Inhibidores de la fosfodiesterasa 4 para la enfermedad pulmonar obstructiva crónica
Referencias
Referencias de los estudios incluidos en esta revisión
Referencias de los estudios excluidos de esta revisión
Referencias de los estudios en espera de evaluación
Referencias de los estudios en curso
Referencias adicionales
Referencias de otras versiones publicadas de esta revisión
Characteristics of studies
Characteristics of included studies [ordered by study ID]
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 102 centres in Canada, Mexico, and the USA Participants: 647 (15 mg cilomilast: 431, placebo: 216) Baseline characteristics: mean age 65 years, 62% male, mean FEV₁ % predicted 49.7%, mean smoking history 59.9 pack‐years for cilomilast and 56.1 pack‐years for placebo, or current smokers (44% and 47%, respectively) Inclusion criteria: FEV₁/FVC ≤ 0.7, FEV₁ 30% to 70% with smoking history > 10 pack‐years or current smokers Exclusion criteria: active tuberculosis, lung cancer, bronchiectasis Total numbers of participant withdrawals: 137 (32%) and 52 (24%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind. Placebo tablets to assess suitability
Concomitant medication
| |
| Outcomes | Primary outcomes: lung function; change in FEV₁; SGRQ averaged over 24 weeks Secondary outcomes: incidence rate of COPD exacerbations; adverse events; FVC at trough; 6MWT; post‐exercise dyspnoea | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that the randomisation process was adequate due to pharma sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | "The primary reasons for the withdrawal of subjects from the study prior to randomisation were the failure to meet inclusion/exclusion criteria (15.4%) and the presence of adverse effects, including COPD exacerbations (8.5%). More subjects receiving cilomilast than placebo withdrew from the double‐blind phase of study (31.8% (n = 137) versus 24.1% (n = 52)" |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website |
| Other bias | Low risk | Baseline anticholinergic, beta₂‐agonist, or corticosteroid use 54% in cilomilast, 58% placebo used ipratropium; 99% in cilomilast, 100% placebo used albuterol; 9% in cilomilast, 12% placebo used salmeterol; 7% in cilomilast, 8% placebo used triamcinolone; 6% in cilomilast, 7% placebo used beclomethasone |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 98 centres in Australia and New Zealand, Germany, Spain, South Africa, and the UK Participants: 700 (15 mg cilomilast: 474, placebo: 226) Baseline characteristics: mean age 64.6 years, 80% male, mean FEV₁ % predicted 49% with 5.1% reversibility, DLCO 71% predicted, also with higher rates of chronic bronchitis 80.1%. 45% active smokers Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7, FEV₁ 30% to 70% with smoking history > 10 pack‐years Exclusion criteria: active tuberculosis, lung cancer, bronchiectasis Total numbers of participant withdrawals: 122 (26%) and 51 (23%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: lung function; change in FEV₁; SGRQ averaged over 24 weeks Secondary outcomes: incidence rate of COPD exacerbations; summary symptom score; FVC at trough; 6MWT; post‐exercise dyspnoea | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that the randomisation method was adequate due to pharma sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 51 (23%) placebo, 122 (26%) cilomilast, primarily due to adverse events, of which most were not from COPD exacerbations |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Analysis was done on per‐protocol population | |
| Participants | Setting: not stated Participants: 59 (15 mg cilomilast: 29, placebo: 30) Baseline characteristics: mean age 61 to 62 years, 81% male, 53% active smokers, mean 46 pack‐years, 53% to 58% FEV₁ predicted Inclusion criteria: aged 40 to 80 years, fixed airflow obstruction, smoking history > 10 pack‐years Exclusion criteria: not stated Total numbers of participant withdrawals: 4 (14%) and 2 (7%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind with placebo
Concomitant medication
Used alongside SABA (available to all) and anticholingeric drugs (offered to 24%) | |
| Outcomes | Primary outcome: change in neutrophil percentage in induced sputum Secondary outcomes: FEV₁; numbers of subepithelial CD8+ cells, CD 68+ cells, epithelial, and subepithelial neutrophils | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that the randomisation process was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Unclear risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | "One patient was lost to follow‐up 3 days after randomisation and another withdrawn for non‐compliance 32 days after randomisation. Four patients were withdrawn after adverse events" |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 110 centres in Belgium, Finland, France, Italy, the Netherlands, Norway, Portugal, Spain, and the UK Participants: 711 (15 mg cilomilast: 469, placebo: 242) Baseline characteristics: mean age 64.6 years, 86% male, mean FEV₁ % predicted 53% with 5.0% reversibility, 38% active smokers Inclusion criteria: FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years Exclusion criteria: active tuberculosis, lung cancer, bronchiectasis Total numbers of participant withdrawals: 121 (26%) and 63 (26%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: lung function; change in FEV₁; SGRQ averaged over 24 weeks Secondary outcomes: incidence rate of COPD exacerbations; summary symptom score; FVC at trough; 6MWT; post‐exercise dyspnoea | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that the randomisation process was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 63 (26%) placebo, 121 (26%) cilomilast, primarily due to adverse events, of which most were not due to COPD exacerbations |
| Selective reporting (reporting bias) | Unclear risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 103 centres in the USA Participants: 613 (15 mg cilomilast: 296, placebo: 317) Baseline characteristics: mean age 63.2 years placebo, 63.1 years cilomilast, 47% male placebo, 46% male cilomilast. Mean FEV₁ % predicted not available Inclusion criteria: aged ≥ 40 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, ≤ 70% post‐albuterol reversibility, ≤ 15% or ≤ 200 mL (or both) post‐albuterol FEV₁ ≤ 70% of predicted normal, ≥ 1 COPD exacerbation within 12 months before screening Exclusion criteria: not stated Total numbers of participant withdrawals: 105 (35%) and 76 (24%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcomes: change from baseline to endpoint in trough pre‐bronchodilator FEV₁; change in total SGRQ score averaged over 24 weeks Secondary outcomes: changes from baseline in clinic trough FVC; time to first level 2 or level 3 COPD exacerbation | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Unclear risk | The trial was double‐blinded (participants and investigator). It is not clear if the investigator was administering the treatment |
| Blinding of outcome assessment (detection bias) | Low risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 76 (24%) placebo, 105 (35%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. The trial was registered at clinicaltrials.gov |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Analysis was done on per‐protocol population | |
| Participants | Setting: 10 centres in the USA Participants: 65 (15 mg cilomilast: 31, placebo: 34) Baseline characteristics: mean age 64.4 years placebo and 66.1 years cilomilast, 67% male placebo and 84% male cilomilast, mean FEV₁ % predicted not available Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, post‐salbutamol reversibility ≤ 15% or 200 mL, post‐salbutamol FEV₁ ≥ 1.0 L and between 30% and 70% predicted Exclusion criteria: not stated Total numbers of participant withdrawals: 1 (3%) and 1 (3%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcome: change from baseline at endpoint in neutrophils as a percentage of total cells in induced sputum Secondary outcomes: FVC at trough; sputum macrophages, eosinophils, and lymphocytes as a percentage of total cells in induced sputum; total cell counts in induced sputum | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Low risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 1 (3%) placebo, 1 (3%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 32 centres in the USA, Canada, and Australia Participants: 156 (15 mg cilomilast: 79, placebo: 77) Baseline characteristics: mean age 64.2 years placebo and 65 years cilomilast, 66% male placebo and 65% male cilomilast, mean FEV₁ % predicted not available Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, post‐salbutamol reversibility ≤ 15% or 200 mL, post‐salbutamol FEV₁ ≥ 1.0 L and between 30% and 70% predicted, baseline RV (from plethysmography) ≥ 120% predicted RV Exclusion criteria: not stated Total numbers of participant withdrawals: 15 (19%) and 14 (18%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcome: change from baseline to endpoint in volume of trapped gas (D) Secondary outcomes: lung volume measurements, including SVC and RV; 6MWT; exertional dyspnoea | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the allocation concealment process because of pharmaceutical sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 14 (18%) placebo, 15 (19%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 22 centres in China Participants: 1018 (15 mg cilomilast: 678, placebo: 340) Baseline characteristics: mean age 63.9 years placebo and 64.6 years cilomilast, 91% male placebo and 93% male cilomilast, mean FEV₁ % predicted not available Inclusion criteria: aged 40 to 75 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, documented history of COPD exacerbations each year for 3 years before screening, ≥ 1 exacerbation in the last year that required oral corticosteroids or antibiotics, post‐salbutamol reversibility ≤ 15% or 200 mL, post‐salbutamol FEV₁ ≥ 1.0 L and between 25% and 70% predicted, % predicted FRC ≥ 120% from plethysmography Exclusion criteria: not stated Total numbers of participant withdrawals: 124 (18%) and 35 (10%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcome: change from baseline to endpoint in trough pre‐bronchodilator FEV₁ Secondary outcomes: time to first level 2 or level 3 COPD exacerbation (level 2 is defined as acute worsening of COPD that requires additional treatment or hospital outpatient visit; level 3 is hospital admission for treatment); change from baseline to endpoint RV and FRC; change from baseline total score on SGRQ | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the randomisation process because of pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that trialists used a robust method to carry out the allocation concealment process because of pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 35 (10%) placebo, 124 (18%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 132 centres in USA and Canada Participants: 825 (15 mg cilomilast: 418, placebo: 407) Baseline characteristics: mean age 64.4 years placebo and 64.5 years cilomilast, 62% male placebo and 56% male cilomilast, > 50% predicted FEV₁ for both groups Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, post‐salbutamol reversibility ≤ 15% or 200 mL, post‐salbutamol FEV₁ ≥ 1.0 L and between 30% and 70% predicted Exclusion criteria: not stated Total numbers of participant withdrawals: 143 (34%) and 96 (24%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcomes: change from baseline to endpoint in trough pre‐bronchodilator FEV₁; change in total SGRQ score averaged over 24 weeks Secondary outcomes: post‐exercise breathlessness; clinic trough FVC; time to first level 2 or level 3 COPD exacerbation | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | The randomisation code was provided by RAMOS (registration and medication ordering system) |
| Allocation concealment (selection bias) | Low risk | No further information on allocation concealment method |
| Blinding of participants and personnel (performance bias) | Low risk | Patients and personnel did not know which treatment had been allocated |
| Blinding of outcome assessment (detection bias) | Low risk | Outcome assessors did not know which treatment had been allocated |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 96 (24%) placebo, 143 (34%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 137 centres from 18 countries Participants: 907 (15 mg cilomilast: 455, placebo: 452) Baseline characteristics: mean age 63.3 years placebo and 64.6 years cilomilast, 73% male placebo and 78% male cilomilast, 42% current smokers Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, poor reversibility of airway obstruction defined by ≤ 10% predicted normal or ≤ 200 mL (or both) increase in FEV₁ after administration of salbutamol 400 µg via MDI at screening, post‐salbutamol FEV₁ between 30% and 70% predicted normal at screening Exclusion criteria: not stated Total numbers of participant withdrawals: 167 (37%) and 121 (27%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcomes: mean change from baseline in trough pre‐bronchodilator FEV₁ averaged over 52 weeks; incidence rate of level 2 (moderate) and level 3 (severe) COPD exacerbations during treatment period Secondary outcomes: time to first level 2 or level 3 COPD exacerbation; quality of life determined by SGRQ | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Randomisation code was generated via the pharmaceutical company's coding memo system in blocks |
| Allocation concealment (selection bias) | Low risk | No further information on allocation concealment method |
| Blinding of participants and personnel (performance bias) | Low risk | Patients and investigator were blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 121 (27%) placebo, 167 (37%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: not stated | |
| Participants | Setting: 42 centres in the USA Participants: 306 (15 mg cilomilast: 203, placebo: 103) Baseline characteristics: mean age 64.3 years placebo and 65.0 years cilomilast, 64% male placebo and 70% male cilomilast Inclusion criteria: pre‐albuterol FEV₁/FVC ≤ 0.7, post‐albuterol FEV₁ between 30% and 70% predicted Exclusion criteria: not stated Total numbers of participant withdrawals: 61 (30%) and 14 (14%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcome: no primary efficacy or safety analyses defined; descriptive statistics of change from baseline in minimum and maximum heart rate via 24‐hour Holter monitoring reported Secondary outcome: no secondary efficacy or safety analyses defined | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | The randomisation sequence was generated by the pharmaceutical company's biometrics unit |
| Allocation concealment (selection bias) | Unclear risk | No further information on allocation concealment method |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 14 (14%) placebo, 61 (30%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 18 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 34 centres in the USA, Canada, and South America Participants: 199 (15 mg cilomilast: 97, placebo: 102) Baseline characteristics: mean age 64.7 years placebo and 63.7 years cilomilast, 76% male placebo and 69% male cilomilast Inclusion criteria: age ≥ 40 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, baseline FEV₁ < 70% predicted normal, moderate to severe chronic dyspnoea defined by BDI focal score ≤ 7, evidence of hyperinflation defined by RFRC ≥ 140% predicted, exercise limitation defined as peak symptom limited VO₂ < 75% Exclusion criteria: not stated Total numbers of participant withdrawals: 24 (25%) and 13 (13%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcome: change from baseline at endpoint in RFRC Secondary outcomes: change from baseline at endpoint in IC during exercise; exertional dyspnoea as measured by the modified Borg Scale | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Patients were randomised via a call to the sponsor's medication ordering system, during which the patient's subject number was confirmed and the patient was provided with a 6‐digit container number for identification of the initial bottle of double‐blind medication |
| Allocation concealment (selection bias) | Unclear risk | No further information on allocation concealment method. |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double blind. Cilomilast and matched placebo tablets were identical in appearance, and only the double‐blind medication included the container number |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 13 (13%) placebo, 24 (25%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 13 weeks Analysis was done on the per‐protocol population | |
| Participants | Setting: 27 centres in Australia, Canada, Finland, Ireland, Lithuania, Norway, Romania, Slovakia, Slovenia, South Africa, Sweden, and the UK Participants: 127 (15 mg cilomilast: 65, placebo: 62) Baseline characteristics: mean age 63.4 years placebo and 61.4 years cilomilast, 76% male placebo and 72% male cilomilast Inclusion criteria: aged 40 to 80 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, post‐bronchodilator FEV₁ between 40% and 80% predicted normal, poor reversibility of ≤ 10% or 200 mL increase in FEV₁ Exclusion criteria: not stated Total numbers of participant withdrawals: 8 (12%) and 6 (10%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcomes: change from baseline at endpoint in CD68+ (macrophages) and CD8+ (cytotoxic T lymphocytes) per unit area of tissue Secondary outcomes: change from baseline in numbers of subepithelial cells per unit area in biopsy for neutrophil elastase‐positive (ne+) cells, CD4+, IL‐8 mRNA‐positive cells, TNF‐alpha mRNA‐positive cells | |
| Notes | Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | A central randomisation schedule that was balanced at site level. An interactive voice response system was used to generate a random number to assign eligible participants |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind. Participants and personnel were blind to which treatment they were assigned to. Cilomilast and matched placebo tablets were identical in appearance, and only the double‐blind medication included the container number |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants withdrawn 6 (10%) placebo, 8 (12%) cilomilast |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website only |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 6 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 60 centres in Austria, France, Germany, the Netherlands, and the UK Participants: 424 (5 mg cilomilast: 109, 10 mg cilomilast: 102, 15 mg cilomilast: 107, placebo: 106) Baseline characteristics: mean age 62 to 63 years, 75% to 78% male, mean FEV₁ % predicted 46.8%, mean smoking history 36 to 43 (SD 22.4) pack‐years Inclusion criteria: FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years Exclusion criteria: asthma, poorly controlled COPD needing hospital visit 6 weeks before study, recent COPD exacerbations, recent corticosteroid use Total numbers of participant withdrawals: 18 (17%) and 17 (16%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 2 weeks, single‐blind, placebo tablets to assess compliance
