The objective of this study was to systematically review the evidence for the efficacy of pharmacologic and nonpharmacologic treatments in alleviating dyspnea in patients with terminal cancer.

Databases searched were Cochrane Library up to 2007, MEDLINE (PubMed) (1966–2007), American Society of Clinical Oncology conference proceedings, and references of all included documents. In addition to databases, the search included the reference lists of key studies, the reference lists of 16 review articles on the topic, reference lists from 16 textbooks, and seven websites. Authors (15) of main investigations were contacted, and all members of the Association of Palliative Care and users of the www.palliativedrugs.com bulletin board were contacted for additional information and unpublished data.

Search keywords were opiate, opioid, morphine, benzodiazepine, furosemide, steroids, corticosteroids, oxygen, nonpharmacological, acupuncture, nursing, cancer, carcinoma, malignancy, dyspnea and breathlessness. 

Studies were included in the review if they were a randomized controlled trial assessing dyspnea in patients with terminal cancer in which any intervention for dyspnea relief was compared with no intervention, placebo, or another intervention.

Studies were excluded if they were nonrandomized studies or trials in which only a minority of the patients had a cancer diagnosis.

Literature evaluated included 37 studies, plus one abstract initially reviewed. A final set of 18 studies was included; 7 assessed opioids, 6 assessed oxygen- or helium-enriched air, 1 assessed furosemide, and 4 assessed nonpharmacologic interventions. Meta-analysis was not completed due to the paucity of studies and heterogeneous outcome measures.

Sample Size Across Studies:

• Opioid intervention = 256 patients
• Oxygen or helium = 148 patients
• Furosemide intervention = 7 patients
• Nonpharmacologic intervention = 403 patients

Sample Range Across Studies:

• Opioids = 9–101 patients
• Oxygen or helium = 12–51 patients
• Nonpharmacologic = 34–203 patients

With respect to gender, age, and diagnosis within the sample, the opioids subgroup included both genders. The median age range was 56–73 years. The majority had primary lung cancer, and both opioid-tolerant and opioid-naïve participants were included.

The oxygen or helium subgroup included both genders. The median age range was 64–72 years. The majority had primary lung cancer.

No comment was available on gender or age for the nonpharmacologic subgroup, but the primary diagnosis was lung cancer.

The primary outcome was subjective dyspnea relief according to the visual analog scale (VAS) or dyspnea intensity according to the modified Borg scale. The secondary outcome was oxygen saturation and adverse effects.

Opioid Intervention:

• The administration of subcutaneous morphine resulted in significant reduction in dyspnea according to the VAS compared with placebo.
• Nebulized morphine versus placebo failed to demonstrate a significant effect of nebulized morphine. No difference in dyspnea VAS score was observed in one trial when nebulized morphine was compared with subcutaneous morphine, although patients preferred the nebulized route.
• In one trial, the addition of benzodiazepines (midazolam) to morphine was significantly more effective than morphine alone, without additional adverse effects.

Oxygen Intervention:

• Oxygen was not superior to medical air for alleviating dyspnea, except for patients with hypoxemia.

Furosemide Intervention:

• One small trial assessed the use of nebulized furosemide with a trend toward worsening dyspnea.

Nonpharmacologic Interventions:

• Nurse-led interventions improved breathlessness.
• Acupuncture was not beneficial.
• Nurse-led interventions encompassed routine follow-up by nurses performing methods of counseling and relaxation and teaching coping strategies.
• The nurse-led breathlessness rehabilitation techniques and education and advice regarding coping with the psychological aspects of the symptoms assessed the primary outcome of dyspnea relief after weeks.
• All nurse-led interventions proved to be beneficial, improving breathlessness and quality of life in terms of physical, psychological, and emotional aspects.

