Efficacy of a Combination Therapy of Montelukast and Antihistamines in Allergic Rhinitis: A Systematic Review and Network Meta-Analysis

Abstract

Allergic rhinitis (AR) significantly impairs quality of life and often necessitates combination therapies for optimal symptom control. This study aimed to evaluate the efficacy of montelukast–antihistamine combination therapy in patients with AR by using a network meta-analysis. A comprehensive search was conducted using PubMed, Embase, MEDLINE, Scopus, the Cochrane Library, and Google Scholar up to April 2025. The treatment strategies included montelukast alone, antihistamine monotherapies (loratadine, desloratadine, levocetirizine, and fexofenadine), their respective combinations with montelukast, including bilastine. Outcomes included daytime and nighttime symptom scores, Rhinoconjunctivitis Quality of Life Questionnaire (RQLQ), and individual symptoms. Both pairwise and network meta-analyses were conducted. Thirty studies (4,486 patients) were included. Montelukast combinations with desloratadine (standardized mean difference [SMD] = −0.51), levocetirizine (SMD = −0.44), and loratadine (SMD = −0.31) significantly improved daytime nasal symptoms compared to montelukast alone. Only montelukast–levocetirizine improved nighttime symptoms (SMD = −0.21) and RQLQ (SMD = −0.48). The combinations with desloratadine or levocetirizine were superior for nasal obstruction, sneezing, and itching, while nasal discharge improved only with montelukast–levocetirizine. No treatment significantly improved eye symptoms. Surface under the cumulative ranking curve rankings generally favored combination therapies, though trends varied by outcome. Desloratadine monotherapy ranked highest for nasal itching. Although some comparisons require cautious interpretation, montelukast-based combination therapy demonstrated greater efficacy than monotherapy for multiple AR symptoms. These results highlight the importance of selecting therapeutic strategies based on the predominant symptom profile of individual patients.

INTRODUCTION

Allergic rhinitis (AR) is a prevalent chronic inflammatory disease of the upper airway, affecting up to 30% of the global population.¹ Recent epidemiological trends show a growing burden of AR, particularly in industrialized regions.² AR is characterized by nasal congestion, rhinorrhea, sneezing, and nasal itching, often accompanied by ocular symptoms, all of which substantially impair quality of life, sleep, and work productivity.³ The management of AR aims to control symptoms, improve patient well-being, and prevent long-term complications, including comorbid asthma and sinusitis.

Oral pharmacological therapy remains the mainstay of AR management, with oral antihistamines and leukotriene receptor antagonists (LTRAs) representing the most prescribed agents. Second-generation antihistamines are particularly effective in relieving histamine-mediated symptoms, such as sneezing, itching, and rhinorrhea, with minimal sedation and other adverse effects; however, LTRAs are considered more effective for nasal congestion and nocturnal symptoms by targeting leukotriene-driven inflammation.⁴ The Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines recommend a stepwise approach to pharmacological treatment based on symptom severity and control.³ Although the ARIA guidelines primarily emphasize intranasal corticosteroid (INCS) therapy, they offer limited recommendations on oral antihistamine–LTRA combinations, despite their widespread use in clinical settings. This discrepancy underscores a critical gap between guideline-based recommendations and real-world clinical practice, where a combination therapy of oral antihistamines and LTRAs is frequently employed to manage persistent or multidimensional AR symptoms, particularly in cases where INCSs are contraindicated or not preferred. Current standard care for AR incorporates individualized treatment strategies tailored to symptom severity, treatment response, and patient preference.⁵

Although antihistamines and LTRAs act through complementary mechanisms targeting distinct inflammatory pathways, clinical evidence on the efficacy of combination therapy has been inconsistent. Previous meta-analyses have reported modest benefits of combination therapy over monotherapy.⁶,⁷ However, these studies evaluated a narrow symptom spectrum, without analyzing treatment effects across individual nasal and ocular domains or comparing specific antihistamine–montelukast combinations. Moreover, traditional pairwise meta-analyses are confined to direct comparisons, restricting the ability to simultaneously evaluate multiple treatment strategies. Network meta-analysis (NMA) addresses these limitations by integrating both direct and indirect evidence, enabling comprehensive comparisons and probabilistic ranking across all included interventions.⁸ This method provides better insights into the relative efficacy of multiple pharmacological options across distinct symptom domains, thereby informing individualized treatment decisions.

In this study, we conducted an NMA to compare ten oral pharmacologic strategies used in the treatment of AR: montelukast monotherapy; 4 antihistamine monotherapies (loratadine, desloratadine, levocetirizine, and fexofenadine); and 5 montelukast-based combination therapies (montelukast combined with loratadine, desloratadine, levocetirizine, fexofenadine, or bilastine). Our aim was to systematically evaluate and rank these treatments in terms of their efficacy for both overall symptom burden and individual nasal/ocular symptoms, as well as quality of life measures, based on outcomes commonly reported in the AR literature. Through this comprehensive evidence synthesis, we sought to identify the most effective and symptom-targeted treatment options for AR to support evidence-based clinical decision-making.

