Morphine Versus Midazolam as Upfront Therapy to Control Dyspnea Perception in Cancer Patients While Its Underlying Cause Is Sought or Treated
Article Outline
Abstract
Context
Cancer patients with dyspnea may be able to have the symptom pharmacologically controlled while its underlying cause is sought or treated.
Objectives
This study was done to determine whether symptom control can be achieved while the cause is evaluated or treated and whether morphine or midazolam would be more suitable in this setting.
Methods
Sixty-three ambulatory patients with advanced cancer and dyspnea were clinically characterized and then randomized to receive either oral morphine or oral midazolam. A fast in-clinic drug titration scheme was implemented followed by an ambulatory five-day period in which the patients received the effective dose that relieved their dyspnea. During this period, the patients were followed daily while the underlying causes of dyspnea were sought out or treated.
Results
Thirty-one patients with dyspnea entered the morphine arm and 32 patients entered the midazolam one. During the initial in-clinic phase, dyspnea was alleviated by at least 50% in all patients, whether they received morphine or midazolam. During the ambulatory phase, midazolam was superior to morphine in controlling baseline and breakthrough dyspnea. Both treatments were well tolerated, with mild somnolence being the most common adverse event. Neither morphine nor midazolam affected the outcome and/or implementation of additional diagnostic and/or therapeutic interventions.
Conclusion
Our results suggest that cancer-related dyspnea in ambulatory patients can be pharmacologically treated while its most probable specific cause is sought and/or while an etiology-oriented intervention is implemented. In this setting, midazolam appeared to be a better option than morphine for the immediate and long-term relief of the symptom.
Key Words: Dyspnea, cancer, morphine, midazolam, opioids, benzodiazepines
Introduction
Dyspnea and pain are among the most common and distressing symptoms experienced by cancer patients in palliative care.1, 2 Both sensory experiences originate from interactions of multiple physiologic, psychological, social, and environmental factors. Although the role of the central processing of pain sensory information has been well characterized, much less is known about the cortical structures involved in the perception of dyspnea.3 The anterior insula and other brain structures are commonly activated during the perception of dyspnea and pain.3, 4, 5, 6, 7 Many of these brain regions that process the perception of pain and dyspnea can be pharmacologically modulated by opioids8, 9 and benzodiazepines.10, 11 However, dyspnea and pain share more than neuronal structures. Both are subjectively perceived physiologic sensations of an unpleasant nature, both motivate adaptive behavior, and both perceptions are alleviated with opioids.12, 13, 14, 15, 16, 17
The mainstay of dyspnea control in cancer patients remains decreasing central perception, and morphine is still the first choice of pharmacological therapy.12, 13, 14, 15, 18 In contrast to pain management, the dose, schedule, and route of administration of morphine to control dyspnea are not well established.13 We have shown that the beneficial effect of morphine in controlling baseline levels of cancer dyspnea may be improved with the addition of midazolam to the treatment.19 The mechanism of action of morphine and midazolam in controlling dyspnea is not well understood, although it may be related to altering central perception.12, 19
It is now generally accepted that, with advanced cancer patients, pain perception should be pharmacologically controlled before its underlying cause is sought or treated.17 In an effort to obtain new alternatives to control cancer-related dyspnea, we designed a clinical trial that articulates a new conceptual management for this symptom. With the idea of translating the highly effective cancer-related pain management model to the realm of dyspnea, we conducted this research to determine whether cancer dyspnea could be pharmacologically controlled while its underlying cause is sought or treated and which drug, morphine or midazolam, would be more suitable in this setting.
Methods
Study Design
The study protocol was approved by the research and ethics committees of the Instituto Angel Roffo of the Universidad de Buenos Aires and was in accordance with the recommendations found in the Helsinki Declaration. The study was designed with an initial assessment and two phases of intervention (Fig. 1). The fast titration phase (FTP) took place in the ambulatory clinic. Eligible patients were randomized to receive either morphine or midazolam for up to three dosing steps. The FTP was designed to complete in less than two hours. The main objective was to find the dose of either drug for a particular patient that alleviates 50% or more of the dyspnea perception (“effective dose”) and could be used for the follow-up phase (FUP). The FUP was carried out for five days when the patients were followed by daily clinic visits. The patients were treated with the effective dose for the allocated drug, with adjustments made based on dyspnea control (baseline and breakthrough components) and adverse events (AEs). Additional tests and procedures to control dyspnea were allowed during this phase in accordance with standard practices. All the tests and procedures, as well as the possible interference of the medication with them, were registered and taken into account for the analysis.