• No evidence supports subcutaneous morphine as effective in treating dyspnea in patients with advanced cancer.
• Use of oxygen to alleviate dyspnea in nonhypoxic patients with cancer cannot be recommended. Supplemental oxygen is expensive and can restrict mobility with possible decrease in quality of life. Use of medical air (78.9% nitrogen, 21.1% oxygen) was shown to be effective in reducing the sensation of dyspnea. However, this intervention is not used routinely in care settings.
• This review recommended integration of pharmacologic and nonpharmacologic interventions, such as those used in nurse-led programs to relieve dyspnea.

Acknowledging the paucity of evidence from randomized controlled trials to support the interventions is important.

Limitations of this review were

• Few randomized controlled trials
• Small studies
• Short follow-up in opioid studies
• Lack of consistency regarding opioid doses.

A major research opportunity exists to further document outcomes from nurse-led dyspnea interventions.

The objective of this meta-analysis and systematic review was to evaluate the role of different interventions to alleviate dyspnea.

• Databases used were CENTRAL, EMBASE and PubMed.
• Search keywords were opiate, benzodiazepine, furosemide, steroids, oxygen, or pharmacological and dyspnea or breathlessness.
• Studies were included if they were randomized controlled trials assessing patients with terminal cancer with dyspnea in which an intervention was compared with no intervention, placebo, or an alternative intervention.
• No exclusion critera were stated.

A total of 829 references were retrieved. The specific method of evaluation was not described, but the small sample size of most studies was noted.

• The final number of studies included was 18.
• Of the specifics included, sample sizes ranged from 9–38.
• All studies involved patients with end-stage disease.

Patients were undergoing end-of-life care.

• Meta-analysis of three trials comparing opiods with placebo yielded a weighted mean difference of -1.31 (95% CI, -2.49, -0.13), showing a statistically significant benefit with opioids. 
• Studies tended to show better results with nebulized opiods rather than subcutaneous administration. 
• Analysis of six trials of oxygen showed lack of benefit. 
• Standard mean difference was -0.3 (95% CI, 1.06, 0.47).
• Two studies compared midazolam to morphine and the combination of both drugs.
• One study showed that addition of midazolam to morphine improved results.
• Two small studies concluded that furosemide did not improve dyspnea.

• Opioids are effective in reducing dyspnea and may work better if nebulized. 
• The addition of midazolam to opioid may improve results. 
• Oxygen and furosemide were not effective in reducing dyspnea.

• Relatively few studies used each intervention. 
• Opioid trials were limited by lack of dosage information and various types of opioids used. 
• In studies evaluating oxygen, hypoxemia was not routinely evaluated. 
• Method of measuring dyspnea varied across studies.

Findings provide guidance regarding effectiveness of interventions for dyspnea in patients with cancer. These results demonstrate the effectiveness of opioids. Findings also confirm those of others that palliative oxygen is of no benefit for this symptom. Some reviews continue to suggest the use of palliative oxygen. This is not supported by evidence, and home oxygen therapy is generally not covered by insurance for patients who do not have hypoxemia. Unnecessary use can be costly to the patient. Evidence is limited regarding the effects of the addition of hypnotics to opioids in managing dyspnea. This is an area that could benefit from additional research.

To compare efficacy of low-frequency, low-intensity electrotherapy and manual lymphatic drainage in treatment of upper-limb lymphedema

Patients were randomized to two groups. Group A underwent electrotherapy therapy for 10 sessions followed by 10 sessions of manual drainage. Group B underwent manual drainage first and then received electrotherapy. There was a month washout period between treatments. Patients were assessed after every 10 treatment sessions. Electrotherapy was delivered with a system that provides massage with low-frequency electrical stimulation.

• The study sample (N = 30) was comprised of female patients with breast cancer.
• Mean age was 68 years.
• All patients had unilateral arm lymphedema.
• Seventy-eight percent of patients had had both chemotherpy and radiation therapy.
• Average duration of lymphedema was 73 months.
• Eighty-eight percent used compression garments.

The study took place in an outpatient setting in Spain.

The study has clinical applicability for late effects and survivorship.

The study used a randomized crossover trial design.