MATERIALS AND METHODS

Study design

This systematic review and NMA was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines for systematic review and meta-analysis. The research question was formulated using the PICOS framework, focusing on the comparative efficacy of oral pharmacotherapy for AR. Eligible studies included randomized controlled trials (RCTs) and comparative studies—either prospective or retrospective—that evaluated the effectiveness of oral pharmacotherapy defined as second-generation oral antihistamines and LTRAs in patients with AR. Patients were diagnosed with AR based on clinical history and physical findings, with or without allergen sensitization. Studies focusing primarily on allergic conjunctivitis or allergic rhinoconjunctivitis without predominant nasal symptoms were excluded. Only oral pharmacotherapies were considered; intranasal sprays, injectables, or combination products including non-oral routes were excluded. The primary outcomes of interest were changes in continuous symptom-specific scores (daytime nasal symptoms score, nighttime nasal symptoms score, nasal obstruction, rhinorrhea, sneezing, itching, and eye symptoms) and quality of life measures (Rhinoconjunctivitis Quality of Life Questionnaire, RQLQ). All outcomes included were continuous; binary outcomes were not analyzed in this study. The study protocol was developed prior to data extraction and followed a pre-specified analysis plan (Supplementary Table S1).

Search strategy

A comprehensive literature search was conducted across PubMed, Embase, MEDLINE, Scopus, the Cochrane Library, and Google Scholar, covering all studies published up to April 2025. The search strategy was developed by a medical librarian with over 10 years of experience in clinical statistics and information retrieval. Search terms included “allergic rhinitis,” “oral pharmacotherapy,” “antihistamines,” “leukotriene receptor antagonists,” and specific drug names such as montelukast, cetirizine, loratadine, desloratadine, fexofenadine, bilastine, and levocetirizine. No language restrictions were applied. The complete search strategy for each database is provided in Supplementary Table S2. Additional manual searches of the reference lists of articles and relevant systematic reviews included were also performed to identify additional eligible studies.

Study selection and data extraction

All studies identified were imported into EndNote (Clarivate Analytics) for reference management, and duplicates were removed. Two reviewers independently screened titles and abstracts for relevance according to the eligibility criteria. The full-text articles of potentially eligible studies were retrieved and assessed for inclusion. Any discrepancies in study selection were resolved through discussion with a third reviewer. The process of study selection is illustrated in the PRISMA 2020 flow diagram (Fig. 1).

Fig. 1. Study flow diagram.

Data extraction was independently performed by 2 reviewers using a pre-designed standardized form. The following variables were recorded: the first author, publication year, country, study design, sample size, patient age, sex, AR type (e.g., seasonal, perennial, or unspecified), treatment arms, and outcome measures. Outcome data included continuous symptom scores for individual symptoms (nasal obstruction, rhinorrhea, sneezing, nasal itching, and eye symptoms), daytime and nighttime total nasal symptom scores, and the RQLQ scores. The clinical relevance and validity of these outcomes were supported by their consistent use in previous studies and their established role as core indicators of AR severity.

Available data on adverse events were also extracted from each included study. However, the quantitative synthesis of safety outcomes was not conducted due to inconsistencies in reporting, including lack of severity grading, unclear attribution to study drugs, and incomplete reporting of event frequencies by treatment arm. When multiple publications originated from the same research group, potential overlap was assessed by comparing study periods, institutions, and participant characteristics. Only the most comprehensive or recent dataset was included to avoid duplication. Importantly, after cross-verification, no overlapping patient populations were identified.

Risk of bias assessment

The methodological quality of RCTs was assessed using the Cochrane Risk of Bias 2.0 tool (Supplementary Table S3).⁹ For non-randomized controlled studies, the Newcastle–Ottawa Scale was used to evaluate study quality (Supplementary Table S4).¹⁰

Statistical analysis

This NMA was conducted using the netmeta package in R version 3.5.0 (R Foundation for Statistical Computing, Vienna, Austria). A random-effects model within a frequentist framework was applied.¹¹ Standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated for all outcomes, including individual nasal symptoms (sneezing, itching, nasal congestion, and rhinorrhea), eye symptoms, total daytime and nighttime nasal symptom scores, and RQLQ scores. Between-study heterogeneity was evaluated using the I² and τ² statistics. Subgroup and sensitivity analyses were performed according to rhinitis phenotype (seasonal or perennial), age group (adult or pediatric), and treatment duration (< 6 or ≥ 6 weeks) to examine the stability of the estimates. To ensure the assumption of transitivity, potential effect modifiers such as mean age, baseline symptom severity, and treatment duration were compared across studies using descriptive and statistical assessments.

