Intranasal Corticosteroids Versus Oral H1 Receptor Antagonists

Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from

Intranasal corticosteroids versus oral H1 receptor antagonists in allergic rhinitis: systematic review of randomised controlled trials John M Weiner, Michael J Abramson, Robert M Puy

Department of Abstract showed that in Tasmania the prevalence of hay fever is Medicine, Monash 4 University, 41%, and that hay fever is the second most frequently Objective To determine whether intranasal 5 Melbourne, self reported condition in Australia. corticosteroids are superior to oral H receptor Victoria, 3181, 1 Apart from local disease, allergic rhinitis can cause Australia antagonists (antihistamines) in the treatment of considerable morbidity including chronic sinusitis and John M Weiner, allergic rhinitis. honorary senior otitis. The condition can also cause irritability and Meta-analysis of randomised controlled trials lecturer Design impaired sleep which can affect quality of life by comparing intranasal corticosteroids with oral Department of leading to poor performance at school or work, antihistamines. Epidemiology and absenteeism from school or work, and chronic Preventive Setting Randomised controlled trials conducted Medicine, Monash tiredness. It can also have detrimental effects on worldwide and published between 1966 and 1997. 16 University emotional and social wellbeing. Subjects 2267 subjects with allergic rhinitis in 16 Michael J Treatment of allergic rhinitis includes avoiding Abramson, randomised controlled trials. allergens (when possible), intranasal corticosteroids, associate professor Main outcome measures Nasal blockage, nasal short term decongestants, oral or topical H receptor Department of discharge, sneezing, nasal itch, postnasal drip, nasal 1 antagonists (antihistamines), intranasal cromoglycate, Allergy and Clinical discomfort, total nasal symptoms, nasal resistance, and Immunology, Alfred anticholinergic agents, and allergen immunotherapy.1 Hospital, Monash eye symptoms and global ratings. Outcomes measured Topical intranasal corticosteroids are said to be University on different scales were combined to determine pooled more effective than oral antihistamines in controlling Robert M Puy, odds ratios (categorical outcomes) or standardised visiting physician nasal blockage and discharge.17 Furthermore, oral mean differences (continuous outcomes). Assessment antihistamines are said to be better at treating nasal Correspondence to: of heterogeneity between studies, and subgroup Dr Weiner itch, sneezing, and eye symptoms.17There is also a per- jmweiner@allergynet. analyses of eye symptoms, were undertaken. ception, especially in popular reviews on allergic rhini- com.au Results Intranasal corticosteroids produced tis, that intranasal corticosteroids do not improve eye significantly greater relief than oral antihistamines of 8 BMJ 1998;317:1624–9 symptoms. nasal blockage (standardised mean difference − 0.63, To address these issues we reviewed published ran- 95% confidence interval − 0.73 to − 0.53), nasal domised controlled trials comparing intranasal discharge ( − 0.5, − 0.6 to − 0.4), sneezing ( − 0.49, − 0.59 to − 0.39), nasal itch ( − 0.38, − 0.49 to − 0.21), corticosteroids with oral antihistamines, and per- http://www.bmj.com/ postnasal drip ( − 0.24, − 0.42 to − 0.06), and total formed a meta-analysis on the efficacy of these nasal symptoms ( − 0.42, − 0.53 to − 0.32), and global interventions on relevant clinical outcomes. ratings gave an odds ratio for deterioration of symptoms of 0.26 (0.08 to 0.8). There were no Selection criteria significant differences between treatments for nasal We restricted our review to randomised controlled discomfort, nasal resistance, or eye symptoms. The trials, which provide the strongest evidence for the effi- effects on sneezing, total nasal symptoms, and eye cacy of any medical treatment. We included only stud- on 24 September 2021 by guest. Protected copyright. symptoms were significantly heterogeneous between ies that focused on allergic rhinitis, and we did not