• Limb volume was calculated using six circumference measurements.
• Patients completed the Functional Assessment of Cancer Therapy Questionnaire for Breast Cancer Version 4 (FACT-B4).
   

 There were no significant differences in outcomes between the two treatments.

There was no difference in benefits from manual lymphatic drainage and low-frequency, low-intensity electrotherapy.

• The study had a small sample size, with less than 100 participants.
• The study has a risk of bias because it had no blinding.

 Findings suggest there is no difference in efficacy of these two treatment approaches for management of arm lymphedema in patients with breast cancer.

The object of the systemic review was to assess the effectiveness of ketamine as an adjuvant therapy in treating cancer pain.

Medline (1996-2002); EMBASE (1980-2002); CancerLit (1966-2002); Cochrane Controlled Trials Registar and Database of Systematic Reviews (Cochrane Library, Issue 1, 2002); Specialized Registar of the Cochrane Pain, Palliative and Supportive Care Group (2001), PARDLARS (inhouse datadase of Pfizer UK, February 2002), hand searching reference lists

Two RCT met inclusion criteria. Two other RCT appeared to meet the criteria but were considered methodologically flawed. There were 32 case reports or open-label uncontrolled trials that could not be included.

Thirty patients, 17 men and 13 women, ages 21-69. Cancers included stomach, cervix, liver, lung, colon, pancreas, bladder, rectum, histiocytoma, and uterus. Study 1 with 20 had patients whose pain had been treated successfully with opioids. Study 2 had 10 whose pain had been unrelieved by their dose of morphine. Study 2 defined the pain as “neuropathic,” and the patients had a Karnofsky of 50 or more.

There was a reduction in pain scores for those receiving the ketamine. Four patients experienced hallucinations (most common adverse effect associated with ketamine). Two of these four patients also experienced a sensation of insobriety. All of these effects were relieved by diazepam.

Treatments evaluated:

• Ketamine 0.25 mg/kg adjuvant to morphine given IV
• Ketamine 0.5 mg/kg adjuvant to morphine given  IV
• Ketamine 1 mg/kg co-administered intrathecally with   morphine (reducing the morphine dose)

Outcomes Measured:

The primary outcome measure was patient-reported pain intensity and pain relief, using validated measures on movement and at rest ( e.g., visual analog scale and verbal rating scale).

Secondary outcomes included: total opioid consumption, rescue medication, adverse events, study withdrawals, and dropouts.

Because of the small number of RCTs , the small sample sizes, and the lack of other acceptable research, there is insufficient evidence for the use of ketamine as an adjuvant to opioids to make a recommendation for practice.

Suggested for research are crossover designs, larger patient groups, comparisons with other opioids, comparisons of routes of administration, and clearly defined outcomes. Research into ketamine’s role as an NMDA antagonist also would be welcomed.

DATABASES USED: MEDLINE (1996–2002); EMBASE (1980–2002); CANCERLIT (1966–2002); Cochrane Controlled Trials Register and Database of Systematic Reviews (Cochrane Library, Issue 1, 2002); Specialized Register of the Cochrane Pain, Palliative and Supportive Care Group (2001); PARDLARS (in-house database of Pfizer UK, February 2002); hand-searching reference lists

EVALUATION METHOD AND COMMENTS ON LITERATURE USED: Two RCTs met inclusion criteria. Two other RCTs appeared to meet the criteria but were considered methodologically flawed. Thirty-two case reports or open-label uncontrolled trials could not be included.

FINAL NUMBER STUDIES INCLUDED = 2

TOTAL PATIENTS INCLUDED IN REVIEW: 30

KEY SAMPLE CHARACTERISTICS: Seventeen men and 13 women aged 21–69 years. Cancers included stomach, cervix, liver, lung, colon, pancreas, bladder, rectum, histiocytoma, and uterus. The first study included 20 patients whose pain had been treated successfully with opioids. The second study included 10 patients whose pain had been unrelieved by their dose of morphine. In the second study, pain was defined as “neuropathic,” and the patients had a Karnofsky score of 50 or more.

