National Cancer Institute
Last Modified: October 26, 2012
Prevention and control of nausea and vomiting (emesis) (N&V) are paramount in the treatment of cancer patients. N&V can result in serious metabolic derangements, nutritional depletion and anorexia, deterioration of patients' physical and mental status, esophageal tears, fractures, wound dehiscence, withdrawal from potentially useful and curative antineoplastic treatment, and degeneration of self-care and functional ability. (See Table 1 for criteria on grading severity.) Despite advances in pharmacologic and nonpharmacologic management, N&V remain two of the more distressing and feared side effects to cancer patients and their families, and incidence may be underestimated by physicians and nurses. 1 2 3 4 5
In this summary, unless otherwise stated, evidence and practice issues as they relate to adults are discussed. The evidence and application to practice related to children may differ significantly from information related to adults. When specific information about the care of children is available, it is summarized under its own heading.
Nausea is a subjective phenomenon of an unpleasant, wavelike sensation experienced in the back of the throat and/or the epigastrium that may culminate in vomiting (emesis). Vomiting is the forceful expulsion of the contents of the stomach, duodenum, or jejunum through the oral cavity. Retching is gastric and esophageal movements of vomiting without expulsion of vomitus and is also referred to as dry heaves.
Various classifications of N&V have been used, 1 6 including acute, delayed, late or persistent, chronic, anticipatory, breakthrough, or refractory, as well as distinctions related to type of treatment (e.g., chemotherapy induced or radiation induced) and clinical course of disease (e.g., advanced or terminal disease). 7 8 Despite this variety, the most commonly described types of N&V are acute, delayed, and anticipatory chemotherapy-induced N&V and chronic N&V in advanced cancer patients. Although there are no standard definitions, the following are commonly used to classify the different types.
|Nauseab||1||Loss of appetite without alteration in eating habits|
|2||Oral intake decreased without significant weight loss, dehydration, or malnutrition|
|3||Inadequate oral caloric or fluid intake; tube feeding, TPN, or hospitalization indicated|
|4||Grade not available|
|5||Grade not available|
|Vomitingc||1||12 episodes (separated by 5 min) in 24 h|
|2||35 episodes (separated by 5 min) in 24 h|
|3||6 episodes (separated by 5 min) in 24 h; tube feeding, TPN, or hospitalization indicated|
|4||Life-threatening consequences; urgent intervention indicated|
|N&V = nausea and vomiting (emesis); TPN = total parenteral nutrition.|
Progress has been made in understanding the neurophysiologic mechanisms that control nausea and vomiting (emesis) (N&V). Both are controlled or mediated by the central nervous system but by different mechanisms. Nausea is mediated through the autonomic nervous system. Vomiting results from the stimulation of a complex reflex that is coordinated by a putative true vomiting center, which may be located in the dorsolateral reticular formation near the medullary respiratory centers. The vomiting center presumably receives convergent afferent stimulation from several central neurologic pathways, including the following: 1 2
The CTZ is located in the area postrema, one of the circumventricular regions of the brain on the dorsal surface of the medulla oblongata at the caudal end of the fourth ventricle. Unlike vasculature within the blood-brain diffusion barrier, the area postrema is highly vascularized with fenestrated blood vessels, which lack tight junctions (zonae occludentes) between capillary endothelial cells. The CTZ is anatomically specialized to readily sample elements present in the circulating blood and cerebrospinal fluid (CSF). 3 4
Currently, evidence indicates that acute emesis following chemotherapy is initiated by the release of neurotransmitters from cells that are susceptible to the presence of toxic substances in the blood or CSF. Area postrema cells in the CTZ and enterochromaffin cells within the intestinal mucosa are implicated in initiating and propagating afferent stimuli that ultimately converge on central structures corresponding to a vomiting center. The relative contribution from these multiple pathways culminating in N&V symptoms is complex and is postulated to account for the variable emetogenicity (intrinsic emetogenicity and mitigating factors [i.e., dosage, administration route, and exposure duration]) and emetogenic profile (i.e., time to onset, symptom severity, and duration) of agents.
