Table 1

Study designs and characteristics of participants in 29 randomised trials of CPAP

Study (first author/year)Total, nMean ageMale (%)Study duration (months)Delta BMI controlDelta BMI CPAPDelta weight controlDelta weight CPAPAHI (events/h)Design*Placebo- controlledCPAP compliance (hours)Diet counselling†Physical activity counselling†
Arias (2005)w12752.0100.030.0006−0.28910.0833−0.857144CYes6NoNo
Barbe (2001)w25454.090.01.50.080.430.310.855.4PYes5NoNo
Barbe (2012)w372351.985.648−0.45−0.25−1.47−0.4538.5PNo5YesNo
Craig (2012)w434657.880.06−0.20.2−0.50.613.5PNo3NoNo
Drager (2007)w52445.5100.04−0.09−0.10−0.35−0.1659PNo6NoNo
Drager (2011)w63643.0100.030.100.19056.5PNo5.2NoNo
Durán-Cantolla (2010)w734052.481.03−0.2−0.3−1043.5PYes4.5NoNo
Hoyos (2012)w86549.0100.030.15650.15330.46960.5939.9PYes3.6YesNo
Hui (2006)w94650.880.430.480.351.110.8531.2PYes5.1NoNo
Ip (2004)w102742.7100.01−0.220.27−0.580.5446PNo4.2NoNo
Jenkinson (1999)w1110149.0100.010.0735−0.2320.135−0.630.7PYes5.4NoNo
Kritikou (men) (2013)w122254.3100.020.0510.4180.11.19242.1CYes6.01NoNo
Kritikou (women) (2013)w122057.60.020.3580.240.9760.77832.10CYes6.00NoNo
Lam (2007)w13§6746.079.02.5−0.22−0.42−0.3−1.1921.4PNo4.2NoNo
Lozano (2010)w146459.268.830.340.370.91.152.67PNo5.6YesNo
Mansfield (2004)w154057.394.530.10.30.40.8328.2PNo5.6YesYes
Montserrat (2001)w164554.091.01.50.440.391.241.0654PYes4.25YesNo
Pedrosa (2013)w173556.077.06−0.0610.21−0.0490.49229PNo6.01NoNo
Pepperell (2002)w1811850.5100.04−0.2320.1215−0.710.3535PYes4.9NoNo
Quan (2013)w1981252.564.66−0.0340.037−0.1070.13640.5PYes4.8NoNo
Robinson (2006)w203554.091.21−0.210.12−0.80.728.1CYes5.2NoNo
Ryan (2005)w211859.088.81−0.250.06−1.10.143.6PNo6.2NoNo
Ryan (2011)w224461.779.510.03610.66190.14622.22835.9PNo4.96NoNo
Sivam (2012)w232747.096.32−0.3380.313−0.8880.80437.2CYes4.6NoNo
Weinstock (2012)w245054.042.02−0.170.31−0.630.5144CYes4.8YesNo
West (2007)w254056.2100.030.0006−0.28910.0833−0.857136.1PYes3.3YesYes
  • Values presented are mean or percent unless indicated.

  • *For crossover studies, we used only data prior to the crossover to avoid carryover bias.

  • †We did not exclude studies that reported diet and physical activity counselling only. We only excluded articles that have an active protocol of diet and/or physical activity and/or drug combined to CPAP therapy.

  • ‡Kritikou and colleagues reported two data sets (independent analysis in men and women).

  • §Lam and colleagues compared three groups: conservative measurements, CPAP and oral appliance. Only the first two groups were used for this analysis.

  • AHI, Apnoea–Hypopnoea Index; BMI, body mass index; C, crossover study design; P, parallel study design.