Pain scores were reduced in those patients receiving ketamine. Four patients experienced hallucinations (most common adverse effect associated with ketamine). Two of those four patients also experienced a sensation of insobriety. All of these effects were relieved by diazepam.

The following treatments were evaluated.

• Ketamine 0.25 mg/kg adjuvant to morphine given IV
• Ketamine 0.5 mg/kg adjuvant to morphine given  IV
• Ketamine 1 mg/kg co-administered intrathecally with morphine (reducing the morphine dose)

The primary outcome measure was patient-reported pain intensity and pain relief, using validated measures on movement and at rest (e.g., visual analog scale and verbal rating scale). Secondary outcomes included total opioid consumption, rescue medication, adverse events, study withdrawals, and dropouts.

The objective of the systematic review was to assess the effectiveness of ketamine as an adjuvant therapy in treating cancer pain. Because of the small number of RCTs, small sample sizes, and lack of other acceptable research, evidence is insufficient to make the use of ketamine as an adjuvant to opioids a recommendation for practice.

Suggested for research are crossover designs, larger patient groups, comparisons with other opioids, comparisons of routes of administration, and clearly defined outcomes. Research into ketamine’s role as an NMDA antagonist also would be welcomed.

To determine the efficacy and adverse effects of ketamine as an adjuvant to opioids in the treatment of cancer pain

Databases searched were CENTRAL, MEDLINE, EMBASE, PubMed, and the Pfizer Product Information Database.

Studies were included in the review if they

• Were randomized clinical trials (RCTs).
• Involved adult patients with cancer pain being treated with opioids.
• Studied patients receiving ketamine at any dose or route for cancer pain or a placebo or active control.

Studies were excluded if they had fewer than 10 participants.

A total of 120 references were retrieved. Studies were selected based on criteria, assessed independently, reviewed by two reviewers, and chosen for inclusion by three independent reviewers.

• The final number of studies included was two.
• The total number of patients included in the review across studies was 30.
• Of these patients, 20 were hospitalized with cancer pain unrelieved effectively with opioids who received two different doses of IV ketamine (0.25 mg or 0.5 mg) and 10 were patients with neuropathic pain unrelieved by opioids.

• Patients were at the end-of-life phase of care.
• This study has clinical applicability for palliative care.

• For IV ketamine, the 0.5 mg dose demonstrated a significant reduction in pain intensity and the relief was sustained throughout the three-hour measurement period.
• For intrathecal ketamine, pain was controlled on reduced doses of intrathecal opioids or on ketamine alone.

Current evidence is insufficient to assess the benefits and harms of ketamine as an adjuvant to opioids in the relief of cancer pain because of the small number of trials with low sample sizes.

• The sample size was small, with only 30 patients.
• Few RCTs exist on this topic.
• Doses and routes of administration of ketamine were different, measurement of pain was variable, the type of pain was variable (one study was of neuropathic pain but not the other), and the time for measurement was not consistent.

Although studies suggest that ketamine may be a helpful adjuvant therapy to improve pain and decrease opioid requirements in patients with cancer, more studies are needed with larger sample sizes and control groups. Evidence is lacking to recommend ketamine in practice.

• N = 67
• MEAN AGE = 59.6 years
• MALES: 41.5%, FEMALES: 58.5%
• KEY DISEASE CHARACTERISTICS: Patients had hematologic cancer (e.g., leukemia, lymphoma, myelodysplastic syndrome) or other oncologic illness and received care for at least five weeks
• OTHER KEY SAMPLE CHARACTERISTICS: 61.5% were spousal caregivers, and 16.9% were “offspring” caregivers; patients agreed to have their main caregiver involved in the study, and patients had approval for study participation from the doctor/psychological dyad

• SITE: Six acute care units, five different settings  
• SETTING TYPE: Multi-site  
• LOCATION: Clermont-Ferrand, France

• PHASE OF CARE: Active treatment

• Randomized, controlled trial
  • Patient-caregiver dyads were assigned to the experimental or control group using randomization software