Not all cancer patients will experience nausea, vomiting (emesis), or both. The most common causes are emetogenic chemotherapy drugs and radiation therapy to the gastrointestinal (GI) tract, liver, or brain. Several patient characteristics have also been identified. These include incidence and severity of nausea and vomiting (N&V) during past courses of chemotherapy, history of chronic alcohol use, age, and gender. Patients with poor control of N&V during prior chemotherapy cycles are likely to experience N&V in subsequent cycles. N&V is less likely in patients with a history of chronic high intake of alcohol 1 and more likely in women 2 3 and in patients younger than 50 years. 2
Other possible causes include fluid and electrolyte imbalances such as hypercalcemia, volume depletion, or water intoxication; tumor invasion or growth in the GI tract, liver, or central nervous system, especially the posterior fossa; constipation; certain drugs such as opioids; infection or septicemia; or uremia. The psychological variables of state anxiety (level of anxiety during chemotherapy infusions) and pretreatment expectations for N&V (self-fulfilling prophecy) have also been investigated as predictors of posttreatment nausea. 4 5 6 7 8 9 At present, studies have found mixed results that vary due to different research methods. Better designed studies, however, have found state anxiety and patient expectations for nausea to be predictors of posttreatment nausea, even after controlling for known physiological predictors (susceptibility to nausea during pregnancy and motion sickness) and emetogenic potential of the chemotherapy drugs. 6 7 8 10 11 It is important to note, however, that patients' fears and expectations about chemotherapy can be variable and change over time. 12 In a longitudinal study, 12 patients' anticipatory fears of vomiting decreased significantly from pretreatment to a period 3 to 6 months later, particularly when their chemotherapy included antiemetic medications.
Clinicians treating N&V must be alert to all potential causes and factors, especially in cancer patients who may be receiving combinations of several treatments and medications. (Refer to the PDQ® summary on Pain for more information on opioid-induced N&V.)
The prevalence of anticipatory nausea and vomiting (emesis) (ANV) has varied, owing to changing definitions and assessment methods. 1 However, anticipatory nausea appears to occur in approximately 29% of patients receiving chemotherapy (about one of three patients), while anticipatory vomiting appears to occur in 11% of patients (about one of ten patients). 2 With the introduction of new pharmacologic agents (5-HT3 receptor antagonists), it was anticipated that the prevalence of ANV might decline; however, studies have shown mixed results. One study found a lower incidence of ANV, 3 and three studies found comparable incidence rates. 2 4 5 It appears that the 5-HT3 agents reduce postchemotherapy vomiting but not postchemotherapy nausea, 2 5 and the resulting impact on ANV is unclear.
Although other theoretical mechanisms have been proposed, 6 ANV appears to be best explained by classical conditioning (also known as Pavlovian or respondent conditioning). 7 In classical conditioning, a previously neutral stimulus (e.g., smells of the chemotherapy environment) elicits a conditioned response (e.g., ANV) after a number of prior pairings or learning trials. In cancer chemotherapy, the first few chemotherapy infusions are the learning trials. The chemotherapy drugs are the unconditioned stimuli that elicit postchemotherapy nausea and vomiting (N&V) (in some patients). The drugs are paired with a variety of other neutral, environmental stimuli (e.g., smells of the setting, oncology nurse, chemotherapy room). These previously neutral stimuli then become conditioned stimuli and elicit ANV in future chemotherapy cycles. ANV is not an indication of psychopathology but is rather a learned response that, in other life situations (e.g., food poisoning), results in adaptive avoidance. A variety of correlational studies provide empirical support for classical conditioning. For example, the prevalence of ANV prior to any chemotherapy is very rare, and few patients ever experience ANV without prior postchemotherapy nausea. 8 Also, most studies have found (1) a higher probability of ANV with increasing numbers of chemotherapy infusions and (2) the intensity of ANV increasing as patients get closer to the actual time of their infusion. 9 In one experimental study, it was shown that a novel beverage could become a conditioned stimulus to nausea when paired with several chemotherapy treatments. 10
Many variables have been investigated as potential factors that correlate with the incidence of ANV in hopes of developing a list of risk factors. There is currently no agreement on which factors predict ANV. A patient with fewer than three of the first eight characteristics listed below, however, is unlikely to develop ANV, and screening following the first chemotherapy infusion could identify those patients at increased risk. 11
Antiemetic drugs do not seem to control ANV once it has developed; 2 however, a variety of behavioral interventions have been investigated. 19 These include progressive muscle relaxation with guided imagery, 20 hypnosis, 21 systematic desensitization, 22 electromyography and thermal biofeedback, 23 and distraction via the use of video games. 24 25 Progressive muscle relaxation with guided imagery, hypnosis, and systematic desensitization has been studied the most and is the recommended treatment. Referral to a psychologist or other mental health professional with specific training and experience in working with cancer patients is recommended when ANV is identified. The earlier ANV is identified, the more likely treatment will be effective; thus, early screening and referral are essential. In addition, physicians and nurses underestimate the incidence of chemotherapy-induced nausea and vomiting. 26[Level of evidence: II]
In addition to emetogenic potential, the dose and schedule used are also extremely important factors. For example, a drug with a low emetogenic potential given in high doses may cause a dramatic increase in the potential to induce N&V. Standard doses of cytarabine rarely produce N&V, but these are often seen with high doses of this drug. Another factor to consider is the use of drug combinations. Because most patients receive combination chemotherapy, the emetogenic potential of all of the drugs combined and individual drug doses needs to be considered.
