INTRODUCTION — Nausea and vomiting are common symptoms near the end of life that can cause substantial physical and psychological distress for patients and their families and significantly impact quality of life (QOL). Nausea, the unpleasant sensation of being about to vomit, can occur alone or can accompany vomiting, dyspepsia, or other gastrointestinal symptoms. Vomiting is the expulsion of gastric contents through the mouth caused by forceful and sustained contraction of the abdominal muscles and diaphragm. Retching differs from vomiting in the absence of expulsion of gastric contents.
There are many potential causes of nausea and vomiting in palliative care populations, and the etiology is often multifactorial, particularly in patients with cancer. In such patients, nausea and vomiting might be caused by anti-tumor treatment (radiation therapy [RT] or chemotherapy), represent an acute complication of the malignancy (hypercalcemia, malignant bowel obstruction), be due to gastroparesis (eg, from autonomic dysfunction), or be from drug-induced constipation (opioids, type 3 serotonin receptor [5-HT3] antagonists). These various causes of nausea and vomiting may occur simultaneously or sequentially in any individual patient.
This topic review will cover the prevalence, etiology/pathophysiology, assessment, and management of nausea and vomiting in palliative care populations. Prevention and treatment of chemotherapy- and radiation therapy-induced nausea and vomiting are covered in detail elsewhere. (See "Prevention of chemotherapy-induced nausea and vomiting in adults" and "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting" and "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment" and "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults".)
PREVALENCE
Cancer patients — Symptom research in patients who have a serious and/or life-threatening illness has largely been conducted in those with cancer. Cancer treatment is one of the more common causes of nausea and vomiting in such patients. Despite adherence to published guidelines on prophylactic antiemetics, breakthrough nausea and vomiting affects up to 40 percent of patients treated with either chemotherapy or radiation therapy (RT).
Radiation therapy — The incidence and severity of radiotherapy-induced nausea and vomiting (RINV) are both treatment related (irradiated site and volume, single and total dose, fractionation schedule, techniques) and patient related. The most important factor appears to be the radiation field (table 1). Between 90 and 100 percent of patients receiving total body irradiation, 30 to 90 percent of those undergoing upper abdominal or craniospinal irradiation, and 10 to 30 percent of those receiving brain, head and neck, thorax, or pelvic irradiation will develop RINV. Irradiation to the breast or extremities produces a less than 10 percent risk. This subject is discussed in detail elsewhere. (See "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment", section on 'Risk classification'.)
Chemotherapy — Chemotherapy is a well-known cause of nausea and vomiting. Three distinct types of chemotherapy-induced nausea and vomiting (CINV) are defined (see "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Types of emesis'):
●Acute emesis, which most commonly begins within one to two hours of chemotherapy and usually peaks in the first four to six hours
●Delayed emesis, occurring more than 24 hours after chemotherapy
●Anticipatory emesis, occurring prior to treatment as a conditioned response in patients who have developed significant nausea and vomiting during previous cycles of chemotherapy
The single most important factor predicting the likelihood that nausea and/or emesis will develop during chemotherapy is the intrinsic emetogenicity of the chemotherapy agent(s). Chemotherapy agents are divided into four emetogenic levels, which are defined by the expected frequency of emesis in the absence of effective antiemetic prophylaxis (see "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Chemotherapy agent'):
●High – >90 percent risk of emesis
●Moderate – >30 to 90 percent risk of emesis
●Low – 10 to 30 percent risk of emesis
●Minimal – <10 percent risk of emesis
The predicted acute emetogenicity potential of individual parenteral and oral chemotherapy agents is provided (table 2 and table 3). For combination regimens, the emetogenic level is determined by identifying the most emetogenic agent in the combination and then assessing the relative contribution of the other agents. As an example, cyclophosphamide and doxorubicin are both moderately emetogenic agents, but when given together, the regimen is considered to be highly emetogenic. (See "Pathophysiology and prediction of chemotherapy-induced nausea and vomiting", section on 'Chemotherapy agent'.)
Chemotherapy agents with the greatest potential for acute emesis in the absence of effective prophylaxis (high-dose cisplatin and regimens that combine cyclophosphamide and doxorubicin) are also the regimens that are most likely to cause delayed emesis. Other moderately emetogenic agents associated with delayed emesis include doxorubicin ≥40 mg/m2 as a single agent, epirubicin ≥75 mg/m2 as a single agent, combinations of cyclophosphamide ≥600 mg/m2 in combination with drugs other than doxorubicin, carboplatin ≥300 mg/m2, and oxaliplatin. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Regimens with a high risk of delayed emesis'.)
Despite the improved control of both acute and delayed CINV when prophylaxis is provided according to published guidelines, breakthrough CINV, defined as nausea or emesis that occurs despite adequate antiemetic prophylaxis, remains a significant patient problem. The incidence of breakthrough symptoms is as high as 40 percent [1-5]. In particular, conventional antiemetics are more successful at preventing emesis than in preventing nausea, particularly delayed nausea. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Prevention of delayed emesis' and "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults".)
Unrelated to cancer therapy — Nausea and vomiting that is unrelated to cancer therapy is also common in cancer patients at the end of life. In several reports, the prevalence of nausea and/or vomiting in terminal cancer patients who are no longer receiving anti-tumor treatment ranges from 30 to 60 percent [6-13]. Although nausea is frequently reported, longitudinal studies measuring symptom intensity with the Edmonton Symptom Assessment Scale (ESAS) suggest that it is generally not severe at the end of life; fewer than 20 percent of affected patients have a score ≥4 (moderate intensity or worse) during the last seven days of life [14]. A population based cohort of over 10,000 patients also found nausea intensity to be low (mean ESAS <1.5) with a stable, flat trajectory over the last six months of life [15]. Effective, readily available pharmacologic antiemetics may in part explain the low ESAS scores for nausea.
Other palliative care patients — There are fewer studies describing the prevalence of nausea and vomiting in patients with an advanced serious and/or life-threatening illness other than cancer. The following data are available:
●A review of the literature [6] comparing the prevalence of symptoms among end-stage patients with cancer, AIDS, chronic obstructive pulmonary disease (COPD), chronic heart failure (HF), or chronic kidney disease (CKD) found that rates of nausea ranged from 43 to 49 percent in patients with AIDS, 17 to 48 percent in HF, and 30 to 43 percent in patients with CKD. No data could be obtained for the prevalence of nausea among patients with COPD. However, a subsequent study of patients with advanced COPD found a prevalence of 18 percent for nausea and 4 percent for emesis [16].
●A second systematic review of 39 articles found similar results as regards to the heterogeneity and prevalence of symptoms in patients with various end-stage diseases [17]. Patients with CKD and HF were again found to have the highest prevalence of nausea (about 40 and 25 percent, respectively), although coexisting morbidities were considerable in some of the studies and the definition, intensity, and etiology of the nausea were unclear.
