Figure 20 depicts the results of a meta-analysis for the comparison of intranasal corticosteroid versus leukotriene receptor antagonist. Four trials reported on the outcome of total nasal symptom scores at two weeks. A random effects model using inverse variance weighting was used. Standardized mean differences are reported for individual trials and for the overall pooled effect. In the first trial, Lu (Trial 1), 2009, there were 172 patients in the intranasal corticosteroid group and 111 patients in the leukotriene receptor antagonist group. The standardized mean difference was 0.64 favoring intranasal corticosteroid with a 95 percent confidence interval of -0.88 to -0.39. The weight of this trial was 16.8 percent. In the second trial, Martin, 2006, there were 364 patients in the intranasal corticosteroid group and 366 patients in the leukotriene receptor antagonist group. The standardized mean difference was 0.37 favoring intranasal corticosteroid with a 95 percent confidence interval of -0.52 to -0.23. The weight of this trial was 28.8 percent. In the third trial, Nathan, 2005, there were 291 patients in the intranasal corticosteroid group and 282 patients in the leukotriene receptor antagonist group. The standardized mean difference was 0.26 favoring intranasal corticosteroid with a 95 percent confidence interval of -0.43 to -0.10. The weight of this trial was 26.1 percent. In the fourth trial, Ratner, 2003, there were 353 patients in the intranasal corticosteroid group and 352 patients in the leukotriene receptor antagonist group. The standardized mean difference was 0.41 favoring intranasal corticosteroid with a 95 percent confidence interval of -0.56 to -0.26. The weight of this trial was 28.3 percent. The overall result included 1180 patients in the intranasal corticosteroid group and 1111 patients in the leukotriene receptor antagonist group. The overall pooled standardized mean difference was 0.40 favoring intranasal corticosteroid with a 95 percent confidence interval of -0.52 to -0.27. The figure shows a forest plot of these results. The x-axis spans minus 1 at the left favoring intranasal corticosteroid to positive 1 at the right favoring leukotriene receptor antagonist. There is a vertical line at zero that that extends up from the x-axis. The mean difference for each trial is represented by a green box sized according to the weight of the trial. The confidence interval is represented by a horizontal black line centered on the box. The overall pooled result is represented by a black diamond, the ends of which represent the confidence interval for the pooled result. The green boxes for all trials and the black diamond fall to the left of the vertical line. Tests for heterogeneity are Tau squared of 0.01, chi squared of 6.43 with 3 degrees of freedom and a p-value of 0.09, and I squared of 53 percent. The Z test for overall effect was 6.28 with a p-value less than 0.00001.

Figure 20Total nasal symptom score at 2 weeks: meta-analysis of 4 trials–intranasal corticosteroid versus oral leukotriene receptor antagonist

From: Results

Cover of Treatments for Seasonal Allergic Rhinitis
Treatments for Seasonal Allergic Rhinitis [Internet].
Comparative Effectiveness Reviews, No. 120.
Glacy J, Putnam K, Godfrey S, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.