Concomitant medication
| |
| Outcomes | Primary outcomes: lung function: change in FEV₁; SGRQ Secondary outcomes: peak expiratory flow and FVC; first dose effect of active treatment on FEV₁ | |
| Notes | Post‐bronchodilator results not given, so pre‐bronchodilator values used in analysis. Funded by GlaxoSmithKline | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Randomised. Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded. Participants were not aware of which treatment they were receiving because cilomilast and matched placebo tablets were identical in appearance |
| Blinding of outcome assessment (detection bias) | Unclear risk | The trial was double‐blinded, but it is unclear who assessed the outcomes |
| Incomplete outcome data (attrition bias) | Low risk | "14 patients (13%) taking cilomilast 15 mg had adverse events leading to patient withdrawal, as did 12 each in the 5 and 10 mg groups (11 and 12%, respectively) and eight (8%) in the placebo group" |
| Selective reporting (reporting bias) | Unclear risk | Unclear whether outcomes were assessed as planned; it was not possible to find the trial in the GSK registry |
| Other bias | Low risk | 102 (24%) participants had been taking long‐acting β₂‐agonists (e.g. salmeterol, formoterol). 331 (78%) individuals had taken other medications for their COPD, the most common being inhaled steroids; 229 (54%) took beclomethasone, budesonide, or fluticasone |
| Study characteristics | ||
| Methods | 14 double‐blind and placebo‐controlled studies (Roflumilast FK1 101; Roflumilast FK1 103; Roflumilast IN‐108; Roflumilast M2‐107; Roflumilast M2‐110; Roflumilast M2‐111; Roflumilast M2‐112; Roflumilast M2‐118; Roflumilast M2‐119; Roflumilast M2‐121; Roflumilast M2‐124; Roflumilast M2‐125; Roflumilast M2‐127; Roflumilast M2‐128) | |
| Participants | See individual studies | |
| Interventions | Roflumilast 500 µg once daily Roflumilast 250 µg once daily Placebo once daily | |
| Outcomes | ||
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Randomised |
| Allocation concealment (selection bias) | Unclear risk | See individual studies |
| Blinding of participants and personnel (performance bias) | Low risk | The trials were double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Unclear risk | See individual studies |
| Selective reporting (reporting bias) | Unclear risk | See individual studies |
| Other bias | Unclear risk | See individual studies |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: not stated | |
| Participants | Setting: 1 pulmonary medicine ward in India Participants: 61 (intervention: 31; control: 30) Baseline characteristics: mean age 58 years, mean FEV₁ predicted 0.93, current smokers 33% Inclusion criteria: Indian ethnicity, age ≥ 40 years with moderate to severe COPD, current or past smokers, other co‐existing conditions Exclusion criteria: bronchial asthma, other lung diseases, lower respiratory tract infection, pregnant or breastfeeding Total numbers of participant withdrawals: not stated; assumed 1 person was not included in the analysis | |
| Interventions | Run‐in: not stated
Concomitant medication
| |
| Outcomes | Primary outcomes: lung function (FEV₁ and FVC); change in mean FEV₁ after treatment Secondary outcomes: adverse events in the roflumilast treatment group | |
| Notes | Funding not stated | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | High risk | Reported as a "randomised control study". No further information about randomisation method. Also, groups are not balanced with regards to baseline characteristics. For example, the placebo group includes a high percentage of patients with diabetes |
| Allocation concealment (selection bias) | Unclear risk | Allocation concealment method was not reported |
| Blinding of participants and personnel (performance bias) | High risk | Assumed that there was no blinding of participants or personnel |
| Blinding of outcome assessment (detection bias) | High risk | Assumed that there was no blinding of the outcome assessor |
| Incomplete outcome data (attrition bias) | Low risk | Only 1 participant in the roflumilast group was lost; however, reasons for attrition were not reported |
| Selective reporting (reporting bias) | High risk | Outcomes not reported in the methods, so unclear whether outcomes reported were what they intended to assess. Adverse event data were not reported for the control group, so it is unclear whether there were no events in this group. Outcome data for FEV₁ were not clear, as no units were reported. If it is assumed that trial authors reported litres, then those in the intervention group improved by 660 mL, which is large in COPD terms, as it indicates 28% improvement, yet in the discussion, trial authors mention that it is similar to the 60 mL reported in the Fabbri study |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: single hospital in Yan'an, China Participants: 120 (roflumilast 500 µg: 60, placebo: 60) Baseline characteristics: COPD stage II to IV according to GOLD criteria, mean age 65 years, FEV₁ % predicted < 50%, 72% male, smoking history 37 pack‐years, 66% current smokers Inclusion criteria: aged ≥ 40 years, post‐bronchodilator FEV₁ < 50% predicted, FEV1:FVC ratio 70%, post‐bronchodilator FEV₁ with 30% to 80% predicted, COPD history > 12 months, no medication change for past 3 months Exclusion criteria: asthma, other lung disease, systemic glucocorticosteroids, SABA 1 month before study, severe mental disorder, pregnant or breastfeeding Total numbers of participant withdrawals: 5 (8%) and 6 (10%) from treatment and control groups, respectively | |
| Interventions | Run‐in: not stated
Concomitant medication
23% of participants were on home oxygen therapy | |
| Outcomes | Primary outcome: change from baseline in lung function (FEV₁, FVC, and FEF25%-75%) Secondary outcomes: quality of life (SGRQ); adverse events | |
| Notes | Funding not stated | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Randomisation was achieved via a computerised number programme generated by a statistician who was blinded to treatment allocation |
| Allocation concealment (selection bias) | Unclear risk | Unclear how the allocation sequence was concealed from patients (i.e. no mention of concealed envelopes or any other method) |
| Blinding of participants and personnel (performance bias) | Low risk | Participants were reported to be blinded to treatment allocation; investigators, data analysts, and outcome assessors were blinded to treatment allocation |
| Blinding of outcome assessment (detection bias) | Low risk | Participants were reported to be blinded to treatment allocation; investigators, data analysts, and outcome assessors were blinded to treatment allocation |
| Incomplete outcome data (attrition bias) | Low risk | Overall, the number of withdrawals was 9%; however there was 1 more withdrawal in the placebo group compared to the intervention group |
| Selective reporting (reporting bias) | High risk | Reporting of outcome data for lung function and SGRQ is unclear. Data provided consist of mean difference and 95% CI when standard errors should be provided. A check of the data revealed discrepancies in the numbers. Data for adverse events were also unclear. There was no reference to a protocol, so we do not know whether outcomes were reported as planned. The paper includes some confusing statements about follow‐up at 12 weeks vs 12 months ‐ probably 12 months ‐ but then follow‐up for another 3 months. Abstract states that adverse events were increased, but this is not the same as the data in the paper |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, triple‐blind, placebo‐controlled trial Trial duration: 104 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 12 specialist centres across the USA Participants: 84 (tetomilast 25 µg: 51, placebo: 33) Baseline characteristics: mean age 58 years, 51% male Inclusion criteria: aged 40 to 75 years, rating ≥ 1 on Goddart scale for emphysema, FEV₁:FVC > 70% predicted, ≥ 1 COPD exacerbation in the past 12 months Exclusion criteria: asthma, active tuberculosis/bronchiectasis, respiratory tract infection in past month before screening, cancer in past 5 years, cardiovascular/endocrine blood/nervious system disorder, uncontrolled COPD exacerbation (level 2 or 3), recent systemic ICS or immunosuppressant, anticoagulant | |
| Interventions | Run‐in: 25 µg dose tetomilast for 2 weeks Intervention: 50 µg once daily Comparator: placebo once daily Concomitant medication
| |
| Outcomes | Primary outcomes: change in FEV₁; rate of change in 20th percentile of lung voxels Secondary outcomes: change in trough FEV₁; density mask score based on lung density voxels; rate of change in 20th percentile of lung density voxels expressed in HU units for whole lung; rate of change in emphysema (observed); change in cumulative frequency of HU; change in computed tomography (derived lung volumes); change in trough RV/TLC; change in trough inspiratory capacity; change in trough functional residual capacity; change in carbon monoxide diffusion capacity; changes in mean specific airway resistance and specific conductance; change in 7‐day average total symptom score for dyspnoea, cough, and sputum; change in 7‐day mean number of actuations of rescue medications; percentage of participants with COPD exacerbations by group; percentage of participants experiencing a COPD exacerbation Safety outcomes: adverse events; changes in laboratory parameters, blood pressure, heart rate, physical examination findings, body weight, and BMI | |
| Notes | Clinicaltrials.gov identifier: NCT00874497 Funded by Otsuka Pharmaceutical Development & Commercialization, Inc. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Unclear risk | The trial was double‐blind; participants, care providers, and outcome assessors were blinded |
| Blinding of outcome assessment (detection bias) | Low risk | The trial was double‐blind; participants, care providers, and outcome assessors were blinded |
| Incomplete outcome data (attrition bias) | High risk | 54% in each group did not complete treatment and discontinued the trial early, but there were 5 more adverse events in the tetomilast group vs the placebo group. 4 more people in the intervention arm discontinued because of early termination of the trial. Other factors included loss to follow‐up, withdrawal by either participant, or physician decision. Different numbers were used in different analyses |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Trial information was reported on the GSK website |
| Other bias | Unclear risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 43 centres in mainland China, Hong Kong, and Singapore Participants: 626 (500 µg roflumilast: 313, placebo: 313) Baseline characteristics: mean age 64 years, 91% male, mean FEV₁ % predicted 36%, mean smoking history 37.2 pack‐years for roflumilast and 37.5 pack‐years for placebo or current smokers (24% and 29%, respectively) Inclusion criteria: Chinese, Malaysian, or Indian ethnicity, age 40 to 80 years with severe or very severe COPD, FEV₁/FVC ≤ 0.7, post‐bronchodilator FEV₁ ≤ 50%. Current smokers or ex‐smokers with smoking history > 10 pack‐years or current smokers; 12‐month history of COPD and ≥ 14 puffs of rescue medication during the week before randomisation Exclusion criteria: primary bronchiectasis, cystic fibrosis, bronchiolitis, lung resection, lung cancer, interstitial lung disease, active TB, lower respiratory tract infection, diagnosis of asthma at < 40 years of age, α₁‐antitrypsin deficiency Total numbers of participant withdrawals: 67 (21.4%) and 50 (16%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind. Placebo tablets to assess suitability
Concomitant medication Participants were allowed to continue taking fixed combinations of ICS plus LABA or LAMA monotherapy (e.g. tiotropium) if taken at a stable dose for at least 6 months before the run‐in period. SAMAs (e.g. ipratropium) were allowed at a constant daily dose as concomitant medication if taken on a regular basis for at least 4 weeks before study inclusion. All other COPD treatments were not allowed | |
| Outcomes | Primary outcomes: lung function; change in pre‐bronchodilator FEV₁ Secondary outcomes: changes in post‐bronchodilator FEV₁, FVC, incidence rates of COPD exacerbations, time to first COPD exacerbation, transition dyspnoea index, proportions of participants experiencing a COPD exacerbation, adverse events, changes in body weight, laboratory values, vital signs, and physical examination findings | |
| Notes | Clinicaltrials.gov identifier: NCT01313494 Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Investigators used an automated, interactive voice‐response system to randomly assign participants. The sponsor generated a list of participant numbers using a pseudo‐random number generator |
| Allocation concealment (selection bias) | Low risk | The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded, and tablets were identical in appearance |
| Blinding of outcome assessment (detection bias) | Low risk | The trial was double‐blinded. The investigator or anyone at the study site was prevented from knowing the treatment allocation |
| Incomplete outcome data (attrition bias) | Low risk | Total numbers of participants that discontinued 50 (16%) placebo, 67 (21.4%) roflumilast |
| Selective reporting (reporting bias) | Unclear risk | Outcomes were reported as planned, and the trial was registered at the NCT website |
| Other bias | Low risk | LAMA: 17.9% for placebo; 20.4% for roflumilast SAMA: 18.2% for placebo; 17.3% for roflumilast ICS/LABA: 55.9% for placebo; 59.7% for roflumilast No information available. SABA allowed |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 16 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 18 centres in Denmark, Germany, Poland, Sweden, and United Kingdom Participants: 158 (500 µg roflumilast: 79; placebo: 79) Baseline characteristics: mean age 63 years, 77% male, mean FEV₁ predicted 60%, mean smoking history longer than 20 years or current smokers 54% Inclusion criteria: post‐bronchodilator 30% ≤ FEV₁ ≤ 8% predicted, post‐bronchodilator FEV₁/FVC ratio ≤ 70%, current/former smoker history ≥ 20 pack‐years; aged 40 to 80 years with COPD diagnosed at least 12 months before study inclusion, chronic productive cough for 3 months in each of previous 2 years Exclusion criteria: recent COPD exacerbation, ongoing upper or lower respiratory tract infection, asthma (with or without other lung disease), alpha‐1‐antitrypsin deficiency, bleeding disorder, concomitant glucocorticosteroids, theophylline, lipoxygenase inhibitors, antiplatelet therapy, leukotriene antagonists Total numbers of participant withdrawals: 3 (4%) and 6 (8%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 6 weeks, single‐blind with placebo to assess compliance. ICS and other non‐allowed drugs stopped
Concomitant medication
| |
| Outcomes | Primary outcome: change in numbers of CD8 inflammatory cells in bronchial biopsy samples Secondary outcomes: change in numbers of inflammatory cells measured in submucosa, bronchial epithelium, induced sputum; blood FEV₁, FVC, and FEV₁/FVC ratio Safety outcomes: adverse events; changes in laboratory parameters, blood pressure, heart rate, physical examination findings, body weight, and BMI | |
| Notes | Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | A computerised central randomisation system stratified by concomitant use of LABA was used |
| Allocation concealment (selection bias) | Low risk | Both roflumilast and placebo were given as identical yellow, triangular tablets; blinding was maintained via an interactive voice‐response system and an interactive web‐response system |
| Blinding of participants and personnel (performance bias) | Low risk | Participants and investigators were blinded to group assignment |
| Blinding of outcome assessment (detection bias) | Low risk | Outcome assessors were blinded |
| Incomplete outcome data (attrition bias) | Low risk | Overall the total % of participants who discontinued was 5.7%. In the roflumilast group, 3.7% discontinued compared to 7.59% in the placebo group. The number of adverse events was the same in each group; the numbers not completing were 6 in the placebo group and 3 in the roflumilast group |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. Some of the outcome data were reported in the NCT and EU clinical trials registers |
| Other bias | Low risk | No information on baseline anticholinergic, beta₂‐agonist, or corticosteroid use |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 203 centres in 21 countries (see online appendix) Participants: 1945 (500 µg roflumilast: 969; placebo: 966) Baseline characteristics: mean age 65 years, 75% male, mean FEV₁ predicted 35%, mean smoking history 48 pack‐years for roflumilast and 48 pack‐years for placebo or current smokers (42% and 45%, respectively) Inclusion criteria: ≥ 40 years of age with a smoking history ≥ 20 pack‐years and a diagnosis of COPD with severe airflow limitation (confirmed by post‐bronchodilator FEV₁/FVC ratio < 0.70 and post‐bronchodilator FEV₁ ≤ 50% predicted), symptoms of chronic bronchitis, history of ≥ 2 exacerbations in the previous year. Participants must have been taking an ICS–LABA combination for 12 months before the study and a constant dose of an ICS–LABA fixed combination for at least 3 months before enrolment, with placebo tablet compliance of 80% to 125% during the 4‐week baseline observation period, and with a total cough and sputum score ≥ 14 (in which the score was a sum of daily scores on 4‐point scales for cough and sputum) recorded in a daily diary during the week preceding the randomisation visit Exclusion criteria: COPD exacerbation that was ongoing during the baseline period, diagnosis of asthma or other major lung disease Total numbers of participant withdrawals: 269 (28%) and 192 (20%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks, single‐blind. Placebo tablets to assess suitability