Some of the challenge inherent to studying nausea and emesis in palliative care populations with life-limiting illness is that there may not be a clear distinction between nausea and vomiting. The distinction is important, since nausea and vomiting may have different pathophysiologic mechanisms and also different responses to interventions.
PATHOPHYSIOLOGY AND ETIOLOGY — In the setting of palliative care, identifying the underlying cause(s) of nausea and vomiting may be useful for selecting the therapeutic option. However, the success of etiology-based therapy is limited by the multifactorial etiology of nausea and vomiting, particularly in cancer patients, and the lack of prospective data demonstrating the superiority of this approach over empirically chosen antiemetic therapy (with the exception of chemotherapy- or radiation-related nausea and vomiting and treatment for malignant bowel obstruction). Nevertheless, etiology-based management of nausea and vomiting in palliative care patients is widely practiced. (See 'Pharmacologic approaches' below.)
Pathophysiologic pathways — An important distinction must be made between the mechanisms producing emesis (vomiting, retching) and those causing nausea. Emesis is not merely a more severe degree of nausea, since the neural circuits responsible for nausea appear to be anatomically distinct from those that generate emesis.
Emesis is mediated centrally by two separate “centers”: the chemoreceptor trigger zone (CTZ) in the area postrema in the floor of the fourth ventricle, and the vomiting center (VC), an anatomically indistinct area that includes the nucleus of the tractus solitarius (NTS) and the reticular formation of the medulla oblongata (figure 1). Neural input from the CTZ, vestibular system, higher cortical centers, and thoracoabdominal organs is integrated in the VC. Chemosensitive receptors of the CTZ detect emetic agents in the blood and relay this information to the adjacent VC. The CTZ lacks a blood-brain barrier, and this allows toxins to activate the CTZ via dopamine 2 receptors. Peripheral signals from abdominal vagal afferents and splanchnic nerves that detect intestinal luminal contents and gastric tone via mechanoreceptors or by the release of serotonin from gut enterochromaffin cells also terminate in the NTS and can lead to emesis [18]. The vomiting reflex is coordinated by the VC in the brainstem and does not require input from the cortex. As such, autonomic activation, sepsis, and motion may induce vomiting in decorticate individuals.
On the other hand, the mechanisms underlying the symptom of nausea are not well understood. However, unlike vomiting, nausea requires conscious awareness and cerebral function [19,20]. Studies of motion sickness in humans suggest involvement of gastric myoelectrical activity and endogenous neuroendocrine responses (ie, increases in plasma cortisol, beta-endorphins, epinephrine and norepinephrine). (See "Approach to the adult with nausea and vomiting", section on 'Nausea'.)
Furthermore, the concept that nausea is mediated through different pathways than emesis is supported by the observation that nausea and vomiting seem to be pharmacologically distinct. Among patients receiving emetogenic chemotherapy, neurokinin-1 (NK1) receptor antagonists and type 3 serotonin (5-HT3) receptor antagonists appear to control emesis more effectively than they control nausea, particularly delayed nausea [21]. (See "Prevention of chemotherapy-induced nausea and vomiting in adults".)
Etiology and differential diagnosis — In palliative care patients, particularly those with advanced cancer, the etiology of nausea and vomiting may be multifactorial [22]. Although the differential diagnosis is broad (table 4), the most common causes of nausea and vomiting in this population are outlined in the table (table 5). (See "Approach to the adult with nausea and vomiting".)
PATIENT ASSESSMENT — In many cases, the cause of the nausea and vomiting can be determined from the history and physical examination. As noted above, the most common causes of nausea and vomiting in palliative care patients are outlined in the table (table 5). Most patients have two or more contributing factors [22].
Additional laboratory investigations or imaging may be required depending on the clinical presentation and prognosis of the patient. However, even in advanced disease, when the burden of diagnostic testing may outweigh the benefit, additional investigations (eg, abdominal computed tomography [CT]) may be justified if they have the potential for identifying potentially modifiable etiologies that might influence management.
For patients near death, empiric treatment for nausea (eg, with haloperidol, metoclopramide, or chlorpromazine) is reasonable without a search for a specific etiology [23]. (See "Palliative care: The last hours and days of life", section on 'Nausea'.)
History and physical examination — The history should include a complete list of medications, both prescription and over-the-counter. The potential for drug-induced nausea should always be considered since the offending medications can often be discontinued or switched to an alternative. Opioids, tramadol, nonsteroidal antiinflammatory drugs (NSAIDs), iron supplements, selective serotonin reuptake inhibitors (SSRIs), bupropion, antibiotics, and chemotherapy or radiation can all contribute to nausea in palliative care patients (table 5).
The following clinical features are especially important for assessing potential etiologies:
●Abdominal pain with vomiting often indicates an organic etiology (eg, bowel obstruction). (See "Acute calculous cholecystitis: Clinical features and diagnosis".)
●Patients receiving chemotherapy and/or radiation therapy may have breakthrough emesis despite adequate prophylaxis according to published guidelines. However, it is important to exclude other disease- and medication-related causes such as:
•The use of opioid analgesics or certain antibiotics (eg, erythromycin)
•Central nervous system metastases
•Gastrointestinal obstruction or gastroparesis
•Hypercalcemia
Patients who are anxious and have experienced severe and frequent nausea or vomiting after prior chemotherapy are at high risk of developing anticipatory nausea and vomiting (ANV) [24]. ANV is a conditioned or learned response that develops in some patients, particularly those who are <50, are susceptible to motion sickness or have expectations of post-treatment nausea. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Anticipatory emesis'.)
●Abdominal distension and tenderness with high-pitched bowel sounds suggests malignant bowel obstruction (MBO). MBO should be suspected in a patient with progressively worsening colicky pain with diminishing stool and flatus. Among patients with advanced cancer, female sex [25] and tumor types such as gastrointestinal malignancies [26] and gynecologic malignancies are associated with increased risk of nausea and vomiting and also a greater risk of MBO.
A proximal obstruction (gastric and small bowel) may be distinguished from a distal obstruction (distal small bowel and colon) by large-volume postprandial emesis. A distal obstruction is more likely to be accompanied by small-volume, malodorous (feculent) emesis and abdominal distension [27].
Feculent vomiting can also be associated with a gastrocolic fistula. (See "Etiologies, clinical manifestations, and diagnosis of mechanical small bowel obstruction in adults" and "Acute colonic pseudo-obstruction (Ogilvie's syndrome)".)
●The relationship between eating and vomiting may also provide clues to the etiology. Vomiting of food eaten several hours earlier and a succussion splash detected on abdominal examination suggest gastric outlet obstruction or gastroparesis. (See "Gastroparesis: Etiology, clinical manifestations, and diagnosis" and "Gastric outlet obstruction in adults".)
