Literature Search Strategy
MEDLINE (1966 to July, week 1, 2006), EMBASE (1980 through 2006, week 27), CINAHL (1982 through July, week 1 2006), HealthSTAR (1975 through May 2006), DARE (2006, 2nd Quarter), and the Cochrane Library (2006, Issue 2) databases were searched. Relevant articles were identified by combining the Medical subject heading (MeSH) "dyspnea" with each of the following phrases used as text words: "breathlessness", "shortness of breath", "respiratory distress", "breath and shortness", "breath and difficult". These terms were then combined with the following MeSH terms and with various text words for pharmacological agents: "narcotics", "analgesics, opioid", "phenothiazines", "antipsychotic agents, phenothiazine", "benzodiazepines", "chlordiazepoxide", "clorazepate dipotassium", "estazolam", "medazepam", "midazolam", "triazolam", "benzazepines" "anti-anxiety agents", "benzodiazepine", "adrenal cortex hormones", "hydrocortisone", "cortisone", "betamethasone", "dexamethasone", "prednisolone", "methylprednisolone", "triamcinolone", "prednisone", "fludrocortisone". These terms were then combined with the search terms for the following study designs and publication types: practice guidelines, systematic reviews, meta-analyses, reviews, randomized controlled trials, controlled clinical trials, comparative studies, case-control studies, cohort studies, and cross-sectional studies. In addition, abstracts published in the conference proceedings of the American Society of Clinical Oncology (1995-2006) were searched for reports of new or ongoing trials. The reference lists from relevant articles were searched for additional trials, as were the reference lists from a sample of relevant review articles. The Canadian Medical Association Infobase (http://mdm.ca/cpgsnew/cpgs/index.asp) and the National Guideline Clearinghouse (http://www.guideline.gov/) were also searched for existing evidence-based practice guidelines.
Study Selection Criteria
Inclusion Criteria
Due to the lack of trials examining solely palliative care cancer populations, the inclusion criteria were expanded to include non-cancer patient populations. Articles were selected for inclusion in this systematic review of the evidence if they met all of the following criteria:
- The study population included adult patients with any advanced disease experiencing dyspnea where the underlying cause of the dyspnea was not treatable or its specific treatment did not relieve the dyspnea.
- One or more of the interventions included an opioid, phenothiazine, or benzodiazepine, administered by any route, or a systemic corticosteroid.
- One of the outcomes reported was dyspnea as measured by visual analogue scale (VAS), Borg score, or some other patient-reported scale. Other outcomes of interest were exercise tolerance, quality of life (QOL), arterial oxygen saturation (SaO2), measures of respiratory depression, forced expiratory volume in 1 second (FEV1) and symptomatic adverse effects.
- The article was a systematic review, meta-analysis, evidence-based practice guideline, or a fully published or abstract report of a randomized or non-randomized controlled study. Comparative studies, including prospective and retrospective cohort, case control, and cross sectional studies were eligible for inclusion.
Exclusion Criteria
Articles reporting studies of the management of dyspnea by treating specific underlying causes, such as by radiotherapy, chemotherapy, drainage of pleural fluid or treatment of bronchospasm, were excluded.
- Studies of systemic corticosteroids in non-cancer populations were excluded since the mechanisms of action of these agents in those populations are unlikely to be generalizable to cancer patients.
- Since translation resources were not available, articles in languages other than English were not included in the review nor searched for systematically.
- Letters, comments and editorials were not considered.
- Redundant publications of the same study were excluded. The most complete report of such a study was included, or the earliest report if the publications were identical.