studies. Other combined outcomes were consider studies on the treatment of nasal polyps. homogeneous between studies. Subgroup analysis of For the purposes of our review, intranasal the outcome of eye symptoms suggested that the corticosteroids included beclomethasone dipropion- duration of assessment (averaged mean score over the ate, budesonide, flunisolide, fluocortin, fluticasone pro- study period versus mean score at end of study pionate, mometasone, and triamcinolone acetonide. period) might have accounted for the heterogeneity. All forms of delivery vehicle (aqueous and non- Conclusion The results of this systematic review, aqueous) were considered. We included studies of together with data on safety and cost effectiveness, comparisons with any form of oral antihistamine, but support the use of intranasal corticosteroids over oral excluded studies that used topical antihistamines or antihistamines as first line treatment for allergic rhinitis. topical mast cell stabilisers. Studies were also excluded if they were not randomised or not double blinded. For a study to be included in our review at least one Introduction of the following clinical outcomes had to be reported: Allergic rhinitis is a common disease characterised by nasal symptoms (including total nasal symptom nasal itch, sneezing, watery and mucous rhinorrhoea, scores), eye symptoms, global symptoms, drug require- and nasal obstruction.1 The condition is often ments for treating the rhinitis, nasal function accompanied by allergic conjunctivitis. In the past 30 (including measurements of nasal resistance), and years there has been a dramatic increase in the preva- assessment of quality of life. lence of allergic rhinitis, and studies from England, We excluded studies that reported only nasal chal- Sweden, and Australia have confirmed a doubling of lenge with specific allergens or non-clinical outcomes, prevalence over this time.2–4 Studies from Australia such as in vitro results of inflammatory mediators. We

1624 BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from considered studies published in languages other than Characteristics of included studies. All studies were double blind, double dummy,* English if the translated abstract indicated that the parallel group, randomised controlled trials except where indicated study was a randomised controlled trial of intranasal No of corticosteroids for rhinitis, and a translator was sought. participants with seasonal Age range Search strategy Study allergic rhinitis (mean age) Interventions Outcomes 10 We conducted Medline and Embase searches for Munch 61 16-56 (29) Budesonide 400 ìg Nasal symptoms randomised controlled trials of topical corticosteroids Dexchlorpheniramine 12 mg Eye symptoms 11 ì and rhinitis published between 1966 and 1997. Beswick 49 14-64 (28) Beclomethasone 400 g Nasal symptoms Terfenadine 120 mg Eye symptoms Specifically, studies were retrieved that were indexed as Global rating randomised controlled trials with treatments compris- 12 Wood 74 12-75 (28) Beclomethasone 400 ìg Nasal symptoms ing the following intranasal corticosteroids: beclo- Astemizole 10 mg Eye symptoms methasone dipropionate, budesonide, flunisolide, fluo- Lancer13 18 Not reported Beclomethasone 400 ìg Global rating cortin, fluticasone propionate, mometasone, and Terfenadine 120 mg Nasal resistance triamcinolone acetonide. Review articles identified in Juniper14 90 18-70 (41) Beclomethasone 400 ìg Nasal symptoms this process were surveyed for additional and earlier Astemizole 10 mg Eye symptoms citations. We also used Healthgate and Winspirs Robinson†15 20 15-58 (31) Beclomethasone 400 ìg Nasal symptoms software to search Medline for more recently published Terfenadine 120 mg Eye symptoms studies. Where relevant abstracts were identified in con- Bunnag16 69‡ 16-70 (30) Budesonide 200 ìg Nasal symptoms ference proceedings, Medline searches were conducted Astemizole 10 mg 17 and inquiries made of the authors or sponsoring Darnell 214 13-69 (28) Fluticasone 200 ìg Nasal symptoms companies to identify any subsequent full publications. Terfenadine 