Gastroparesis can be distinguished from mechanical obstruction by bowel movements and clinical response to prokinetic medications.
●In patients with gastroparesis or obstruction, nausea is usually relieved after emesis; by contrast, patients suffering from drug, chemical, or metabolic causes have persistent nausea that is unrelieved by vomiting.
●Heartburn with nausea often indicates gastroesophageal reflux disease (GERD), and GERD can present as chronic nausea without typical reflux symptoms [28]. (See "Clinical manifestations and diagnosis of gastroesophageal reflux in adults".)
●Vertigo and nystagmus are typical of vestibular neuritis (labyrinthitis) and other causes of vestibular dysfunction. (See "Vestibular neuritis and labyrinthitis" and "Causes of vertigo".)
●Neurogenic vomiting may be positional, as well as projectile, and is usually associated with other neurologic signs or symptoms. (See "Overview of the clinical features and diagnosis of brain tumors in adults".)
Patients who have accompanying personality changes or confusion associated with vomiting may also have metabolic abnormalities (eg, hyponatremia, hypercalcemia, adrenal insufficiency). Intracranial disease is more likely to present with headache and motor sensory deficits including hemiparesis and gait disturbance, whereas metabolic abnormalities are more likely to be associated with sedation or delirium. Delirium accompanied by nausea may also be due to sepsis.
Projectile vomiting may also be associated with gastric obstruction.
Focal neurologic signs or papilledema would especially suggest raised intracranial pressure as an etiology for emesis. (See "Overview and differential diagnosis of papilledema".)
●Constipation is a frequent cause of nausea. Patients should be questioned about the consistency and the frequency of their bowel movements, and a digital rectal examination should be performed to assess the presence of fecal impaction. The combination of autonomic dysfunction, opioids, use of type 3 serotonin (5-HT3) antagonists as antiemetics, and primary gastrointestinal tumor or metastases increases the risk for severe constipation that can exacerbate nausea.
●Although dehydration is thought to be associated with nausea in palliative care patients [29], symptomatic improvement from parenteral hydration has not been shown in patients with advanced cancer. (See "Stopping nutrition and hydration at the end of life", section on 'Impact on target symptoms'.)
●Patients who have hepatomegaly and/or ascites may have nausea and vomiting due to extrinsic compression on the stomach. For patients with malignant ascites, one prospective study of 24 patients demonstrated that placement of a permanent peritoneal port for periodic drainage improved nausea scores by two or more points on a 0 to 10 scale [30].
●Comorbid conditions such as myocardial ischemia, adrenal insufficiency, pancreatitis, and hepatitis should be considered in the differential diagnosis, as they may also cause nausea and vomiting in palliative care patients.
Investigations — Potentially useful laboratory investigations for diagnosis and management include serum sodium, calcium, albumin, potassium, creatinine, and blood urea nitrogen/creatinine. Adrenal insufficiency may present with hyponatremia and hyperkalemia and can be diagnosed if a serum cortisol level is <4 mcg/dL. Usually, however, additional testing will be required for a definitive diagnosis of adrenal insufficiency [31]. (See "Clinical manifestations of adrenal insufficiency in adults" and "Diagnosis of adrenal insufficiency in adults".)
As noted above, even in advanced disease, where the burden of diagnostic testing may outweigh the benefit, additional investigations may be justified if they have the potential for identifying potentially modifiable etiologies that might influence management. As examples:
●A plain abdominal radiograph can assist in the identification of constipation, ileus, or gastrointestinal tract obstruction.
●Abdominal CT is more sensitive and specific than plain radiographs for diagnosing a bowel obstruction. However, the ability of CT to detect individual peritoneal implants is quite poor (<20 percent) when the tumor size is <1 cm [32].
●Magnetic resonance imaging (MRI) of the brain might be useful if no recent imaging has been done in a patient with a clinical presentation suggesting an intracranial etiology for nausea and vomiting.
MANAGEMENT
Overview
Pharmacologic approaches — Dopamine, histamine, serotonin (type 3 [5-HT3]), and muscarinic (acetylcholine) receptor antagonists are all conventional, established classes of antiemetic and antinausea agents that have been used in the palliative care setting (table 6). These and other occasionally used agents vary in their selectivity and some (eg, chlorpromazine and olanzapine) bind to multiple receptors. With the exception of metoclopramide (which has serotonin 5HT4 receptor agonist activity), antiemetics with affinity to other receptors (eg, cannabinoids ghrelin, and serotonin 5HT1A receptor agonists) have not been used as frequently in patients with advanced cancer but may prove to be useful in the future.
In general, it is conventional teaching in palliative care to base the choice of the antiemetic on the presumptive etiology of the nausea and vomiting (the so-called etiology or mechanistic-based approach as opposed to the empiric approach) even though outside of the setting of chemotherapy-induced nausea and vomiting (CINV), the evidence to prove benefit for this approach is weak:
●In a prospective study, the management of nausea and vomiting in 121 patients with advanced cancer admitted to hospice based upon suspected etiology was only moderately effective (after one week, vomiting was controlled in 89 percent of patients, but nausea was only controlled in 56 percent); about one-fourth of patients were thought to have more than one cause for their nausea [33].
●The value of etiology-based as compared with empiric antiemetic therapy was addressed in a systematic review of 93 studies of treatment of nausea and vomiting in cancer unrelated to chemotherapy or radiation therapy (RT) that was conducted by the Multinational Association of Supportive Care in Cancer (MASCC) [13]. The studies included one systematic review and 18 randomized trials, many of which were low quality based on either lack of blinding, lack of description of the method of randomization, concealment, and/or attrition. The authors concluded that there was no evidence that antiemetic choice based on the presumed etiology of the nausea and vomiting was any better than a single empirically chosen antiemetic.
Despite the paucity of evidence, an etiology-based approach to pharmacologic management of nausea and vomiting in palliative care is widely practiced. Medications and doses for specific causes of nausea and vomiting that may be encountered in palliative care patients are listed in the table and discussed in more detail below (table 6).
Cannabinoids and cannabis — Synthetic oral cannabinoids should be limited to the management of breakthrough nausea and vomiting caused by chemotherapy. (See 'Management of breakthrough symptoms' below.)
We suggest not pursuing the use of medical marijuana for refractory nausea and vomiting in palliative care patients. The use of medical marijuana for refractory nausea and emesis in patients with a terminal illness is very controversial. Medical use of marijuana is legal in several countries, including the Netherlands and Canada. Despite legalization by several states, marijuana use is still illegal in the United States at the federal level and there are no controlled clinical studies to support marijuana’s efficacy for symptom management [34]. Furthermore, there can be disturbing central nervous system side effects, particularly among older individuals. Because of medical and legal concerns, the use of medical marijuana is not included in the guidelines for CINV from the National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), or MASCC [13,34,35]. This subject is discussed in detail elsewhere. (See "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults", section on 'Cannabinoids and medical marijuana'.)