Fig. 1
Flowchart of the study design. The protocol was spatially and temporally divided in three parts, including the initial assessment that took place in the ambulatory clinic, where the dyspnea was graded and characterized and specific treatment and tests were ordered; the FTP that took place in a procedure room in the ambulatory clinic, where patients were randomized and treated accordingly; and the FUP that took place as programmed visits to the ambulatory clinic (where the maintenance medication was adjusted and dyspnea graded) and in other places, where additional tests and/or procedures were performed.
Patients were randomly assigned (using a random number generator in 1:1 ratio in blocks of six) to one of the two treatment groups. Numbered envelopes that were used to implement the randomization were concealed until interventions were assigned. The researchers had final responsibility for patient enrollment.
Inclusion Criteria
Ambulatory patients who could provide informed consent and were 18 years or older, with a documented diagnosis of advanced cancer, Mini-Mental Status Examination score >23 out of 30, moderate or severe dyspnea at rest, and a performance status of ≤3 were eligible.
Exclusion Criteria
Exclusion criteria included active or uncontrolled chronic obstructive pulmonary disease (COPD), noncompensated congestive heart failure, severe renal or hepatic failure (clinically and/or biochemically detected), and uncontrolled symptoms (numerical rating scale [NRS] >3 out of 10), and hemoglobin saturation by pulse oximetry (SaO2) <
85%. Unstable patients who required immediate treatment or admission also were excluded.
Treatment Regimen
Drug administrations were performed in a single-blind fashion. Patients and caregivers were blinded. During the FTP (Fig. 1), eligible patients were randomized to receive oral morphine or oral midazolam. The starting doses were 2mg for midazolam and 3mg for morphine, with incremental steps of 25% of the preceding dosing every 30 minutes. Unless the dyspnea intensity was reduced by at least 50%, they received the next step dose. The dose that reduced the intensity of dyspnea by at least 50% was considered the “effective dose.” If the dyspnea intensity was not reduced by at least 50% after two incremental steps or AEs appeared, the patient was considered to be a therapeutic failure. For patients already receiving opioids for reasons other than dyspnea, the total daily opioid dose was calculated and converted to oral morphine equivalents. If the daily equivalent dose of morphine (DEDM) was lower than 5mg, then patients were considered opioid-naive patients for the sole purpose of determining the starting dose (i.e., they received 3mg as starting dose). If the DEDM was equal to, or higher than, 5mg, patients received an increase in dose equal to 25% of their respective DEDM for the initial step. No adjustment was made for patients already receiving benzodiazepines other than midazolam and for reasons other than dyspnea. The dosing interval was chosen so that the clinical effect of morphine and midazolam will have reached approximately 50% of its maximum by about the time the next dose is given. Maximum plasma concentrations for both drugs after an oral dose are reached on average within 60 minutes.20, 21
During the FUP (Fig. 1), the patients received their respective “effective dose” of either morphine or midazolam every four hours (excepting the sleep hours) on an “around-the-clock” schedule. The patients were followed daily for five days while the underlying causes of their dyspnea were sought or treated. Only treatment interventions intended to eradicate or control the causes or main cause for their dyspnea were allowed. In case of breakthrough dyspnea (BD), the patients received rescues with their respective medication. For all cases, rescue doses were administered with an interval equal to or greater than 30 minutes apart. In each follow-up daily visit, the daily dose of morphine or midazolam was adjusted accordingly to the efficacy in controlling the dyspnea, the need for rescues, and the appearance of AEs related to the medication. A dose was skipped for patients who developed somnolence Grade 2 or higher at the point of receiving the corresponding dose of medication. Patients who received morphine were systematically premedicated with laxatives.
All drugs were given orally in the form of a solution. The generic study drugs were purchased from the market and prepared by the institutional pharmacy as 0.2% water solutions.