Inconsistency across the network was assessed using both global and local approaches.¹² The global inconsistency was tested using the design-by-treatment interaction model, whereas node-splitting analyses were performed to evaluate local inconsistencies in specific treatment comparisons.¹³ Depending on data availability and the connectivity of the network, node-splitting analyses were conducted at both outcome-specific and treatment-pair levels. Outcome-specific consistency was examined for domains with sufficient data (e.g., RQLQ and nasal obstruction), while pair-specific consistency was assessed for treatment comparisons with limited outcome overlap.

To rank treatment effectiveness, the surface under the cumulative ranking curve (SUCRA) and the mean rank method were employed. SUCRA values range from 0 to 1, with 1 representing the most effective treatment and 0 the least effective.

RESULTS

We ultimately analyzed 4,486 subjects evaluated in 30 studies.¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴,²⁵,²⁶,²⁷,²⁸,²⁹,³⁰,³¹,³²,³³,³⁴,³⁵,³⁶,³⁷,³⁸,³⁹,⁴⁰,⁴¹,⁴²,⁴³ The studies are summarized in Table 1. The network included ten interventions: montelukast monotherapy; desloratadine, levocetirizine, loratadine, fexofenadine, and bilastine monotherapy; and their respective combinations with montelukast. Among these, bilastine combined with montelukast was evaluated in only 2 studies and was therefore included only in a limited subset of outcomes in the NMA. The structure of the treatment networks for each outcome, including total and individual symptom domains, is illustrated in Supplementary Fig. S1.

Table 1. Summary of the studies included in our network meta-analysis.

Studies	Nation	Type	Total number	Age (yr)	Rhinitis type	Treatment duration (wk)	Comparison	Outcomes
Meltzer et al.15 (2000)	USA	RCT, placebo-controlled, multicenter, parallel-group	460	15–75	SAR	2	Mon vs. Lora vs. Lora + Mon	Total daytime nasal symptoms, total nighttime nasal symptoms, daytime eye symptoms
Wilson et al.14 (2002)	UK	RCT, double-blind, placebo-controlled	37	Mean, 37.0 (SE, 2.0)	SAR	2	Fexof vs. Lora + Mon	Individual nasal symptoms, total daytime nasal symptoms, daytime eye symptoms
Nayak et al.16 (2002)	USA	RCT, double-blind, multicenter, parallel group	758	15–82	SAR	2	Mon vs. Lora vs. Lora + Mon	Individual nasal symptoms, total daytime nasal symptoms, total nighttime nasal symptoms, daytime eye symptoms
Ciebiada et al.17 (2006)	Poland	RCT, double-blind, placebo-controlled	40	18–65	Persistent AR	6	DesL vs. Mon vs. DesL + Mon/LevoC vs. Mon vs. LevoC + Mon	Individual nasal symptoms, total daytime nasal symptoms, daytime eye symptoms
Ciebiada et al.18 (2008)	Poland	RCT, double-blind, placebo-controlled	40	18–65	Persistent AR	6	DesL vs. Mon vs. DesL + Mon/LevoC vs. Mon vs. LevoC + Mon	Total nighttime nasal symptoms, HRQL score
Li et al.19 (2009)	Hong Kong	RCT, placebo-controlled	44	6–18	Persistent AR	16	Fexof vs. Fexof + Mon	Individual nasal symptoms, total daytime nasal symptoms, total nighttime nasal symptoms
Lu et al.20 (2009)	Belgium	RCT, parallel-group, phase 2	876	15–85	SAR	2	Mon vs. Lora vs. Lora + Mon	Total daytime nasal symptoms
Gupta and Matreja21 (2010)	India	RCT, open-label, parallel-group	95	18–60	AR	6	LevoC vs. LevoC + Mon	Total daytime nasal symptoms, total nighttime nasal symptoms, daytime eye symptoms
Ciebiada et al.22 (2011)	Poland	RCT, double-blind, placebo-controlled	40	18–65	Persistent AR	6	DesL vs. Mon vs. DesL + Mon/LevoC vs. Mon vs. LevoC + Mon	Individual nasal symptom (congestion)
Nayak et al.23 (2013)	India	RCT, open-label, comparative	118	18–75	AR	2	Fexof + Mon vs. LevoC + Mon	Individual nasal symptoms, total daytime nasal symptoms, daytime eye symptoms
Son et al.24 (2013)	Korea	Retrospective chart review	60	Mean, 29.3 (SD, 2.9)	Persistent AR for mite	Mean, 6.7	Fexof vs. Fexof + Mon	Individual nasal symptom
Chawla et al.25 (2014)	India	RCT, open-label, parallel-group	54	18–70	AR	2	LevoC vs. DesL	Total daytime nasal symptoms, RQLQ
Erdoğan et al.26 (2014)	Turkey	RCT	40	17–44	Persistent AR	6	DesL vs. DesL + Mon	RQLQ, total nighttime nasal symptoms
Florincescu-Gheorghe et al.27 (2014)	Romania	Prospective	70	15–53	Moderate-severe AR	8	DesL vs. Mon	Total nasal symptom score, Individual nasal symptom
Mahatme et al.28 (2016)	India	RCT, parallel group	65	18–65	AR	4	LevoC + Mon vs. Fexof + Mon	Total daytime nasal symptoms
Andhale et al.29 (2016)	India	RCT	75	15–75	Persistent AR	2	LevoC + Mon vs. Mon	Individual nasal symptoms, daytime eye symptoms
Kaur et al.30 (2017)	India	RCT	100	10–55	AR	6	LevoC vs. Fexof vs. DesL vs. Mon	Total nighttime nasal symptoms
Jia et al.31 (2017)	China	RCT	57	12–56	Persistent severe AR	4	Lora vs. Mon	Total daytime nasal symptoms, Individual nasal symptom
Kim et al.32 (2018)	Korea	RCT, double-blind, multicenter	210	> 15	Asthma and AR	4	LevoC + Mon vs. Mon	Individual nasal symptoms, total daytime nasal symptoms, total nighttime nasal symptoms
Suchita et al.33 (2020)	India	RCT, open-label, parallel-group	80	18–60	SAR	3	LevoC + Mon vs. Fexof + Mon	Total daytime nasal symptoms, daytime eye symptoms, RQLQ
Zhou et al.34 (2020)	China	RCT	104	18–80	AR	4	Lora + Mon vs. Mon	Total daytime nasal symptoms
Panchal et al.35 (2021)	India	RCT	248	18–60	SAR	2	LevoC + Mon vs. Mon vs. LevoC	Total daytime nasal symptoms, total nighttime nasal symptoms, daytime eye symptoms, RQLQ
Pullerits et al.36 (2002)	Estonia	RCT, double-blind, placebo-controlled, parallel-group	31	15–50	SAR	7	Mon vs. Lora + Mon	Total daytime nasal symptoms, total nighttime nasal symptoms
Sadredini et al.37 (2023)	Iran	RCT, cross-over study	70	43.03	AR	8	Fexof vs. DesL	Total daytime nasal symptoms
Sinha et al.38 (2023)	India	RCT, double-blind, parallel group	202	18–65	AR	4	Bila + Mon vs. LevoC + Mon	Total daytime nasal symptoms
Garza-Beltrán et al.39 (2023)	Mexico	RCT, double-blind, multicenter	86	33.9 (estimated)	PAR	6	DesL+ Mon vs. Lora + Mon	Total daytime nasal symptoms
Ghanbari et al.40 (2024)	Iran	RCT, open label	45	6–14	Persistent moderate to severe AR	8	DesL vs. Mon vs. DesL + Mon	Individual nasal symptoms, total daytime nasal symptoms
Prasad et al.43 (2024)	India	RCT, open label	182	18–60	AR	8	Bila + Mon vs. LevoC + Mon	Total daytime nasal symptoms
Kim et al.41 (2024)	Korea	RCT, open-label, multicenter	147	6–14	PAR	4	LevoC + Mon vs. Mon	Individual nasal symptoms, total daytime nasal symptoms, total nighttime nasal symptoms, RQLQ
Lee et al.42 (2024)	Korea	RCT, open label	52	6–14	PAR	4	LevoC + Mon vs. Mon	Total daytime nasal symptoms