Nonpharmacologic approaches — A variety of environmental and psychologic factors can contribute to nausea. Simple preventive measures such as minimizing the sights, sounds, and smells associated with nausea may be effective in the palliative care setting. Although avoidance of fatty, spicy, highly salted foods is frequently suggested [36], there are no published studies on the benefits of dietary modification in reducing nausea.
Complementary and integrative medicine measures (eg, acupuncture, ginger, guided imagery, progressive muscle relaxation, music therapy) have some evidence for benefit in control of nausea and vomiting; however, almost all studies have been conducted in patients with CINV [9]. Very few studies have examined the benefits of any of these complementary therapies, including acupuncture or acupressure, in patients with chronic nausea unrelated to chemotherapy, and the benefits in palliative care populations remain uncertain [37]. (See 'Complementary and alternative therapies for patients with advanced cancer' below and "Complementary and alternative therapies for cancer", section on 'Nausea and vomiting' and "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults", section on 'Complementary therapies'.)
Cancer patients — The management of nausea and vomiting in patients with cancer should be individualized and targeted toward the specific etiology. For example, patients with nausea as a result of bowel obstruction will be managed differently than those with nausea as a result of hypercalcemia. In addition, patients with CINV or radiotherapy-induced nausea and vomiting (RINV) require a different treatment strategy than those who have advanced cancer and are no longer receiving disease-specific therapy.
Active antitumor therapy
Chemotherapy — The medications used for CINV include 5-HT3 antagonists, corticosteroids, and neurokinin-1 (NK1) antagonists for highly emetogenic therapy, while 5-HT3 antagonists and corticosteroids are combined for moderately emetogenic treatment, and corticosteroids alone are recommended for regimens with a low emetogenic risk. Guidelines for prevention of CINV based upon emetic risk category are outlined in the table (table 7) and discussed in more detail elsewhere.
For patients with nausea and emesis that occurs despite adequate antiemetic prophylaxis (“breakthrough” nausea and emesis), it is important to exclude other disease- and medication-related causes before considering changes to the antiemetic regimen. Examples include:
●The use of opioid analgesics or certain antibiotics (eg, erythromycin)
●Central nervous system metastases
●Gastrointestinal obstruction or gastroparesis
●Hypercalcemia
Assuming these factors are excluded, it is important to ensure that the patient is receiving the antiemetic appropriate for the drug(s) being given (table 2 and table 3) and the correct dose. If the patient was receiving chemotherapy with a low emetic risk and experiencing poor emesis control, the antiemetic regimen could be adjusted to one typically used for a higher-risk group.
Management of breakthrough symptoms — The majority of patients who have breakthrough emesis have derived some benefit from the original antiemetic regimen employed, and additional agents can be added, such as lorazepam or alprazolam, olanzapine [38-40], or a dopaminergic antagonist (eg, prochlorperazine, thiethylperazine, haloperidol). A systematic review and meta-analysis comparing olanzapine with other standard antiemetics for either prevention (10 randomized controlled trials) or breakthrough treatment (three randomized controlled trials) found olanzapine inclusion improved CINV prophylaxis and was superior to standard therapy for rescue treatment [41]. Another approach is substituting high-dose intravenous (IV) metoclopramide for the 5-HT3 antagonist (30 to 40 mg/day, titrating to no more than 60 to 100 mg/day as needed). Another alternative is to switch to a different 5-HT3 antagonist, since there might be slight differences in action between agents.
Antiseizure medications such as gabapentin may decrease nausea in a variety of conditions including delayed CINV [42].
Another option is the use of synthetic oral cannabinoids such as dronabinol or nabilone (not available in the United States), although antiemetic efficacy is modest at best, rigorous comparisons of either drug with the most effective antiemetic therapies for CINV are lacking, and adverse effects tend to be more intense and more frequent than with other rescue agents such as neuroleptics. The modest antiemetic activity of this class of agents combined with their relatively unfavorable side effect profile (vertigo, xerostomia, hypotension, dysphoria), especially in older patients, has limited their clinical utility. Nevertheless, guidelines from the NCCN [43], ASCO [35], and the MASCC [44] state that cannabinoids can be considered for refractory nausea and vomiting and as a rescue antiemetic. (See "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults", section on 'Cannabinoids and medical marijuana'.)
As noted above, the use of medical marijuana for refractory nausea and emesis is very controversial. Because of medical and legal concerns, the use of medical marijuana is not recommended. (See 'Cannabinoids and cannabis' above.)
For patients with anticipatory nausea and vomiting (ANV), behavioral interventions such as guided imagery, biofeedback, and hypnosis have all shown benefit [24]. Benzodiazepines are the most useful pharmacologic therapy for ANV and have been combined with behavioral therapies or other antiemetics. This subject is discussed in detail elsewhere. (See "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Anticipatory emesis'.)
Radiation therapy — The incidence and severity of RINV are related to both treatment (irradiated site and volume, single and total dose, fractionation schedule, techniques) and patient factors. Risk categories for emesis based upon the site of irradiation stratify a patient’s risk for RINV into four categories (table 1).
Specific guidelines for prophylaxis based upon these risk categories are available from ASCO (table 8) [35] and largely follow published guidelines from the MASCC/European Society of Medical Oncology (ESMO) [45]. For patients being treated with concomitant RT and chemotherapy, antiemetic prophylaxis should be based upon the higher chemotherapy or RT emesis risk category. Notably, in contrast to CINV, the evidence base from randomized trials to support these specific guidelines is more limited. (See "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment", section on 'Prevention and treatment' and "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Estimating the risk of nausea and vomiting'.)
Few studies have examined the benefit of specific rescue strategies for treatment of breakthrough nausea in patient receiving prophylactic therapy for RINV. A 5-HT3 receptor antagonist (eg, ondansetron) is an option. Alternatively, olanzapine is another reasonable option given its activity in delayed CINV. (See "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment", section on 'Rescue therapy' and "Prevention of chemotherapy-induced nausea and vomiting in adults", section on 'Olanzapine'.)
Advanced cancer not receiving antitumor therapy — There are multiple etiologies of nausea in patients with advanced cancer who are not receiving chemotherapy or RT, including constipation, medications and metabolic abnormalities, central nervous system malignancy, bowel obstruction, gastroparesis, hepatomegaly, or ascites. A summary of the approach to management of non-chemotherapy or RT-related nausea and vomiting in patients with terminal cancer is provided in the figure (algorithm 1). A detailed history and physical examination are required as they may provide clues to the etiology; investigations such as computed tomography (CT) scans or abdominal radiographs may also be required in selected patients. (See 'Patient assessment' above.)