Concomitant medication All participants used a fixed‐dose ICS–LABA combination during baseline and treatment periods. If a participant had an exacerbation that needed additional treatment during the study, the investigator could give up to 40 mg prednisolone, administered systemically, per day for 7 to 14 days. In the case of purulent sputum or suspected bacterial infection, additional antibiotic therapy was allowed. Use of the following treatments was not allowed: oral and parenteral glucocorticosteroids (except to treat acute exacerbations), LABA or ICS monotherapy, SAMA, and any SABA (with the exception of salbutamol) or oral β₂‐agonists. Participants already taking inhaled tiotropium bromide (a LAMA) were allowed to continue this treatment | |
| Outcomes | Primary outcomes: rate of moderate or severe COPD exacerbations per patient per year Secondary outcomes: change from baseline in post‐bronchodilator FEV₁, rate of severe COPD exacerbations per patient per year, rate of COPD exacerbations per patient per year (all categories), percentage of participants experiencing ≥ 1 COPD exacerbation, time to first COPD exacerbation (all categories), time to second moderate or severe COPD exacerbation, time to third moderate or severe COPD exacerbation, number of participants needed to treat to avoid 1 moderate or severe COPD exacerbation derived from exacerbation per patient per year, number of moderate or severe COPD exacerbation days, duration of moderate or severe COPD exacerbations per participant, change from baseline in post‐bronchodilator FVC, change from baseline in post‐bronchodilator FEF (25% to 75% vital capacity), change from baseline in post‐bronchodilator FEV₆, change from baseline in post‐bronchodilator FEV₁/FVC, change from baseline in rescue medication use, change from baseline in COPD symptom score, percentage symptom‐free days, percentage rescue medication‐free days, change from baseline in CAT total score, percentage participants with CAT score improvement, time to mortality (all‐cause and COPD exacerbation‐related), time to withdrawal (all‐cause and COPD exacerbation‐related), percentage of participants with major adverse cardiovascular event, time to first major adverse cardiovascular event, percentage of participants with hospitalisation (all‐cause), time to first hospitalisation, time to withdrawal due to adverse event, percentage of participants experiencing ≥ 1 adverse event (treatment‐related), change from baseline in body weight, change from baseline in BMI | |
| Notes | Clinicaltrials.gov identifier: NCT01329029 Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Enrolled participants were randomly assigned in a 1:1 ratio, with a block size of 4, by a computerised central randomisation system, the Interactive Voice Response System–Interactive Web Response System (PPD Global Limited, Cambridge, UK) |
| Allocation concealment (selection bias) | Low risk | All parties involved in the study were masked to treatment assignment |
| Blinding of participants and personnel (performance bias) | Low risk | Roflumilast and placebo were supplied as identical yellow triangular tablets in wallet cards containing 40 tablets; all parties involved in the study were masked to treatment assignment |
| Blinding of outcome assessment (detection bias) | Low risk | All parties involved in the study were masked to treatment assignment |
| Incomplete outcome data (attrition bias) | Low risk | 269 participants (28%) in the roflumilast group discontinued from the study and 192 (20%) from the placebo group discontinued |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. The trial was registered at the NCT website |
| Other bias | Unclear risk | LAMA: 69% for placebo; 70% for roflumilast. No group differences stated; however 1900 (98%) of 1935 participants were using a combination of ICS–LABA according to the protocol |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: single centre in USA Participants: 27 (500 µg roflumilast: 11, placebo: 16) Baseline characteristics: mean age 62 years, 64% male, mean FEV₁ % predicted 45%, mean smoking history 44 pack‐years for roflumilast and 47 pack‐years for placebo or current smokers (63% and 55%, respectively) Inclusion criteria: > 40 years old with a diagnosis of moderate to severe COPD as defined by GOLD criteria, current or former cigarette smokers with more than 10 pack‐years of total consumption, chronic bronchitis defined by chronic cough and sputum production lasting ≥ 3 months for 2 consecutive years Exclusion criteria: asthma as defined by ATS/ERS guidelines, clinically significant bronchiectasis, known sensitivity to roflumilast, use of other methylxanthines (specifically theophylline) within 1 month of screening, changes to maintenance COPD therapy within 1 month of screening Total numbers of participant withdrawals: 1 (9%) and 1 (6%) from treatment and control groups, respectively | |
| Interventions | Run‐in: no run‐in
Concomitant medication Allowed, except for theophylline. For roflumilast and placebo groups, respectively: LAMA was used by 8 (50%) and 6 (55%); ICS or LABA/ICS was used by 10 (63%) and 6 (55%) | |
| Outcomes | Primary outcome: change in induced sputum AcPGP at 12 weeks post randomisation in an intention‐to‐treat analysis Secondary outcomes: changes in plasma AcPGP, sputum neutrophil counts, additional sputum biomarkers, 6MWT, Breathlessness Cough and Sputum Scale, SGRQ scores, changes in post‐bronchodilator FEV₁ at 12‐week visit | |
| Notes | Clinicaltrials.gov identifier NCT01572948. Funded by Forest Laboratories Inc., University of Alabama at Birmingham | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | The trial was reported as block randomised with a 1:1 allocation, stratified by current smoking status and ICS use, but no information about the sequence generation was provided |
| Allocation concealment (selection bias) | Low risk | Sealed envelopes were used to conceal allocation |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was triple‐blinded (participant, care provider, and investigator) |
| Blinding of outcome assessment (detection bias) | Low risk | The trial was triple‐blinded (participant, care provider, and investigator) |
| Incomplete outcome data (attrition bias) | Low risk | At follow‐up, only 1 participant in each group was lost because of refusal to attend the final visit or inability to be contacted for the final visit |
| Selective reporting (reporting bias) | Low risk | Outomes were reported as planned; the trial was registered at clinicaltrials.gov |
| Other bias | Unclear risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 26 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: not stated Participants: 516 (roflumilast 250 µg: 175, roflumilast 500 µg: 169, placebo: 172) Baseline characteristics: median age 61 to 62 years, 72% male, mean 38 to 63 pack‐years, 53% current smokers Inclusion criteria: aged 40 to 75 years, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, reversibility < 12% or 200 mL, post‐bronchodilator FEV₁ 35% to 75% predicted Exclusion criteria: not stated Total numbers of participant withdrawals: not stated | |
| Interventions | Run‐in: 2 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: post‐bronchodilator FEV₁ and FEF between 25% and 75% of vital capacity Secondary outcomes: numbers of moderate or severe COPD exacerbations that required treatment with OCS | |
| Notes | Funding not reported | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Reported as randomised. No further information |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | Reported as double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | No available information |
| Incomplete outcome data (attrition bias) | Unclear risk | No available information |
| Selective reporting (reporting bias) | Unclear risk | Unpublished study; no available information |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: not stated Participants: 518 (roflumilast 500 µg: 200, placebo: 186) Baseline characteristics: mean age 60 years, 75% male, 62% current smokers, average 35 pack‐years Inclusion criteria: aged 40 to 75 years, FEV₁/FVC ≤ 0.7, post‐bronchodilator FEV₁ 35% to 75% predicted, FEV₁ reversibility ≤ 12% and ≤ 200 mL, pre‐bronchodilator FEV₁/FVC ≤ 70% Exclusion criteria: not stated Total numbers of participant withdrawals; not stated | |
| Interventions | Run‐in: 2 weeks with placebo
Concomitant medication
Used alongside short‐acting β₂‐agonists (available to all) | |
| Outcomes | Primary outcomes: results for 12/24‐week post‐bronchodilator FEV₁ | |
| Notes | Funding not reported | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Method of randomisation not stated |
| Allocation concealment (selection bias) | Unclear risk | No information available |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | No information available |
| Incomplete outcome data (attrition bias) | Unclear risk | No information available |
| Selective reporting (reporting bias) | Unclear risk | No information available |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 26 months Intention‐to‐treat analysis: stated Responder analysis for the most part | |
| Participants | Setting: 2 centres Participants: 41 (500 µg roflumilast: 30, placebo: 11) Baseline characteristics: not stated Inclusion criteria: not stated Exclusion criteria: not stated Total numbers of participant withdrawals: not stated | |
| Interventions | Run‐in: not stated
Concomitant medication: not stated | |
| Outcomes | Primary outcomes: post bronchodilation: spirometry, body plethysmography, 6MWT, patient‐reported outcomes Secondary outcomes: not stated | |
| Notes | Clinicaltrials.gov identifier NCT01480661 Funded by Takeda | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Reported as block randomised in a 3:1 ratio of roflumilast to placebo, respectively; no further information about sequence generation |
| Allocation concealment (selection bias) | Unclear risk | No information available |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was reported as triple‐blind (participant, care provider, and investigator) |
| Blinding of outcome assessment (detection bias) | Low risk | The trial was reported as triple‐blind (participant, care provider, and investigator) |
| Incomplete outcome data (attrition bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Selective reporting (reporting bias) | Low risk | Outcomes reported as intended; trial registered at clinicaltrials.gov |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: not stated | |
| Participants | Setting: 5 centres in India Participants: 118 recruited (roflumilast 500 µg: 47, roflumilast 200 µg: 46, placebo: 25) Baseline characteristics: mean age 60 years, 98% male, 41% current smokers, post‐bronchodilator FEV₁ 57% to 61%, average 25 pack‐years Inclusion criteria: not stated Exclusion criteria: not stated Total numbers of participant withdrawals: roflumilast 500 µg: 13 (28%); roflumilast 200 µg: 7 (15%); control 10 (40%) | |
| Interventions | Run‐in: none
Concomitant medication
| |
| Outcomes | Primary outcome: post‐bronchodilator FEV₁ | |
| Notes | Funded by Forest Laboratories | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Reported as randomised. No further information available |
| Allocation concealment (selection bias) | Unclear risk | No information available |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was reported as double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Data as above |
| Selective reporting (reporting bias) | Unclear risk | No information available |
| Other bias | Unclear risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: not stated | |
| Participants | Setting: Japan Participants: 600 (roflumilast 250 µg: 205, roflumilast 500 µg: 204, placebo: 191) Baseline characteristics: mean age 70 years, 96% male, post‐bronchodilator FEV₁ not stated, average 56 pack‐years, 37% current smokers Inclusion criteria: not stated Exclusion criteria: not stated Total numbers of participant withdrawals: not stated | |
| Interventions | Run‐in: single‐blind 4 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: pulmonary function (FEV₁ pre‐bronchodilator, FVC pre‐ and post‐bronchodilator, MMEF pre‐ and post‐bronchodilator) Secondary outcomes: number of COPD exacerbations, number of days to first COPD exacerbation, adverse events (all‐cause and drug‐related), serious adverse events (all‐cause and drug‐related) | |
| Notes | Funded by Mitsubishi‐Tanabe | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Reported as randomised; no further information about randomisation process |
| Allocation concealment (selection bias) | Unclear risk | No information available |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was reported as double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Unclear risk | Not described |
| Selective reporting (reporting bias) | High risk | No further information available about the trial. No trial registry information found |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 159 centres in Australia, Austria, Belgium, Canada, France, Germany, Hungary, Ireland, South Africa, Spain, and the UK Participants: 1411 (roflumilast 250 µg: 576, roflumilast 500 µg: 555, placebo: 280) Baseline characteristics: median age 64 years, 74% male, post‐bronchodilator FEV₁ 51% for both groups, average 42 pack‐years, 45% current smokers Inclusion criteria: aged ≥ 40 with history of COPD > 12 months, FEV₁ < 50% predicted, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, reversibility < 12% or 200 mL, mean post‐bronchodilator FEV₁ 30% to 80% predicted Exclusion criteria: asthma, lung cancer or bronchiectasis, long‐term oxygen treatment, recent exacerbation that required a course of systemic corticosteroids, emergency room treatment or hospital admission within 4 weeks before run‐in period Total numbers of participant withdrawals: 124 (22%) and 32 (11%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: post‐bronchodilator FEV₁; SGRQ total score Secondary outcomes: change from baseline in pre‐bronchodilator FEV₁; post‐bronchodilator FVC; post‐bronchodilator FEV in 6 seconds and FVC; FEF rate between 25% and 75% of vital capacity; number of moderate or severe COPD exacerbations | |
| Notes | Funded by ALTANA Pharma AG | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | The randomisation sequence was generated by the sponsor in a blind manner |
| Allocation concealment (selection bias) | Unclear risk | No information on allocation concealment method, but "no person involved in data analysis had knowledge of the randomisation sequence" |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind. Roflumilast and placebo tablets and packaging were identical, so neither participants nor study personnel were aware of either medication allocation |
| Blinding of outcome assessment (detection bias) | Low risk | "No person involved in data analysis had knowledge of the randomisation sequence" Roflumilast and placebo tablets and packaging were identical, so the investigator was not aware of either medication allocation |
| Incomplete outcome data (attrition bias) | Low risk | 100 participants discontinued from the roflumilast 250 µg group, 124 from the roflumilast 500 µg group, and 32 from the placebo group |
| Selective reporting (reporting bias) | High risk | There was inconsistency in the quoting of statistical errors. Within the text and in Table 2, data are quoted as "least squares means and SD"; however in Figures 2 and 3, SE bars are shown. It is more likely that results represented SE, not SD. Trial registration was not found |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: not stated | |
| Participants | Setting: 36 centres in Argentina, Canada, Columbia, Mexico, Peru, and the USA Participants: 909 participants randomised (roflumilast 500 µg: 449; placebo: 460) Baseline characteristics: aged between 55 and 74 years (mean age 64.2 years in the roflumilast group and 64.6 years in the placebo group), 88% participants were white, roflumilast group included 51% males and the placebo group 55% males Inclusion criteria: clinical diagnosis of COPD based on ATS criteria, post‐bronchodilator FEV₁/FVC ≤ 70%, post‐bronchodilator FEV₁ ≥ 30% and ≤ 80% predicted, post‐bronchodilator FEV₁ increase ≤ 12% or ≤ 200 mL compared to pre‐bronchodilator value, score grade ≥ 1 on the MRC Dyspnea Scale, currently stable COPD with no change in COPD treatment in the prior 4 weeks Exclusion criteria: clinical diagnosis of asthma, poorly controlled COPD, regular need for daily oxygen therapy Total numbers of participant withdrawals: roflumilast group: 15.4% withdrew due to adverse events, 10.5% withdrew consent, 2.9% withdrew due to lack of efficacy; placebo group: 7.6% withdrew due to adverse events, 8.5% withdrew consent, 3% withdrew due to protocol violation | |
| Interventions | Run‐in: 4‐week single‐blind period during which respiratory medication (including ICS, LABA, and long‐acting anticholinergics) was withdrawn
Concomitant medication
| |
| Outcomes | Primary outcomes: pulmonary function tests (FEV₁, FVC, FEF, PEF, FIV₁, FVCin) | |
| Notes | ClinicalTrials.gov Identifier: NCT00062582. Funded by ALTANA Pharma AG | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Reported as randomised. Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Low risk | Assumed that the allocation concealment method was adequate due to pharmaceutical company sponsorship |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was reported as double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Unclear risk | Roflumilast group: 15.4% withdrew due to adverse events, 10.5% withdrew consent, 2.9% withdrew due to lack of efficacy; placebo group: 7.6% withdrew due to adverse events, 8.5% withdrew consent, 3% withdrew due to protocol violation |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. The trial was registered at clinicaltrials.gov |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: M2‐111 was conducted at 188 centres in 6 countries, and M2‐112 at 159 centres in 14 countries Inclusion criteria: aged ≥ 40 years, post‐bronchodilator FEV₁ < 50% predicted, reversibility < 15%, mean post‐bronchodilator FEV₁ 42%, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, 40% current smokers, 60% ex‐smokers, average 46 to 48 pack‐years Exclusion criteria: history of asthma, lung cancer, or bronchiectasis; need for long‐term oxygen therapy; known α₁‐antitrypsin deficiency, clinically significant cardiopulmonary comorbidity Total numbers of participant withdrawals: data combined with M2‐112 showing 433 (33%) and 348 (26%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