120 mg Eye symptoms Global rating Simpson18 143 >15 (27) Budesonide 400 ìg Nasal symptoms Methods Terfenadine 120 mg Eye symptoms Van Bavel19 232 >12 (40) Fluticasone 200 ìg Nasal symptoms Inclusion of studies in the review was decided by a sim- Terfenadine 120 mg Global rating ple majority of all three reviewers, who independently Schoenwetter20 298 >12 (31) Triamcinolone 220 ìg Nasal symptoms read the methods sections of papers identified by the Loratadine 10 mg Eye symptoms search strategy and applied the stated criteria. Quality Bernstein21 239 Adults (36) Triamcinolone 220 ìg Nasal symptoms assessment was performed by two reviewers (RMP and Astemizole 10 mg Eye symptoms JMW), who independently assessed the concealment of Global rating allocation following the guidelines of the Cochrane Bronsky22 348 >12 (30) Fluticasone 200 ìg Nasal symptoms Collaboration.9 We calculated the mean daily cost of Terfenadine 120 mg Global rating intranasal corticosteroids and of non-sedating oral Nasal resistance 23 antihistamines available in Australia. Brooks 60 Not reported Beclomethasone 336 ìg Nasal symptoms Loratadine 10 mg Eye symptoms Global rating http://www.bmj.com/ Statistical considerations Géhanno24 114 13-80 (39) Fluticasone 200 ìg Nasal symptoms We compared the effectiveness of intranasal cortico- Loratadine 10 mg Global rating steroids versus oral antihistamines on nasal symptoms, Vervloet25 238 12-75 (29) Fluticasone 200 ìg Nasal symptoms eye symptoms, and nasal resistance, whenever the Cetirizine 10 mg results were reported. *All patients took tablets and used nasal sprays, but only one preparation was active. For the purpose of statistical analysis outcome data †Double blind double dummy crossover randomised controlled trial. were extracted and entered into RevMan 3.1 (Update ‡Participants had perennial allergic rhinitis.

Software, Oxford). Categorical outcomes (global Fixed effects models were used to obtain summary on 24 September 2021 by guest. Protected copyright. ratings) were analysed as odds ratios and 95% statistics for the overall efficacy of intranasal cortico- confidence intervals, calculated by Peto’s method for steroids on both categorical and continuous outcomes, individual studies. The odds ratio was calculated by and ÷2 tests were performed to assess heterogeneity expressing the odds for deterioration or no change in between studies. In this context, a P value of < 0.05 the treatment group divided by the odds for deteriora- indicates significant differences between studies, tion or no change in the control group. The and raises doubts whether the results can be meaning- convention of the Cochrane Collaboration is to fully combined. Sensitivity and subgroup analyses consider odds ratios > 1.0 as indicating clinically were undertaken to identify the sources of such undesirable outcomes. heterogeneity. Continuous outcomes (symptom scores, nasal resistance) were also extracted from tables. Continuous Description of studies outcomes were analysed as standardised mean We identified 16 studies that complied with the differences. The standardised mean difference is a sta- inclusion criteria, and from which we were able to tistic which expresses the difference in means between obtain sufficient data either directly, or after corre- corticosteroid groups and control groups after sponding with the sponsoring companies or treatment in units of the pooled SD. In individual stud- authors.10–25 These studies totalled 2267 subjects (mean ies, scores for nasal itch and other symptoms were self age 32 years, range 12 to 75 years), of whom 1247 reported by patients. Standardised mean differences (55%) were men. Two further studies met the inclusion allow the scores from different assessment scales to be criteria but they contained insufficient data to allow combined. It was thus possible for us to combine meta-analysis, and we were unable to obtain further symptom scores measured on ordinal scales and visual information despite several attempts.26 27 The table analogue scales. summarises the characteristics of the included studies.

BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com 1625 Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from We excluded 16 studies from the meta-analysis. Nasal blockage Favours Favours Standardised mean Reasons for exclusion were use of: non-random alloca- Study steroid antihistamine Weight (%) difference (95% CI) tion,28 29 a single blind protocol,30 31 combined intra- Bunnag16 4.0 -0.835 (-1.335 to -0.334) nasal corticosteroid and oral antihistamines in the 18 Simpson 3.7 -0.830 (-1.353 to -0.308) 32 33 Brooks23 2.4 -0.830 (-1.479 to -0.181) comparison arm, topical antihistamines in the 34 35 Bernstein21 12.7 -0.797 (-1.079 to -0.515) comparison arm, decongestant in the comparison 20 Schoenwetter 16.8 -0.778 (-1.024 to -0.532) arm,36–38 non-clinical challenge or outcome,39 40 and the Vervloet25 14.8 -0.678 (-0.940 to -0.416) Bronsky22 13.5 -0.606 (-0.879 to -0.332) publication of an abstract only without reporting Munch10 3.8 -0.503 (-1.018 to 0.012) detailed results.41–43 Géhanno24 7.3 -0.471 (-0.843 to -0.098) Juniper14 3.9 -0.444 (-0.956 to 0.069) Darnell17 8.0 -0.428 (-0.783 to -0.073) Beswick11 2.6 -0.377 (-1.006 to 0.252) Methodological quality Wood12 4.7 -0.256 (-0.718 to 0.205) All included studies were of high calibre incorporating 15 Robinson 1.7 -0.155 (-0.925 to 0.615) the features of clearly stated objectives, defined 100 -0.628 (-0.729 to -0.527) diagnostic criteria, stated source of subjects, randomi- χ2=11.76, df=13, NS sation, double blindedness, well defined treatments, Nasal discharge and a description of withdrawals and dropouts. Schoenwetter20 16.5 -0.770 (-1.016 to -0.524) Methods of concealing allocation to the treatment Robinson15 1.6 -0.756 (-1.556 to -0.045) 14 arms were identified and classified according to the cri- Juniper 3.7 -0.579 (-1.097 to -0.062) 9 Bronsky22 13.4 -0.565 (-0.838 to -0.292) teria of the Cochrane Collaboration : A, adequate; B, Géhanno24 7.1 -0.547 (-0.921 to -0.173) allocation method unclear; and C, inadequate. Two 23 Brooks 2.5 -0.526 (-1.157 to 0.106) 13 19 Bernstein21 13.1 -0.514 (-0.789 to -0.238) studies were classified as A and the remainder were Vervloet25 14.9 -0.498 (-0.757 to -0.240) classified as B. Simpson18 3.9 -0.439 (-0.945 to 0.068) Beswick11 2.5 -0.378 (-1.007 to 0.251) Munch10 3.8 -0.369 (-0.880 to 0.142) Bunnag16 4.3 -0.361 (-0.845 to 0.122) Results Darnell17 8.0 -0.290 (-0.642 to 0.063) Wood12 4.7 0.178 (-0.283 to 0.639) Nasal symptoms 100 -0.501 (-0.601 to -0.401) Figure 1 gives an overview of the results for nasal symptoms. Scores for nasal blockage, nasal discharge, and χ2=15.87, df=13, NS sneezing were reported by 14 studies each. Intranasal Sneezing corticosteroids produced significantly greater relief of Schoenwetter20 16.5 -0.820 (-1.066 to -0.573) Géhanno24 7.0 -0.706 (-1.085 to -0.327) nasal blockage than did oral antihistamines (combined Bronsky22 13.3 -0.634 (-1.909 to -0.360) standardised mean difference − 0.63 (95% confidence 23 Brooks 2.5 -0.625 (-1.262 to 0.011) interval − 0.73 to − 0.53)). Intranasal corticosteroids Bernstein21 13.1 -0.584 (-0.861 to -0.307) Robinson15 1.6 -0.583 (-1.371 to 0.205) produced significantly greater relief of nasal discharge Juniper14 3.8 -0.542 (-1.058 to -0.026) ( − 0.5, − 0.6 to − 0.4) than did oral antihistamines. 25

Vervloet 15.1 -0.450 (-0.708 to -0.192) http://www.bmj.com/ Simpson18 3.9 -0.400 (-0.905 to 0.106) These effects were homogeneous between studies Munch10 3.8 -0.391 (-0.902 to 0.121) (÷2 = 11.8, 15.9 NS). Intranasal corticosteroids were also Bunnag16 4.3 -0.350 (-0.832 to 0.133) − − 17 more effective in relieving sneezing ( 0.49, 0.59 to Darnell 8.1 -0.248 (-0.600 to 0.105) − Beswick11 2.6 0.144 (-0.480 to 0.768) 0.39). However, there was significant heterogeneity Wood12 4.4 0.751 (0.275 to 1.228) (÷2 = 42.4, P < 0.0005) with one study showing that oral 100 -0.488 (-0.588 to -0.387) antihistamines produced greater relief of sneezing than 12 χ2=42.40, df=13, P<0.0005 did intranasal corticosteroids. Nasal itch Nasal itch scores were reported by 11 studies. Intra- on 24 September 2021 by guest. Protected copyright. Schoenwetter20 18.9 -0.646 (-0.889 to -0.403) nasal corticosteroids produced significantly greater Bronsky22 