This network meta-analysis and systematic review, begun in December 2021 and completed in May 2022, taking approximately 5 months, were conducted, and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020.3 The search was performed by two reviewers using PubMed, Embase, OVID, the Cochrane Library and ClinicalTrials.gov with a search period from the creation date of each database to April 2022. The keywords used in the search strategy included “allergic rhinitis”, “allergic rhinitides”, “hay fever”, “antihistamine”, “histamine antagonist”, “loratadine”, “desloratadine”, “cetirizine”, “levocetirizine”, “rupatadine”, “fexofenadine”, “ebastine”, “bilastine”, “terfenadine”, “olopatadine”, “acrivastine”. The reference lists of the obtained literature were also searched to ensure that no available studies were missed.

Studies included in this review were required to meet the following criteria: (1) The type of studies was randomized controlled trial; (2) The content of studies was the comparison among different oral H1 antihistamine treatments or between oral H1 antihistamine treatment and placebo in patients with AR, whose age, gender, race and disease type of patients are not limited; (3) The reductions of patients’ symptom scores, which included Total Symptom Score (TSS), nasal congestion score, rhinorrhea score, nasal itching score, sneezing score and/or ocular symptom (tearing/itching/redness) score, after treatment were provided or could be calculated as the outcome measures in each study.

Studies meeting the following criteria were excluded: (1) The studies were reported in case report format or review article format; (2) The data of studies were missing, or the data could not be extracted from the published results, or the raw data could not be obtained.

All data were extracted from the involved studies by two reviewers separately and cross-checked afterwards. The extracted data included mean and Standard Deviation (SD) of the reductions in symptom scores after treatment, as well as the number, age, gender, and disease type of patients in each intervention group. The GetData Graph Digitizer software (version 2.26) was used to extract data in figures. The data reported with Standard Error (SE) or 95% Confidence Interval (95% CI) were converted to SD referring to the Cochrane Handbook for Systematic Reviews of Interventions (version 6.2).4 The risk of material bias of the studies was assessed by using the Cochrane risk-of-bias tool for randomized trials (version 2),5 and each study was assessed as “low risk of bias”, “some concerns” or “high risk of bias” in each domain.

Meta-analysis was performed by using Stata 16.0. The net league command was used to calculated Relative Risk (RR) with 95% CI between two treatments. For ranking all the treatments, the Surface Under the Cumulative Ranking Curve (SUCRA) from 0% to 100% was applied, with 0% being the statistically worst and 100% being the statistically best.6 The inconsistency between direct and indirect evidence was evaluated by the loop-specific method to reflect heterogeneity in each loop. Inconsistency indicated difference between direct and indirect evidence with 95% CI for Inconsistency Factor (IF) excluding 0. The potential publication bias was assessed by comparison-adjusted funnel plots.

6 networks involving 6 major outcome measures were established, and each network plot involved 10 antihistamine treatments and placebo. The network plots for all the outcome measures are shown in Fig. 2. The number of studies on rupatadine 10 mg was the largest among all the active treatments.

Figure 2.

Network plots for all the outcome measures. The node sizes are weighted by the sample of treatments, and the line widths are weighted by the number of studies involved. (A) Total Symptom Score (TSS) reduction; (B) Nasal congestion score reduction; (C) Rhinorrhea score reduction; (D) Nasal itchinig score reduction; (E) Sneezing score reduction; (F) Ocular symptom score reduction. (a) Rupatadine 10 mg; (b) Rupatadine 20 mg; (c) Ebastine 10 mg; (d) Ebastine 20 mg; (e) Fexofenadine 120 mg; (f) Fexofenadine 180 mg; (g) Levocetirizine 5 mg; (h) Cetirizine 10 mg; (i) Desloratadine 5 mg; (j) Loratadine 10 mg; (k) Placebo.

(0.53MB).

With regard to TSS reduction, all the antihistamine treatments performed better than placebo, and the effect size was the largest for rupatadine 20 mg and rupatadine 10 mg. In comparisons among the antihistamine treatments, rupatadine 20 mg was more effective than the others, and in contrast, loratadine 10 mg was inferior to the other treatments. The network meta-analysis results for TSS reduction are presented in the lower left triangular area of Fig. 3.

Figure 3.

