Comparison of the Risk of Drowsiness and Sedation between Levocetirizine and Desloratadine

Abstract

Desloratadine and levocetirizine are histamine H(1) receptor antagonists (antihistamines) that were launched in the UK in 2001. Our objective was to compare the frequency with which drowsiness and sedation were reported for desloratadine and levocetirizine within the first 30 days of observation, as monitored using the observational cohort technique of prescription-event monitoring (PEM). Exposure data were derived from dispensed prescriptions written by primary care physicians and outcome data were derived from questionnaires that were posted to prescribers at least 6 months after the date of the first prescription for each patient. The odds ratio (OR) was calculated using unconditional logistic regression modelling. The effect of age, sex, reported prescribing indication (allergic rhinitis with asthma/wheezing, allergic rhinitis without asthma/wheezing, 'other'), pattern of use and reported previous antihistamine use on the OR was examined. A time-to-event analysis was performed. The cohorts comprised >24,000 patients in total. Cohort demographics were similar (both cohorts: median age 37 years; 60% women); the most frequently reported prescribing indication for both drugs was allergic rhinitis without asthma/wheezing (54%). The incidence of first reports of drowsiness/sedation for levocetirizine or desloratadine was low (46 [0.37%] and 9 [0.08%], respectively) and statistically different (p < 0.0001). These events tended to occur earlier for desloratadine than levocetirizine (50% at 7 or 14 days of observation, respectively; p = 0.6487), but the cumulative time to event differed, with more events observed for levocetirizine than expected (p < 0.0001; 46 vs 28.09). The final estimates of risk were the sex-adjusted ORs for each prescribing indication category: allergic rhinitis with asthma/wheezing (3.51; 95% CI 0.71, 17.43; n = 3357), allergic rhinitis without asthma/wheezing (6.75; 95% CI 2.37, 19.22; n = 12,627) and 'other' (3.11; 95% CI 0.86, 11.31; n = 6725). Although the reporting rates of drowsiness and sedation are low for both drugs, patients prescribed levocetirizine are more likely to experience drowsiness and sedation in the first month of observation (after starting treatment) than patients prescribed desloratadine. For patients with allergic rhinitis without asthma/wheezing, the sex-adjusted odds of drowsiness/sedation were over six times greater in patients using levocetirizine than desloratadine in the first month of observation, with the OR being statistically significant. For the other two indication categories, allergic rhinitis with asthma/wheezing and 'other', the OR was not statistically significant. Although the risk of drowsiness/sedation is low, conditions such as allergic rhinitis are common, which makes any impact on patient cognitive function important. Doctors should be aware of this when prescribing these products to patients where daytime sedation is undesirable. However, essential components of the comparative benefit-risk evaluation of these two products include assessment of efficacy and patient preference (neither of which forms part of this study).