15.1 -0.546 (-0.818 to -0.274) relief of itch than did oral antihistamines (combined Beswick11 2.8 -0.524 (-1.159 to 0.110) − − − Vervloet25 16.9 -0.392 (-0.649 to -0.135) standardised mean difference 0.38, 0.49 to 0.21). Robinson15 1.9 -0.376 (-1.153 to 0.401) In two studies there was a modest but still significant Simpson18 4.4 -0.323 (-0.826 to 0.181) 21 effect of intranasal corticosteroids on postnasal drip Bernstein 15.0 -0.297 (-0.570 to -0.025) − − − Géhanno24 8.2 -0.235 (-0.604 to 0.133) ( 0.24, 0.42 to 0.06). Both of these effects were Darnell17 9.1 -0.093 (-0.444 to 0.258) homogeneous. Only one study reported nasal discom- Bunnag16 4.9 -0.042 (-0.521 to 0.437) Brooks23 2.9 0.069 (-0.551 to 0.689) fort, and there was no significant difference between the two treatments.25 100 -0.379 (-0.485 to -0.273) Total nasal symptom scores were reported by nine χ2=13.72, df=10, NS studies (fig 2). Intranasal corticosteroids produced sig- Postnasal drip nificantly greater relief of total nasal symptoms than Schoenwetter20 53.1 -0.306 (-0.545 to -0.068) did oral antihistamines ( − 0.42, − 0.53 to − 0.32). How- Bernstein21 46.9 -0.150 (-0.421 to 0.122) ever, there was significant heterogeneity (÷2 = 26.8, 100 -0.238 (-0.417 to-0.059) P < 0.001), with Wood12 showing greater (albeit not sig- χ2=0.72, df=1, NS nificantly) relief of symptoms with oral antihistamines Nasal discomfort than with intranasal corticosteroids. 25 Vervloet 100 0.093 (-0.162 to 0.348) -1.5 -1.0 -0.5 0 0.5 1.0 Eye symptoms

Fig 1 Comparison of effectiveness of intranasal corticosteroids and oral H1 receptor Eye symptoms were reported by 11 studies (fig 3). antagonists (antihistamines) on nasal symptoms There was no significant difference between intranasal corticosteroids and oral antihistamines on eye

1626 BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from symptoms. There was significant heterogeneity (÷2 = 32.4, P < 0.0005), and both Beswick11 and Wood12 Total nasal symptom score Favours Favours Standardised mean showed that oral antihistamines were more efficacious Study steroid antihistamine Weight (%) difference (95% CI) than intranasal corticosteroids. When these two Géhanno24 8.1 -0.677 (-1.055 to -0.299) outliers were removed, homogeneity was achieved. Bronsky22 15.4 -0.645 (-0.919 to -0.370) Munch10 4.3 -0.645 (-1.165 to -0.124) To further identify the source of the heterogeneity, Schoenwetter20 19.7 -0.606 (-0.848 to -0.364) subgroup analyses were conducted. Stratification by Van Bavel19 11.0 -0.498 (-0.822 to -0.174) Bernstein21 15.4 -0.427 (-0.701 to -0.152) intranasal corticosteroids showed that most of the 11 2 Beswick 2.9 -0.386 (-1.015 to 0.244) heterogeneity (÷ = 18.7, P < 0.001) occurred in trials Vervloet25 17.8 -0.062 (-0.317 to 0.193) with beclomethasone. Trials that used intranasal Wood12 5.4 0.389 (-0.076 to 0.853) fluticasone, triamcinolone, or budesonide all showed 100 -0.423 (-0.531 to -0.315) no difference from oral antihistamines. This was a -1.5 -1.0 -0.5 0 0.5 1.0 χ2 homogeneous finding. Stratification by antihistamine =26.82, df=8, P<0.001 showed significant heterogeneity in trials that used ter- Fig 2 Comparison of effects of intranasal corticosteroids and oral H1 receptor antagonists fenadine (÷2 = 14.6, P < 0.01) or astemizole (÷2 = 12, (antihistamines) on total nasal symptom scores P < 0.01). Stratification by the period of data extraction showed a small homogeneous benefit from intranasal corticosteroids ( − 0.17, − 0.35 to − 0.05) in those trials Ocular symptoms Standardised mean reporting eye symptoms as a single end point. There Study Favours steroid Favours antihistamine Weight (%) difference (95% CI) 2 was, however, significant heterogeneity (÷ = 20.2, Robinson15 2.1 -0.646 (-1.438 to 0.146) P < 0.0005) when eye symptoms had been averaged Brooks23 3.3 -0.382 (-1.008 to 0.244) Bronsky22 18.1 -0.338 (-0.608 to -0.069) over the duration of the trial. Bunnag16 5.7 -0.216 (-0.696 to 0.265) Schoenwetter20 23.3 -0.149 (-0.386 to 0.088) Bernstein21 17.8 0.000 (-0.271 to 0.271) Other outcomes Darnell17 10.7 0.022 (-0.329 to 0.373) 18 13 23 Simpson 5.2 0.030 (-0.470 to 0.530) Global ratings were reported by only two studies. Juniper14 5.1 0.224 (-0.284 to 