RCT, randomized controlled trial; SAR, seasonal allergic rhinitis; Mon, montelukast; Lora, loratadine; SE, standard error; Fexof, fexofenadine; AR, allergic rhinitis; DesL, desloratadine; LevoC, levocetirizine; HRQL, Health-Related Quality of Life; SD, standard deviation; RQLQ, Rhinoconjunctivitis Quality of Life Questionnaire; Bila, bilastine; PAR, perennial allergic rhinitis.

Overall symptom scores and quality of life

In terms of the daytime nasal symptom score, the combination therapies of montelukast and either desloratadine (SMD = −0.51, 95% CI, −0.96 to −0.06), levocetirizine (SMD = −0.44, 95% CI, −0.73 to −0.16), or loratadine (SMD = −0.31, 95% CI, −0.56 to −0.05) resulted in statistically significant improvement compared to montelukast monotherapy (Fig. 2). In contrast, the combination with fexofenadine (SMD = −0.10, 95% CI, −0.56 to 0.37) and all antihistamine monotherapies—desloratadine (SMD = 0.30, 95% CI, −0.08 to 0.67), levocetirizine (SMD = −0.07, 95% CI, −0.43 to 0.30), loratadine (SMD = −0.15, 95% CI, −0.42 to 0.12), or fexofenadine (SMD = 0.20, 95% CI, −0.26 to 0.66)—did not show statistically significant differences.

Fig. 2. Forest plots of SMDs for network meta-analysis of nasal symptom scores and quality of life. (A) Daytime nasal symptoms score. (B) Nighttime nasal symptoms score. (C) Rhinoconjunctivitis Quality of Life Questionnaire.

SMD, standardized mean difference; Mon, montelukast; CI, confidence interval; Bila, bilastine; DesL, desloratadine; Fexof, fexofenadine; LevoC, levocetirizine; Lora, loratadine.

