Assessments of Safety
Clinical adverse events and laboratory abnormalities occurring between the screening visit and the end of study drug treatment were recorded and monitored. Exacerbations of asthma were not assessed as adverse events, but rather recorded in detail as reflections of both safety and efficacy of study drug.
The Cochran-Mantel-Haenszel test was used to compare sex and ethnic origin across treatment groups. Age and results of pulmonary function tests at the screening visit were compared across groups using the F test. Among-treatment changes in FEVi and PEF were tested using analysis of variance F tests cialis professional canadian health care mall. Nonparametric methods were used to evaluate patient-rated symptoms and area under the curve (AUC). The two-sided Fisher’s Exact Test was used to test differences in adverse event frequency.
Because salmeterol was known to have a duration of action of at least 12 h or more in many patients, no true unmedicated baseline could be determined at the beginning of treatment day 8. Therefore, for statistical analysis purposes, the value obtained at treatment day 1 was used for the baseline value.
Of the 160 patients enrolled in the study, 154 completed treatment day 8. Four withdrawals occurred in the placebo group and one each in the salmeterol 42 fig and 84 fig groups; four of the six patients were withdrawn for protocol violations, and two patients in the placebo group withdrew as a result of an asthma exacerbation and an adverse event, respectively. Data from three other patients who completed the study were not included in the final efficacy analysis due to protocol violations (two patients took theophylline during the treatment period and one patient did not take the study medication as prescribed); thus, data from a total of 151 patients were included in the analyses of efficacy at treatment day 8. The background characteristics and baseline FEVi results of the patients enrolled are shown in Table 1. The treatment groups were generally comparable, but the placebo and 21 ng groups tended to be younger, to have fewer women, and to have higher baseline FEVi than the other treatment groups.
Table 1—Background Characteristics of Patients at Screening
|10.5 Mg (n=29)||21 Mg (n=32)||42 Mg (n=33)||84 Mg (n=31)|
|Prestudy theophylline (%)||21 (60)||17 (59)||22 (69)||20(61)||22 (71)|