0.732) Patients randomised to receive intranasal cortico- Wood12 5.6 0.869 (0.387 to 1.351) 11 steroids were 0.26 (95% confidence interval 0.08 to 0.8) Beswick 3.0 0.908 (0.251 to 1.566) times more likely to deteriorate than those patients 100 -0.043 (-0.157 to 0.072) randomised to receive oral antihistamines. There was -1.0 -0.50 0.5 1.0 1.5 thus a significant and homogeneous benefit from χ2=32.4, df=10, P<0.0005 intranasal corticosteroids. Nasal resistance was only Fig 3 Comparison of effects of intranasal corticosteroids and oral H1 receptor antagonists reported by one trial, which did not find any difference (antihistamines) on eye symptoms between treatments.13 None of the studies included in this review separately reported drug scores or quality of life.

Two studies met the inclusion criteria, but we were http://www.bmj.com/ unable to obtain sufficient data for analysis.26 27 Both of Discussion these studies favoured intranasal corticosteroids for treating allergic rhinitis, and inclusion of these studies Our systematic review of the effectiveness of intranasal was unlikely to have altered the combined outcomes. corticosteroids versus oral H1 receptor antagonists The various studies, however, measured symptom (antihistamines) for allergic rhinitis identified 18 scores on different scales. For example in the study by randomised controlled trials that met the inclusion cri- Géhanno and Desfougeres the benefit from intranasal

teria. The meta-analysis of 16 evaluable trials corticosteroids was equivalent to an additional 1.8 days on 24 September 2021 by guest. Protected copyright. confirmed that intranasal corticosteroids were signifi- symptom free per week.24 We believe that effects of this cantly more effective at relieving nasal blockage, magnitude are clinically important. discharge, and itch, and postnasal drip than were oral The results of the pooled data on eye symptoms are antihistamines. Furthermore, all these results were surprising as there was no difference between the homogeneous between studies. This indicates that an effectiveness of intranasal corticosteroids and oral anti- analysis of pooled data from clinical trials strongly sup- histamines, although there was significant heterogen- ports the clinical suspicion that intranasal cortico- eity. Subgroup analysis suggested that this heterogen- steroids are more effective than oral antihistamines for eity was not due to the use of different intranasal such nasal symptoms.1 corticosteroids and oral antihistamines in the various Intranasal corticosteroids were also more effective trials. In some trials, however, the effect of each at relieving sneezing and at reducing total nasal symp- treatment was expressed as a mean score over the toms than oral antihistamines, but there was significant whole study period of 6 to 8 weeks and in others as the heterogeneity between studies. Some heterogeneity mean score in the last 2 weeks of an intervention could be accounted for by differences in scoring symp- period of 8 weeks. The stratified analysis indicated that toms, although only one of the 13 studies showed that much of the heterogeneity resulted from those studies oral antihistamines produced greater relief of sneezing where eye symptoms had been averaged over the than did intranasal corticosteroids, and none of the entire duration of treatment. A possible explanation nine studies showed that oral antihistamines signifi- for this observation is the difference in the onset of cantly improved total nasal symptom scores. Despite action between intranasal corticosteroids and oral the heterogeneity, we suggest that the pooled data antihistamines. However, the difference in onset may favour the use of intranasal corticosteroids for be much smaller than commonly believed.44 Although relieving nasal symptoms. the effect of oral antihistamines on the suppression of

BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com 1627 Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from histamine induced wheal and flare reactions is rapid, Key messages the clinical onset in seasonal allergic rhinitis may take 45 up to 5 hours. Furthermore, although intranasal x Allergic rhinitis is increasing in prevalence and corticosteroids were previously thought to take 3-10 is a common cause of morbidity days before a beneficial effect was observed, recent x studies have shown significant relief of nasal symptoms Intranasal corticosteroids produce greater relief in 12-24 hours.46 47 In addition, continuing treatment from most nasal symptoms than do oral H1 with intranasal corticosteroids may lead to a significant receptor antagonists (antihistamines) inhibition of the early nasal response as well as almost x Intranasal corticosteroids and oral total inhibition of the late nasal response.48 Briefly, we antihistamines show no difference in the relief believe that differences in onset between the intranasal of the eye symptoms of allergic rhinitis corticosteroids and oral antihistamines might explain x Intranasal corticosteroids are more cost effective the observed heterogeneity of the subgroup analysis, than oral antihistamines but we are not convinced that these differences in onset x Intranasal corticosteroids are recommended for of action translate into important clinical differences, first line treatment of allergic rhinitis for the reasons outlined. Despite these reservations the results do not support the widely held view that oral antihistamines allergic rhinitis. There may be a role for oral antihista- are superior to intranasal corticosteroids for control- mines as ancillary treatment, particularly if eye ling eye symptoms in allergic rhinitis.7 We calculated symptoms or nasal itch are not controlled by intranasal that there was no difference between these treatment corticosteroids. modalities when eye symptoms were measured. Intranasal corticosteroids may improve eye symptoms We thank the following people who provided unpublished data by increasing nasolacrimal drainage, or there may be from clinical trials: Eric Lehl (Astra Australia), Margaret Ward an effect from absorption of the corticosteroid. Curran (Glaxo-Wellcome), and Katy Williams-Day and Steven Francom (Pharmacia and Upjohn). Michael Bailey prepared the Intranasal corticosteroids are considered safe. Local illustrations. adverse effects are usually mild (mucosal irritation, Contributors: JMW initiated the study and contributed to epistaxis), and nasal septal perforation is exceptionally reviewing the data and writing the paper; he will act as guaran- rare.49 Clinical and histopathological examination of tor for the paper. MJA conducted the statistical analyses and contributed to reviewing the data and writing the paper. RMP nasal mucosa up to 5.5 years of continuous budesonide contributed to reviewing the data and to writing the paper. 50 use have failed to show significant changes. Intranasal Funding: Astra Pharmaceuticals (Australia) provided library corticosteroids can result in systemic bioavailability,51 but and research assistance and support for presentation at a scien- studies have failed to show significant effects on serum tific meeting. Astra did not provide any other direct financial markers of bone metabolism,52 short term bone support. 53 Conflict of interest: Each of the authors is involved in clinical growth, or cortisol concentrations after stimulation by practice and prescribes both intranasal corticosteroids and oral 54

adrenocorticotrophic hormone. antihistamines for patients. While the support of Astra Pharma- http://www.bmj.com/ First generation oral antihistamines are safe, but ceuticals (manufacturers of budesonide brand nasal spray) is sedative and anticholinergic effects may be trouble- acknowledged, the authors produced this review independently 55 and it was not subject to any editorial review or changes by some. Second generation (non-sedating or low- Astra. The authors believe that no conflict of interest arose dur- sedating) oral antihistamines do not have these effects ing the production of this paper. and are well tolerated. Near fatal and fatal arrhythmias 55 1 International Rhinitis Management Working Group. International have been reported with terfenadine and astemizole, consensus report on the diagnosis and management of rhinitis. Allergy and these drugs are contraindicated in patients with 1994;49(suppl 19):1-34. 2 Fleming DM, Crombie LD. Prevalence of hay fever in England and Wales.

