Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Inhaled corticosteroid dose


There is little more rewarding than to see good evidence used to make some assessment of the balance of benefit and harm in a treatment in terms we can understand. One such comes from Australia, where there was concern about the common use of high doses of inhaled corticosteroids in asthma. Using evidence from Cochrane reviews made this possible.


This review of reviews sought systematic reviews from the Cochrane Database only, and found six that compared inhaled corticosteroids with either placebo or different doses of inhaled corticosteroids for chronic asthma. Efficacy outcomes were peak expiratory flow, FEV1, night waking and rescue beta-agonist use, together with withdrawal because of worsening asthma. Adverse events were also examined.


The response of efficacy and harm with increasing dose of inhaled corticosteroids could only be assessed for fluticasone. For neither beclamethasone nor budesonide were there were insufficient studies covering a range of doses.

For fluticasone, efficacy increased with dose. There were small but clinically useful improvements in peak expiratory flow (by 30 to 45 L/minute more than placebo) and FEV1 (by 0.3 to 0.5 litres more than placebo) as the daily dose increased from 100 μg/day to 1000 μg/day. There were also small improvements in night waking.

Numbers needed to treat could be calculated for withdrawal because of worsening asthma. These were 2.9 at 100 μg/day improving to 2.0 at 500 to 1000 μg/day (Table 1).

Table 1: NNT for efficacy and NNH for harm with increasing daily doses of inhaled corticosteroid

Fluticasone (µg/day)
Withdrawal because of worsening asthma
Hoarseness or dysphonia
Oral candidiasis
2.9 (2.4 to 3.4)
150 (40 to 1140)
90 (27 to 750)
2.4 (2.2 to 2.8)
131 (50 to 420)
61 (22 to 260)
2.0 (1.7 to 2.3)
23 (15 to 52)
21 (14 to 46)
2.1 (1.8 to 2.4)
17 (11 to 35)
23 (14 to 75)
11 (6 to 100)
6 ( 4 to 17)

At low daily doses of 100 or 200 μg/day, neither dysphonia nor oral candidiasis were much of a problem, affecting perhaps an additional one person per hundred treated, and with NNTs of about 100.

At daily doses of 500 μg and above, numbers needed to harm (NNH) fell to levels of about 20 or below, indicating that for every 100 patients treated with these doses, about an additional 5 would experience dysphonia and 5 would experience oral candidiasis.


The paper concludes that use of lower doses of inhaled corticosteroids would be associated with lower adverse events without much loss of efficacy. What Bandolier found interesting when doing a search was that there was little obvious literature about oral candidiasis and inhaled corticosteroids. Yet adverse events like this can have important impact on cost of treatment and patient acceptability.

The real pleasure comes from an intelligent use of evidence to improve outcomes and, because of fewer adverse events, reduce costs.


  1. H Powell, PG Gibson. Inhaled corticosteroid doses in asthma: an evidence-based approach. Medical Journal of Australia 2003 178: 223-225.

previous story