Regarding the nighttime nasal symptom score, only the combination of montelukast and levocetirizine achieved a statistically significant improvement compared to montelukast monotherapy (SMD = −0.21, 95% CI, −0.37 to −0.05). Other combinations of montelukast and desloratadine (SMD = −0.00, 95% CI, −0.49 to 0.48), fexofenadine (SMD = −0.20, 95% CI, −0.96 to 0.57), or loratadine (SMD = 0.01, 95% CI, −0.14 to 0.17), as well as all antihistamine monotherapies, did not show significant differences.

For the RQLQ outcome, a significant improvement was observed only in the montelukast–levocetirizine combination therapy (SMD = −0.48, 95% CI, −0.93 to −0.03). All of the other combination therapies or monotherapies failed to show any statistically significant benefit when compared to montelukast monotherapy.

Individual symptom domains

In the domain of nasal obstruction, montelukast combined with either desloratadine (SMD = −0.74, 95% CI, −1.27 to −0.22) or levocetirizine (SMD = −0.54, 95% CI, −0.90 to −0.19) showed significant efficacy compared to montelukast monotherapy (Fig. 3). Other treatments, including combinations with fexofenadine or loratadine, as well as all antihistamine monotherapies, did not differ significantly from montelukast monotherapy.

Fig. 3. Forest plots of SMDs for network meta-analysis in individual nasal and ocular symptoms domains. (A) Nasal obstruction. (B) Nasal itching. (C) Nasal discharge. (D) Nasal sneezing. (E) Eye symptoms.

SMD, standardized mean difference; Mon, montelukast; CI, confidence interval; DesL, desloratadine; Fexof, fexofenadine; LevoC, levocetirizine; Lora, loratadine.

With respect to nasal itching, montelukast combined with desloratadine (SMD = −0.54, 95% CI, −0.99 to −0.09) or levocetirizine (SMD = −0.27, 95% CI, −0.49 to −0.05), as well as desloratadine monotherapy (SMD = −0.55, 95% CI, −0.92 to −0.17), were significantly more effective than montelukast monotherapy. The remaining combination therapies did not demonstrate superiority.

In the evaluation of nasal discharge, only the montelukast–levocetirizine combination showed a statistically significant improvement (SMD = −0.64, 95% CI, −1.07 to −0.21) compared to montelukast monotherapy. None of the other treatments, whether combination or monotherapy, showed significant differences.

For sneezing, both montelukast combined with desloratadine (SMD = −0.86, 95% CI, −1.43 to −0.29) and montelukast with levocetirizine (SMD = −0.54, 95% CI, −0.92 to −0.17) demonstrated significantly greater efficacy than montelukast monotherapy. None of the other treatments reached statistical significance.

Regarding eye symptoms, none of the combination or monotherapy regimens showed statistically significant improvement compared to montelukast monotherapy. The wide CI and absence of consistent trends suggest that treatment effects on ocular symptoms may be limited or variable across combination therapies.

Treatment ranking and SUCRA results

The SUCRA-based ranking analysis indicated that all combination therapies generally ranked higher than monotherapies across total symptom scores and RQLQ (Table 2). Among them, the combination of montelukast and levocetirizine consistently ranked at the top for nighttime nasal symptoms, RQLQ, eye symptoms, and nasal discharge (Table 3). Similarly, montelukast combined with desloratadine ranked highest for daytime nasal symptoms, nasal congestion, and sneezing. Although monotherapies such as montelukast, levocetirizine, loratadine, and fexofenadine showed lower rankings in most outcomes, desloratadine monotherapy ranked first for nasal itching, suggesting potential superiority in this individual domain. On the other hand, bilastine combined with montelukast was not included in the SUCRA-based rankings for individual symptom domains due to insufficient direct or indirect comparisons within the network.

Table 2. Ranked probabilities of standardized mean differences for network meta-analysis in daytime and nighttime symptom-summated scores and RQLQ.

Treatment	Daytime nasal symptoms score		Nighttime nasal symptoms score		RQLQ
	SUCRA	Rank	SUCRA	Rank	SUCRA	Rank
Mon	0.3107	8	0.4961	5	0.5027	5
Bila + Mon	0.6544	4
DesL	0.0711	10	0.2435	9	0.1982	7
DesL + Mon	0.8710	1	0.5040	4	0.6180	3
Fexof	0.1490	9	0.3096	8
Fexof + Mon	0.4556	6	0.6991	2	0.0725	8
LevoC	0.4097	7	0.5747	3	0.6326	2
LevoC + Mon	0.8561	2	0.8865	1	0.9137	1
Lora	0.5104	5	0.3189	7	0.4913	6
Lora + Mon	0.7120	3	0.4675	6	0.5709	4

RQLQ, Rhinoconjunctivitis Quality of Life Questionnaire; SUCRA, surface under the cumulative ranking curve; Mon, montelukast; Bila, bilastine; DesL, desloratadine; Fexof, fexofenadine; LevoC, levocetirizine; Lora, loratadine.