Assessment
The initial assessment included complete medical history, physical examination, symptom characterization and grading, oxymetry, and chest radiography. Dyspnea intensity was assessed by NRS (0–10) for breathlessness, anchored at one end (0) with “no breathlessness” and at the other end (10) with “worst possible breathlessness.” During the initial assessment, the researchers were asked to identify active possible causes of dyspnea for each patient, define dyspnea syndromes, and recommend a specific treatment approach or additional tests. This was carried out using a semistructured questionnaire. Researchers also were asked to record the descriptor that the patient used for the first time to characterize their dyspnea.
During the FTP, dyspnea relief was assessed using a dyspnea relief five-category scale. These scales are more sensitive than intensity scales in detecting immediate drug treatment effects.22 The relief experienced by the patient was tabulated in a percentage scale as follows: none (0%), slight (25%), moderate (50%), a lot (75%), and complete (100%). Dyspnea relief was assessed 30 minutes after each dose step.
During the FUP, dyspnea was assessed daily using an NRS for intensity, registering the number of BD episodes per day and recording the need for rescues. In contrast to the FTP, the use of an NRS in the FUP allowed the patients to directly estimate the dyspnea without the need to compare with their memory of initial dyspnea, as relief scales require.
The National Cancer Institute Common Toxicity Criteria (NCI CTC) v3.0 was used to score the treatment AEs. For scoring somnolence, we considered the time that the patients spent sleeping during the daytime hours, and the following criteria were used: Grade 1 (less than three hours), Grade 2 (3–5 hours), Grade 3 (6–11 hours), and Grade 4 (12 or more hours).
Statistical Analysis
Results are presented as mean with the 95% confidence interval (CI) or median with the median absolute deviation (MAD). Unless otherwise noted, Wilcoxon signed rank test was used for intragroup comparisons and Wilcoxon rank sum test was used for intergroup comparisons. The uncorrected two-sample proportion test was used for proportions comparison between baseline and subsequent values and between groups. The P-values cited were two sided, and P-values less than 0.05 were judged as statistically significant. Sample size was calculated for the FUP of the study using previous reports of improvement in dyspnea with morphine and midazolam in our population with a predicted standard deviation of 10%. An estimated total of 34 patients would provide 80% power to detect a one-point true difference between treatments in the dyspnea scale at a two-sided 5% significance level at any time during the FUP of the protocol. All calculations were done using Statistix 7.0 (Analytical Software, Tallahassee, FL).
Results
Initial Assessment
Sixty-three ambulatory advanced (incurable) cancer patients with moderate to severe dyspnea were considered eligible. Overall, 16 of 63 (25.4%) patients had lung cancer, 15 of 63 (23.8%) had breast cancer, six of 63 (9.5%) had head and neck cancer, and 26 of 63 (41.3%) had other cancers. Fig. 2a shows a number of proximal factors that were highly suspected of being active contributors to the presence of dyspnea in a particular patient. Risk factors for dyspnea such as history of smoking, COPD, and exposure to asbestos or coal dust were not taken into account. At least one factor was suspected for each patient (Fig. 2), and in 15 of 63 cases (23.8%), the researchers asked for additional tests and/or procedures (Fig. 3). Although two patients had a SaO2<90%, hypoxemia was not considered as an active dyspnea contributor in those cases. Interstitial lung disease was the cause that most frequently associated with other causes (Fig. 2a). Although patients presenting with lung metastases constituted 30% of the population, only 4.7% presented with lung metastases as the unique contributor for dyspnea. Although pulmonary microembolism was a suspected contributor in 36.5% of the patients, none of them were considered as candidates for anticoagulation, mainly because they were previously refractory to anticoagulants, because they present additional causes for dyspnea, and/or had risks from anticoagulation therapy that surpassed its benefits. The descriptors used by the patients to indicate their dyspnea are shown in Fig. 2b.
