ABSTRACT: The optimal management of opioid-related side effects is hampered by a lack of comparative studies of management strategies. The prevalence of such side effects is influenced by the extent of disease, the patient’s age, the presence of coexistent renal and hepatic disease, pulmonary disease, and cognitive dysfunction, a prior opioid history, use of polypharmacy, dose of opioid drug being administered, and the route of administration. The most common opioid-related side effects are constipation, sedation, nausea, vomiting, and cognitive disturbance. Less frequent side effects include urinary retention, perceptual distortion, respiratory depression, and myoclonus. In an era emphasizing quality of life in cancer care, clinicians need to be aware of (1) factors that influence the prevalence of opioid-related side effects, (2) effective management strategies, and (3) how to recognize when symptoms are opioid related as opposed to caused by other etiologies, such as the patient’s disease process or treatment approaches. The use of validated instruments and repeated assessment enhances such an evaluation and subsequent treatment. This article delineates the current optimal management of opioid-related nausea and vomiting, constipation, cognitive side effects, myoclonus, and respiratory depression. [ONCOLOGY 15(1):61-82, 2001]
Opioid use is associated with a wide range of side effects, the most common of which are constipation, sedation, nausea, vomiting, and cognitive disturbance. Less frequent side effects include urinary retention, perceptual distortion, respiratory depression, and myoclonus. Adding to the complexity of the assessment and management of these side effects is the multifactorial etiology of many of these symptoms in the cancer patient.
In day-to-day clinical practice, the management of opioid-related side effects is heuristic, based on clinical experience rather than research. Few studies have compared the incidence of the side effects associated with one narcotic vs another or have explored the role of the route of administration in provoking side effects.[1-9] Several studies of opioids are summarized in Table 1. Much of the available comparative data, however, comes from surveys that lack information on the doses used.
A prospective study of 100 cancer patients evaluated by a pain and palliative care service found that alterations in the opioid regimen, including changes in the drug and/or route of administration, were made in 80% of patients. Of these alterations, 17.7% were instituted to simultaneously improve pain control and decrease opioid toxicity, while 25% were aimed at diminishing side effects in the setting of controlled pain.
About 31% of the alterations in pain management approaches were made for the convenience of the patient in the setting of controlled pain—for example, the conversion of a patient’s opioid medication from a parenteral patient-controlled analgesia delivery system to an oral or transdermal route of administration in order to simplify a hospital discharge plan. An additional 19% of changes were made to reduce the invasiveness of therapy in the setting of controlled pain, such as conversion from parenteral to oral opioid medication regimens.
The personal impact of an individual side effect often depends on the stage of a patient’s illness and the goals of care. For example, opioid-related sedation may cause minimal distress to a patient in the last days of life, and in fact, may be desirable. However, its occurrence earlier in the course of a patient’s illness may be a source of distress to the patient and family, leading to noncompliance with a given medication. Having a clear understanding of the goals of care for the patient will dictate the decision to aggressively work-up an individual symptom or side effect or to change the management approach, switching, for example, to a different opioid or route of administration.
Unfortunately, in a patient who is receiving opioid therapy, nausea, vomiting, constipation, sedation, cognitive and perceptual disturbances, respiratory depression, and myoclonus are too often dismissed as being opioid related, without further assessment for other possible etiologies. In addition to missing potentially serious and treatable diagnoses, this approach may result in a decreased dose or premature discontinuation of the opioid and, therefore, increased pain and suffering for the patient. A patient may also be converted unnecessarily from a convenient, economical route of drug delivery to a more cumbersome and costly route.
Our understanding of the mechanisms underlying opioid-related side effects is hampered by a deficit of studies in this area. These side effects are known to occur in patients with advanced disease, the elderly, those with coexisting renal, hepatic, pulmonary, and cognitive dysfunction, and patients who have been receiving opioids for an extended period of time (Table 2).[11-23] In addition, clinical observations have associated the route of opioid administration with the occurrence of side effects. For example, nausea, vomiting, and constipation may occur more frequently when opioid medications are administered orally.