Table 3. Ranked probabilities of standardized mean differences for network meta-analysis in individual symptoms scores.

Treatment	Congestion		Itching		Discharge		Sneezing		Eye symptoms
	SUCRA	Rank	SUCRA	Rank	SUCRA	Rank	SUCRA	Rank	SUCRA	Rank
Mon	0.3438	6	0.1394	8	0.4293	7	0.1719	8	0.2939	9
Bila + Mon
DesL	0.2001	9	0.8765	1	0.0817	9	0.1369	9	0.3284	8
DesL + Mon	0.9056	1	0.8508	2	0.4912	4	0.9041	1	0.5043	4
Fexof	0.4574	5	0.5253	5	0.4720	5	0.5593	5	0.5259	3
Fexof + Mon	0.7707	3	0.6535	3	0.8077	2	0.7309	2	0.7021	2
LevoC	0.2685	7	0.0942	9	0.4440	6	0.2025	7	0.4571	6
LevoC + Mon	0.8163	2	0.6103	4	0.9197	1	0.7153	3	0.7734	1
Lora	0.2587	8	0.3111	7	0.3153	8	0.5057	6	0.4198	7
Lora + Mon	0.4790	4	0.4389	6	0.5391	3	0.5733	4	0.4951	5

SUCRA, surface under the cumulative ranking curve; Mon, montelukast; Bila, bilastine; DesL, desloratadine; Fexof, fexofenadine; LevoC, levocetirizine; Lora, loratadine.

Assessment of consistency

Global inconsistency was evaluated using the design-by-treatment interaction model. No significant inconsistency was identified for most outcomes, including daytime nasal symptom score (P = 0.1630), nighttime nasal symptom score (P = 0.9383), nasal itching (P = 0.4239), nasal discharge (P = 0.9256), sneezing (P = 0.1254), and eye symptoms (P = 0.6087). However, significant global inconsistency was observed for RQLQ (P = 0.0018) and nasal obstruction (P = 0.0127), indicating that findings for these outcomes should be interpreted with caution.

Local inconsistency was further assessed by using node-splitting and comparison-specific approaches. For the comparison between montelukast combined with levocetirizine and levocetirizine monotherapy in daytime nasal symptom score, the direct and indirect estimates were discordant (direct SMD = 0.60, indirect SMD = −0.43; P = 0.0208), suggesting local inconsistency. Similarly, for nasal obstruction, a statistically significant inconsistency was observed between the direct and indirect estimates in the comparison of montelukast combined with levocetirizine versus levocetirizine monotherapy (SMD = −0.54 vs. −0.98; P = 0.0072), as well as in the comparison of montelukast monotherapy versus levocetirizine monotherapy (SMD = 1.75 vs. −1.92; P = 0.0077). These findings suggest potential study-level variability related to baseline congestion severity, outcome definitions, or sample size differences. Sensitivity analyses excluding levocetirizine-related comparisons confirmed that the overall treatment ranking and direction of effect remained unchanged (Spearman ρ = 0.96; mean ΔSUCRA = 3.0%) (Supplementary Table S5).

Validation of heterogeneity and transitivity assumptions

Between-study heterogeneity was moderate across the network (global I² = 44.7%, τ² = 0.030). Subgroup analyses by rhinitis phenotype (seasonal or perennial), age group (adult or pediatric), and treatment duration (< 6 or ≥ 6 weeks) showed comparable effect estimates, with I² values consistently below 50% (Supplementary Table S6).

Potential effect modifiers, including mean age, baseline symptom severity, and treatment duration, were evenly distributed across the included studies (P > 0.3). These findings indicate that the transitivity assumption was satisfied. No statistical imbalance was observed between monotherapy and combination regimens, further supporting the comparability of study populations and outcome measures.

Safety

Adverse event data were inconsistently reported across studies, with varying levels of detail, definitions, and outcome measures. Most studies reported only mild, transient adverse events such as headache, fatigue, or dry mouth; however, serious adverse events were rarely reported. Importantly, none of the included studies reported statistically significant differences in adverse event incidence between combination therapy and monotherapy groups, or between treatment arms within individual studies. Owing to the heterogeneity and lack of standardized reporting, a formal meta-analysis of adverse events was not conducted. A descriptive summary of the available adverse event data is provided in Supplementary Table S7.

Assessment of publication bias

In all comparison-adjusted funnel plots (Supplementary Fig. S2), the scatter of treatment effects appeared visually symmetrical, indicating a low likelihood of publication bias across all of the evaluated outcomes, including daytime and nighttime nasal symptom scores, RQLQ, and individual symptoms. Furthermore, Egger’s linear regression test for funnel plot asymmetry revealed no significant publication bias (P > 0.05).