• SF-36 health survey (patients and caregivers self-administered)—used to assess eight health dimensions
• Zarit Burden Scale (caregivers self-administered)—used to assess three dimensions of burden
• Questionnaire for satisfaction of hospitalized patients (patient self-administered)—used to assess eight factors related to scope of hospital experience

No difference was seen in quality of life scores between the control (N = 34) and experimental (N = 33) patient-caregiver dyad groups at the first testing. Evaluation (pre- and post-test scores) supported improved quality-of-life scores in the experimental group patient-caregiver dyads as compared to those in the control group. Patients in the experimental group showed significant improvement on two of six dimensions (physical role [p = 0.039] and general health [p = 0.037] on the quality-of-life measure. Experimental group caregivers showed significant improvement on six of eight dimensions (physical role, emotional role, vitality, mental health, social functioning, and general health, all p < 0.05) on the quality-of-life measure. The average score of burden of care in the experimental group (M = 16, SD = 9.9) was significantly lower than that of control group (M = 31.4, SD = 14.9) at last evaluation (p < 0.001). Evolution scores for burden also showed significantly (p = 0.004) less perceived burden among experimental group caregivers as compared to control group caregivers. Patient satisfaction scores of patients in the experimental and control groups showed no difference, except patients in the experimental group showed a significant improvement in relationship with nurses (p = 0.037) because of program involvement.

Nurse-provided educational interventions tailored to caregiver needs and patient expectations may minimize caregiver burden and improve patient and caregiver quality of life. These interventions, if provided during patients’ acute setting care, may pave the way for more effective patient home care by informed and involved caregivers.

• Small sample (less than 100)
• Risk of bias (no blinding)
• Risk of bias (no appropriate attentional control condition)
• Findings not generalizable
• Other limitations/explanation: Authors acknowledge strict inclusion and exclusion criteria that influenced the loss of 110 of 265 participants initially assessed for the study. Author conclusions note a need for a less rigorous program to meet patient and caregiver needs for those who do not meet more stringent criteria found in an RCT used in this study. Whether nurses who were members of the research team delivered the intervention and could connect knowledge of individual patients/caregivers to study results is unclear. Which of the four teaching categories influenced caregiver quality of life and burden the most also is unclear.

Nurses who deliver evidence-based, tailored educational interventions to family cancer caregivers may increase patient-caregiver quality of life and minimize caregiver burden during cancer care. Stronger nurse-patient relationships, based on interaction during caregiver task teaching, may facilitate patient care satisfaction and coping during the cancer experience.

To investigate the effects of acupressure on anxiety in patients with cancer

Patients were randomly assigned to acupressure, sham acupressure, or control groups. Control group patients received usual care. The acupressure group had pressure applied at nine points for two minutes each by a research assistant. Sham acupressure was also given at nine different points on the body. Patients received 10 sessions of sham or actual acupressure. Study measures were obtained at baseline, after session 5, and after session 10.

• N = 85
• MEAN AGE = 48 years
• MALES: 54.1%, FEMALES: 45.9%
• KEY DISEASE CHARACTERISTICS: 51% had stage 4 disease
• OTHER KEY SAMPLE CHARACTERISTICS: The majority had prior surgery and chemotherapy

• SITE: Single site  
• SETTING TYPE: Inpatient  
• LOCATION: Iran

Double-blinded, sham-controlled, randomized trial

• Spielberger State/Trait Anxiety Inventory (STAI)

Mean anxiety scores in the acupressure group declined over the three study measurement periods (p = 0.001). Anxiety scores in the sham group also declined somewhat. Anxiety levels in the control group increased over time. An analysis of variance showed that acupressure was associated with a significant reduction in state anxiety scores and a significant difference in comparison to the control group patients (p < 0.0001). There was no significant difference between acupressure and sham acupressure changes.

Both sham and actual acupressure were associated with reductions in anxiety scores.

