Viola, R., Kiteley, C., Lloyd, N.S., Mackay, J.A., Wilson, J., Wong, R.K., & Supportive Care Guidelines Group of the Cancer Care Ontario Program in Evidence-Based Care. (2008). The management of dyspnea in cancer patients: A systematic review. Supportive Care in Cancer, 16(4), 329-337.

The objective of this study was to evaluate the effectiveness of four drug classes: opioids, phenothiazines, benzodiazepines, and systemic.

Databases searched were HealthSTAR, MEDLINE, CINAHL, EMBASE, Cochrane Library and Database of Abstracts and Reviews of Effects Issue 2, American Society of Clinical Oncology conference proceedings (1995-2006), Canadian Medical Association Infobase, and National Guidelines Clearing House. Reference lists from relevant articles were searched for additional trials

Search keywords were dyspnea, breathlessness, shortness of breath, respiratory distress, breath and shortness, and breath and difficult combined with terms for pharmacologic agnets, study designs, and publication types.

Inclusion criteria included

• Systematic reviews
• Meta-analyses
• Evidence-based practice guidelines
• Fully published or abstract reports of randomized or nonrandomized controlled studies of opioids, phenothiazines, or benzodiazepines administered by any route involving adult patients with dyspnea
• Subjects with any advanced disease
• Studies involving corticosteroids, only if the primary advanced disease was cancer
• Studies in which one of the outcomes reported was dyspnea, measured by a patient-reported scale.
 

Exclusion criteria included
• Studies in languages other than English
• Stuides eported in letters or editorials.

• The search identified two practice guidelines, three systematic reviews, 23 published randomized controlled trials (RCTs), two abstracts of RCTs, and three published nonrandomized trials, for a total of 33 references.
• The review did not identify the number of excluded items from the initial search. 
• Study quality was evaluated formally using the Jadad scale.

The total sample across  29 trials was 600 patients, with individual sample sizes ranging from 4-101. Trials included involved

• 6 trials of opiods in only patients with cancer
• 10 trials of systemic opioids including patients who did not have cancer
• 7 trials of nebulized opioids including patients with and without cancer
• 4 trials of benzodiazepines
• 2 trials of phenothiazines.
   

• Search sources and criteria were not reported in either of the two practice guidelines. One indicated that both corticosteroids and opioids were options for managing dyspnea but that the evidence was poor. The other, a Finnish guideline, recommended opioids, steroids, and benzodiazepines, but evidence was only cited for opiods.
• Opioids studied included morphine orally, subcutaneously, or via nebulizer; dihydrocodeine; diamorphine; and promethazine with morphine. All but three studies examined the effects of a single dose on dyspnea via use of a visual analogue scale or exercise tolerance.
• In opioid trials involving only patients with cancer, four examined systemic opioids, one used nebulized opiods, and one included both systemic and nebulized administration. One trial used a combination of morphine and midazolam. Systemic opioid studies tended to show significant decrement in mean dyspnea and respiratory rate with morphine. In the trial that included midazolam, more patients on the combined regimen reported relief from dyspnea at 24 and 48 hours and had fewer episodes of breakthrough dyspnea. However, no differences were seen in mean dyspnea scores and exercise tolerance between groups overall.
• Nebulized opioids did not show significant differences compared to systemic morphine in one trial.
• One benzodiazepine trial involved patients with cancer. In trials with other patients, none of the studies demonstrated a significant reduction in dyspnea when compared to placebo.
• No trials were on phenothiazines in patients with cancer. One study showed a benefit with promethazine compared to placebo on dyspnea and exercise tolerance.
• Adverse effects reported across trials included drowsiness, nausea and vomiting, and constipation in opioid trials. Results of opioids on oxygen level were mixed.
• Results of benzodiazepines and phenothiazines on oxygen and carbon dioxide levels were mixed. Drowsiness was the most frequent adverse effect reported with benzodiazepine.

• Overall evidence favors a beneficial effect of systemic opioids on dyspnea and exercise tolerance.
• None of the studies comparing nebulized morphine with placebo or systemic opioids found it to be beneficial. 
• Whether studies with opioids indicate a drug class effect is not clear because only a few drugs have been studied.
• Studies of benzodiazepines did not suggest any benefit.
• Studies of phenothiazines gave conflicting results.
• Overall evidence in this area demonstrate conflicting results, and this systematic review also gives conflicting results and conclusions within the article.
• Most studies had very small sample sizes, and doses, dose schedules, routes, and outcome measures varied greatly, making overall conclusions difficult.
• While the stated purpose of the review was to determine effects within a cancer population, most of the research reviewed was not in this population.
• While use of opioids may be of benefit for patients with cancer in reducing the sensation of dyspnea, this effect needs to be balanced with the adverse effects that can be expected with such treatment, including symptoms of constipation, drowsiness, and nausea and vomiting. Interventions to prevent or manage these effects would also be essential.