Delayed (or late) N&V occurs more than 24 hours after chemotherapy administration. Delayed N&V is associated with cisplatin, cyclophosphamide, and other drugs (e.g., doxorubicin and ifosfamide) given at high doses or given on 2 or more consecutive days.
Antiemetic agents are the most common intervention in the management of treatment-related nausea and vomiting (emesis) (N&V). The basis for antiemetic therapy is the neurochemical control of vomiting. Although the exact mechanism is not well understood, peripheral neuroreceptors and the chemoreceptor trigger zone (CTZ) are known to contain receptors for serotonin, histamine (H1 and H2), dopamine, acetylcholine, opioids, and numerous other endogenous neurotransmitters. 1 2 Many antiemetics act by competitively blocking receptors for these substances, thereby inhibiting stimulation of peripheral nerves at the CTZ and possibly at the vomiting center. Most drugs with proven antiemetic activity can be categorized into one of the following groups:
Although all routes of administration are listed for each of the following drugs, the intramuscular (IM) route should be used only when no other access is available. IM delivery is painful, is associated with erratic absorption of drug, and may lead to sterile abscess formation or fibrosis of the tissues. This is particularly important when more than one or two doses of a drug are to be given.
Phenothiazines act on dopaminergic receptors at the CTZ, possibly at other central nervous system (CNS) centers, and peripherally. With the exception of thioridazine, many phenothiazines possess antiemetic activity, including chlorpromazine given in the 10- to 50-mg dose range orally, IM, intravenously (IV), and rectally (pediatric dose for patients >12 years: 10 mg every 68 hours; for patients <12 years: 5 mg every 68 hours); thiethylperazine given in the 5- to 10-mg dose range orally, IM, and IV; and perphenazine. The primary consideration in selecting among phenothiazines are differences in their adverse effect profiles, which substantially correlate with their structural classes. Generally, aliphatic phenothiazines (e.g., chlorpromazine, methotrimeprazine) produce sedation and anticholinergic effects, while piperazines (e.g., prochlorperazine, thiethylperazine, perphenazine, fluphenazine) are associated with less sedation but greater incidence of extrapyramidal reactions (EPRs).
Prochlorperazine is perhaps the most frequently (and empirically) used antiemetic and, in low doses, is generally effective in preventing nausea associated with radiation therapy and in treating N&V attributed to very low to moderately emetogenic chemotherapeutic drugs. Prochlorperazine is a phenothiazine and can be given orally, IM, IV, and rectally. It is usually given in the 10- to 50-mg dose range (pediatric dose for children who weigh >10 kg or who are >2 years: orally or rectally, 0.4 mg/kg/dose tidqid; or IM, 0.10.15 mg/kg/dose tidqid, maximum 40 mg/d). Higher prochlorperazine doses (e.g., 0.20.6 mg/kg/dose) are also given IV for patients receiving chemotherapy with high emetogenic potential. 3; 4[Level of evidence: I] Phenothiazines may be of particular value in treating patients who experience delayed N&V (postacute phase symptoms) on cisplatin regimens. 5[Level of evidence: I]
As with other dopaminergic antagonists, the most common side effects of prochlorperazine are EPRs (acute dystonias, akathisias, neuroleptic malignant syndrome [uncommon], and rarely, akinesias and dyskinesias) and sedation. Marked hypotension may also result if IV prochlorperazine is administered rapidly at high doses. Administration over at least 30 minutes appears adequate to prevent hypotensive episodes. 6 7 8
Droperidol and haloperidol represent another class of dopaminergic (D2 subtype) receptor antagonists that are structurally and pharmacologically similar to the phenothiazines. While droperidol is used primarily as an adjunct to anesthesia induction, haloperidol is indicated as a neuroleptic antipsychotic drug; however, both agents have some antiemetic activity. Droperidol is administered IM or IV, typically from 1 to 2.5 mg every 2 to 6 hours, but higher doses (up to 10 mg) have been safely given. 9 10 Haloperidol is administered IM, IV, or orally, typically from 1 to 4 mg every 2 to 6 hours. 11 Results of a small, uncontrolled, open-label study showed some efficacy for haloperidol in palliative care patients. 12 Both agents may produce EPRs, akathisia, hypotension, and sedation.