• Small sample (< 100)
• Other limitations/explanation: The timing of postintervention anxiety measurements were not clear, but it appears this was done immediately after the sham or acupressure interventions. The actual clinical relevance of an immediate short-term change is not clear.

The findings of this study suggest that acupressure may be of some benefit in the short-term management of anxiety among patients with cancer. The findings regarding the effects of sham acupressure suggest there may be a placebo effect involved. Additional well-designed research would be helpful to demonstrate the role of acupressure in the management of anxiety.

To investigate the effects of adenosine 5'-triphosphate (ATP) infusions on quality-of-life (QOL) parameters in patients with preterminal cancers of mixed tumor types.

Patients were randomized to the usual care control group with standard nutritional advice or to the intervention of usual care, nutritional advice, and ATP infusion. An eight- to 10-hour ATP infusion was given weekly, with a maximum dose of 50 µ/kg/minute.

• In total, 83 patients (23 women, 60 men) (ATP, n = 44; control, n = 39) were included.
• Age ranged from 54 to 78 years.
• Patients were undergoing curative treatment for lung, colon, gastrointestinal, prostate, and other types of cancer.
• Patients reported fatigue and had weight loss greater than 5% or anorexia.

• The study was conducted at multiple sites in The Netherlands.
• The initial dose was given at the daycare centers of participating hospitals and then patients were infused at home.

The study was a randomized, controlled trial.

• Quality of Life Questionnaire
• Groningen Activity Restriction Scale (GARS)
• Dutch Short Fatigue Questionnaire
• Jamar Hydraulic Hand Dynamometer
• MicroFET 2 Dynamometer

• Between the ATP and control groups, no statistically significant differences were observed in QOL, functional status, or fatigue.
• Elbow flexor muscle strength had an almost significant increase in the control group (1.98).

Better survival was observed during the eight-week intervention in the ATP group. The rationale was that palliative cancer drugs cause serious side effects. The authors concluded that that may be an advantage of the ATP treatment.

• The study had a small sample size, with less than 100 patients.
• The increased drop-out may have limited the power of the study.
• Large heterogeneity existed between patients.
• The eight-week intervention was shorter than other studies of 12 weeks.

The study provided little useful information for nurses.

To determine whether an exercise-based intervention was safe and effective for breast cancer survivors

• N = 67  
• AVERAGE AGE = 55.1 years
• FEMALES: 100%
• KEY DISEASE CHARACTERISTICS: Breast cancer 

• SITE: Single-site    
• SETTING TYPE: Home    
• LOCATION: United States

• PHASE OF CARE: Late effects and survivorship

Quasiexperimental

• Height and weight
• Physical activity outside of weight lifting
• Perometry
• National Cancer Institute's Common Terminology Criteria for Adverse Events–Lymphedema (NCI-CTCAE-L) version 3
• Muscular strength 
• Semistructured interviews 

The number of lymphedema symptoms reduced from baseline to the 12-month follow-up visit (2.6 [SD = 2.6] to 1.7 [SD = 2.1], p = .002), but there was no statistically significant reduction in the severity of lymphedema symptoms from baseline to the 12-month follow up visit (p = .58). In addition, there was no difference in arm volume, measured as a percent of interlimb difference, at the 12-month follow-up visit (p = .98). Within this sample, four patients (8%) had new-onset lymphedema at the 12-month measurement and five (19%) had a flare-up of existing lymphedema.

The PAL intervention reduced the number of lymphedema symptoms but not the severity of symptoms experienced by participants. In addition, the intervention did not reduce the limb volume of participants. Participants were able, however, to complete the intervention, which took place mostly in the home setting, making it a feasible intervention for women with breast cancer.

• Small sample (< 100)
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Unintended interventions or applicable interventions not described that would influence results
• Other limitations/explanation: It is unclear what other exercises participants were doing at home.

When available, home-based exercise interventions like the ones described here may be beneficial for patients who are motivated to participate in and complete exercises at home. Nurses should assess patients for readiness to participate in a home exercise program and help patients find resources for planning an appropriate home-based program.