Potentially treatable cause — Potentially reversible causes should be sought for and remedied.
Medications including opioids — Medications such as opioids (particularly in the first 72 hours after initiation of therapy), antibiotics, tramadol, nonsteroidal antiinflammatory drugs (NSAIDs), and antidepressants should be excluded as causes of nausea or vomiting (table 5).
Opioids have three potentially emetogenic mechanisms: a direct effect on the chemoreceptor trigger zone (CTZ), enhanced vestibular sensitivity, and delayed gastric emptying. Refractory constipation and stool impaction may be contributory. When present, constipation should be managed before addressing opioid-induced nausea. (See "Prevention and management of side effects in patients receiving opioids for chronic pain", section on 'Opioid bowel dysfunction'.)
Chronic nausea from opioid therapy usually responds to the same group of drug therapies that are used for acute nausea and vomiting. Opioid rotation or a change in route of administration could also be considered, although the data to support benefit from any of these strategies are limited [13]. A systematic review [46] concluded that while there was some weak evidence for changing the opioid or the route of administration, the data were too limited to give evidence-based recommendations for the use of antiemetics for opioid-induced nausea and vomiting in patients with cancer.
For patients who do not respond to opioid rotation, it seems reasonable to initiate metoclopramide and to switch to an atypical antipsychotic such as olanzapine or risperidone in refractory cases. Although both metoclopramide and prochlorperazine can cause extrapyramidal symptoms, metoclopramide has the advantage of causing less sedation and promoting gastric motility. In a trial of prophylactic antiemetics among cancer patients starting oral oxycodone, nausea and vomiting outcomes were similar between those receiving prochlorperazine and placebo, although there was more somnolence in the prochlorperazine group [47]. On the other hand, 5-HT3 receptor antagonists have the disadvantage of exacerbating constipation in patients who might already be experiencing opioid-induced gastrointestinal dysmotility. This subject is discussed in detail elsewhere. (See "Prevention and management of side effects in patients receiving opioids for chronic pain".)
Adrenal insufficiency — The abrupt discontinuation of corticosteroids or megestrol acetate can precipitate nausea due to adrenal insufficiency for which cortisol treatment is indicated. (See "Causes of secondary and tertiary adrenal insufficiency in adults" and "Treatment of adrenal insufficiency in adults".)
Constipation — Severe constipation is another easily reversible cause of nausea that needs to be considered especially in patients receiving opioids. Some patients are able to improve bowel function by dietary modifications, including increased consumption of fluids and dietary fiber. Fiber should be discontinued, however, if the patient is debilitated, bowel obstruction is suspected, or hydration has been difficult to maintain. Patients who have passed no stool in several days and who have no evidence of bowel obstruction or ileus are likely to be impacted. It is often possible to clear the rectal vault and lower sigmoid colon with a mineral oil enema followed by an irritant enema, but manual disimpaction may be required. Once an impaction has been ruled out or cleared, laxative therapy may be started.
There are numerous options for laxative therapy (table 9) and no data to suggest that any one approach is superior to any other. Refractory cases may require the use of a peripherally acting opioid antagonist such as methylnaltrexone. However, concerns have been raised about severe abdominal pain and bowel perforation in patients with advanced cancer who were receiving methylnaltrexone [48]. These concerns led the FDA to issue a warning for clinicians to use caution in administering methylnaltrexone to patients with known or suspected lesions in the intestinal wall, and to stop the drug immediately for worsening of gastrointestinal symptoms [49]. (See "Prevention and management of side effects in patients receiving opioids for chronic pain", section on 'Opioid bowel dysfunction'.)
Raised intracranial pressure — Nausea and vomiting are present in up to 40 percent of patients with primary or secondary brain tumors [50]. Corticosteroids are first-line therapy for associated symptoms of nausea, given their ability to decrease intracerebral edema. Antiseizure medications such as carbamazepine may decrease nausea in a variety of conditions including leptomeningeal carcinomatosis [51]. (See "Management of vasogenic edema in patients with primary and metastatic brain tumors" and "Treatment of leptomeningeal disease from solid tumors", section on 'Hydrocephalus and increased intracranial pressure'.)
Gastroduodenal dysmotility — Early satiety and vomiting of food eaten several hours earlier suggests gastroparesis. Dietary modification and pharmacologic management using prokinetics such as metoclopramide are discussed elsewhere. (See "Treatment of gastroparesis" and "Gastroparesis: Etiology, clinical manifestations, and diagnosis".)
Gastrointestinal obstruction — Patients with cancer may develop upper or lower gastrointestinal tract obstruction due to intrinsic or extrinsic compression, adhesions, or postradiation fibrosis. Tumors can also impair bowel motility by infiltration of mesentery, nerves, celiac plexus, and bowel muscle. (See "Etiologies, clinical manifestations, and diagnosis of mechanical small bowel obstruction in adults".)
Gastroduodenal obstruction from either primary gastric or duodenal cancers or from extrinsically compressing tumors causes severe nausea and vomiting. Stenting usually provides adequate and sustained palliation. (See "Enteral stents for the palliation of malignant gastroduodenal obstruction".)
Malignant large and/or small bowel obstruction (malignant bowel obstruction [MBO]) is a well-recognized complication in patients with advanced intraabdominal or pelvic tumors. Most of these patients are inoperable, and their survival is generally short. Clinical management of MBO requires a specific and individualized approach that is based on disease prognosis and the objectives of care. In most cases, there are few or no data addressing the relative value of palliative surgery versus medical management, and a decision to proceed to surgical intervention requires careful weighing of risks and benefits, including an assessment of the estimated life expectancy and patient goals and preferences. This subject is discussed in detail elsewhere. (See "Palliative care of bowel obstruction in cancer patients".)
No obvious cause — Metoclopramide is a reasonable initial choice for palliative care patients with nausea and vomiting for whom a cause is not identified based upon the initial evaluation. This recommendation is supported by published guidelines from the Multinational Association of Supportive Care in Cancer/European Society of Medical Oncology (MASCC/ESMO) [13]. Other options are also described below.
While treatment of CINV and RINV is supported by well-performed randomized controlled trials, high-quality evidence to support any specific antiemetic approach for patients with terminal cancer who have non-chemotherapy and non-RT-related nausea and no other obvious cause for the symptom is lacking. A systematic review of the treatment of nausea and/or emesis in cancer patients unrelated to chemotherapy and radiation included 93 studies and 14 randomized trials; the authors concluded that there was moderate-quality evidence for benefit from first-line therapy with metoclopramide [52]. Formal meta-analysis was not possible because of heterogeneity in patient populations, antiemetics, and antiemetic doses in the clinical trials.