Used alongside corticosteroids, anticholinergics, and rescue short‐acting β₂‐agonists 54% overall (available to all) | |
| Outcomes | Primary outcomes: change from baseline to endpoint in post‐bronchodilator FEV₁; number of moderate or severe exacerbations per patient per year Secondary outcomes: change from baseline in SGRQ total score; change from baseline in pre‐bronchial FEV₁, post‐bronchodilator FEV in 6 seconds and in FVC; FEF rate between 25% and 75% vital capacity; number of moderate or severe COPD exacerbations requiring systemic corticosteroid treatment per patient per year | |
| Notes | NCT00076089/BY217/M2‐111. Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | The randomisation sequence was generated by a multiplicative congruent pseudo‐random numbers generator programme (programme RANDOM, based on Fishman and Moore) |
| Allocation concealment (selection bias) | Low risk | "Each study participant who qualified was assigned a number in sequential order. Code labelling prevented the investigator and the patient from knowing which drug was administered" |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Data combined with M2‐112 |
| Selective reporting (reporting bias) | Low risk | Trial registered at clinicaltrials.gov; outcomes reported as planned. M2‐111 and M2‐112 data combined |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | As described in separate studies above and below | |
| Participants | ||
| Interventions | ||
| Outcomes | ||
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | See individual trials |
| Allocation concealment (selection bias) | Low risk | See individual trials |
| Blinding of participants and personnel (performance bias) | Low risk | See individual trials |
| Blinding of outcome assessment (detection bias) | Low risk | See individual trials |
| Incomplete outcome data (attrition bias) | Low risk | See individual trials |
| Selective reporting (reporting bias) | Low risk | See individual trials |
| Other bias | Low risk | |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 159 centres in 14 countries Participants: 1514 (roflumilast 500 µg: 761, placebo: 753) Baseline characteristics: severe COPD according to GOLD criteria grades III and IV, mean age 65 years, 75% male Inclusion criteria: aged ≥ 40 years, post‐bronchodilator FEV₁ < 50% predicted, reversibility < 15%, mean post‐bronchodilator FEV₁ 41%, FEV₁/FVC ≤ 0.7 with smoking history > 10 pack‐years, 37% current smokers, 63% ex‐smokers, average 44 pack‐years Exclusion criteria: history of asthma, lung cancer, or bronchiectasis; need for long‐term oxygen therapy; known α₁‐antitrypsin deficiency or clinically significant cardiopulmonary comorbidity Total numbers of participant withdrawals: 217 (29%) and 163 (22%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
Used alongside corticosteroids, anticholinergics, and rescue short‐acting β₂‐agonists 54% overall (available to all) | |
| Outcomes | Primary outcomes: change from baseline to endpoint in post‐bronchodilator FEV₁ and in the number of moderate or severe exacerbations per patient per year Secondary outcomes: change from baseline in SGRQ total score; change from baseline in pre‐bronchial FEV₁; post‐bronchodilator FEV in 6 seconds and FVC; FEF rate between 25% and 75% of vital capacity; number of moderate or severe COPD exacerbations requiring systemic corticosteroid treatment per patient per year | |
| Notes | NCT00430729/BY217/M2‐112. Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "The randomisation list was generated using a multiplicative congruent pseudo‐random number generator (program RANDOM, based on Fishman and Moore)" |
| Allocation concealment (selection bias) | Low risk | "Each study participant who qualified was assigned a number in sequential order. Code labelling prevented the investigator and the patient from knowing which drug was administered" |
| Blinding of participants and personnel (performance bias) | Low risk | The trial was double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | "Over 70% of patients completed the study. The reasons for withdrawal were similar between groups except for adverse events, which occurred more frequently with roflumilast" "Withdrawal due to COPD exacerbations was reported in 3.5 and 3.2% of patients in roflumilast and placebo groups, respectively" |
| Selective reporting (reporting bias) | Low risk | None |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 22 centres in 4 countries Participants: 250 (roflumilast 500 µg: 127, placebo: 123) Baseline characteristics: mean age 60 years, 73% (roflumilast) vs 84% (placebo) male, post‐bronchodilator FEV₁ 55% predicted, average 41 pack‐years, 53% current smokers Inclusion criteria: clinically stable patients ≥ 40 years of age with smoking history > 10 pack‐years and 12‐month history of COPD. Other inclusion criteria included post‐bronchodilator FEV₁ 30% to 80% predicted, FEV₁/forced vital capacity (FVC) < 0.7, and set plethysmographic FRC and peak oxygen uptake requirements Exclusion criteria: asthma or lung disease other than COPD, α₁‐antitrypsin deficiency, participation in pulmonary rehabilitation programme within 2 months, supplemental oxygen therapy, significant medical comorbidity that may influence exercise tolerance Total numbers of participant withdrawals: 16 (13%) and 12 (10%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 2‐ to 3‐week baseline period consisting of 2 familiarisation visits during which a symptom‐limited constant work rate cycle exercise test was performed at 75% of maximum incremental work rate. If a constant work rate endurance time was not produced within 2 minutes at both visits, a third visit was performed. If reproducibility was not achieved at the third visit, the patient was not randomised
Concomitant medication
| |
| Outcomes | Primary outcomes: activity‐related dyspnoea (TDI); spirometry and body plethysmography; symptom‐limited exercise tests | |
| Notes | Funded by Nycomed GmbH (Konstanz, Germany) | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Reported as randomised. Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | Trial reported as double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | Of 250 randomised participants, 16 from the roflumilast group and 12 from the placebo group discontinued prematurely |
| Selective reporting (reporting bias) | Unclear risk | Outcomes reported, but no trial protocol found on trial registry websites |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 32 centres in 5 countries Participants: 410 (roflumilast 500 µg: 203, placebo: 207) Baseline characteristics: mean age 68 years, 93% male, post‐bronchodilator FEV₁ 50.5% predicted, average 44 pack‐years, 69% current smokers Inclusion criteria: former or current smokers with pack‐year history ≥ 10, aged ≥ 40 years, post‐bronchodilator FEV₁/FVC ≤ 0.7, FEV₁ 30% to 80% predicted, clinically stable COPD within 4 weeks before baseline Exclusion criteria: history of asthma or other relevant lung disease, COPD exacerbation within 4 weeks before baseline, need for long‐term oxygen therapy, known α₁‐antitrypsin deficiency, clinically significant cardiopulmonary comorbidity Total numbers of participant withdrawals: 40 (20%) and 18 (9%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcome: mean change in post‐bronchodilator FEV₁ from baseline Secondary outcomes: mean change in pre‐bronchodilator FEV₁ from baseline; change in other lung function measures, time to COPD exacerbation; proportion of participants experiencing exacerbations; time to study withdrawal; adverse effects | |
| Notes | Clinicaltrials.gov identifier: NCT00242320; BY217/M2‐119. Funded by Nycomed GmbH, Konstanz, Germany | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Computer‐generated randomisation list used |
| Allocation concealment (selection bias) | Unclear risk | Not stated |
| Blinding of participants and personnel (performance bias) | Low risk | Double‐blinded |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | High risk | Of the 411 randomised participants, 41 from the roflumilast group and 18 from the placebo group discontinued during the treatment period (20% compared with 8%, respectively) |
| Selective reporting (reporting bias) | Low risk | Outcomes were reported as planned. The protocol was registered at clincaltrials.gov |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 12 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 16 centres in 6 countries Participants: 600 participants (full analysis set; roflumilast 500 µg: 301, placebo: 299) Baseline characteristics: median age 65 years, 74% male, FEV₁ 46% predicted, 48 mean pack‐years Inclusion criteria: history of COPD ≥ 12 months as defined by GOLD criteria, age ≥ 40 years, FEV₁/FVC ratio (post‐bronchodilator) ≤ 70%, FEV₁ (post‐bronchodilator) ≤ 65% predicted, FRC (post‐bronchodilator) ≤ 120% predicted Exclusion criteria: COPD exacerbation indicated by treatment with systemic glucocorticosteroids not stopped ≥ 4 weeks before baseline visit; non‐smoker, current smoker, or ex‐smoker (smoking cessation ≥ 1 year ago) with smoking history < 10 pack‐years; any concomitant disease that might interfere with study procedures or evaluation Total numbers of participant withdrawals: 32 participants withdrew due to COPD exacerbations | |
| Interventions | Run‐in: 4‐week single‐blind placebo tablet once daily in the morning and all disallowed concomitant medications withdrawn
Concomitant medication
| |
| Outcomes | Primary outcome: lung function parameters indicative of hyperinflation in people with COPD Secondary outcomes: mean change from randomisation to endpoint in additional pre‐ and post‐bronchodilator spirometric and lung volume parameters; measurement of quality of life parameters; dyspnoea | |
| Notes | ClinicalTrials.gov Identifier: NCT00108823; BY217/M2‐121. Funded by AstraZeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Reported as randomised. Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | Trial reported as double‐blind |
| Blinding of outcome assessment (detection bias) | Unclear risk | No available information |
| Incomplete outcome data (attrition bias) | Low risk | Of 600 randomised participants, 13 from the roflumilast group and 19 from the placebo group withdrew due to exacerbations |
| Selective reporting (reporting bias) | Unclear risk | A publication was not found for this trial; however, study results were obtained from the trial registry website |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 246 centres in 10 countries Participants: 1513 (roflumilast 500 µg: 760, placebo: 753) Baseline characteristics: mean age 64 years, 71% male, post‐bronchodilator FEV₁ 37.6% predicted, average 47 pack‐years, 48% current smokers Inclusion criteria: former or current smokers with ≥ 20 pack‐year history, aged ≥ 40 years, post‐bronchodilator FEV₁/FVC ≤ 0.7, chronic cough and sputum production, post‐bronchodilator FEV₁ < 50% predicted, ≥ 1 recorded COPD exacerbation requiring systemic glucocorticosteroids or treatment in hospital in previous year Exclusion criteria: available in the online web appendix (p 11) Total numbers of participant withdrawals: 264 (34%) and 234 (31%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: mean change in pre‐bronchodilator FEV₁; mean rate of COPD exacerbations requiring oral or parenteral glucocorticosteroids or requiring hospitalisation or leading to death (per patient per year) Secondary outcomes: mean change in post‐bronchodilator FEV₁; time to mortality for any reason; natural log‐transformed CRP (mg/L); mean TDI focal score | |
| Notes | Clinicaltrials.gov identifier: NCT00297102. Funded by AstraZeneca Adverse event data are pooled with numbers from study M2‐125, which followed an identical study design | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | A randomisation list was generated via a pseudo‐random numbers generator, and an automated interactive voice‐response system was used to randomly assign participants |
| Allocation concealment (selection bias) | Low risk | All individuals involved in the studies were unaware of treatment assignment |
| Blinding of participants and personnel (performance bias) | Low risk | "All individuals involved in the studies were unaware of treatment assignment. Tablets were identical in appearance. The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling" |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | 264 participants discontinued from the roflumilast group and 234 discontinued from the placebo group |
| Selective reporting (reporting bias) | Low risk | Outcomes reported as planned. Trial registered at clinicaltrials.gov |
| Other bias | High risk | 44% of participants in both roflumilast and placebo groups received corticosteroid pre‐treatment |
| Study characteristics | ||
| Methods | As described in separate studies above and below | |
| Participants | ||
| Interventions | ||
| Outcomes | ||
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | See individual studies |
| Allocation concealment (selection bias) | Low risk | See individual studies |
| Blinding of participants and personnel (performance bias) | Unclear risk | See individual studies |
| Blinding of outcome assessment (detection bias) | Unclear risk | See individual studies |
| Incomplete outcome data (attrition bias) | Unclear risk | See individual studies |
| Selective reporting (reporting bias) | Unclear risk | See individual studies |
| Other bias | Unclear risk | See individual studies |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 221 centres in 8 countries Participants: 1571 (roflumilast 500 µg: 773, placebo: 798) Baseline characteristics: mean age 64 years, 80% male, average 48 pack‐years, 35% current smokers Inclusion criteria: former or current smokers with pack‐year history ≥ 20 years, aged ≥ 40 years, post‐bronchodilator FEV₁/FVC ≤ 0.7, chronic cough and sputum production, post‐bronchodilator FEV₁ < 50% predicted, ≥ 1 recorded COPD exacerbation requiring systemic glucocorticosteroids or treatment in hospital in previous year Exclusion criteria: available in the online web appendix (p 11) Total numbers of participant withdrawals: 246 (32%) and 248 (31%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo
Concomitant medication
| |
| Outcomes | Primary outcomes: mean change in pre‐bronchodilator FEV₁; mean rate of COPD exacerbations (moderate or severe) requiring oral or parenteral glucocorticosteroids or requiring hospitalisation or leading to death (per patient per year) Secondary outcomes: mean change in post‐bronchodilator FEV₁; time to mortality for any reason; natural log‐transformed CRP (mg/L); mean TDI focal score during treatment period | |
| Notes | Clinicaltrials.gov identifier: NCT00297115; BY217/M2‐125. Funded by AstraZeneca Adverse event data are pooled with numbers from study M2‐124, which followed an identical study design | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | A randomisation list was generated via a pseudo‐random numbers generator, and an automated interactive voice‐response system was used to randomly assign participants |
| Allocation concealment (selection bias) | Low risk | All individuals involved in the studies were unaware of treatment assignment |
| Blinding of participants and personnel (performance bias) | Unclear risk | "All individuals involved in the studies were unaware of treatment assignment. Tablets were identical in appearance. The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling" |
| Blinding of outcome assessment (detection bias) | Unclear risk | Assumed that this would be low risk; however, no available information |
| Incomplete outcome data (attrition bias) | Low risk | 246 patients discontinued from the roflumilast group and 248 discontinued from the placebo group |
| Other bias | High risk | 40% of participants in both roflumilast and placebo groups received corticosteroid pre‐treatment |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 135 centres in 10 countries Participants: 1221 (roflumilast 500 µg: 467, placebo: 468) Baseline characteristics: mean age 65 years, 71% male, post‐bronchodilator FEV₁ 54.7% and 55.3% predicted (roflumilast and placebo), average 43 pack‐years, 39% current smokers Inclusion criteria: former or current smokers with ≥ 1 year smoking cessation and a pack‐year history ≥ 10, aged ≥ 40 years, post‐bronchodilator FEV₁/FVC ≤ 0.7, post‐bronchodilator FEV₁ 40% to 70% predicted, partial reversibility to albuterol with increase from baseline FEV₁ ≤ 12% or 200 mL Exclusion criteria: available in the online web appendix (p 10) Total numbers of participant withdrawals: 107 (23%) and 82 (18%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo once a day
Concomitant medication
| |
| Outcomes | Primary outcomes: change in mean pre‐bronchodilator FEV₁ from baseline to each post‐randomisation visit | |
| Notes | ClinicalTrials.gov identifier NCT00313209; BY217/M2‐127; 2005‐005080‐28 (EudraCT Number). Funded by Nycomed GmbH, Konstanz, Germany | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "The sponsor generated a randomisation list of patient random numbers using a pseudo‐random number generator. The investigator used an automated, interactive voice response system to randomly assign patients" |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | All individuals involved in the studies were unaware of treatment assignment |
| Blinding of outcome assessment (detection bias) | Low risk | All individuals involved in the studies were unaware of treatment assignment |
| Incomplete outcome data (attrition bias) | Low risk | 107 participants discontinued from the roflumilast group and 82 discontinued from the placebo group |
