Randomised controlled trials that compared rehabilitation programmes of upper extremity exercise to a control group or upper extremity exercise plus lower extremity exercise compared with lower extremity exercise alone in patients with chronic obstructive pulmonary disease were eligible for inclusion. Eligible interventions need to be at least three weeks duration and report data on the primary outcome of dyspnoea using the Borg score or Borg-modified score during activities of daily living and exercise training. Studies with age- or gender-matched control groups of healthy participants or patients with other respiratory conditions were excluded from the review.
The studies were published between 1988 and 2012. Most patients were male, elderly and diagnosed with moderate to severe chronic obstructive pulmonary disease. Upper extremity interventions consisted of unsupported arm exercise, gravity-resisted exercise and proprioceptive neuromuscular facilitation exercise. Control groups participated in walking training, lower limb training (including endurance training), bronchia hygiene therapy, general exercises and sham training of upper limb flexibility and stretching. Where reported, exercise time ranged from 20 to 30 minutes for three to seven days per week. Intervention durations ranged from three to eight weeks. Dyspnoea in the studies was measured using the simulated activities of daily living tests, endurance tests, standard daily physical activities test, six-minute pegboard and ring test and endurance arm crank test.
Two reviewers performed the study selection; any disagreements were resolved by a third reviewer.