Network meta-analysis results for Total Symptom Score (TSS) reduction and nasal congestion score reduction. The data in the boxes are Relative Risks (RRs) with 95% Confidence Intervals. The lower left triangular area represents the results for TSS reduction, and RR > 1 suggests that the corresponding column treatment is superior to the corresponding row treatment, and the opposite is true for RR < 1. The upper right triangular area represents the results for nasal congestion score reduction, and RR > 1 suggests that the corresponding row treatment is superior to the corresponding column treatment, and the opposite is true for RR < 1. Statistically significant results are shown in italic.

(0.75MB).

With regard to nasal congestion score reduction, all the antihistamine treatments were superior to placebo, and the effect size was the largest for rupatadine 20 mg and rupatadine 10 mg. In comparisons among the antihistamine treatments, rupatadine 20 mg performed better than the others. Fexofenadine 180 mg was more effective than fexofenadine 120 mg, cetirizine 10 mg and loratadine 10 mg. There was no significant difference in efficacy between levocetirizine 5 mg and loratadine 10 mg, while loratadine 10 mg was inferior to the other treatments. The network meta-analysis results for nasal congestion score reduction are presented in the upper right triangular area of Fig. 3.

With regard to rhinorrhea score reduction, all the antihistamine treatments performed better than placebo, and the effect size was the largest for rupatadine 20 mg and rupatadine 10 mg. In comparisons among the antihistamine treatments, rupatadine 20 mg was more effective than ebastine 10 mg, ebastine 20 mg, fexofenadine 120 mg, desloratadine 5 mg and loratadine 10 mg. Both of fexofenadine 180 mg and cetirizine 10 mg performed better than fexofenadine 120 mg and loratadine 10 mg. Rupatadine 10 mg, ebastine 20 mg, fexofenadine 120 mg and desloratadine 5 mg also showed better effect than loratadine 10 mg. The network meta-analysis results for rhinorrhea score reduction is presented in the lower left triangular area of Fig. 4.

Figure 4.

Network meta-analysis results for rhinorrhea score reduction and nasal itching score reduction. The data in the boxes are Relative Risks (RRs) with 95% Confidence Intervals. The lower left triangular area represents the results for rhinorrhea score reduction, and RR > 1 suggests that the corresponding column treatment is superior to the corresponding row treatment, and the opposite is true for RR < 1. The upper right triangular area represents the results for nasal itching score reduction, and RR > 1 suggests that the corresponding row treatment is superior to the corresponding column treatment, and the opposite is true for RR < 1. Statistically significant results are shown in italic.

(0.75MB).

With regard to nasal itching score reduction, each antihistamine treatment was superior to placebo, and the effect size was the largest for rupatadine 20 mg and rupatadine 10 mg. In comparisons among the antihistamine treatments, both of rupatadine 20 mg and fexofenadine 180 mg performed better than fexofenadine 120 mg, cetirizine 10 mg and loratadine 10 mg. Ebastine 20 mg also performed better than loratadine 10 mg. The network meta-analysis results for nasal itching score reduction are presented in the upper right triangular area of Fig. 4.

With regard to sneezing score reduction, all the antihistamine treatments were superior to placebo, and the effect size was the largest for levocetirizine 5 mg and rupatadine 20 mg. In comparisons among the antihistamine treatments, rupatadine 10 mg, rupatadine 20 mg and ebastine 20 mg were more effective than fexofenadine 120 mg and loratadine 10 mg; moreover, rupatadine 20 mg was more effective than fexofenadine 180 mg. Cetirizine 10 mg was superior to fexofenadine 120 mg and loratadine 10 mg. The network meta-analysis results for sneezing score reduction are presented in the lower left triangular area of Fig. 5.

Figure 5.

Network meta-analysis results for sneezing score reduction and ocular symptom score reduction. The data in the boxes are Relative Risks (RRs) with 95% Confidence Intervals. The lower left triangular area represents the results for sneezing score reduction, and RR > 1 suggests that the corresponding column treatment is superior to the corresponding row treatment, and the opposite is true for RR < 1. The upper right triangular area represents the results for ocular symptom score reduction, and RR > 1 suggests that the corresponding row treatment is superior to the corresponding column treatment, and the opposite is true for RR < 1. Statistically significant results are shown in italic.

(0.75MB).