Metoclopramide is a substituted benzamide, which, prior to the introduction of serotonin (5-HT3) receptor antagonists, was considered the most effective single antiemetic agent against highly emetogenic chemotherapy such as cisplatin. Although metoclopramide is a competitive antagonist at dopaminergic (D2) receptors, it is most effective against acute vomiting when given IV at high doses (e.g., 0.53 mg/kg/dose), probably because it is a weak competitive antagonist (relative to other serotonin antagonists) at 5-HT3 receptors. It may act on the CTZ and the periphery. Metoclopramide also increases lower esophageal sphincter pressure and enhances the rate of gastric emptying, which may factor into its overall antiemetic effect. It can be administered IV at the U.S. Food and Drug Administration (FDA)approved dose of 1 to 2 mg/kg every 2 hours (or less frequently) for three to five doses. Metoclopramide has also been safely given by IV bolus injection at higher single doses (up to 6 mg/kg) and by continuous IV infusion, with or without a loading bolus dose, with efficacy comparable to multiple intermittent dosing schedules. 13 14 15
Metoclopramide is associated with akathisia and dystonic extrapyramidal effects; akathisia is seen more frequently in patients older than 30 years, and dystonic extrapyramidal effects are seen more commonly in patients younger than 30 years. Diphenhydramine, benztropine mesylate, and trihexyphenidyl are commonly used prophylactically or therapeutically to pharmacologically antagonize EPRs. 7 16 While cogwheeling rigidity, acute dystonia, and tremor are responsive to anticholinergic medications, akathisiathe subjective sense of restlessness or inability to sit stillis best treated by the following:
Four serotonin receptor antagonistsondansetron, granisetron, dolasetron, and palonosetronare available in the United States. Tropisetron, while not approved by the FDA, is available internationally. Agents in this class are thought to prevent N&V by preventing serotonin, which is released from enterochromaffin cells in the gastrointestinal (GI) mucosa, from initiating afferent transmission to the CNS via vagal and spinal sympathetic nerves. 17 The 5-HT3 receptor antagonists may also block serotonin stimulation at the CTZ and other CNS structures.
Several studies have demonstrated that ondansetron produces an antiemetic response that equals or is superior to high doses of metoclopramide, but ondansetron has a superior toxicity profile compared with dopaminergic antagonist agents. 18 19 20 21 22[Level of evidence: I] 23 24 Ondansetron (0.15 mg/kg) is given IV 15 to 30 minutes prior to chemotherapy and is repeated every 4 hours for two additional doses. Alternatively, for patients older than 18 years, a large multicenter study determined that a single 32-mg dose of ondansetron is more effective in treating cisplatin-induced N&V than a single 8-mg dose and is as effective as the standard regimen of three doses at 0.15 mg/kg given every 4 hours starting 30 minutes before chemotherapy. 25[Level of evidence: I] A single-center retrospective chart review has reported ondansetron-loading doses of 16 mg/m2 (maximum, 24 mg) IV to be safe in infants, children, and adolescents. 26
Currently, the oral and injectable ondansetron formulations are approved for use without dosage modification in patients older than 4 years, including elderly patients and patients with renal insufficiency. Oral ondansetron is given 3 times daily starting 30 minutes before chemotherapy and continuing for up to 2 days after chemotherapy is completed. Patients older than 12 years should receive 4 mg per dose. Ondansetron is not approved for use in children younger than 4 years. Ondansetron clearance is diminished in patients with severe hepatic insufficiency; therefore, such patients should receive a single injectable or oral dose no higher than 8 mg. There is currently no information available evaluating the safety of repeated daily ondansetron doses in patients with hepatic insufficiency. Other effective dosing schedules such as a continuous IV infusion (e.g., 1 mg/h for 24 h) or oral administration have also been evaluated. 25
