Table 1. SMART Results
SMART Patients Asthma-Related Deaths in Salmeterol Group
n (%*) Asthma-Related
Deaths in
Placebo Group
n (%*) Relative Risk of Asthma-Related Death
(95% Confidence Interval) Excess Deaths Expressed per 10,000 Patients+
(95% Confidence Interval)
All Patients§
salmeterol: n = 13,176
placebo: n = 13,179 13 (0.10%) 3 (0.02%) 4.37 (1.25, 15.34) 8 (3, 13)
Caucasian Patients
Salmeterol: n = 9,281
Placebo: n = 9,361 6 (0.07%) 1 (0.01%) 5.82 (0.70, 48.37) 6 (1, 10)
African American Patients
Salmeterol: n = 2,366
Placebo: n = 2,319 7 (0.31%) 1 (0.04%) 7.26 (0.89, 58.94) 27 (8, 46)
* 28‑week estimate, adjusted according to actual lengths of exposure to study treatment to account for early withdrawal of patients from the study.
+ Estimate of the number of additional asthma‑related deaths in patients treated with salmeterol in SMART, assuming 10,000 patients received salmeterol for a 28‑week treatment period. Estimate calculated as the difference between the salmeterol and placebo groups in the rates of asthma-related death multiplied by 10,000.
§ The Total Population includes Caucasian, African American, Hispanic, Asian, and Other and not reported. No asthma‑related deaths occurred in the Hispanic (salmeterol n = 996, placebo n = 999), Asian (salmeterol n = 173, placebo n = 149), or Other (salmeterol n = 230, placebo n = 224) subpopulations. One asthma‑related death occurred in the placebo group in the subpopulation whose ethnic origin was not reported (salmeterol n = 130, placebo n = 127).
The SNS was a 16-week, double-blind study that compared the addition of salmeterol or albuterol to standard asthma therapy in 25,180 asthma patients who were 12 years of age and older. In the study, there was an increase in the number of respiratory and asthma-related deaths in the salmeterol group (0.07% [12 out of 16,787 patients]) compared to the albuterol group (0.02% [2 out of 8,393 patients] relative risk of 3.0, p=0.105).
In preparation for the December 2008 Advisory Committee, FDA conducted a meta-analysis of 110 studies evaluating the use of LABAs in 60,954 patients with asthma. The meta-analysis used a composite endpoint to measure severe exacerbation of asthma symptoms (asthma-related death, intubation, and hospitalization). The results of the meta-analysis suggested an increased risk for severe exacerbation of asthma symptoms in patients using LABAs compared to those not using LABAs. The largest risk difference per 1000 treated patients was seen in children 4-11 years of age, see table 2 below. The results of the meta-analysis were primarily driven by asthma-related hospitalizations. Other meta-analyses evaluating the safety of LABAs in the treatment of asthma have not shown a significant increase in the risk for severe asthma exacerbations.
Table 2. Meta-Analysis Results: Number of Patients Experiencing an Event*
Patient Population LABA Patients experiencing an event Non-LABA Patients experiencing an event Risk Difference Estimate per 1000 treated patients 95% Confidence Interval