TABLE 2

Comparison of Moderate and Low Doses of ICSs in Terms of Clinical and Functional Benefits

StudyICSModerate vs Low Doses, μg/dImprovements from baseline
PEFAM, L/min or % PredictedaPEFPM, L/min or % PredictedaFEV1, L or % PredictedbSymptom Scoreβ2-Agonist Use, Puffs per d
Mean, SESMD (95% CI)Mean (SE)SMD (95% CI)Mean (SE)SMD (95% CI)Mean (SE)WMD (95% CI)Mean (SE)WMD (95% CI)
Allen et al15 (1998)FLU Diskhaler200 vs 1000.25 (0.03) vs 0.20 (0.02)0.20 (−007 to 0.47)
Katz et al14 (1998)FLU DPI200 vs 10057 (3.9) vs 50 (5.0)0.17 (−0.13 to 0.47)53 (3.9) vs 44 (4.0)0.24 (−0.06 to 0.54)0.25 (0.03) vs 0.17 (0.03)0.29 (−0.02 to 0.59)−0.44 (0.06) vs −0.43 (0.08)−0.01 (−0.21 to 0.19)−1.14 (0.19) vs −0.73 (0.16)−0.41 (−0.91 to 0.09)
Peden et al16 (1998)FLU Diskus200 vs 10040 (5.0) vs 34 (3.0)0.16 (−0.14 to 0.45)34 (4.0) vs 26 (3.0)0.24 (−0.05 to 0.54)0.24 (0.03) vs 0.22 (0.03)0.08 (−0.24 to 0.39)−0.41 (0.07) vs −0.36 (0.07)−0.05 (−0.24 to 0.14)−1.04 (0.19) vs 0.08 (0.23)−1.12 (−1.71 to −0.52)
Peden et al16 (1998)FLU Diskhaler200 vs 10042 (4.0) vs 41 (5.0)0.02 (−0.28 to 0.32)36 (4.0) vs 36 (4.0)0.0 (−0.29 to 0.29)0.23 (0.04) vs 0.24 (0.03)−0.03 (−0.35 to 0.29)−0.36 (0.07) vs −0.41 (0.07)0.05 (−0.14 to 0.25)−0.90 (0.23) vs −1.02 (0.18)0.12 (−0.45 to 0.69)
Shapiro et al12 (2001)BUD Turbuhaler400 vs 2001.3 (1.3) vs 2.9 (1.3)−0.13 (−0.42 to 0.16)2.4 (1.3) vs 1.8 (1.2)0.05 (−0.24 to 0.34)2.7 (1.6) vs 2.3 (1.6)0.03 (−0.26 to 0.32)
Nayak et al13 (2002)BDP HFA Autohaler160 vs 8010.1 (1.2) vs 9.1 (1.1)0.08 (−0.18 to 0.33)−0.58 (0.12) vs −0.26 (0.18)−0.32 (−0.74 to 0.10)
Skoner et al17 (2008)CIC HFA MDI160 vs 400.15 (0.01) vs 0.13 (0.01)0.11 (−0.08 to 0.29)
Pooled result0.05 (−0.09 to 0.20)0.13 (−0.02 to 0.28)0.11 (0.01 to 0.21)−0.003 (−0.12 to 0.11)−0.42 (−0.87 to 0.03)
  • FLU indicates fluticasone; BUD, budesonide; HFA, hydrofluoroalkane; CIC, ciclesonide; —, not available/applicable.

  • a PEFAM and PEFPM were measured as percent predicted by Shapiro et al12 and L/min by Katz et al14 and Peden et al.16

  • b FEV1 was measured as percent predicted by Nayak et al13 and Shapiro et al12 and liters by Allen et al,15 Katz et al,14 Peden et al,16 and Skoner et al.17