Certain opioid side effects may be mediated by specific opioid receptors. For example, activation of the mu-1 receptor may be associated with pruritus, whereas activation of the mu-2 receptor may mediate cardiovascular effects, delayed gastrointestinal tract transit time, and respiratory depression. Most mu receptors are found in the areas of the brain responsible for opioid-induced analgesia. None of the opioid agonists—morphine, meperidine, hydromorphone(Drug information on hydromorphone), oxymorphone (Numorphan), levorphanol (Levo-Dromoran), fentanyl(Drug information on fentanyl), sufentanil(Drug information on sufentanil), alfentanil(Drug information on alfentanil) (Alfenta), and methadone—selectively activate the mu-1 receptor without activating the mu-2 receptor. The opioid agonist-antagonists—butorphanol (Stadol), nalbuphine, and pentazocine(Drug information on pentazocine) (Talwin)—are weak mu-receptor antagonists and weak delta-receptor agonists. Both butorphanol and pentazocine act as agonists at the kappa receptor but nalbuphine(Drug information on nalbuphine) is a partial kappa-receptor antagonist. The analgesia obtained from the agonist-antagonist medications mainly derives from kappa-receptor activation.
Because the opioid agonist-antagonist medications (ie, butorphanol, nalbuphine, pentazocine) can cause withdrawal symptoms in patients who exhibit tolerance to the effects of pure opioid agonist medications (ie, morphine, meperidine, hydromorphone, oxymorphone, levorphanol, fentanyl, sufentanil, alfentanil, methadone(Drug information on methadone)), they are not commonly used in patients with cancer pain who are likely to need chronic opioid therapy. Importantly, when there is a prevalence of side effects, kappa-receptor activation appears to be associated with sedation and psychotomimetic effects, including dysphoria and hallucinations, miosis, and mild respiratory depression.
The assessment and management of opioid-related nausea and vomiting, constipation, cognitive changes, myoclonus, and respiratory depression are reviewed below.
Nausea and vomiting are not uncommon at the start of opioid therapy. Tolerance to this effect typically occurs within days to weeks. Opioid-associated nausea usually does not require treatment or requires only temporary treatment, although a minority of patients will require opioid rotation or a change in route of administration for management. Anecdotal experience suggests that this side effect occurs more frequently with lower doses of morphine(Drug information on morphine) than with higher doses.
Survey data show small differences in the overall prevalence of nausea from one opioid to another. There is currently no evidence to suggest that mu-receptor agents such as codeine(Drug information on codeine), dextropropoxyphene(Drug information on dextropropoxyphene), morphine, hydromorphone, meperidine, levorphanol, fentanyl, and methadone are less emetogenic than other opioids. In a study of 260 patients treated with opioids for cancer pain, 8.3% of those who received buprenorphine(Drug information on buprenorphine) (Buprenex) developed nausea at the onset of therapy, compared with 18.3% of patients treated with morphine, 16.2% treated with codeine, and 10% treated with oxycodone(Drug information on oxycodone). At 72 hours, these percentages rose to 22.7%, 28%, 29.7%, and 18%, respectively. Refractory nausea and vomiting led to the use of an alternative opioid in 22% of patients after 72 hours. This study did not control for starting dose.
Opioid-induced nausea may be triggered by afferent input to the medullary emesis center from gastrointestinal receptors, vestibular centers, the cerebral cortex, and the chemoreceptor trigger zone. Nausea that occurs primarily with movement can be assumed to be generated by input from the vestibular center. Nausea and vomiting that is postprandial and associated with a feeling of bloating and satiety suggests a gastrointestinal etiology.
Studies of opioid-treated animals have demonstrated reversal of the emetic response following administration of the mu-opioid antagonist naltrexone(Drug information on naltrexone) (ReVia, Depade).
Clinicians should investigate nausea and vomiting as separate issues, since either symptom may occur in isolation. Assessing the severity of nausea/vomiting is often difficult. Observer-rated measures of nausea tend to underestimate its frequency. Indirect measures, such as the patient’s appetite and food intake, tend to be confounded by disease stage. The patient’s visual analog scale reports of nausea, especially in those with advanced disease, may be compromised by cognitive impairment occurring in conjunction with the symptom or its treatment.
Furthermore, some of the therapies available for the treatment and/or prevention of nausea and vomiting (eg, phenothiazine, butyrophenones, benzodiazepines) may exacerbate cognitive impairment, further limiting assessment. The most practical quantification of emesis is observer assessment of the number of episodes and duration of emesis. The association of nausea or vomiting with an opioid may be validated by eliciting a temporal relationship between the two.
The extent of the work-up in a patient complaining of nausea and/or vomiting who is receiving opioid therapy for pain is dictated by the clinical setting and the goals of care (Figure 1). Given the potential for intestinal obstruction in many of these patients, the work-up, at a minimum, includes examination for abdominal distension, organomegaly, and bowel sounds. A neurologic examination should also be performed, given the association between refractory nausea and intracranial pathology.