●Metoclopramide – The appropriate dose is 10 mg orally every four hours. Metoclopramide is more effective as a controlled-release formula and also may be more effective after eight days of therapy than on day one [53]. In patients with advanced cancer, controlled release metoclopramide is well tolerated during short-term use, with no increase in frequency or severity of side effects compared with placebo [54]. Unfortunately, there is no controlled-release formula available in the United States. Patients not responding to a divided dose of 40 to 60 mg/day may obtain relief with a continuous subcutaneous or IV infusion of up to 4 to 5 mg/hr (total = 100 to 120 mg/day) [55,56]. It is also available as a nasal spray, dosed as one spray (15 mg) in one nostril four times daily (30 minutes prior to each meal and at bedtime). Metoclopramide has been used successfully as a rescue antiemetic in a large trial [57].
Adding or combining antiemetics with different mechanisms of action (eg, 5-HT3 receptor blockers, dopamine [D2] receptor antagonists such as chlorpromazine or haloperidol) has proven useful for preventing and treating CINV; however, the available evidence suggests that the addition of other agents to metoclopramide may not provide further benefit in controlling chronic nausea in patients with advanced cancer:
•A randomized controlled trial comparing the combination of dexamethasone and metoclopramide versus metoclopramide alone found no benefit for combination therapy [53].
•In another prospective study of patients with advanced cancer, the addition of a 5-HT3 receptor antagonist to metoclopramide was more effective at controlling nausea after three days than was the combination of metoclopramide plus dexamethasone [58]. However, this study was limited by the relatively low dose of metoclopramide (only 10 mg three times/day); ideally, metoclopramide should be given every four hours because of its short half-life.
Concerns about the risk of tardive dyskinesia (TD) have decreased the use of metoclopramide [59]; however, a literature review suggests the risk of TD from metoclopramide use is likely to be <1 percent, much less than the estimated 1 to 10 percent risk previously proposed in national guidelines [60]. However, prolonged use for more than three months increases the risk of TD, so the risks and benefits of continued use after three months should be carefully evaluated and discussed with patients.
●Other options
•For terminally ill patients who have contraindications to or do not respond well to metoclopramide, a number of other centrally acting antiemetic agents (eg, haloperidol, chlorpromazine [or, where available levomepromazine], olanzapine) can be used. Haloperidol is often used as a second-line drug or when patients have gastrointestinal tract obstruction, and there have been few trials evaluating its use in palliative care demonstrating some efficacy [57,61,62]. There are no trials evaluating the use of chlorpromazine/levomepromazine; nevertheless, these recommendations are consistent with recommendations from MASCC [13].
A particularly well-tolerated option is olanzapine [40,63]. In a trial of 30 patients with advanced cancer and persistent nausea and vomiting lasting for at least one week, unrelated to chemotherapy or radiation therapy (baseline nausea severity 8 to 10 on a 10-point scale), olanzapine 5 mg daily for seven days was well tolerated and resulted in an 8-point greater reduction in nausea scores when compared with placebo [63]. Patients in the olanzapine arm also reported less emesis, less use of other antiemetic drugs, better appetite, less sedation, less fatigue, and better wellbeing. Although larger trials are preferred to demonstrate benefit, olanzapine appears to be a reasonable alternative to metoclopramide in patients with advanced cancer and no clear etiology for the chronic nausea and vomiting.
•In the postoperative setting, 5-HT3 receptor antagonists can be useful and may be considered in other clinical scenarios of refractory nausea. Tropisetron and levosulpiride are considered third-line antiemetics in patients with advanced cancer [13].
•Corticosteroids are commonly used in palliative care for a wide variety of nonspecific indications, including nausea. However, there are few studies assessing the effect of corticosteroids on nausea and vomiting not related to chemotherapy, RT, or surgery [53,58,64], and, at least in the setting of advanced cancer, there is very low-quality evidence from randomized trials that neither supports nor refutes the utility of corticosteroids in this setting [65]. The utility of corticosteroids in medical treatment of MBO is discussed elsewhere. (See "Palliative care of bowel obstruction in cancer patients", section on 'Pharmacologic management'.)
For patients who are unable to take oral medications, metoclopramide, dexamethasone, and haloperidol can be safely administered intravenously and subcutaneously. Metoclopramide is also available in a nasal formulation. The rectal route can occasionally be used but may be uncomfortable for patients because of the need for frequent administration given the short duration of action of antiemetics. (See "Palliative care: The last hours and days of life", section on 'Nausea'.)
Complementary and alternative therapies for patients with advanced cancer — Acupuncture or acupressure may be of benefit for CINV, but the available evidence is insufficient to support use in other palliative care populations. However, given the overall safety of this approach, interested patients may be referred for a trial of acupuncture if symptoms of moderate to severe nausea persist despite other forms of therapy.
Very few complementary therapies have been tested in patients with advanced cancer with non-chemotherapy-induced nausea and vomiting:
●A small crossover study of six hospice patients receiving acupuncture, acupressure, or sham wrist band reported no difference between the interventions [66].
●A non-blinded observational study found acupressure by "motion sickness bands" improved nausea and vomiting in 29 of 33 patients [67]. Eighty-five percent of the patients had a diagnosis of cancer and 18 of the patients had complete resolution of their symptoms.
●Another small study of 10 terminal cancer patients comparing acupressure wrist bands with sham wrist bands also found benefit from acupressure in decreasing episodes of emesis and reducing the need for additional doses of antiemetics.
These preliminary findings have not yet been confirmed in any larger randomized trials. There are no data on acupuncture or acupressure for nausea and vomiting in other palliative care populations.
With regard to CINV, a systematic review and meta-analysis [68] found benefit from acupuncture. However, for RT-induced nausea, penetrating acupuncture was no more effective than sham acupuncture, although 95 percent of patients in both groups believed the treatment had been successful for nausea [69]. This subject is addressed in detail elsewhere. (See "Complementary and alternative therapies for cancer", section on 'Nausea and vomiting' and "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults", section on 'Acupuncture and related therapies'.)
Ginger is another complementary therapy that may benefit patients with cancer and nausea and vomiting, although as with acupressure, the data are mixed. (See "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults", section on 'Ginger'.)
Topical gels containing lorazepam, diphenhydramine, and haloperidol without (ABH) or with metoclopramide (ABHR) have been used in patients with refractory nausea and vomiting [70]; however, there is no evidence that they are absorbed through the skin [71], and at least one placebo-controlled trial has failed to show benefit [72]. We do not recommend their use.
Non-cancer conditions — For life-limiting illnesses other than cancer, the assessment should be based on the approach to the adult with nausea and vomiting. (See "Approach to the adult with nausea and vomiting".)
Where there is evidence for management that is specific to the condition, it has been noted in the text. There are few pharmacologic therapies that have been evaluated for treating nausea or vomiting in non-cancer palliative care patients.