Fig. 2
Suspected active contributing factors and patient's dyspnea descriptors. a) Each column represents one patient (n
=63), whereas each row indicates one cause (on the left) with its corresponding frequency (on the right). Patients are ordered from left to right in an increasing number of contributing factors presented. The patients presenting one, two, three, or four contributing factors are shown in the bottom as percentages. The specific causes of dyspnea and dyspnea syndromes in the questionnaire included interstitial lung disease (including lymphangitis carcinomatosis), thoracic vessel involvement (pulmonary veno-occlusion and pulmonary embolism), pulmonary metastasis, pleural effusion, pericardial effusion without heart failure, major airway external compression (adenomegaly and tumor mass), chest wall infiltration, lung parenchymal and/or pleural tumor, respiratory muscle weakness, abnormal diaphragm mechanics (cancer cachexia, steroid myopathy, paraneoplastic syndrome, phrenic nerve paralysis, ascites, and hepatomegaly), pneumonitis (radiation and chemotherapy related), pneumonia (microaspiration and permanent tracheoesophageal fistula), anemia, metabolic acidosis, and others. b) Patients' dyspnea descriptors (n=63). The number of patients (and percentages) is shown for each descriptor.
Fig. 3
Patients who received additional tests and/or procedures. Each pie represents the total number of patients evaluated over the five days of the protocol, whereas the dotted portion represents the number of patients who received additional tests or procedures to control their dyspnea in each group. The number of tests and procedures is listed for each group. More than one test or procedure per patient were prescribed.
Fast Titration Phase
The 63 eligible patients were randomized to receive morphine (31 patients) or midazolam (32 patients). The groups exhibit a similar distribution of demographic, disease-specific, and dyspnea-related characteristics (Table 1). During this phase, the patients received oral morphine or oral midazolam. The starting doses were 2mg for midazolam and 3mg for morphine, with incremental steps of 25% of the preceding dose every 30 minutes. Unless the dyspnea intensity was reduced by at least 50%, they received the next step dose. The number of patients in each dosing step that represents the patients who were alleviated 50% or more in their dyspnea is shown in Fig. 4. The values for SaO2 (%) for the morphine and midazolam groups were, respectively, baseline: 94.4% (±2.2) vs. 94.8% (±1.2), 30 minutes: 93.9% (±1.9) vs. 94.3% (±1.4), 60 minutes: 93.7% (±1.7) vs. 94.6% (±0.8), and 90 minutes: 94.1% (±1.3) vs. 94.7% (±1).
Table 1. Characteristics of the Randomized Patients (n=63)
| Characteristic | Midazolam | Morphine |
|---|---|---|
| Patients randomized | 32 | 31 |
| Age, median (range) | 59 (36–82) | 55 (30–80) |
| Diagnoses | ||
| Lung (%) | 8 (25) | 8 (25.8) |
| Breast (%) | 7 (22) | 8 (25.8) |
| Head & neck (%) | 3 (9.3) | 3 (9.7) |
| Other (%) | 14 (43.87) | 12 (38.7) |
| Performance status (median, MAD) | 2 (0) | 2 (0) |
| Hemoglobin saturation, mean (95% CI) | 94.8% (±1.2) | 94.4% (±2.2) |
| Dyspnea NRS, mean (95% CI) | 8.8 (±0.3) | 8.7 (±0.3) |
| Dyspnea causes, mean (95% CI) | 2 (±0.3) | 2 (±0.3) |
| Dyspnea causes, number of patients | ||
| One (%) | 7 (22) | 6 (19.4) |
| Two (%) | 14 (43.7) | 15 (48.4) |
| Three (%) | 8 (25) | 10 (32.2) |
| Four or more (%) | 3 (9.3) | 0 |
| Previous morphine, n (%) | 12 (37.5) | 11 (35.5) |
| Previous benzodiazepines, n (%) | 4 (12.5) | 3 (9.7) |
Fig. 4
Flowchart of the FTP. Patients were randomized in two groups (morphine and midazolam). Individualized doses were determined for each patient in the trial. The number of patients in each dosing step represents the patients whose dyspnea was alleviated 50% or more. One patient from each group withdrew their consent to continue with the protocol because they were unable or unwilling to comply with the programmed follow-up visits.