The appearance of nausea in conjunction with a recent dose escalation or the initiation of opioid therapy, particularly when the nausea responds to conservative measures, should not prompt a work-up for alternative etiologies. However, nausea and vomiting occurring during a time of stable opioid dosing should be investigated further, especially if these symptoms are refractory to therapy or associated with abnormalities on physical examination.
Assessment of electrolytes and calcium levels may be indicated. If the physical examination is suggestive of focal pathology, computed tomography (CT) scans of the head or abdomen should be performed, as dictated by the goals of care. Any nonessential medications with emetogenic potential (eg, serotonin-reuptake inhibitors, antibiotics, and nonsteroidal anti-inflammatory drugs [NSAIDs]) should be discontinued or their doses adjusted.
An immediate clinical decision concerns whether to treat the side effect or switch opioids and/or the route of administration with the goal of eliminating or decreasing the symptom. Because nausea and vomiting is not uncommon at the start of opioid therapy and tolerance to this effect typically develops within days to weeks, only a minority of patients will require opioid rotation or a change in the route of drug administration. The following discussion focuses on the treatment of the symptom rather than on changing the opioid or route of administration.
Treating the Symptom: In patients with dyspepsia who are being treated with NSAIDs, consideration should be given to discontinuing the NSAID. If the NSAID is an important part of the patient’s therapy, adding a histamine-2-receptor antagonist such as ranitidine(Drug information on ranitidine), nizatidine(Drug information on nizatidine) (Axid) or prostaglandin analogs such as misoprostol(Drug information on misoprostol) (Cytotec) may be appropriate. In patients who are constipated, optimizing their bowel regimen may improve or eliminate the symptom. Patients with bloating and postprandial nausea may benefit from small, more frequent meals, and patients with a compromised biliary system may find that their symptoms improve with low-fat meals.
Most patients will respond to antiemetics that are active at the chemoreceptor trigger zone. These include the phenothiazine antiemetics, such as prochlorperazine(Drug information on prochlorperazine) and chlorpromazine(Drug information on chlorpromazine), and the butyrophenones, haloperidol(Drug information on haloperidol), and droperidol(Drug information on droperidol). These agents should be administered at the standard dose initially, with the dose doubled as indicated by clinical response. There is wide individual variability in response to these agents and, therefore, assessment and reassessment of treatment efficacy is essential. Haloperidol appears to be the most active of the dopamine(Drug information on dopamine)-receptor antagonists at the chemoreceptor trigger zone.
Metoclopramide and domperidone(Drug information on domperidone) (Motilium, investigational) have less dopamine-receptor antagonist function. Motility agents such as metoclopramide(Drug information on metoclopramide) and domperidone are indicated in patients with postprandial fullness and a high likelihood of autonomic neuropathy (eg, patients with intra-abdominal disease, multiple surgeries, or recent chemotherapy). Because it penetrates the blood barrier to a lesser extent, domperidone is associated with fewer and less severe extrapyramidal side effects and central nervous (CNS) system toxicity than metoclopramide.
Metoclopramide is thought to act by increasing gastrointestinal cholinergic activity through activation of a 5-HT3 receptor. Antiemetics should be administered on an around-the-clock dosing schedule if a trial of intermittent as-needed dosing is ineffective. Metoclopramide, prochlorperazine, and chlorpromazine are available in suppository formulations. Haloperidol can be given subcutaneously, whereas prochlorperazine and chlorpromazine cause subcutaneous irritation and should not be administered subcutaneously.
If nausea and vomiting fail to respond to the above measures and the pattern of nausea and vomiting suggests that it is movement related, a trial of an anticholinergic medication (eg, scopolamine or atropine(Drug information on atropine)) or an antihistamine (eg, meclizine or cyclizine(Drug information on cyclizine) [Marezine]) should be considered. Scopolamine and cyclizine can be administered subcutaneously.
Refractory nausea may be treated with a combination of antiemetics from different drug classes (Table 3). However, because of their cost, the newer, more selective 5-HT3-receptor antagonist medications (eg, ondansetron(Drug information on ondansetron) [Zofran], dolasetron(Drug information on dolasetron) [Anzemet], and granisetron(Drug information on granisetron) [Kytril]) are considered second- and third-line drugs in the management of opioid-related nausea and vomiting. A trial of these drugs may be useful, for example, if the symptom persists or if the patient develops dystonic reactions or uncontrolled and distressing anticholinergic side effects from metoclopramide or prochlorperazine therapy.