The major adverse effects include headache (which can be treated with mild analgesics), constipation or diarrhea, fatigue, dry mouth, and transient asymptomatic elevations in liver function tests (alanine and aspartate transaminases), which may be related to concurrent cisplatin administration. 27 Ondansetron has been etiologically implicated in a few case studies involving thrombocytopenia, renal insufficiency, and thrombotic events. 28 In addition, a few case reports have implicated ondansetron in causing EPRs. However, it is not clear in some cases whether the events described were in fact EPRs; in other reports, the evidence is confounded by concurrent use of other agents that are known to produce EPRs. Nevertheless, the greatest advantage of serotonin receptor antagonists over dopaminergic receptor antagonists is that they have fewer adverse effects. Despite prophylaxis with ondansetron, many patients receiving doxorubicin, cisplatin, or carboplatin will experience acute and delayed-phase N&V. 29[Level of evidence: II] A randomized, double-blind, placebo-controlled trial suggests that the addition of aprepitant, a neurokinin-1 (NK-1) receptor antagonist, may mitigate N&V. 30[Level of evidence: I] The optimal dose of aprepitant may be 125 mg on day 1 followed by 80 mg on days 2 to 5. 31[Level of evidence: I]
Granisetron has demonstrated efficacy in preventing and controlling N&V at a broad range of doses (e.g., 1080 g/kg and empirically, 3 mg/dose). In the United States, granisetron injection, transdermal patch, and oral tablets are approved for initial and repeat prophylaxis for patients receiving emetogenic chemotherapy, including high-dose cisplatin. Granisetron is pharmacologically and pharmacokinetically distinct from ondansetron; however, clinically it appears equally efficacious and equally safe. 32 33 34 35[Level of evidence: I] Both granisetron formulations are given before chemotherapy, as either a single IV dose of 10 g/kg (0.01 mg/kg) or 1 mg orally every 12 hours.
Both granisetron formulations and ondansetron injection share the same indication against highly emetogenic chemotherapy. In contrast, the oral ondansetron formulation has been approved only for use against N&V associated with moderately emetogenic chemotherapy.
Currently, granisetron injection is approved for use without dosage modification in patients older than 2 years, including elderly patients and patients with hepatic and renal insufficiency. Oral granisetron has not yet been approved for use in pediatric patients.
Both oral and injection formulations of dolasetron are indicated for the prevention of N&V associated with moderately emetogenic cancer chemotherapy, including initial and repeat courses. Oral dolasetron should be dosed as 100 mg within 1 hour before chemotherapy. Dolasetron should be given IV or orally at 1.8 mg/kg as a single dose approximately 30 minutes before chemotherapy.
The effectiveness of oral dolasetron in the prevention of chemotherapy-induced nausea and vomiting (CINV) has been proven in a large randomized, double-blind, comparative trial of 399 patients. 36[Level of evidence: I] Oral dolasetron was administered in the range of 25 to 200 mg 1 hour prior to chemotherapy. The other study arm consisted of oral ondansetron (8 mg) administered 1.5 hours before chemotherapy and every 8 hours after chemotherapy for a total of three doses. Complete response (CR) rates improved with increasing doses of dolasetron. Both dolasetron 200 mg and ondansetron had significantly higher CR rates as compared with dolasetron 25 or 50 mg. (CR was defined as no emetic episodes and no use of escape antiemetic medications.) Dolasetron injection has also been proven effective in the prevention of CINV. 37[Level of evidence: I]
Palonosetron is a 5-HT3 receptor antagonist (second generation) that has antiemetic activity at both central and gastrointestinal sites. In comparison to the older 5-HT3 receptor antagonists, it has a higher binding affinity to the 5-HT3 receptors, a higher potency, a significantly longer half-life (approximately 40 hours, four to five times longer than that of dolasetron, granisetron, or ondansetron), and an excellent safety profile. 38[Level of evidence: I] A dose-finding study demonstrated that the effective dose was 0.25 mg or higher. 38 In two large studies of patients receiving moderately emetogenic chemotherapy, CR (no emesis, no rescue) was significantly improved in the acute and the delayed period for patients who received 0.25 mg of palonosetron alone compared with either ondansetron or dolasetron alone. 39; 40[Level of evidence: I] Dexamethasone was not given with the 5-HT3 receptor antagonists in these studies, and it is not yet known whether the differences in CR would persist if d