| Selective reporting (reporting bias) | Low risk | Outcomes reported as planned. Trial registered at clincialtrials.gov |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 85 centres in 7 countries Participants: 910 (roflumilast 500 µg: 372, placebo: 372) Baseline characteristics: mean age 64 years, 71% male, post‐bronchodilator FEV₁ 56.0% and 56.2% predicted (roflumilast and placebo), average 44 pack‐years, 40% current smokers Inclusion criteria: former or current smokers with ≥ 1 year smoking cessation and a pack‐year history ≥ 10, aged ≥ 40 years, post‐bronchodilator FEV₁/FVC ≤ 0.7, post‐bronchodilator FEV₁ 40% to 70% predicted, partial reversibility to albuterol with increase from baseline FEV₁ ≤ 12% or 200 mL Exclusion criteria: available in the online web appendix (p 10) Total numbers of participant withdrawals: 62 (17%) and 39 (11%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 4 weeks with placebo once a day
Concomitant medication
| |
| Outcomes | Primary outcomes: change in mean pre‐bronchodilator FEV₁ from baseline to each post‐randomisation visit | |
| Notes | Clinicaltrials.gov identifier: NCT0042468; BY217/M2‐128. Funded by Nycomed GmbH, Konstanz, Germany | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "The sponsor generated a randomisation list of patient random numbers using a pseudo‐random number generator. The investigator used an automated, interactive voice response system to randomly assign patients" |
| Allocation concealment (selection bias) | Low risk | The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling |
| Blinding of participants and personnel (performance bias) | Low risk | "All individuals involved in the studies were unaware of treatment assignment. The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling. Tablets were identical in appearance" |
| Blinding of outcome assessment (detection bias) | Low risk | "All individuals involved in the studies were unaware of treatment assignment. The investigator or anyone at the study site was prevented from knowing the allocation sequence with code labelling. Tablets were identical in appearance" |
| Incomplete outcome data (attrition bias) | Low risk | 62 participants discontinued from the roflumilast group and 39 discontinued from the placebo group |
| Selective reporting (reporting bias) | Unclear risk | Outcomes reported as planned. Trial protocol registered at clincialtrials.gov and at European trial registry |
| Other bias | Low risk | None |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, double‐blind, placebo‐controlled trial Trial duration: 52 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 338 locations across Australia, Argentina, Canada, Chile, Columbia, Italy, Malaysia, Peru, Phillippines, Romania, Russia, Serbia, Spain, Taiwan, and Ukraine Participants: 2354 (500 µg roflumilast: 1178; placebo: 1176) Baseline characteristics: mean age 64 years, 68% male, mean FEV₁ % predicted 33%, mean smoking history 52.2 pack‐years for roflumilast and 53.1 pack‐years for placebo or current smokers (39% and 40%, respectively) Inclusion criteria: ≥ 40 years with severe or very severe COPD, chronic bronchitis, ≥ 2 exacerbations and/or hospitalisations in the previous year, receiving ICS/LABA with or without LAMA daily for ≥ 3 months Exclusion criteria: within the 4 weeks before enrolment, had a moderate or severe COPD exacerbation and/or a COPD exacerbation treated with antibiotics or systemic corticosteroids or a lower respiratory tract infection. Other exclusionary criteria included diagnoses of other lung diseases, moderate to severe liver impairment (Child‐Pugh B or C), HIV or hepatitis infection, current diagnosis of asthma, cancer in the past 5 years, α₁‐antitrypsin deficiency, clinically significant cardiovascular condition, resting QTc interval > 470 ms, BMI ≥ 45 kg/m² Total numbers of participant withdrawals: 337 (29%) and 254 (21%) from treatment and control groups, respectively | |
| Interventions | Run‐in: 2 weeks, single‐blind. Placebo tablets to assess suitability
Concomitant medication ICS/LABA FDC (fluticasone propionate/salmeterol, 250/50 mg (1 inhalation twice a day), or budesonide/formoterol, 160/4.5 mg (2 inhalations twice a day)). Participants taking fluticasone propionate/salmeterol, 500/50 mg, at study entry were required to switch to the lower dosage (250/50 mg) before entry. Up to 60% of participants were allowed concomitant LAMA (e.g. tiotropium) if administered for ≥ 3 months before screening, with no dose change. Those not on LAMA were allowed a SAMA | |
| Outcomes | Primary outcome: rate of moderate or severe COPD exacerbations per patient per year Secondary outcomes: rate of COPD exacerbations leading to hospitalisation or death (severe COPD exacerbations); rate of moderate or severe exacerbations; rate of moderate or severe COPD exacerbations or COPD exacerbations treated with antibiotics; rate of moderate or severe COPD exacerbations treated with antibiotics during the treatment period; mean change in pre‐dose FEV₁; mean change in pre‐dose FEV₁ from randomisation over 52 weeks; adverse events; mortality (all‐cause); serious adverse events; other adverse events (not including serious events) | |
| Notes | Clinicaltrials.gov identifier: NCT01443845. Funded by Astra Zeneca | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Reported as randomised. Assumed that the randomisation method was adequate due to pharmaceutical company sponsorship |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | Reported as triple‐blind (participant, investigator and outcome assessor) |
| Blinding of outcome assessment (detection bias) | Low risk | Reported as triple‐blind (participant, investigator and outcome assessor) |
| Incomplete outcome data (attrition bias) | Unclear risk | 337 participants (29%) discontinued from the roflumilast group and 254 (22%) discontinued from the placebo group |
| Selective reporting (reporting bias) | Unclear risk | Outcomes reported as planned. Trial registered at clinicaltrials.gov |
| Other bias | Unclear risk | 47% in each treatment group were taking LAMAs. Participants were also using combined LABA/ICS. 65% in the placebo group were taking fluticasone propionate/salmeterol FDC, and 65% in the roflumilast group. 35% of participants in each treatment group were taking budesonide/formoterol FDC |
| Study characteristics | ||
| Methods | Study design: parallel‐group study Randomisation: randomised, triple‐blind, placebo‐controlled trial Trial duration: 24 weeks Intention‐to‐treat analysis: stated | |
| Participants | Setting: 1 specialist respiratory and critical care medicine centre at a hospital in Austria Participants: 80 (roflumilast 500 µg: 40, placebo: 40) Baseline characteristics: median age 64, 52% male, median 50 smoking pack‐years Inclusion criteria: 40 years of age and over, history of ≥ 2 COPD exacerbation requiring systemic corticosteroid treatment or hospitalisation in the last year Exclusion criteria: inability to comply with study medication, history of acute exacerbation, alpha₁‐antitrypsin deficiency, asthma, acute/severe respiratory infection, lung cancer, bronchiectasis, ILD, acute MI, systolic left ventricular dysfunction, CHF, cardiac arrhythmia/heart valve deformation, peripheral arterial occlusive disease, acute or chronic hepatic failure, autoimmune disease, active malignancy, pregnant/breastfeeding, hypersensitivity to study medication or placebo, mental or neurological disorder, history of depression Total numbers of participant withdrawals: 7 (17%) and 6 (15) in the roflumilast and placebo groups, respectively | |
| Interventions | Run‐in: 4 weeks
Concomitant medication
| |
| Outcomes | Primary outcome: change in carotid femoral‐pulse wave velocity Secondary outcomes: change in reactive hyperaemia index; change in augmentation index; change in matrix metalloproteinase‐9; change in asymmetrical dimethylamine; change in tumour necrosis factor‐alpha; change in FEV₁; change in 6‐minute walk test; change in COPD assessment test | |
| Notes | Clincaltrials.gov identifier: NCT01630200. Funded by Ludwig Boltzmann Institute for COPD and Respiratory Epidemiology | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Reported as randomised; no further information |
| Allocation concealment (selection bias) | Unclear risk | No available information |
| Blinding of participants and personnel (performance bias) | Low risk | "All responsible persons, those administering interventions or assessing the outcomes, and elementally all experimental and control patients were blinded to group assignment" |
| Blinding of outcome assessment (detection bias) | Low risk | "All responsible persons, those administering interventions or assessing the outcomes, and elementally all experimental and control patients were blinded to group assignment" |
| Incomplete outcome data (attrition bias) | Low risk | 40 randomised in each group. In the roflumilast group 33/40 completed (82.5%), in the placebo arm 34/40 completed (85%), so similar attrition. Similar numbers of non‐fatal and serious fatal adverse events in each group |
| Selective reporting (reporting bias) | Low risk | Study authors reported outcomes as planned; methods and results were published on EU trials registry |
| Other bias | Unclear risk | Criteria for COPD not well defined apart from exacerbations |
6MWT: 6‐minute walk test; AcPGP: plasma acetyl‐proline‐glycine‐proline; ATS: American Thoracic Society; BDI: Baseline Dyspnoea Index; BMI: body mass index; BORG Scale: rating of perceived exertion; CAT: COPD Assessment Test; CHF: congestive heart failure; COPD: chronic obstructive pulmonary disease; CRP: C‐reactive protein; DLCO: diffusing capacity of the lung for carbon monoxide; EU: European Union; ERS: European Respiratory Society; FDC: fixed dose combination; FEF: forced expiratory flow; FEV₁: forced expiratory volume in one second; FEV₆: forced expiratory volume in six seconds; FIV₁: forced expiratory volume in one second; FRC: functional residual capacity; FVC: forced vital capacity; GOLD: Global Initiative for Chronic Obstructive Lung Disease; HU: Hounsfield unit; HIV: human immunodeficiency virus; IC: inspiratory capacity; ICS: inhaled corticosteroid; ILD: interstitial lung disease; LABA: long‐acting beta₂‐agonist; LAMA:long‐acting muscarinic antagonist; MDI: metered‐dose inhaler; MI: myocardial infarction; MMEF: maximal mid‐expiratory flow; NCT: national clinical trial; PEF: peak expiratory flow; QTc: corrected Q wave and T wave; RFRC: resting functional residual capacity; RV: residual volume; SABA: short‐acting beta₂‐agonist; SAMA: short‐acting muscarinic antagonist; SD: standard deviation; SE: standard error; SGRQ: St George's Respiratory Questionnaire; SOBQ: Shortness of Breath Questionnaire; SVC: slow vital capacity; TDI: transition dyspnoea index; TLC: total lung capacity VO₂: oxygen uptake.
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
| Insufficient data; only RR of QoL improvement provided | |
| No placebo group | |
| No placebo group | |
| Integrated results from four 24‐week cilomilast trials | |
| Analysis focused on participants with baseline SGRQ score ≥ median SGRQ score only | |
| Endpoint: first dose bronchodilator effects only | |
| Treatment Bayer BAY 19‐8004; 11 participants; only 1 week in duration | |
| Endpoint: first dose bronchodilator effects only | |
| Cross‐over design | |
| Phase 2 trial; no primary outcome measure investigating lung function; only 1 trial to date | |
| No study ID or group numbers identified | |
| No SD or SE given | |
| Combining results from 2 pivotal European phase 3 cilomilast trials | |
| Different regimens of roflumilast; no placebo group | |
| Different regimens of roflumilast; no placebo group | |
| Patients were diagnosed with AECOPD | |
| Different regimens of roflumilast; no placebo group | |
| Study 038: insufficient data available for changes in lung function and exacerbation rates | |
| Treatment QAK423 (Novartis), discontinued. Only 1 trial available | |
| Editorial letter | |
| Combined results; individual studies already included in review | |
| Systematic review; relevant individual studies already included in review | |
| JP108 is an extension study of APTA‐2217‐06. After the key open of APTA‐2217‐06, administration to placebo group would be terminated. Not all participants enrolled in JP106 continued onto the JP108 study | |
| No placebo group | |
| Open‐label; no placebo group | |
| Open‐label study; men or women with COPD who successfully completed study 042 or 091 in which participants received cilomilast 15 mg twice daily or placebo for 24 weeks in study 042 and 26 weeks in study 091 without tolerability problems. Concomitant COPD medication use allowed; given placebo or placebo/Ariflo during study period | |
| Open‐label study; men or women with COPD who successfully completed study 039 in which participants received cilomilast 15 mg twice daily or placebo for 24 weeks without tolerability problems. Concomitant COPD medication use allowed; given placebo or placebo/Ariflo during study period | |
| Abstract only; unable to contact study author | |
| No study identification or group numbers identified | |
| Treatment UK‐500,001 (Pfizer); discontinued | |
| Treatment UK‐500,001 (Pfizer); discontinued | |
| Although quoted as significant, mean and SD figures not provided | |
| Inhaled therapy | |
| Different regimens of roflumilast |
AECOPD: acute exacerbation of COPD; COPD: chronic obstructive pulmonary disease; QoL: quality of life; RR: risk ratio; SD: standard deviation; SE: standard error; SGRQ: St George's Respiratory Questionnaire.
Characteristics of studies awaiting classification [ordered by study ID]
| Methods | international, 16‐week, randomised, double‐blind, placebo‐controlled, parallel‐group study investigating effects of roflumilast 500 µg once daily vs placebo on inflammatory parameters in bronchial biopsy tissue specimens, sputum, and blood serum |
| Participants | 150 participants with COPD and chronic bronchitis for at least 12 months will be recruited into the study and randomised in a 1:1 ratio to receive either roflumilast or placebo |
| Interventions | Roflumilast and placebo |
| Outcomes | Primary endpoint will be number of CD8+ cells in bronchial biopsy tissue specimens (submucosa) Key secondary endpoint will be number of CD68+ cells assessed by indirect immunohistochemistry |
| Notes | Completed; awaiting results |
| Methods | Randomised controlled trial |
| Participants | Participants with history of moderate to severe COPD for at least 12 months |
| Interventions | Roflumilast and placebo |
| Outcomes | Primary outcome variable will be mean change in post‐bronchodilator FEV₁ from baseline |
| Notes | No data provided; awaiting results |
| Methods | Single‐blind, randomised, placebo‐controlled study carried out in the Department of Respiratory Medicine at National Institute of Diseases of the Chest and Hospital (NIDCH), Dhaka, Bangladesh |
| Participants | 130 participants were recruited initially and were randomly distributed into Group A, where they received conventional therapy (inhaled salmeterol + fluticasone and tiotropium) and roflumilast (0.5 mg once daily), and Group B, where participants were given placebo with conventional therapy Study duration was 3 months |
| Interventions | As above |
| Outcomes | Primary outcome variable was change in mean FEV₁ Secondary outcome variable was change in mean CAT score from baseline |
| Notes | No data provided; study authors contacted |
| Methods | Multi‐centre double‐blind randomised controlled trial over 12 weeks across USA investigating the safety and efficacy of various doses of oglemilast |
| Participants | 427 participants with COPD, post‐bronchodilator FEV₁/FVC < 70%, post‐bronchodilator FEV₁ > 30% and < 80% |
| Interventions | Oglemilast and placebo |
| Outcomes | Primary outcome variable will be pre‐bronchodilator morning (trough) FEV₁ at 12 weeks Secondary endpoint will be pre‐bronchodilator FVC at 12 weeks |
| Notes | No data provided; awaiting results |
| Methods | Single‐centre, double‐blind randomised controlled trial over 12 weeks in Spain |
| Participants | 150 participants with a diagnosis of COPD, FEV₁ < 70%; current and former smokers |
| Interventions | Roflumilast and placebo |
| Outcomes | Primary outcome variable will be endothelial function at 12 weeks Secondary endpoints include arterial stiffness; serum and plasma inflammation markers; serum oxidative stress markers; serum endothelial dysfunction markers at 12 weeks |
| Notes | No data provided; awaiting results |
| Methods | Triple‐blind randomised controlled trial for 26 weeks investigating whether roflumilast could improve metabolic profiles and reduce visceral adiposity in people with COPD |
| Participants | 14 participants with moderate to severe COPD, assigned to either 500 µg roflumilast or placebo for 26 weeks |
| Interventions | Roflumilast and placebo |
| Outcomes | Primary outcome variables will be change in body mass index; change in waist circumference; change in hip‐to‐waist ratio; change in metabolic profiles; change in body composition; change in subcutaneous adiposity; change in liver fat |
| Notes | No data provided; awaiting results |
CAT: COPD Assessment Test; COPD: chronic obstructive pulmonary disease; FEV₁: forced expiratory volume in one second; FVC: forced vital capacity.
Characteristics of ongoing studies [ordered by study ID]
| Study name | Evaluation of the effect of roflumilast in hyperinflated COPD patients using functional respiratory imaging |
| Methods | Parallel RCT |
| Participants | 40 people who are stable on LABA/LAMA therapy and who are prone to dynamics hyperinflation |
| Interventions | Roflumilast and placebo |
| Outcomes | Radiological (CT) changes in airway measures Changes in spirometry and body plethysmography |
| Starting date | September 2015 |
| Contact information | University Hospital of Antwerp |
| Notes | Other Study ID Numbers: FLUI‐2014‐134, EudraCT Estimated study completion date: January 2017 |
| Study name | A multicenter randomised double‐blind clinical study evaluated the safety, pharmacokinetic and pharmacodynamic characteristics of roflumilast in COPD patients |
| Methods | Parallel RCT |
| Participants | People with COPD in China |
| Interventions | Roflumilast and placebo |
| Outcomes | Area under the plasma concentration after vs drug dose Percentage of participants with adverse events of interest Change in pre‐bronchodilator FEV₁ during the down‐titration period |
| Starting date | March 2016 |
| Contact information | Contact: Zheng Jinping |
| Notes | Estimated enrolment: 120 Estimated study completion date: August 2017 |
COPD: chronic obstructive pulmonary disease; CT: computed tomography; FEV₁: forced expiratory volume in one second; LABA: long‐acting beta₂‐agonist; LAMA:long‐acting muscarinic antagonist; RCT: randomised controlled trial.
