgery [21]. Risk factors for PDNV include those that predict PONV [189] and, in addition, administration of opioids in the PACU and nausea in the PACU. A simplified risk scoring system has been developed for PDNV, including the following risk factors (figure 3):

●Female sex

●Age <50

●History of PONV

●Opioids administered in the PACU

●Nausea in the PACU

Zero, one, two, three, four, and five risk factors are associated with a 7, 20, 28, 53, 60, and 90 percent incidence of PDNV, respectively [21].

Preventive strategies for PDNV include risk stratification followed by a multimodal prophylactic approach, similar to prophylaxis for PONV. Some options may be limited by availability of the newer antiemetics and formulary issues. The addition of either long-acting antiemetics (eg, aprepitant, palonosetron, transdermal scopolamine) or repeat, scheduled dosing of shorter-acting medications in the postoperative period (eg, ondansetron oral disintegrating tablet [ODT]) may reduce PDNV.

In one study, high-risk patients were randomly assigned to standard prophylaxis with ondansetron 4 mg intravenously (IV), or a study regimen including dexamethasone 8 mg IV and ondansetron 4 mg IV intraoperatively, and ODT 8 mg to be taken on discharge and in the morning of postoperative days 1 and 2 [190]. The study regimen reduced the incidence of postdischarge nausea (20 percent versus 57 percent) and vomiting (3 percent versus 20 percent) compared with standard prophylaxis.

The incidence of PDNV in children may be lower than in adults. In a prospective observational study of 1041 children who underwent ambulatory surgery, the incidence of PDNV was 14 percent [191]. Risk factors for PDNV included perioperative opioid administration, inhalation anesthetics, and the use of opioid medications at home; no patient who received total intravenous anesthesia (TIVA) had PDNV. There was no association between PDNV and age, gender, duration of anesthesia, use of nitrous oxide (N₂O) for maintenance of anesthetic, intraoperative antiemetic administration, airway management strategy, or the length of time from recovery room discharge to first oral intake.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Postoperative nausea and vomiting".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Nausea and vomiting after surgery (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Incidence – Postoperative nausea and vomiting (PONV) occurs in approximately 30 percent of adults and children after general anesthesia, and in up to 80 percent of high-risk patients. (See 'Risk factors' above.)

●Risk stratification – The preventive strategy is based on risk stratification that takes into account patient, anesthesia, and procedure specific factors. (See 'Risk factors' above.)

•Adults – We use Apfel's simplified risk score to evaluate adults. This risk score includes four factors: female gender, nonsmoking, history of motion sickness or PONV, and expected administration of postoperative opioids (figure 3). (See 'Risk score for postoperative nausea and vomiting for adults' above.)

•Children – We use an alternative risk scoring system for children, which includes preoperative, intraoperative, and postoperative risk factors (table 1). (See 'Risk score for postoperative vomiting for children' above.)

●Prophylaxis for all patients – Prophylaxis for all patients should include multimodal opioid-sparing postoperative pain control. Additional prophylactic interventions may include the following:

•Modification of the anesthetic (eg, avoiding general anesthesia, use of total intravenous anesthesia with propofol) (see 'Reduction of baseline risk' above)

•Antiemetics (see 'Antiemetics' above)

•Acupuncture (see 'Acupuncture' above)

●Prophylaxis for adults – For all patients with one or more risk factors for PONV, we recommend prophylaxis. (Grade 1B). We suggest multimodal prophylaxis, with antiemetics from different drug classes, and possibly modification of the anesthetic (algorithm 1 and table 3) (Grade 2A).

•Adults with ≥3 risk factors – For these patients we use three or more interventions, including at least three antiemetics from different drug classes. We use regional anesthesia if possible or TIVA with propofol if general anesthesia is required, opioid-sparing anesthesia techniques, and we administer transdermal scopolamine 1.5 mg, dexamethasone 4 to 8 mg intravenous (IV), and ondansetron 4 mg IV for prophylaxis. (See 'Adults with ≥3 risk factors' above and 'Options for prophylaxis in adults and children' above.)

•Adults with one or two risk factors – For these patients we use ≥2 interventions, including at least two antiemetics from different drug classes. We usually administer ondansetron 4 mg IV and dexamethasone 4 to 8 mg IV and may also choose to modify the anesthetic technique (ie, regional anesthesia or TIVA with propofol) or perform acupuncture.

•Adults without risk factors – For adults without risk factors who receive inhalation anesthesia or TIVA with opioids, we administer two antiemetics. We usually administer ondansetron 4 mg IV and dexamethasone 4 to 8 mg IV in this setting.

For adults without risk factors who receive regional anesthesia or TIVA with propofol without perioperative opioids, we do not administer PONV prophylaxis, though others do.

●Prophylaxis for children (algorithm 2 and table 3) (see 'Children' above)

•Children with ≥3 risk factors – We offer regional anesthesia with sedation to older high risk children if possible. If general anesthesia is required, we use TIVA with propofol. We administer dexamethasone 0.25 mg/kg IV, maximum 4 mg, and ondansetron 0.1 mg IV, maximum 4 mg for prophylaxis.

•Children with one or two risk factors – We use the same approach as for high risk children, without using TIVA rather than inhalation agents for general anesthesia.

•Children without risk factors – We administer ondansetron 0.1 mg/kg IV, maximum 4 mg, for prophylaxis, though it is reasonable to administer no prophylaxis.

●Antiemetics – Antiemetics commonly used for prophylaxis and treatment of PONV are shown in a table (table 3). (See 'Antiemetics' above.)

General principles for their use include the following (see 'General principles' above):

•The most commonly used antiemetics reduce the risk of PONV by approximately 25 percent (table 2).

•The absolute benefit of an antiemetic depends on the degree of baseline risk; higher-risk patients benefit more than low-risk patients.

•The effects of drugs that act on different receptors are additive.

•All antiemetics have side effects, some more serious than others.

●Rescue therapy – Rescue therapy for PONV in the post-anesthesia care unit (PACU) or operating room should include a drug from a different class than those that have already been administered, unless the effect of the first drug has worn off or a potentially inadequate dose has been administered. (See 'Rescue therapy' above and 'Our strategy' above.)

●Postdischarge nausea and vomiting – Postdischarge nausea and vomiting (PDNV) may occur in as many as 37 percent of adults who undergo same-day surgery. Risk factors for PDNV include nausea in the PACU and opioids administered in the PACU, in addition to the risk factors for PONV (figure 3). Prophylaxis may include administration of longer-acting antiemetics (eg, aprepitant, palonosetron, transdermal scopolamine) or repeat, scheduled doses of the shorter-acting antiemetics. (See 'Postdischarge nausea and vomiting' above.)