Chronic obstructive pulmonary disease (COPD) — There are no studies specifically evaluating nausea in patients with end-stage COPD. One study evaluating symptoms specifically related to gastroesophageal reflux disease (GERD) such as indigestion, abdominal pain, and reflux found that patients with COPD had more gastrointestinal symptoms and impaired psychological wellbeing than those without COPD [73]. The severity of COPD did not correlate with a higher gastrointestinal symptom burden, although other studies have described a tendency toward an inverse correlation between GERD symptoms and the forced expiratory volume in one second (FEV1) [74,75]. No specific therapy can be recommended for patients with COPD. (See "Medical management of gastroesophageal reflux disease in adults".)
Much of the research reporting nausea or vomiting in patients with COPD is concerned with medication side effects, particularly the use of methylxanthines, which are associated with nausea and vomiting [76]. Use of oral corticosteroids does not appear to increase the risk for peptic ulcer disease or gastrointestinal symptoms in patients with COPD [73].
Chronic kidney disease — Patients with chronic kidney disease (CKD) are at increased risk of adverse events with many medications because of impaired renal excretion. Guidelines for drug prescribing in patients with kidney failure are available [77]. For end-stage patients with uremia-induced nausea, evidence-based guidelines by a United Kingdom Expert Consensus group [78] suggest haloperidol at a 50 percent dose reduction. If symptoms persist, levomepromazine is recommended as an alternative antiemetic (not available in the United States). The consensus group was concerned about the risk of extrapyramidal reactions with metoclopramide and suggested that it should be used with caution and at a maximum dose of 30 mg/24 hours. (See "Kidney palliative care: Principles, benefits, and core components".)
Motility disorders, such as delayed emptying or gastroparesis, may be associated with uremia, thereby leading to anorexia, nausea, and vomiting; this disorder is a frequent reason to initiate dialysis and usually resolves after renal replacement therapy has begun. Gastroparesis that is unresponsive to dialysis is sometimes caused by an autonomic neuropathy, particularly that due to diabetes. These patients usually respond to prokinetic drugs such as metoclopramide and erythromycin. (See "Unique aspects of gastrointestinal disease in dialysis patients".)
HIV/AIDS — Prior to potent antiretroviral therapy (ART), nausea and vomiting in patients with HIV was largely due to opportunistic infections of the gastrointestinal tract or central nervous system. These opportunistic infections still need to be considered as an etiology, especially when T cell counts are <200. (See "AIDS-related cytomegalovirus gastrointestinal disease" and "Cyclospora infection" and "Toxoplasmosis in patients with HIV".)
A post-ART multicenter study found nausea and vomiting is still prevalent and ranged from 42 to 57 percent for nausea and 28 to 32 percent for vomiting. The proportion rating nausea and vomiting as "severe and distressing" ranged from 26 to 40 percent for nausea and 8 to 40 percent for vomiting, and there was a perception among patients that their symptoms were not adequately treated [79]. Nausea can be caused by most nucleoside reverse transcriptase inhibitors (NRTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), and protease inhibitors (PIs). The mechanisms of nausea as a result of these drugs can be varied, including mitochondrial toxicity, lactic acidosis, and pancreatitis. (See "Mitochondrial toxicity of HIV nucleoside reverse transcriptase inhibitors".)
No specific antiemetic can be recommended based upon evidence. In patients with advanced AIDS, there is limited evidence for the use of oral cannabinoids; however, they have a narrow therapeutic window and a risk for side effects [80].
Heart failure — A retrospective study showed that almost one-half of patients with heart failure (HF) complained of nausea during their last six months of life [81]; however, other symptoms such as anorexia, fatigue, and dyspnea [82] are more distressing. In this population, medications (eg, digoxin) are a particularly important cause of nausea and vomiting. For patients who do not have an identifiable cause for nausea, no evidence-based recommendations for a specific antiemetic can be made. However, the potential for prolongation of the corrected QT interval should be considered in drug selection. (See "Palliative care for patients with advanced heart failure: Decision support and management of symptoms", section on 'Symptom management' and "Secondary pharmacologic therapy in heart failure with reduced ejection fraction (HFrEF) in adults".)
Liver failure — Patients with end-stage liver disease have high symptom burden, with pain (65 percent) and nausea (58 percent) being most common [83]. Nausea may be due to various causes, including gastroparesis or increased abdominal pressure, or may be centrally mediated by circulating toxins and bilirubin.
Patients with chronic liver disease are at increased risk of adverse events with many medications because of impaired hepatic metabolism. Many medications require dose adjustments or should be avoided altogether; however, specific recommendations are often lacking. A comprehensive review of managing medications in patients with chronic liver disease became available in 2013, but it did not address dosing of antiemetics or prokinetic agents [84].
An earlier literature review suggested dosage guidelines for palliative care patients with liver disease [85], including a limit of 8 mg ondansetron per 24 hours in severe impairment (Child-Pugh C cirrhosis (table 10)). This recommendation is also included in the US Food and Drug Administration (FDA)-approved manufacturer’s information for ondansetron. A small placebo-controlled trial of eight cirrhotic patients with mild encephalopathy demonstrated that metoclopramide was effective and safe up to a total dose of 60 mg per 24 hours [86]. Dose adjustment information is not provided in the FDA-approved manufacturer’s product information.
Refractory symptoms — Palliative sedation is an option for the rare patient who has intractable nausea and vomiting at the end of life. (See "Palliative sedation".)
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: Palliative care".)
SUMMARY AND RECOMMENDATIONS
●Nausea and vomiting are common symptoms near the end of life that can cause substantial physical and psychological distress for patients and their families and significantly impact quality of life. (See 'Introduction' above.)
Cancer patients
●Among patients with cancer, chemotherapy and radiation therapy (RT) are common causes of nausea and vomiting. Despite adherence to published guidelines, breakthrough nausea and vomiting affects up to 40 percent of patients treated with either modality. (See 'Cancer patients' above.)
Chronic nausea is also a significant problem in patients with terminal cancer who are not receiving active cancer treatment and in other palliative care populations with chronic kidney disease (CKD), AIDS, and heart failure (HF). (See 'Unrelated to cancer therapy' above and 'Other palliative care patients' above.)
●In many cases, the cause can be determined from the history and physical examination. The most common causes of nausea and vomiting in palliative care patients are outlined in the table (table 5); most patients have two or more contributing factors.
Additional laboratory investigations or imaging may be required depending on the clinical presentation and prognosis of the patient. However, even in advanced disease, when the burden of diagnostic testing may outweigh the benefit, investigations may be justified if they could identify potentially modifiable etiologies and influence management. (See 'Pathophysiology and etiology' above.)
●An effort should be made to identify and correct reversible underlying causes. This may include correction of metabolic abnormalities, aggressive bowel care for constipated patients, opioid rotation in patients with suspected opioid-induced nausea and vomiting, and, in patients with cancer who have brain metastases, the administration of glucocorticoids followed by local therapy (radiation or surgery). (See 'Potentially treatable cause' above.)