Data and analyses
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 FEV₁ (by drug) Show forest plot | 29 | 20815 | Mean Difference (IV, Fixed, 95% CI) | 49.33 [44.17, 54.49] |
| Analysis 1.1  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 1: FEV₁ (by drug) | ||||
| 1.1.1 Tetomilast 50 µg | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| 1.1.2 Roflumilast 500 μg | 18 | 14384 | Mean Difference (IV, Fixed, 95% CI) | 55.18 [48.65, 61.71] |
| 1.1.3 Roflumilast 250 μg | 3 | 1033 | Mean Difference (IV, Fixed, 95% CI) | 56.88 [24.38, 89.38] |
| 1.1.4 Cilomilast 15 mg | 10 | 5322 | Mean Difference (IV, Fixed, 95% CI) | 38.15 [29.41, 46.90] |
| 1.2 FVC Show forest plot | 17 | 22108 | Mean Difference (IV, Fixed, 95% CI) | 86.98 [74.65, 99.31] |
| Analysis 1.2  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 2: FVC | ||||
| 1.3 PEF Show forest plot | 5 | 4245 | Mean Difference (IV, Fixed, 95% CI) | 6.54 [3.95, 9.13] |
| Analysis 1.3  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 3: PEF | ||||
| 1.3.1 Roflumilast 500 μg | 4 | 3685 | Mean Difference (IV, Fixed, 95% CI) | 5.46 [2.74, 8.17] |
| 1.3.2 Roflumilast 250 μg | 1 | 347 | Mean Difference (IV, Fixed, 95% CI) | 7.00 [‐4.05, 18.05] |
| 1.3.3 Cilomilast 15 mg | 1 | 213 | Mean Difference (IV, Fixed, 95% CI) | 34.00 [20.14, 47.86] |
| 1.4 SGRQ total score Show forest plot | 11 | 7645 | Mean Difference (IV, Fixed, 95% CI) | ‐1.06 [‐1.68, ‐0.43] |
| Analysis 1.4  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 4: SGRQ total score | ||||
| 1.4.1 Roflumilast 500 μg | 2 | 722 | Mean Difference (IV, Fixed, 95% CI) | ‐1.87 [‐3.80, 0.06] |
| 1.4.2 Roflumilast 250 μg | 2 | 2229 | Mean Difference (IV, Fixed, 95% CI) | ‐0.64 [‐2.02, 0.74] |
| 1.4.3 Cilomilast 15 mg | 8 | 4694 | Mean Difference (IV, Fixed, 95% CI) | ‐1.06 [‐1.81, ‐0.31] |
| 1.5 SGRQ symptom score Show forest plot | 2 | 1048 | Mean Difference (IV, Fixed, 95% CI) | ‐1.53 [‐4.11, 1.06] |
| Analysis 1.5  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 5: SGRQ symptom score | ||||
| 1.5.1 Roflumilast | 1 | 835 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐3.78, 1.78] |
| 1.5.2 Cilomilast | 1 | 213 | Mean Difference (IV, Fixed, 95% CI) | ‐4.80 [‐11.73, 2.13] |
| 1.6 Number of participants with 1 or more exacerbations (by drug) Show forest plot | 27 | 20382 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.73, 0.84] |
| Analysis 1.6  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 6: Number of participants with 1 or more exacerbations (by drug) | ||||
| 1.6.1 Roflumilast 500 μg | 16 | 14778 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.73, 0.86] |
| 1.6.2 Cilomilast | 10 | 5528 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.76 [0.67, 0.85] |
| 1.6.3 Tetomilast 50 µg | 1 | 76 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.45 [0.26, 23.13] |
| 1.7 Exacerbation rate (inverse variance) Show forest plot | 9 | Rate Ratio (IV, Fixed, 95% CI) | 0.88 [0.83, 0.93] | |
| Analysis 1.7  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 7: Exacerbation rate (inverse variance) | ||||
| 1.7.1 Roflumilast | 8 | Rate Ratio (IV, Fixed, 95% CI) | 0.87 [0.82, 0.92] | |
| 1.7.2 Cilomilast | 1 | Rate Ratio (IV, Fixed, 95% CI) | 0.95 [0.78, 1.17] | |
| 1.8 Borg Scale Show forest plot | 6 | 2860 | Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.33, ‐0.05] |
| Analysis 1.8  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 8: Borg Scale | ||||
| 1.8.1 Cilomilast | 6 | 2860 | Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.33, ‐0.05] |
| 1.9 Shortness of Breath Questionnaire Show forest plot | 2 | 1633 | Mean Difference (IV, Fixed, 95% CI) | ‐1.09 [‐2.47, 0.28] |
| Analysis 1.9  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 9: Shortness of Breath Questionnaire | ||||
| 1.10 Summary symptom score Show forest plot | 5 | 6186 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.02 [‐0.07, 0.03] |
| Analysis 1.10  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 10: Summary symptom score | ||||
| 1.10.1 Roflumilast | 2 | 4287 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.02 [‐0.08, 0.04] |
| 1.10.2 Cilomilast | 3 | 1899 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.04 [‐0.13, 0.06] |
| 1.11 Breathlessness Cough and Sputum Scale (BCSS) (tetomilast 50 µg) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| Analysis 1.11  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 11: Breathlessness Cough and Sputum Scale (BCSS) (tetomilast 50 µg) | ||||
| 1.11.1 Breathlessness | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | ‐0.07 [‐0.77, 0.63] |
| 1.11.2 Cough | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | 0.23 [‐0.54, 1.00] |
| 1.11.3 Sputum | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | ‐0.16 [‐0.97, 0.65] |
| 1.12 6‐minute walk test Show forest plot | 6 | 2055 | Mean Difference (IV, Fixed, 95% CI) | 3.50 [‐5.84, 12.85] |
| Analysis 1.12  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 12: 6‐minute walk test | ||||
| 1.12.1 Roflumilast | 2 | 107 | Mean Difference (IV, Fixed, 95% CI) | 52.61 [‐0.21, 105.42] |
| 1.12.2 Cilomilast | 4 | 1948 | Mean Difference (IV, Fixed, 95% CI) | 1.92 [‐7.58, 11.41] |
| 1.13 Number of participants experiencing an adverse event Show forest plot | 30 | 21310 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.30 [1.22, 1.38] |
| Analysis 1.13  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 13: Number of participants experiencing an adverse event | ||||
| 1.13.1 Roflumilast 500 μg | 15 | 14684 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.34 [1.24, 1.43] |
| 1.13.2 Cilomilast 15 mg | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.21 [1.08, 1.36] |
| 1.13.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.62 [0.25, 1.57] |
| 1.14 Number of participants experiencing an adverse event (roflumilast 500 μg vs 250 μg) Show forest plot | 4 | 1977 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.21 [1.01, 1.46] |
| Analysis 1.14  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 14: Number of participants experiencing an adverse event (roflumilast 500 μg vs 250 μg) | ||||
| 1.15 Diarrhoea Show forest plot | 29 | 20623 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.10 [2.74, 3.50] |
| Analysis 1.15  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 15: Diarrhoea | ||||
| 1.15.1 Roflumilast | 14 | 13997 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.65 [3.10, 4.28] |
| 1.15.2 Cilomilast | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.47 [2.05, 2.98] |
| 1.15.3 Tetomilast | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.68 [0.31, 9.24] |
| 1.16 Nausea Show forest plot | 27 | 20949 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.79 [3.24, 4.43] |
| Analysis 1.16  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 16: Nausea | ||||
| 1.16.1 Roflumilast 500 μg | 12 | 13467 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.25 [2.60, 4.07] |
| 1.16.2 Roflumilast 250 μg | 1 | 856 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.97 [0.91, 17.39] |
| 1.16.3 Cilomilast 15 mg | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 4.37 [3.49, 5.47] |
| 1.16.4 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.00 [0.20, 20.09] |
| 1.17 Vomiting Show forest plot | 12 | 5986 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.95 [2.78, 5.60] |
| Analysis 1.17  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 17: Vomiting | ||||
| 1.17.1 Roflumilast | 2 | 993 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.32 [0.53, 10.23] |
| 1.17.2 Cilomilast | 10 | 4993 | Odds Ratio (M‐H, Fixed, 95% CI) | 4.06 [2.83, 5.82] |
| 1.18 Dyspepsia Show forest plot | 13 | 6247 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.17 [2.33, 4.30] |
| Analysis 1.18  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 18: Dyspepsia | ||||
| 1.18.1 Roflumilast | 1 | 626 | Odds Ratio (M‐H, Fixed, 95% CI) | 7.07 [0.36, 137.40] |
| 1.18.2 Cilomilast | 12 | 5621 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.13 [2.30, 4.27] |
| 1.19 Weight loss Show forest plot | 12 | 12462 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.72 [3.09, 4.47] |
| Analysis 1.19  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 19: Weight loss | ||||
| 1.19.1 Roflumilast | 11 | 12378 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.80 [3.15, 4.58] |
| 1.19.2 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.12 [0.01, 2.63] |
| 1.20 Withdrawals due to adverse events Show forest plot | 31 | 21358 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.89 [1.73, 2.07] |
| Analysis 1.20  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 20: Withdrawals due to adverse events | ||||
| 1.20.1 Roflumilast 500 μg | 16 | 14729 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.90 [1.70, 2.12] |
| 1.20.2 Cilomilast 15 mg | 14 | 6545 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.90 [1.61, 2.24] |
| 1.20.3 Tetomilast 50 mg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.20, 3.18] |
| 1.21 Headache Show forest plot | 23 | 19215 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.69 [1.46, 1.94] |
| Analysis 1.21  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 21: Headache | ||||
| 1.21.1 Roflumilast 500 μg | 12 | 13565 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.13 [1.74, 2.59] |
| 1.21.2 Roflumilast 250 μg | 1 | 347 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.24, 3.99] |
| 1.21.3 Cilomilast 15 mg | 11 | 5303 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.32 [1.08, 1.62] |
| 1.22 Abdominal pain Show forest plot | 15 | 8329 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.02 [1.62, 2.52] |
| Analysis 1.22  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 22: Abdominal pain | ||||
| 1.22.1 Roflumilast | 3 | 2641 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.77 [1.38, 5.56] |
| 1.22.2 Cilomilast | 11 | 5604 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.97 [1.55, 2.49] |
| 1.22.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.97 [0.15, 6.13] |
| 1.23 Influenza‐like symptoms Show forest plot | 9 | 11460 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.09 [0.87, 1.36] |
| Analysis 1.23  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 23: Influenza‐like symptoms | ||||
| 1.23.1 Roflumilast 500 μg | 7 | 10147 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.11 [0.87, 1.41] |
| 1.23.2 Roflumilast 250 μg | 1 | 347 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.98 [0.18, 22.00] |
| 1.23.3 Cilomilast 15 mg | 2 | 966 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.88 [0.44, 1.75] |
| 1.24 Upper respiratory tract infection Show forest plot | 21 | 17022 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.91 [0.81, 1.04] |
| Analysis 1.24  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 24: Upper respiratory tract infection | ||||
| 1.24.1 Roflumilast 500 μg | 11 | 11539 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.77, 1.09] |
| 1.24.2 Roflumilast 250 μg | 2 | 1203 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.84 [0.54, 1.31] |
| 1.24.3 Cilomilast 15 mg | 10 | 4280 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.75, 1.13] |
| 1.25 Psychiatric adverse events (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Analysis 1.25  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 25: Psychiatric adverse events (roflumilast) | ||||
| 1.25.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.13 [1.79, 2.54] |
| 1.25.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.56, 1.33] |
| 1.26 Anxiety or anxiety disorder (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Analysis 1.26  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 26: Anxiety or anxiety disorder (roflumilast) | ||||
| 1.26.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.81 [1.26, 2.62] |
| 1.26.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.94 [0.40, 2.21] |
| 1.27 Depression (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Analysis 1.27  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 27: Depression (roflumilast) | ||||
| 1.27.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.59 [1.11, 2.27] |
| 1.27.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.56 [0.20, 1.56] |
| 1.28 Insomnia and sleep disorders (roflumilast) Show forest plot | 4 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| Analysis 1.28  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 28: Insomnia and sleep disorders (roflumilast) | ||||
| 1.28.1 Roflumilast 500 μg | 4 | 15482 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.67 [2.11, 3.38] |
| 1.28.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.48 [0.81, 2.70] |
| 1.29 Serious adverse events Show forest plot | 29 | 19191 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.99 [0.91, 1.07] |
| Analysis 1.29  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 29: Serious adverse events | ||||
| 1.29.1 Roflumilast 500 μg | 14 | 12562 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.02 [0.93, 1.12] |
| 1.29.2 Cilomilast 15 mg | 14 | 6545 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.72, 1.06] |
| 1.29.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.60 [0.23, 1.55] |
| 1.30 Mortality Show forest plot | 27 | 19786 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.77, 1.24] |
| Analysis 1.30  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 30: Mortality | ||||
| 1.30.1 Roflumilast | 13 | 13370 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.79, 1.30] |
| 1.30.2 Cilomilast | 13 | 6332 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.70 [0.34, 1.45] |
| 1.30.3 Tetomilast | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | Not estimable |
| 1.31 FEV₁ (by mean COPD severity) Show forest plot | 22 | 16813 | Mean Difference (IV, Fixed, 95% CI) | 52.78 [46.73, 58.83] |
| Analysis 1.31  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 31: FEV₁ (by mean COPD severity) | ||||
| 1.31.1 GOLD grade I + II (FEV₁ ≥ 50% predicted) | 10 | 4801 | Mean Difference (IV, Fixed, 95% CI) | 51.82 [39.03, 64.60] |
| 1.31.2 GOLD grade III + IV (FEV₁ < 50% predicted) | 12 | 12012 | Mean Difference (IV, Fixed, 95% CI) | 53.06 [46.19, 59.92] |
| 1.32 FEV₁ (roflumilast 500 μg vs 250 μg) Show forest plot | 3 | 1560 | Mean Difference (IV, Fixed, 95% CI) | 22.61 [‐5.95, 51.16] |
| Analysis 1.32  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 32: FEV₁ (roflumilast 500 μg vs 250 μg) | ||||
| 1.32.1 Roflumilast 250 ug | 3 | 1560 | Mean Difference (IV, Fixed, 95% CI) | 22.61 [‐5.95, 51.16] |
| 1.33 FEV₁ (by study duration) Show forest plot | 28 | 19939 | Mean Difference (IV, Fixed, 95% CI) | 49.09 [43.86, 54.32] |
| Analysis 1.33  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 33: FEV₁ (by study duration) | ||||
| 1.33.1 Duration ≤ 12 weeks | 8 | 1191 | Mean Difference (IV, Fixed, 95% CI) | 101.71 [70.96, 132.46] |
| 1.33.2 Duration 24 to 26 weeks | 13 | 8086 | Mean Difference (IV, Fixed, 95% CI) | 46.14 [38.44, 53.84] |
| 1.33.3 Duration 52 weeks | 7 | 10662 | Mean Difference (IV, Fixed, 95% CI) | 48.77 [41.44, 56.10] |
| 1.34 FEV₁ (additional medication) Show forest plot | 28 | 19719 | Mean Difference (IV, Fixed, 95% CI) | 49.08 [43.85, 54.31] |
| Analysis 1.34  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 34: FEV₁ (additional medication) | ||||
| 1.34.1 Long‐acting bronchodilator | 2 | 1645 | Mean Difference (IV, Fixed, 95% CI) | 60.52 [40.57, 80.46] |
| 1.34.2 Corticosteroids | 3 | 2904 | Mean Difference (IV, Fixed, 95% CI) | 42.26 [25.46, 59.05] |
| 1.34.3 PDE₄i treatment only | 20 | 10323 | Mean Difference (IV, Fixed, 95% CI) | 44.80 [37.69, 51.91] |
| 1.34.4 Various concomitant treatments | 3 | 4847 | Mean Difference (IV, Fixed, 95% CI) | 56.58 [46.91, 66.25] |
| 1.35 FEV₁ (random‐effects model) Show forest plot | 29 | 20015 | Mean Difference (IV, Random, 95% CI) | 51.49 [42.87, 60.10] |
| Analysis 1.35  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 35: FEV₁ (random‐effects model) | ||||