DISCUSSION

This NMA revealed that the combination therapy of montelukast and second-generation antihistamines offers generally improved control of AR symptoms compared to monotherapy. While the levocetirizine–montelukast combination therapy showed broad efficacy across multiple domains, the desloratadine–montelukast combination therapy demonstrated benefits primarily for daytime nasal symptoms. Although the combination therapies were generally effective, certain monotherapies also demonstrated notable efficacy in specific domains, specifically desloratadine alone for nasal itching. These findings support the overall benefit of combination therapy, emphasizing the clinical relevance of symptom-specific treatment selection.

Histamine is a principal mediator of the early-phase allergic response in AR.⁴⁴ Upon allergen exposure, activated mast cells rapidly release histamine, which binds to H₁ receptors in the nasal mucosa and induces sneezing, itching, and rhinorrhea within minutes.⁴⁵ Second-generation oral antihistamines selectively block peripheral H₁ receptors and are recommended as first-line therapy for patients with mild to moderate AR due to their rapid onset, favorable safety profile, and minimal sedation.³,⁴⁴,⁴⁶ However, their therapeutic effect is primarily limited to histamine-mediated symptoms and they show little efficacy against nasal obstruction or the late-phase inflammatory response driven by other mediators such as leukotrienes and cytokines.⁴⁷ Cysteinyl leukotrienes (CysLTs), in particular, contribute to sustained inflammation by increasing vascular permeability, promoting mucus secretion, and facilitating eosinophilic infiltration in the nasal mucosa.⁴⁸ Montelukast, a selective CysLT₁ receptor antagonist, has been shown to relieve nasal obstruction by reducing mucosal edema, to suppress sneezing and itching via inhibition of sensory nerve stimulation, and to improve mucociliary clearance by decreasing mucus viscosity.⁴⁷ These complementary mechanisms provide a rationale for combining antihistamines with LTRAs to enhance symptom control by targeting multiple inflammatory pathways. These mechanistic insights align with findings from prior meta-analyses, which demonstrated that a combination therapy of antihistamines and LTRAs offers greater symptom relief than monotherapy.⁶,⁷ However, previous studies have generally evaluated combination therapy as a single category, without differentiating between specific antihistamine–LTRA pairings. The present study expands on previous work by providing domain-specific comparisons through NMA, incorporating both direct and indirect evidence, and further distinguishing treatment efficacy by evaluating the individual contributions of different combination regimens. Our results showed that the desloratadine–montelukast combination therapy was most effective for controlling daytime nasal symptoms (SMD = −0.51 [−0.96 to −0.06]; SUCRA = 90.56%), while the levocetirizine–montelukast combination therapy showed the greatest benefit for nighttime symptoms (SMD = −0.21 [−0.37 to −0.05]; SUCRA = 81.63%) and quality of life (SMD = −0.48 [−0.93 to −0.03]; SUCRA = 84.92%). These findings underscore that therapeutic efficacy varies by symptom domain, and that treatment selection may benefit from considering the specific components of combination regimens.

Montelukast has been considered clinically effective in relieving nasal obstruction due to its leukotriene receptor–mediated anti-inflammatory mechanism.⁴⁷ In our analysis, the combination therapy of montelukast and antihistamines showed a trend toward improvement in nasal obstruction (Fig. 3A). Statistically significant benefit was observed only with the desloratadine–montelukast combination therapy (SMD = −0.74 [−1.27 to −0.22]; SUCRA = 90.56%) and the levocetirizine–montelukast combination therapy (SMD = −0.54 [−0.90 to −0.19]; SUCRA = 81.63%). However, minor local inconsistency was identified in the comparison involving the montelukast and levocetirizine combination therapy (P = 0.0077), suggesting that variability in baseline congestion severity and outcome definitions among studies may have influenced this result. Therefore, the overall improvement in nasal obstruction should be interpreted cautiously, although montelukast-based combinations generally showed favorable effects compared to monotherapy.

For nasal itching, both the desloratadine–montelukast combination therapy (SMD = −0.54 [−0.99 to −0.09]; SUCRA = 82.58%) and levocetirizine–montelukast combination therapy (SMD = −0.27 [−0.49 to −0.05]; SUCRA = 64.32%) showed significant improvement compared to montelukast monotherapy. Interestingly, desloratadine monotherapy outperformed all of the other combination therapies (SMD = −0.55 [−0.92 to −0.17]; SUCRA = 87.70%), suggesting that strong H₁-receptor blockade may be sufficient for histamine-driven symptoms such as itching. Although a combination of fexofenadine and montelukast ranked third in SUCRA for nasal itching, it did not show statistically significant improvement (SMD = −0.28 [−0.74 to 0.18]; SUCRA = 72.94%), emphasizing the importance of interpreting ranking results alongside effect sizes.

For nasal discharge, the only regimen that showed a statistically significant benefit was the levocetirizine–montelukast combination therapy (SMD = −0.64 [−1.07 to −0.21]; SUCRA = 91.97%), indicating that this combination may be particularly useful in patients with rhinorrhea-predominant AR. For sneezing, most combination therapies tended to be more effective than monotherapies; however, not all combinations demonstrated statistically significant superiority over montelukast monotherapy. This reflects a gap between pharmacologic rationale and clinical outcomes, suggesting that more real-world and phenotype-specific data are needed to clarify the therapeutic value of each combination.