There were no serious AEs that required drug discontinuation. The most common AE was somnolence in 18 of 32 (56.2%) patients receiving midazolam and 15 of 31 (48.4%) patients receiving morphine, in all cases with mild intensity (P=0.53) (Table 2). All the patients randomized were alleviated of their dyspnea by at least 50% with either morphine or midazolam; accordingly, all these patients were eligible for the FUP. One patient from each group withdrew their consent to continue with the protocol because they were unable or unwilling to comply with the programmed follow-up visits.
Table 2. Adverse Events
| Adverse event | Midazolam | Morphine |
|---|---|---|
| FTP (none, mild, moderate, severe, and life threatening) | ||
| Somnolence, n (%) | ||
| Mild | 18/32 (56) | 15/31 (48) |
| Agitation, n (%) | ||
| Mild | 2/32 (6.25) | 3/31 (10) |
| Nausea, n (%) | ||
| Mild | 0 | 2/31 (6.5) |
| Moderate | 0 | 1/31 (3.2) |
| FUP (NCI CTC v3.0) | ||
| Somnolence, n (%) | ||
| G2 | 4/31 (12.9)a | 5/30 (16.7)a |
| G3 | 0 | 1/30 (3.3)a |
| Agitation, n (%) | ||
| G1 | 0 | 2/30 (6.5) |
| G2 | 0 | 1/30 (3.3) |
| Nausea, n (%) | ||
| G1 | 0 | 2/30 (6.5) |
| Constipation, n (%) | ||
| G2 | 0 | 2/30 (6.5) |
| Others (1 patient per adverse event) | Cognitive disturbance G1 | Cough G1 |
| Pruritus G2 | ||
| Xerostomia G1 | ||
| Flushing G1 | ||
aOne patient from midazolam group and two patients from morphine group required dose reduction. |
Follow-Up Phase
During this phase, the patients received their respective “effective dose” (i.e., the dose that reduced the intensity of dyspnea by at least 50%) of either morphine or midazolam every four hours (excepting the sleep hours) on an around-the-clock schedule. The patients were followed daily for four more days while undergoing additional tests and/or procedures to control the active contributors to their dyspnea. The median intensity for the chronic component of dyspnea (NRS) at baseline was 9 (MAD=0) and 9 (MAD=1) for the morphine and midazolam groups, respectively (P=0.62). During the second-day evaluation, dyspnea intensity decreased significantly for the morphine and midazolam groups to 6 (MAD=1) and 4.5 (MAD=1.5) (P<0.001, to baseline), respectively. During the following days, the intensity of dyspnea either continued to drop or maintained the control level achieved on the second day (Fig. 5a). For the intergroup comparison, patients receiving midazolam maintained a significantly lower dyspnea intensity level in comparison with the morphine group, during the four days of follow-up (Fig. 5a). These statistically significant differences favoring the midazolam group persisted even after omitting the patients already receiving morphine at randomization (11 patients for the morphine group and 12 patients for the midazolam group) for Days 3–5 (P=0.0002, P=0.0002, and P=0.0001, respectively) with a tendency for Day 2 (P=0.066). The differences were practically unchanged after excluding patients already receiving benzodiazepines. At baseline, 25 patients per group complained of having one or more episodes of BD during the previous day. The proportions of patients complaining of BD dropped for both groups during the following days and were significantly different between groups at Days 3, 4, and 5 (Fig. 5b), favoring midazolam. The mean number of episodes of BD (±95% CI) is shown in Fig. 5c. One patient from the morphine group required admission by Day 4 of the protocol because of uncontrolled BD at rest and was considered as a therapeutic failure. The overall therapeutic failure rate (i.e., the number of patients with NRS≥8 by Day 5) was 20% (6 of 30) and 0% for the morphine and midazolam groups, respectively. The daily mean values for SaO2 (%) for the morphine and midazolam groups were, respectively, Day 2: 93.9% (±1.7) vs. 94.6% (±0.8), Day 3: 91.2% (±6.1) vs. 94.8% (±0.8), Day 4: 94.1% (±1.5) vs. 94.7% (±0.9), and Day 5: 94.6% (±1) vs. 94.6% (±1.1). There were no serious AEs that required drug discontinuation. However, one patient from the midazolam group and two patients from the morphine group required dose reduction because of excessive somnolence. The midazolam dose in that patient was reduced to 3.2mg every eight hours, whereas the morphine dose was reduced to 3mg every eight hours for both patients. The most common AE was somnolence, which presented in four of 31 (12.9%) patients receiving midazolam and in six of 30 (20%) patients receiving morphine (Table 2). Finally, the percentage of patients who received one or more additional tests and/or interventions was 26.7% (8 of 30) and 22.6% (7 of 31) for the morphine and midazolam groups, respectively (Fig. 3).