| 1.36 FEV₁ (published vs unpublished) Show forest plot | 29 | 20015 | Mean Difference (IV, Fixed, 95% CI) | 49.28 [44.05, 54.51] |
| Analysis 1.36  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 36: FEV₁ (published vs unpublished) | ||||
| 1.36.1 Published | 20 | 15398 | Mean Difference (IV, Fixed, 95% CI) | 55.75 [49.45, 62.06] |
| 1.36.2 Unpublished | 9 | 4617 | Mean Difference (IV, Fixed, 95% CI) | 35.05 [25.70, 44.40] |
| 1.37 SGRQ total score (by mean COPD severity) Show forest plot | 8 | 4851 | Mean Difference (IV, Fixed, 95% CI) | ‐1.56 [‐2.39, ‐0.74] |
| Analysis 1.37  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 37: SGRQ total score (by mean COPD severity) | ||||
| 1.37.1 GOLD grade I + II | 3 | 2042 | Mean Difference (IV, Fixed, 95% CI) | ‐1.62 [‐2.80, ‐0.44] |
| 1.37.2 GOLD grade III + IV | 5 | 2809 | Mean Difference (IV, Fixed, 95% CI) | ‐1.51 [‐2.67, ‐0.34] |
| 1.38 SGRQ total score (by duration) Show forest plot | 11 | 7069 | Mean Difference (IV, Fixed, 95% CI) | ‐0.99 [‐1.65, ‐0.33] |
| Analysis 1.38  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 38: SGRQ total score (by duration) | ||||
| 1.38.1 Duration < 12 weeks | 2 | 240 | Mean Difference (IV, Fixed, 95% CI) | ‐4.19 [‐7.60, ‐0.78] |
| 1.38.2 Duration 24 to 26 weeks | 7 | 4600 | Mean Difference (IV, Fixed, 95% CI) | ‐1.18 [‐1.94, ‐0.42] |
| 1.38.3 Duration 52 weeks | 2 | 2229 | Mean Difference (IV, Fixed, 95% CI) | 0.26 [‐1.18, 1.69] |
| 1.39 SGRQ total score (by published vs unpublished) Show forest plot | 11 | 7069 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐1.65, ‐0.34] |
| Analysis 1.39  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 39: SGRQ total score (by published vs unpublished) | ||||
| 1.39.1 Published | 5 | 3079 | Mean Difference (IV, Fixed, 95% CI) | ‐1.98 [‐3.07, ‐0.89] |
| 1.39.2 Unpublished | 6 | 3990 | Mean Difference (IV, Fixed, 95% CI) | ‐0.43 [‐1.26, 0.40] |
| 1.40 Number of participants on roflumilast with 1 or more exacerbations (additional medication) Show forest plot | 15 | 14698 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.73, 0.85] |
| Analysis 1.40  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 40: Number of participants on roflumilast with 1 or more exacerbations (additional medication) | ||||
| 1.40.1 Long‐acting bronchodilators | 3 | 1834 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.54, 0.88] |
| 1.40.2 Corticosteroids | 1 | 2686 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.70, 0.95] |
| 1.40.3 Treatment only | 7 | 5145 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.67, 0.93] |
| 1.40.4 Various concomitant treatments | 4 | 5033 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.72, 0.91] |
| 1.41 FVC ML (roflumilast 500 µg, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.41  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 41: FVC ML (roflumilast 500 µg, endpoint) | ||||
| 1.42 FEV₁ (by unknown COPD severity) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| Analysis 1.42  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 42: FEV₁ (by unknown COPD severity) | ||||
| 1.43 FEV₁ (by duration, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.43  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 43: FEV₁ (by duration, endpoint) | ||||
| 1.44 FEV₁ (random‐effects model, endpoint data) Show forest plot | 1 | 60 | Mean Difference (IV, Random, 95% CI) | 0.43 [0.31, 0.55] |
| Analysis 1.44  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 44: FEV₁ (random‐effects model, endpoint data) | ||||
| 1.45 FEV₁ (by moderate to severe COPD severity, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.45  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 45: FEV₁ (by moderate to severe COPD severity, endpoint) | ||||
| 1.46 FEV₁ (roflumilast 500 µg, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.43 [0.31, 0.55] |
| Analysis 1.46  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 46: FEV₁ (roflumilast 500 µg, endpoint) | ||||
| 1.47 FEV₁ ML (additional medication (PDE₄i only) endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.47  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 47: FEV₁ ML (additional medication (PDE₄i only) endpoint) | ||||
| 1.48 FEV₁ (published, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.48  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 48: FEV₁ (published, endpoint) | ||||
| 1.49 FEV₁ (roflumilast 500 μg by mean COPD severity) Show forest plot | 16 | 13896 | Mean Difference (IV, Fixed, 95% CI) | 55.51 [48.88, 62.14] |
| Analysis 1.49  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 49: FEV₁ (roflumilast 500 μg by mean COPD severity) | ||||
| 1.49.1 GOLD grade I + II (FEV₁ ≥ 50% predicted) | 7 | 3341 | Mean Difference (IV, Fixed, 95% CI) | 69.83 [53.34, 86.33] |
| 1.49.2 GOLD grade III + IV (FEV₁ < 50% predicted) | 9 | 10555 | Mean Difference (IV, Fixed, 95% CI) | 52.75 [45.52, 59.99] |
| 1.50 FEV₁ (unknown additional medication) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| Analysis 1.50  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 50: FEV₁ (unknown additional medication) | ||||
| 1.51 FEV₁ (by moderate to severe COPD severity, roflumilast 500 µg endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| Analysis 1.51  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 51: FEV₁ (by moderate to severe COPD severity, roflumilast 500 µg endpoint) | ||||
| 1.52 FEV₁ (by unknown COPD severity, roflumilast 500 µg ) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| Analysis 1.52  Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 52: FEV₁ (by unknown COPD severity, roflumilast 500 µg ) | ||||
![Funnel plot of comparison: 1 PDE₄ inhibitor versus placebo (2020 update), outcome: 1.1 FEV₁ (by drug) [mL].](/es/cdsr/doi/10.1002/14651858.CD002309.pub6/media/CDSR/CD002309/image_n/nCD002309-FIG-04.svg)
Funnel plot of comparison: 1 PDE₄ inhibitor versus placebo (2020 update), outcome: 1.1 FEV₁ (by drug) [mL].
Methodological quality summary: review authors' judgements about each methodological quality item for each included study.
![Forest plot of comparison: 1 PDE₄ inhibitor versus placebo (2020 update), outcome: 1.1 FEV₁ (by drug) [mL].](/es/cdsr/doi/10.1002/14651858.CD002309.pub6/media/CDSR/CD002309/image_n/nCD002309-FIG-03.svg)
Forest plot of comparison: 1 PDE₄ inhibitor versus placebo (2020 update), outcome: 1.1 FEV₁ (by drug) [mL].
Forest plot of comparison: 1 PDE₄ inhibitor versus placebo (2020 update), outcome: 1.4 SGRQ total score.
In the control group, 33 out of 100 people had an exacerbation of COPD over 40 weeks of treatment, compared to 27 (95% CI 26 to 29) out of 100 people in the active treatment group.
In the control group, 4 out of 100 people had a diarrhoea episode over 39 weeks of treatment, compared to 11 (95% CI 10 to 12) out of 100 people in the active treatment group.
Funnel plot of comparison: 2 PDE₄ inhibitor versus placebo (2020 update), outcome: 2.36 FEV₁ (published versus unpublished).
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 1: FEV₁ (by drug)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 2: FVC
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 3: PEF
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 4: SGRQ total score
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 5: SGRQ symptom score
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 6: Number of participants with 1 or more exacerbations (by drug)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 7: Exacerbation rate (inverse variance)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 8: Borg Scale
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 9: Shortness of Breath Questionnaire
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 10: Summary symptom score
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 11: Breathlessness Cough and Sputum Scale (BCSS) (tetomilast 50 µg)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 12: 6‐minute walk test
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 13: Number of participants experiencing an adverse event
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 14: Number of participants experiencing an adverse event (roflumilast 500 μg vs 250 μg)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 15: Diarrhoea
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 16: Nausea
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 17: Vomiting
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 18: Dyspepsia
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 19: Weight loss
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 20: Withdrawals due to adverse events
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 21: Headache
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 22: Abdominal pain
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 23: Influenza‐like symptoms
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 24: Upper respiratory tract infection
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 25: Psychiatric adverse events (roflumilast)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 26: Anxiety or anxiety disorder (roflumilast)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 27: Depression (roflumilast)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 28: Insomnia and sleep disorders (roflumilast)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 29: Serious adverse events
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 30: Mortality
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 31: FEV₁ (by mean COPD severity)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 32: FEV₁ (roflumilast 500 μg vs 250 μg)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 33: FEV₁ (by study duration)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 34: FEV₁ (additional medication)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 35: FEV₁ (random‐effects model)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 36: FEV₁ (published vs unpublished)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 37: SGRQ total score (by mean COPD severity)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 38: SGRQ total score (by duration)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 39: SGRQ total score (by published vs unpublished)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 40: Number of participants on roflumilast with 1 or more exacerbations (additional medication)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 41: FVC ML (roflumilast 500 µg, endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 42: FEV₁ (by unknown COPD severity)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 43: FEV₁ (by duration, endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 44: FEV₁ (random‐effects model, endpoint data)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 45: FEV₁ (by moderate to severe COPD severity, endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 46: FEV₁ (roflumilast 500 µg, endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 47: FEV₁ ML (additional medication (PDE₄i only) endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 48: FEV₁ (published, endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 49: FEV₁ (roflumilast 500 μg by mean COPD severity)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 50: FEV₁ (unknown additional medication)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 51: FEV₁ (by moderate to severe COPD severity, roflumilast 500 µg endpoint)
Comparison 1: PDE4 inhibitor versus placebo (2020 update), Outcome 52: FEV₁ (by unknown COPD severity, roflumilast 500 µg )
| Phosphodiesterase‐4 inhibitors compared to placebo for chronic obstructive pulmonary disease | ||||||
| Patient or population: people with stable chronic obstructive pulmonary disease | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | №. of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with placebo | Risk with PDE₄ inhibitor | |||||
| Change in FEV₁ | Mean FEV₁ was ‐21.37 mL | MD 49.33 mL higher | ‐ | 20815 | ⊕⊕⊝⊝ | This is an overall analysis of the outcome that includes roflumilast 250 µg, roflumilast 500 µg, cilomilast 15 mg, and tetomilast 50 µg MID for FEV₁ is 100 mL |
| Change in FVC Follow‐up: weighted mean 45 weeks | Mean FVC was ‐42.47 | MD 86.98 higher (74.65 higher to 99.31 higher) | ‐ | 22108 (17 RCTs) | ⊕⊕⊕⊕ | This is an overall analysis of the outcome that includes roflumilast and cilomilast studies |
| Change in PEF Follow‐up: weighted mean 42 weeks | Mean PEF was ‐2.82 | MD 6.54 higher (3.95 higher to 9.13 higher) | ‐ | 4245 (5 RCTs) | ⊕⊕⊝⊝ | This is an overall analysis of the outcome that includes roflumilast and cilomilast studies |
| Change in SGRQ total score | Mean SGRQ total score was ‐2.21 SGRQ units | MD 1.06 SGRQ units lower | ‐ | 7645 | ⊕⊕⊕⊝ | This is an overall analysis of the outcome that includes roflumilast 500 µg, roflumilast 250 µg, and cilomilast 15 mg. Lower scores on the SGRQ represent improved quality of life. The MID for this scale is a change of 4 units. This result does not reach the MID for this scale (0 to 100; higher scores indicate more limitations) |
| Number of participants with 1 or more exacerbations | 33 per 100 | 27 per 100 | OR 0.78 | 20382 | ⊕⊕⊕⊕ | This is an overall analysis of the outcome that includes roflumilast 500 µg, cilomilast 15 mg, and tetomilast 50 µg |
| Number of participants experiencing an adverse event | 63 per 100 | 69 per 100 | OR 1.30 | 21310 | ⊕⊕⊝⊝ | This is an overall analysis of the outcome that includes roflumilast 500 µg, cilomilast 15 mg, and tetomilast 50 µg, and participants who reported COPD exacerbations as an adverse event |
| Gastrointestinal adverse effects: diarrhoea | 4 per 100 | 11 per 100 | OR 3.10 | 20623 | ⊕⊕⊕⊕ | This is an overall analysis of the outcome that includes roflumilast 500 µg, cilomilast 15 mg, and tetomilast 50 µg. Diarrhoea was the most commonly reported gastrointestinal side effect. See Figure 4. Weight loss was more common and may be a result of diarrhoea |
| Psychiatric adverse effects (roflumilast 500 µg) Follow‐up: 6 to 52 weeks | 3 per 100 | 7 per 100 (6 to 8) | OR 2.13 (1.79 to 2.54) | 11168 (14 studies) | ⊕⊕⊕⊝ | Pooled data from FDA website, not individual trial reports |
| Mortality | 1 per 100 | 1 per 100 | OR 0.98 | 19786 | ⊕⊕⊕⊝ | This is an overall analysis of the outcome that includes roflumilast 500 µg, cilomilast 15 mg, and tetomilast 50 µg |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; COPD: chronic obstructive pulmonary disease; FDA: US Food and Drug Administration; FEV₁: forced expiratory volume in 1 second; FVC: forced vital capacity; MD: mean difference; MID: minimally important difference; OR: odds ratio; PEF: peak expiratory flow; PDE₄: phosphodiesterase‐4 inhibitor; RCT: randomised controlled trial; RR: risk ratio; SGRQ: St George's Respiratory Questionnaire. | ||||||
| GRADE Working Group grades of evidence. | ||||||
| aThe outcome was downgraded by 1 point due to moderate heterogeneity across studies (I² = 30% to 60%). bAlthough some publication bias was found on further investigation through a sensitivity analysis, we did not consider the removal of studies suspected of publication bias to have a large enough impact on the overall effect estimate and CIs. Therefore, we did not downgrade for publication bias. cThe outcome was downgraded by 2 points due to substantial heterogeneity across studies (I² = 50% to 90%). dBased on data from the combined COPD safety pool. Individual study data not obtained. eThe outcome was downgraded by 1 point due to a small number of events, leading to wide confidence intervals. | ||||||
| Search date: | No. of references for which we sought full text |
|---|---|
| December 2008 | 53 |
| January 2010 | 5 |
| August 2010 | 12 |
| June 2013 | 20 |
| October 2016 | 28 |
| April 2020 | 42 |
| Study ID | Percentage reduction (treatment vs placebo) | Rate ratio (95% CI) | P value |
| 45 | ‐ | 0.001 | |
| 24.3 | 0.757 (0.601 to 0.952) | 0.0175 | |