With regard to ocular symptom score reduction, all the antihistamine treatments performed better than placebo, and the effect size was the largest for rupatadine 20 mg and rupatadine 10 mg. In comparisons among the antihistamine treatments, rupatadine 10 mg was more effective than ebastine 10 mg, ebastine 20 mg, cetirizine 10 mg, desloratadine 5 mg and loratadine 10 mg. Rupatadine 20 mg was superior to the other treatments except rupatadine 10 mg and levocetirizine 5 mg. Both of fexofenadine 120 mg and fexofenadine 180 mg performed better than cetirizine 10 mg, desloratadine 5 mg and loratadine 10 mg. Cetirizine 10 mg and desloratadine 5 mg performed better than loratadine 10 mg. The network meta-analysis results for ocular symptom score reduction are presented in the upper right triangular area of Fig. 5.

The treatments were ranked for all the outcome measures according to their SUCRAs, as shown in Table 2; the SUCRA of each treatment is presented in Fig. 6. Rupatadine 20 mg and rupatadine 10 mg were the top 2 treatments for reductions of TSS (SUCRA: 99.7%, 76.3%), nasal congestion score (SUCRA: 96.4%, 76.4%), rhinorrhea score (SUCRA: 96.6%, 74.6%) and ocular symptom score (SUCRA: 97.2%, 88.8%); rupatadine 20 mg and levocetirizine 5 mg were the top 2 treatments for reductions of nasal itching score (SUCRA: 84.8%, 83.4%) and sneezing score (SUCRA: 87.3%, 95.4%); loratadine 10 mg was ranked the lowest in all the symptom score reductions besides placebo.

Figure 6.

Surface Under the Cumulative Ranking Curves (SUCRAs) of the treatments for all the outcome measures. (A) Total Symptom Score (TSS) reduction; (B) Nasal congestion score reduction; (C) Rhinorrhea score reduction; (D) Nasal itchinig score reduction; (E) Sneezing score reduction; (F) Ocular symptom score reduction. (a) Rupatadine 10 mg; (b) Rupatadine 20 mg; (c) Ebastine 10 mg; (d) Ebastine 20 mg; (e) Fexofenadine 120 mg; (f) Fexofenadine 180 mg; (g) Levocetirizine 5 mg; (h) Cetirizine 10 mg; (i) Desloratadine 5 mg; (j) Loratadine 10 mg; (k) Placebo.

(0.85MB).

The inconsistency plots are shown in Fig. 7. Almost all the 95% CIs for their respective IF contained 0, which indicated that there was no significant inconsistency across all the closed loops, and the conclusions of consistency model were reliable.

Figure 7.

Inconsistency in closed loops for all the outcome measures. IF, Inconsistency Factor; CI, Confidence Interval; (A) Total Symptom Score (TSS) reduction; (B) Nasal congestion score reduction; (C) Rhinorrhea score reduction; (D) Nasal itchinig score reduction; (E) Sneezing score reduction; (F) Ocular symptom score reduction. (a) Rupatadine 10 mg; (b) Rupatadine 20 mg; (c) Ebastine 10 mg; (d) Ebastine 20 mg; (e) Fexofenadine 120 mg; (f) Fexofenadine 180 mg; (g) Levocetirizine 5 mg; (h) Cetirizine 10 mg; (i) Desloratadine 5 mg; (j) Loratadine 10 mg; (k) Placebo.

(0.83MB).

Comparison-adjusted funnel plots for the outcome measures in the network meta-analysis are shown in Fig. 8. Scatters of the same color were not completely symmetrical in all plots, which meant that publication bias may exist for the outcome measures involved.

Figure 8.

Comparison-adjusted funnel plots for all the outcome measures. (A) Total Symptom Score (TSS) reduction; (B) Nasal congestion score reduction; (C) Rhinorrhea score reduction; (D) Nasal itchinig score reduction; (E) Sneezing score reduction; (F) Ocular symptom score reduction. (a) Rupatadine 10 mg; (b) Rupatadine 20 mg; (c) Ebastine 10 mg; (d) Ebastine 20 mg; (e) Fexofenadine 120 mg; (f) Fexofenadine 180 mg; (g) Levocetirizine 5 mg; (h) Cetirizine 10 mg; (i) Desloratadine 5 mg; (j) Loratadine 10 mg; (k) Placebo.

(0.65MB).