Fig. 5
Dyspnea control (intensity and breakthrough episodes) during the FUP. a) Each box plot represents the central tendency and variability of the data over the days (x axes) for morphine (gray) and midazolam (crosshatched). The box encloses the middle half of the data and is bisected by a line representing the median. The vertical whiskers indicate the range of typical values. Possible outliers are displayed as asterisks. The P-values immediately over each box represent the comparison with the previous day value (pre-post intragroup comparison). The P-values representing the intergroup comparison (morphine vs. midazolam) are shown on the top. Statistically significant values are shown in bold. b) The high of each column represents the total number of patients evaluated over the days (x axes), whereas the colored part represents the number of patients with one or more episodes of BD (in gray for morphine and in crosshatch for midazolam). The line indicates approximately the 50% zone for both groups. The P-values immediately over each column represent the comparison with the previous day percentage (pre-post intragroup comparison). The P-values representing the intergroup comparison (morphine vs. midazolam) are shown on the top. Statistically significant values are shown in bold. c) Numbers of episodes of BD for each group per day. The circle represents the mean, whereas the length of the bars represents the 95% CI.
Discussion
Our results suggest that cancer-related dyspnea in ambulatory patients can be pharmacologically treated while its most proximal cause is sought and/or while an etiology-oriented intervention is implemented. This study indicates that neither morphine nor midazolam affected the outcome and/or implementation of additional diagnostic and/or therapeutic interventions. Nevertheless, the treatment of specific causes did not significantly improve dyspnea control. This could be explained, in part, by the multifactorial nature of dyspnea. Although at least one factor was suspected to actively contribute to dyspnea for each patient, a majority (almost 80%) presented more than one contributing cause. In contrast to other cancer and noncancer populations, in patients with incurable cancer, the decision whether to perform additional tests or to treat specific causes depends on the singularities of each case.
The selection of the drugs suited well our objectives of (ambulatory) fast titration, outpatient management, and utility for the baseline and the breakthrough components of dyspnea. Oral morphine has been shown to be effective and well tolerated for dyspnea control in ambulatory cancer patients.23 Previous results from our group19 and from clinical practice12 have shown that benzodiazepines in general and midazolam in particular may have a role in controlling dyspnea in cancer patients. The pharmacokinetic properties of oral morphine and oral midazolam are similar. The average oral bioavailability of morphine is 30%–40%, the onset of action is 15–30 minutes, the peak plasma concentration occurs within 30 and 90 minutes, and the half-life is 1.4–3.4 hours—obligating administration every four hours.20 Midazolam exhibits an oral bioavailability of 40%–50%, the onset of action is 15 minutes, the peak plasma concentration occurs within 30 and 90 minutes, and the half-life is 1.5–3 hours.21
The fast titration scheme implemented was successful in finding the most effective dose for both morphine and midazolam. More patients receiving midazolam felt relieved with the starting dose, reflecting that this drug is either more effective in the acute treatment of dyspnea or that it possesses a particularly advantageous pharmacological property. There was no difference in the efficacy for either drug in patients previously treated with opioids for other reasons. In this sense, there were not more naive patients alleviated with the starting dose of morphine than in any of the other escalating steps.
The control of dyspnea significantly improved by the second day for both groups. These gains were maintained during the subsequent days, suggesting that morphine and midazolam are well suited for relatively long-term administration and that the titration to a response consistently determined the optimal dose for each patient. Nevertheless, midazolam was more effective in controlling the continuing and breakthrough components of dyspnea during the ambulatory follow-up. In agreement with our previous study,19 we found that it is important for adequate management of the symptom to provide rescue doses of the medication for breakthrough or incidental dyspnea, in addition to the regularly scheduled doses. Continued reassessment was necessary, and a number of dose modifications were often required before optimal dyspnea control was achieved in some patients. Both treatments were well tolerated, with mild somnolence being the most common AE. Few patients required dose reduction because of AEs.