| 17 | 0.82 (0.63 to 1.06) | 0.163 | |
| 8.5 | 0.95 (0.75 to 1.19) | 0.635 |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 FEV₁ (by drug) Show forest plot | 29 | 20815 | Mean Difference (IV, Fixed, 95% CI) | 49.33 [44.17, 54.49] |
| 1.1.1 Tetomilast 50 µg | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| 1.1.2 Roflumilast 500 μg | 18 | 14384 | Mean Difference (IV, Fixed, 95% CI) | 55.18 [48.65, 61.71] |
| 1.1.3 Roflumilast 250 μg | 3 | 1033 | Mean Difference (IV, Fixed, 95% CI) | 56.88 [24.38, 89.38] |
| 1.1.4 Cilomilast 15 mg | 10 | 5322 | Mean Difference (IV, Fixed, 95% CI) | 38.15 [29.41, 46.90] |
| 1.2 FVC Show forest plot | 17 | 22108 | Mean Difference (IV, Fixed, 95% CI) | 86.98 [74.65, 99.31] |
| 1.3 PEF Show forest plot | 5 | 4245 | Mean Difference (IV, Fixed, 95% CI) | 6.54 [3.95, 9.13] |
| 1.3.1 Roflumilast 500 μg | 4 | 3685 | Mean Difference (IV, Fixed, 95% CI) | 5.46 [2.74, 8.17] |
| 1.3.2 Roflumilast 250 μg | 1 | 347 | Mean Difference (IV, Fixed, 95% CI) | 7.00 [‐4.05, 18.05] |
| 1.3.3 Cilomilast 15 mg | 1 | 213 | Mean Difference (IV, Fixed, 95% CI) | 34.00 [20.14, 47.86] |
| 1.4 SGRQ total score Show forest plot | 11 | 7645 | Mean Difference (IV, Fixed, 95% CI) | ‐1.06 [‐1.68, ‐0.43] |
| 1.4.1 Roflumilast 500 μg | 2 | 722 | Mean Difference (IV, Fixed, 95% CI) | ‐1.87 [‐3.80, 0.06] |
| 1.4.2 Roflumilast 250 μg | 2 | 2229 | Mean Difference (IV, Fixed, 95% CI) | ‐0.64 [‐2.02, 0.74] |
| 1.4.3 Cilomilast 15 mg | 8 | 4694 | Mean Difference (IV, Fixed, 95% CI) | ‐1.06 [‐1.81, ‐0.31] |
| 1.5 SGRQ symptom score Show forest plot | 2 | 1048 | Mean Difference (IV, Fixed, 95% CI) | ‐1.53 [‐4.11, 1.06] |
| 1.5.1 Roflumilast | 1 | 835 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐3.78, 1.78] |
| 1.5.2 Cilomilast | 1 | 213 | Mean Difference (IV, Fixed, 95% CI) | ‐4.80 [‐11.73, 2.13] |
| 1.6 Number of participants with 1 or more exacerbations (by drug) Show forest plot | 27 | 20382 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.73, 0.84] |
| 1.6.1 Roflumilast 500 μg | 16 | 14778 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.73, 0.86] |
| 1.6.2 Cilomilast | 10 | 5528 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.76 [0.67, 0.85] |
| 1.6.3 Tetomilast 50 µg | 1 | 76 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.45 [0.26, 23.13] |
| 1.7 Exacerbation rate (inverse variance) Show forest plot | 9 | Rate Ratio (IV, Fixed, 95% CI) | 0.88 [0.83, 0.93] | |
| 1.7.1 Roflumilast | 8 | Rate Ratio (IV, Fixed, 95% CI) | 0.87 [0.82, 0.92] | |
| 1.7.2 Cilomilast | 1 | Rate Ratio (IV, Fixed, 95% CI) | 0.95 [0.78, 1.17] | |
| 1.8 Borg Scale Show forest plot | 6 | 2860 | Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.33, ‐0.05] |
| 1.8.1 Cilomilast | 6 | 2860 | Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.33, ‐0.05] |
| 1.9 Shortness of Breath Questionnaire Show forest plot | 2 | 1633 | Mean Difference (IV, Fixed, 95% CI) | ‐1.09 [‐2.47, 0.28] |
| 1.10 Summary symptom score Show forest plot | 5 | 6186 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.02 [‐0.07, 0.03] |
| 1.10.1 Roflumilast | 2 | 4287 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.02 [‐0.08, 0.04] |
| 1.10.2 Cilomilast | 3 | 1899 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.04 [‐0.13, 0.06] |
| 1.11 Breathlessness Cough and Sputum Scale (BCSS) (tetomilast 50 µg) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.11.1 Breathlessness | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | ‐0.07 [‐0.77, 0.63] |
| 1.11.2 Cough | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | 0.23 [‐0.54, 1.00] |
| 1.11.3 Sputum | 1 | 22 | Mean Difference (IV, Fixed, 95% CI) | ‐0.16 [‐0.97, 0.65] |
| 1.12 6‐minute walk test Show forest plot | 6 | 2055 | Mean Difference (IV, Fixed, 95% CI) | 3.50 [‐5.84, 12.85] |
| 1.12.1 Roflumilast | 2 | 107 | Mean Difference (IV, Fixed, 95% CI) | 52.61 [‐0.21, 105.42] |
| 1.12.2 Cilomilast | 4 | 1948 | Mean Difference (IV, Fixed, 95% CI) | 1.92 [‐7.58, 11.41] |
| 1.13 Number of participants experiencing an adverse event Show forest plot | 30 | 21310 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.30 [1.22, 1.38] |
| 1.13.1 Roflumilast 500 μg | 15 | 14684 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.34 [1.24, 1.43] |
| 1.13.2 Cilomilast 15 mg | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.21 [1.08, 1.36] |
| 1.13.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.62 [0.25, 1.57] |
| 1.14 Number of participants experiencing an adverse event (roflumilast 500 μg vs 250 μg) Show forest plot | 4 | 1977 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.21 [1.01, 1.46] |
| 1.15 Diarrhoea Show forest plot | 29 | 20623 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.10 [2.74, 3.50] |
| 1.15.1 Roflumilast | 14 | 13997 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.65 [3.10, 4.28] |
| 1.15.2 Cilomilast | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.47 [2.05, 2.98] |
| 1.15.3 Tetomilast | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.68 [0.31, 9.24] |
| 1.16 Nausea Show forest plot | 27 | 20949 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.79 [3.24, 4.43] |
| 1.16.1 Roflumilast 500 μg | 12 | 13467 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.25 [2.60, 4.07] |
| 1.16.2 Roflumilast 250 μg | 1 | 856 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.97 [0.91, 17.39] |
| 1.16.3 Cilomilast 15 mg | 14 | 6542 | Odds Ratio (M‐H, Fixed, 95% CI) | 4.37 [3.49, 5.47] |
| 1.16.4 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.00 [0.20, 20.09] |
| 1.17 Vomiting Show forest plot | 12 | 5986 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.95 [2.78, 5.60] |
| 1.17.1 Roflumilast | 2 | 993 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.32 [0.53, 10.23] |
| 1.17.2 Cilomilast | 10 | 4993 | Odds Ratio (M‐H, Fixed, 95% CI) | 4.06 [2.83, 5.82] |
| 1.18 Dyspepsia Show forest plot | 13 | 6247 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.17 [2.33, 4.30] |
| 1.18.1 Roflumilast | 1 | 626 | Odds Ratio (M‐H, Fixed, 95% CI) | 7.07 [0.36, 137.40] |
| 1.18.2 Cilomilast | 12 | 5621 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.13 [2.30, 4.27] |
| 1.19 Weight loss Show forest plot | 12 | 12462 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.72 [3.09, 4.47] |
| 1.19.1 Roflumilast | 11 | 12378 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.80 [3.15, 4.58] |
| 1.19.2 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.12 [0.01, 2.63] |
| 1.20 Withdrawals due to adverse events Show forest plot | 31 | 21358 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.89 [1.73, 2.07] |
| 1.20.1 Roflumilast 500 μg | 16 | 14729 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.90 [1.70, 2.12] |
| 1.20.2 Cilomilast 15 mg | 14 | 6545 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.90 [1.61, 2.24] |
| 1.20.3 Tetomilast 50 mg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.20, 3.18] |
| 1.21 Headache Show forest plot | 23 | 19215 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.69 [1.46, 1.94] |
| 1.21.1 Roflumilast 500 μg | 12 | 13565 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.13 [1.74, 2.59] |
| 1.21.2 Roflumilast 250 μg | 1 | 347 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.24, 3.99] |
| 1.21.3 Cilomilast 15 mg | 11 | 5303 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.32 [1.08, 1.62] |
| 1.22 Abdominal pain Show forest plot | 15 | 8329 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.02 [1.62, 2.52] |
| 1.22.1 Roflumilast | 3 | 2641 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.77 [1.38, 5.56] |
| 1.22.2 Cilomilast | 11 | 5604 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.97 [1.55, 2.49] |
| 1.22.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.97 [0.15, 6.13] |
| 1.23 Influenza‐like symptoms Show forest plot | 9 | 11460 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.09 [0.87, 1.36] |
| 1.23.1 Roflumilast 500 μg | 7 | 10147 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.11 [0.87, 1.41] |
| 1.23.2 Roflumilast 250 μg | 1 | 347 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.98 [0.18, 22.00] |
| 1.23.3 Cilomilast 15 mg | 2 | 966 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.88 [0.44, 1.75] |
| 1.24 Upper respiratory tract infection Show forest plot | 21 | 17022 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.91 [0.81, 1.04] |
| 1.24.1 Roflumilast 500 μg | 11 | 11539 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.77, 1.09] |
| 1.24.2 Roflumilast 250 μg | 2 | 1203 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.84 [0.54, 1.31] |
| 1.24.3 Cilomilast 15 mg | 10 | 4280 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.92 [0.75, 1.13] |
| 1.25 Psychiatric adverse events (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.25.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.13 [1.79, 2.54] |
| 1.25.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.56, 1.33] |
| 1.26 Anxiety or anxiety disorder (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.26.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.81 [1.26, 2.62] |
| 1.26.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.94 [0.40, 2.21] |
| 1.27 Depression (roflumilast) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.27.1 Roflumilast 500 μg | 1 | 11168 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.59 [1.11, 2.27] |
| 1.27.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.56 [0.20, 1.56] |
| 1.28 Insomnia and sleep disorders (roflumilast) Show forest plot | 4 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.28.1 Roflumilast 500 μg | 4 | 15482 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.67 [2.11, 3.38] |
| 1.28.2 Roflumilast 250 μg | 1 | 6288 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.48 [0.81, 2.70] |
| 1.29 Serious adverse events Show forest plot | 29 | 19191 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.99 [0.91, 1.07] |
| 1.29.1 Roflumilast 500 μg | 14 | 12562 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.02 [0.93, 1.12] |
| 1.29.2 Cilomilast 15 mg | 14 | 6545 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.87 [0.72, 1.06] |
| 1.29.3 Tetomilast 50 µg | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.60 [0.23, 1.55] |
| 1.30 Mortality Show forest plot | 27 | 19786 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.77, 1.24] |
| 1.30.1 Roflumilast | 13 | 13370 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.79, 1.30] |
| 1.30.2 Cilomilast | 13 | 6332 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.70 [0.34, 1.45] |
| 1.30.3 Tetomilast | 1 | 84 | Odds Ratio (M‐H, Fixed, 95% CI) | Not estimable |
| 1.31 FEV₁ (by mean COPD severity) Show forest plot | 22 | 16813 | Mean Difference (IV, Fixed, 95% CI) | 52.78 [46.73, 58.83] |
| 1.31.1 GOLD grade I + II (FEV₁ ≥ 50% predicted) | 10 | 4801 | Mean Difference (IV, Fixed, 95% CI) | 51.82 [39.03, 64.60] |
| 1.31.2 GOLD grade III + IV (FEV₁ < 50% predicted) | 12 | 12012 | Mean Difference (IV, Fixed, 95% CI) | 53.06 [46.19, 59.92] |
| 1.32 FEV₁ (roflumilast 500 μg vs 250 μg) Show forest plot | 3 | 1560 | Mean Difference (IV, Fixed, 95% CI) | 22.61 [‐5.95, 51.16] |
| 1.32.1 Roflumilast 250 ug | 3 | 1560 | Mean Difference (IV, Fixed, 95% CI) | 22.61 [‐5.95, 51.16] |
| 1.33 FEV₁ (by study duration) Show forest plot | 28 | 19939 | Mean Difference (IV, Fixed, 95% CI) | 49.09 [43.86, 54.32] |
| 1.33.1 Duration ≤ 12 weeks | 8 | 1191 | Mean Difference (IV, Fixed, 95% CI) | 101.71 [70.96, 132.46] |
| 1.33.2 Duration 24 to 26 weeks | 13 | 8086 | Mean Difference (IV, Fixed, 95% CI) | 46.14 [38.44, 53.84] |
| 1.33.3 Duration 52 weeks | 7 | 10662 | Mean Difference (IV, Fixed, 95% CI) | 48.77 [41.44, 56.10] |
| 1.34 FEV₁ (additional medication) Show forest plot | 28 | 19719 | Mean Difference (IV, Fixed, 95% CI) | 49.08 [43.85, 54.31] |
| 1.34.1 Long‐acting bronchodilator | 2 | 1645 | Mean Difference (IV, Fixed, 95% CI) | 60.52 [40.57, 80.46] |
| 1.34.2 Corticosteroids | 3 | 2904 | Mean Difference (IV, Fixed, 95% CI) | 42.26 [25.46, 59.05] |
| 1.34.3 PDE₄i treatment only | 20 | 10323 | Mean Difference (IV, Fixed, 95% CI) | 44.80 [37.69, 51.91] |
| 1.34.4 Various concomitant treatments | 3 | 4847 | Mean Difference (IV, Fixed, 95% CI) | 56.58 [46.91, 66.25] |
| 1.35 FEV₁ (random‐effects model) Show forest plot | 29 | 20015 | Mean Difference (IV, Random, 95% CI) | 51.49 [42.87, 60.10] |
| 1.36 FEV₁ (published vs unpublished) Show forest plot | 29 | 20015 | Mean Difference (IV, Fixed, 95% CI) | 49.28 [44.05, 54.51] |
| 1.36.1 Published | 20 | 15398 | Mean Difference (IV, Fixed, 95% CI) | 55.75 [49.45, 62.06] |
| 1.36.2 Unpublished | 9 | 4617 | Mean Difference (IV, Fixed, 95% CI) | 35.05 [25.70, 44.40] |
| 1.37 SGRQ total score (by mean COPD severity) Show forest plot | 8 | 4851 | Mean Difference (IV, Fixed, 95% CI) | ‐1.56 [‐2.39, ‐0.74] |
| 1.37.1 GOLD grade I + II | 3 | 2042 | Mean Difference (IV, Fixed, 95% CI) | ‐1.62 [‐2.80, ‐0.44] |
| 1.37.2 GOLD grade III + IV | 5 | 2809 | Mean Difference (IV, Fixed, 95% CI) | ‐1.51 [‐2.67, ‐0.34] |
| 1.38 SGRQ total score (by duration) Show forest plot | 11 | 7069 | Mean Difference (IV, Fixed, 95% CI) | ‐0.99 [‐1.65, ‐0.33] |
| 1.38.1 Duration < 12 weeks | 2 | 240 | Mean Difference (IV, Fixed, 95% CI) | ‐4.19 [‐7.60, ‐0.78] |
| 1.38.2 Duration 24 to 26 weeks | 7 | 4600 | Mean Difference (IV, Fixed, 95% CI) | ‐1.18 [‐1.94, ‐0.42] |
| 1.38.3 Duration 52 weeks | 2 | 2229 | Mean Difference (IV, Fixed, 95% CI) | 0.26 [‐1.18, 1.69] |
| 1.39 SGRQ total score (by published vs unpublished) Show forest plot | 11 | 7069 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐1.65, ‐0.34] |
| 1.39.1 Published | 5 | 3079 | Mean Difference (IV, Fixed, 95% CI) | ‐1.98 [‐3.07, ‐0.89] |
| 1.39.2 Unpublished | 6 | 3990 | Mean Difference (IV, Fixed, 95% CI) | ‐0.43 [‐1.26, 0.40] |
| 1.40 Number of participants on roflumilast with 1 or more exacerbations (additional medication) Show forest plot | 15 | 14698 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.73, 0.85] |
| 1.40.1 Long‐acting bronchodilators | 3 | 1834 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.54, 0.88] |
| 1.40.2 Corticosteroids | 1 | 2686 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.70, 0.95] |
| 1.40.3 Treatment only | 7 | 5145 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.79 [0.67, 0.93] |
| 1.40.4 Various concomitant treatments | 4 | 5033 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.72, 0.91] |
| 1.41 FVC ML (roflumilast 500 µg, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.42 FEV₁ (by unknown COPD severity) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| 1.43 FEV₁ (by duration, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.44 FEV₁ (random‐effects model, endpoint data) Show forest plot | 1 | 60 | Mean Difference (IV, Random, 95% CI) | 0.43 [0.31, 0.55] |
| 1.45 FEV₁ (by moderate to severe COPD severity, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.46 FEV₁ (roflumilast 500 µg, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.43 [0.31, 0.55] |
| 1.47 FEV₁ ML (additional medication (PDE₄i only) endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.48 FEV₁ (published, endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.49 FEV₁ (roflumilast 500 μg by mean COPD severity) Show forest plot | 16 | 13896 | Mean Difference (IV, Fixed, 95% CI) | 55.51 [48.88, 62.14] |
| 1.49.1 GOLD grade I + II (FEV₁ ≥ 50% predicted) | 7 | 3341 | Mean Difference (IV, Fixed, 95% CI) | 69.83 [53.34, 86.33] |
| 1.49.2 GOLD grade III + IV (FEV₁ < 50% predicted) | 9 | 10555 | Mean Difference (IV, Fixed, 95% CI) | 52.75 [45.52, 59.99] |
| 1.50 FEV₁ (unknown additional medication) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
| 1.51 FEV₁ (by moderate to severe COPD severity, roflumilast 500 µg endpoint) Show forest plot | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.52 [0.25, 0.79] |
| 1.52 FEV₁ (by unknown COPD severity, roflumilast 500 µg ) Show forest plot | 1 | 76 | Mean Difference (IV, Fixed, 95% CI) | 82.00 [‐50.84, 214.84] |