Previous meta-analyses that have not differentiated among antihistamines have failed to demonstrate the superiority of combination therapy for itching or nasal obstruction; instead, they have emphasized the role of INCSs.⁷ However, our findings suggest that when specific antihistamines are evaluated individually, certain oral combinations may provide meaningful efficacy. These results offer a clinical rationale for considering such combinations in patients for whom INCSs are contraindicated, such as those with glaucoma, autoimmune conditions, or steroid-sensitive endocrine disorders. In this context, our NMA provides quantitative and symptom-specific evidence that complements current ARIA recommendations, which primarily emphasize INCS therapy, but provide limited guidance on oral antihistamine–LTRA combinations. By identifying the differential efficacy of specific oral combinations, particularly the montelukast–levocetirizine regimen that improved nighttime nasal symptoms and quality of life, this study contributes to refining the clinical applicability of the ARIA guidelines and identifying patient subgroups who may benefit from oral combination therapy when INCSs are not feasible.

Previous meta-analyses have reported that combination therapy with antihistamines and LTRAs does not consistently outperform antihistamine monotherapy for ocular symptoms. In our study, the levocetirizine–montelukast combination ranked highest in SUCRA for eye symptoms, although no combination achieved statistical significance. This highlights ocular symptoms as a persistent challenge in AR management. Several studies have demonstrated the presence of a naso-ocular reflex, whereby nasal allergen provocation induces ocular symptoms even in the absence of direct conjunctival exposure.⁴⁹ Based on this mechanism, various clinical trials have compared INCS, ophthalmic antihistamines, and their combination for managing ocular symptoms in AR.⁵⁰,⁵¹,⁵² One study reported that olopatadine ophthalmic solution, whether administered alone or in combination with intranasal fluticasone furoate, did not substantially alleviate these reflex-mediated symptoms, implying that conjunctival mast cell activation may not be the primary mechanism.⁵³ On the other hand, other studies have reported that INCS may alleviate ocular symptoms, suggesting a broader therapeutic benefit beyond nasal symptom control.⁴⁹,⁵² Together, these findings underscore the complexity of ocular symptom pathophysiology and support the need for symptom-specific and individualized treatment strategies.

This study has several limitations. First, the number of trials for combination therapies, such as a bilastine–montelukast combination therapy, was limited, preventing SUCRA calculation for individual symptoms. Secondly, some SUCRA-based interpretations warranted caution. For instance, although the combination of fexofenadine and montelukast ranked highly for nasal itching, it did not demonstrate statistical significance (SMD = −0.28 [−0.74 to 0.18]), highlighting the need to interpret SUCRA rankings alongside CIs, especially in domains with modest or variable effect sizes. Thirdly, several comparisons were based primarily on indirect evidence and showed local inconsistency, particularly in analyses involving the montelukast–levocetirizine combination therapy, levocetirizine monotherapy, and montelukast monotherapy for nasal congestion and daytime symptoms. These inconsistencies may reflect differences in baseline symptom severity and variability in AR phenotype classification across trials, contributing to clinical heterogeneity. Given the central role of INCSs in AR management, future studies should stratify patient populations by AR phenotype and assess treatment responses across diverse therapeutic modalities, including monotherapy and combination therapies of intranasal sprays, oral antihistamines, and LTRAs. Such efforts may help refine symptom-specific treatment strategies and support a more personalized approach to AR care.

SUPPLEMENTARY MATERIALS

Supplementary Table S1

Protocol summary

Supplementary Table S2

Search strategy

Supplementary Table S3

Individual randomized controlled trial methodological quality

Supplementary Table S4

Quality of individual non-randomized controlled trial methodology

Supplementary Table S5

Sensitivity analyses excluding levocetirizine-related comparisons across all symptom domains

Supplementary Table S6

Validation of heterogeneity, transitivity, and consistency in the network meta-analysis

Supplementary Table S7

AE reporting across included studies

Supplementary Fig. S1

Network geometry of eligible comparisons in the meta-analysis. Panels represent the network structure of direct comparisons included in the network meta-analysis for each outcome: (A) Daytime nasal symptom score, (B) Nighttime nasal symptom score, (C) Rhinoconjunctivitis Quality of Life Questionnaire, (D) Nasal obstruction, (E) Nasal itching, (F) Nasal discharge, (G) Sneezing, (H) Eye symptoms (redness and itching). Each node represents an intervention, and each line represents a direct comparison between 2 treatments. Line thickness corresponds to the number of studies informing that comparison.

Supplementary Fig. S2

Funnel plot for network meta-analysis. (A) Daytime nasal symptom score. (B) Nighttime nasal symptom score. (C) Rhinoconjunctivitis Quality of Life Questionnaire. (D) Nasal obstruction. (E) Nasal itching. (F) Nasal discharge. (G) Sneezing. (H) Eye symptoms (redness and itching).