A potential limitation of our study is the single-blind nature of the design and lack of a placebo control group. In unblinded trials, researchers tend to overestimate the effect of treatment. However, in our trial, we minimized this bias by blinding the patients who were the ones who reported the effect of the intervention. Because there was a difference in efficacy between morphine and midazolam, and because we used several dose levels of these medications and thereby demonstrated the sensitivity of the study methods, we can assert that the placebo effect, if present, did not obscure drug effect.
This study showed that, although oral morphine might be an effective drug in the management of active dyspnea in this population, midazolam could be a better choice or at least an excellent alternative. Our data support the conclusion that oral midazolam alone should be considered among the first-line options for (maximally treated) cancer-related dyspnea in ambulatory patients.
References
- Symptom priority and course of symptomatology in specialized palliative care. J Pain Symptom Manage. 2006;31:199–206
- Impact of cancer-related symptom synergisms on health-related quality of life and performance status. J Pain Symptom Manage. 2008;35:604–616
- . Cortical substrates for the perception of dyspnea. Chest. 2005;128:345–354
- BOLD fMRI identifies limbic, paralimbic, and cerebellar activation during air hunger. J Neurophysiol. 2002;88:1500–1511
- Neuroimaging of cerebral activations and deactivations associated with hypercapnia and hunger for air. Proc Natl Acad Sci U S A. 2001;98:2029–2034
- Brain responses associated with consciousness of breathlessness (air hunger). Proc Natl Acad Sci U S A. 2001;98:2035–2040
- Neuroimaging evidence implicating cerebellum in the experience of hypercapnia and hunger for air. Proc Natl Acad Sci U S A. 2001;98:2041–2046
- Differential opioid action on sensory and affective cerebral pain processing. Clin Pharmacol Ther. 2008;83:577–588
- Acute opioid effects on human brain as revealed by functional magnetic resonance imaging. Neuroimage. 2006;31:661–669
- . Dose-dependent decrease of activation in bilateral amygdala and insula by lorazepam during emotion processing. Arch Gen Psychiatry. 2005;62:282–288
- The anxiolytic effects of midazolam during anticipation to pain revealed using fMRI. Magn Reson Imaging. 2007;25:801–810
- . Clinical management of dyspnoea. Lancet Oncol. 2002;3:223–228
- . Interventions for alleviating cancer-related dyspnea: a systematic review. J Clin Oncol. 2008;26:2396–2404
- . Pharmacological management of dyspnoea. Curr Opin Support Palliat Care. 2007;1:96–101
- The management of dyspnea in cancer patients: a systematic review. Support Care Cancer. 2008;16:329–337
- . Oral morphine for cancer pain. Cochrane Database Syst Rev. 2007;(4):CD003868
- Evidence-based standards for cancer pain management. J Clin Oncol. 2008;26:3879–3885
- . The etiology and management of intractable breathlessness in patients with advanced cancer: a systematic review of pharmacological therapy. Nat Clin Pract Oncol. 2008;5:90–100
- . Midazolam as adjunct therapy to morphine in the alleviation of severe dyspnea perception in patients with advanced cancer. J Pain Symptom Manage. 2006;31:38–47
- . Clinical pharmacokinetics of morphine. J Pain Palliat Care Pharmacother. 2002;16:5–18
- . Midazolam: a review of therapeutic uses and toxicity. J Emerg Med. 1997;15:357–365
- . Which measurement scales should we use to measure breathlessness in palliative care? A systematic review. Palliat Med. 2007;21:177–191
- . Opioids for the palliation of breathlessness in terminal illness. Cochrane Database Syst Rev. 2001;(4):CD002066
The drugs and other study costs were covered by the Instituto Angel H. Roffo of the Universidad de Buenos Aires. The authors declare no conflicts of interest.
PII: S0885-3924(10)00159-4
doi:10.1016/j.jpainsymman.2009.10.003
© 2010 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