Most patients with advanced cancer and chronic nausea will have no obvious etiology other than underlying autonomic dysfunction and/or gastroparesis.
●Dopamine, histamine, serotonin (both type 3 [5-HT3], and type 4 [5HT4]), and muscarinic (acetylcholine) receptor antagonists are all established classes of antiemetic and antinausea agents that have been used in the palliative care setting. Despite the paucity of evidence demonstrating benefit, we suggest an etiology-based rather than an empiric approach to selection of antiemetic agents (ie, drugs are chosen based upon the presumed etiology of the nausea and/or vomiting) (table 6) (Grade 2C). (See 'Pharmacologic approaches' above.)
Synthetic oral cannabinoids should be limited to the management of breakthrough chemotherapy-induced nausea and vomiting (CINV). The use of medical marijuana for refractory nausea and emesis in patients with a terminal illness is very controversial; given the lack of evidence supporting benefit and medical as well as legal concerns, we suggest against its use (Grade 2C). (See 'Cannabinoids and cannabis' above.)
●Guidelines for prevention of CINV are based upon emetic risk category (table 7). (See 'Chemotherapy' above and "Prevention of chemotherapy-induced nausea and vomiting in adults".)
For patients with breakthrough CINV, ensure that the patient is receiving the antiemetic appropriate for the drug(s) being given (table 2 and table 3) and the correct dose. If poor emesis control is documented with an appropriate antiemetic regimen, the regimen should be adjusted to one typically used for a higher risk group. Breakthrough symptoms can be managed by adding agents such as lorazepam or alprazolam, olanzapine, or a dopaminergic antagonist (eg, prochlorperazine, thiethylperazine, haloperidol), by substituting high-dose intravenous (IV) metoclopramide (30 to 40 mg/day, titrating as needed to no more than 60 to 100 mg/day) for the 5-HT3 receptor antagonist, or by switching to a different 5-HT3 receptor antagonist, given incomplete cross-resistance between agents. (See "Management of poorly controlled or breakthrough chemotherapy-induced nausea and vomiting in adults".)
●The risk for radiotherapy-induced nausea and vomiting (RINV) is stratified into four categories depending on radiation field (table 1). Specific recommendations for prophylaxis based upon these risk categories are available (table 8). (See "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment".)
Few studies have examined the benefit of specific rescue strategies for treatment of breakthrough nausea in patients receiving prophylactic therapy for RINV. A 5-HT3 receptor antagonist (eg, ondansetron) is an option. Alternatively, olanzapine is another reasonable option given its activity in delayed CINV. (See "Radiotherapy-induced nausea and vomiting: Prophylaxis and treatment", section on 'Rescue therapy'.)
●For patients whose history suggests gastroparesis (early satiety, vomiting of food eaten several hours earlier), dietary modification and pharmacologic management with prokinetics such as metoclopramide are appropriate. (See "Treatment of gastroparesis".)
●Malignant bowel obstruction (MBO) should be suspected in a patient with progressively worsening colicky pain with diminishing stool and flatus. Most patients are inoperable, and survival is generally short. Clinical management requires a specific and individualized approach that is based on the likely cause, disease prognosis, and the objectives of care. In most cases, there are few or no data addressing the relative value of palliative surgery versus medical management, and a decision to proceed to surgical intervention requires careful weighing of risks and benefits, including an assessment of the estimated life expectancy and patient goals and preferences. Colonic stent placement is an acceptable palliative option for inoperable patients. Although clinical management should be individualized, a suggested algorithmic approach is provided (algorithm 2). (See 'Gastrointestinal obstruction' above and "Palliative care of bowel obstruction in cancer patients".)
●If a potentially reversible cause cannot be identified and bowel obstruction is ruled out, we suggest symptomatic treatment with a prokinetic agent such as metoclopramide (Grade 2B). (See 'No obvious cause' above.)
Prolonged use of metoclopramide for more than three months increases the risk of tardive dyskinesia (TD); risks and benefits of continued use after three months should be carefully evaluated and discussed with patients.
For terminally ill patients who have contraindications to metoclopramide, a number of other centrally acting antiemetic agents (eg, haloperidol, chlorpromazine, olanzapine) can be used. 5-HT3 receptor antagonists, including ondansetron, can be useful in the postoperative setting and may be considered in other clinical scenarios of refractory nausea.
For patients who are unable to take oral medications, metoclopramide, dexamethasone, and haloperidol have been safely administered intravenously and subcutaneously. Metoclopramide is also available in a nasal formulation. The rectal route can occasionally be used but may be uncomfortable for patients because of the need for frequent administration given the short duration of action of antiemetics.
●Acupuncture or acupressure may be of benefit for CINV, but the available evidence is insufficient to support use in other palliative care populations. However, given the overall safety of this approach, interested patients may be referred for a trial of acupuncture if symptoms of moderate to severe nausea persist despite other forms of therapy. (See 'Complementary and alternative therapies for patients with advanced cancer' above.)
Non-cancer conditions
●For life-limiting illnesses other than cancer, the assessment should be based on the approach to the adult with nausea and vomiting. (See "Approach to the adult with nausea and vomiting".)
●Patients with chronic obstructive pulmonary disease (COPD) frequently have a high symptom burden that includes gastrointestinal symptoms, but medications (especially methylxanthines) are often responsible for nausea and vomiting. (See 'Chronic obstructive pulmonary disease (COPD)' above.)
●Among patients with end-stage kidney failure, uremia is often associated with gastroduodenal motility disorders, and these patients often respond to prokinetic agents. For patients with uremia-induced nausea, haloperidol or ondansetron are potentially useful agents. (See 'Chronic kidney disease' above.)
●Among patients with AIDS, nausea may be caused by medications (nucleoside reverse transcriptase inhibitors [NRTIs], nonnucleoside reverse transcriptase inhibitors [NNRTIs], and protease inhibitors [PIs]). No specific antiemetic can be recommended, although there is limited evidence for the use of oral cannabinoid agents with a narrow therapeutic window. (See 'HIV/AIDS' above.)
●Among patients with end-stage HF, medications (particularly digoxin) are an important cause of nausea and vomiting. Evidence-based recommendations for a specific antiemetic cannot be made. However, the potential for prolongation of the corrected QT interval should be considered in drug selection. (See 'Heart failure' above.)
●For patients with end-stage liver disease, nausea may be due to gastroparesis, increased abdominal pressure, or centrally mediated by circulating toxins and bilirubin. For severe hepatic impairment (Child-Pugh C cirrhosis), doses of ondansetron should be limited to no more than 8 mg/24 hours, and metoclopramide should be limited to no more than 60 mg/24 hours. (See 'Liver failure' above.)