heart or liver disease or when there is concomitant BMJ 1987;294: 279-83. on 24 September 2021 by guest. Protected copyright. treatment with drugs that inhibit the hepatic cyto- 3 Aberg N. Asthma and allergic rhinitis in Swedish conscripts. Clin Exp Allergy 1989;19:59-63. chrome P-450 system. 4 Hopper T, Jenkin M, Carlin C, Giles G. Increase in the self-reported The cost effectiveness of intranasal corticosteroids prevalence of asthma and hay fever in adults over the past generation. Aust J Pub Health 1995;19:120-4. versus oral antihistamines was assessed in three 5 Australian Bureau of Statistics, 1989-90. National health survey. Asthma randomised controlled trials on the treatment of aller- and other respiratory conditions, Australia. Commonwealth Government Printer, 1991. gic rhinitis. An American study showed that if a patient 6 Ziering RW. Immediate and late effects of hay fever. Postgrad Med used terfenadine for more than 11 to 22 days, then flu- 1989;85:183-90. ticasone was a more cost effective choice.56 Two cost 7 Calderon-Zapata MA, Davies RJ. Treatment and management of allergic rhinitis. In: Kay AB, ed. Allergy and allergic diseases. Oxford: Blackwell, effectiveness analyses performed in Canada produced 1997. cost effective ratios of 1:2.5 and 1:5.7 in favour of fluti- 8 Prenner BM. Allergies for all seasons! Allergy Asthma 1997;6:8-9. The Cochrane Collaboration handbook 57 9 Sackett D, Oxman A (eds). . Oxford: casone versus terfenadine and loratadine respectively. Update Software, 1996. We were not able to perform such an analysis on our 10 Munch EP, Soborg M, Norreslet TT, Mygind N. A comparative study of dexchlorpheniramine maleate sustained release tablets and budesonide data, but we did compare the mean daily cost of oral nasal spray in seasonal allergic rhinitis. Allergy 1983;38:517-24. antihistamines in Australia (by asking pharmacists in 11 Beswick KB, Kenyon GS, Cherry JR. A comparative study of beclomethasone dipropionate aqueous nasal spray with terfenadine tablets in four Australian states) with the mean daily cost of seasonal allergic rhinitis. Curr Med Res Opin 1985;9:560-7. intranasal corticosteroids (based on Australian 12 Wood SF. Oral antihistamine or nasal steroid in hay fever; a double-blind double-dummy comparative study of once daily oral astemizole vs twice pharmaceutical benefits schedule). The mean daily daily nasal beclomethasone dipropionate. Clin Allergy 1986;16:195-201. cost of oral antihistamines was 4.5 times that of 13 Lancer JM, Jones AS, Stevens JC, Beckingham E. A comparison by rhinomanometry of beclomethasone and terfenadine in the treatment of intranasal corticosteroids. We believe that the results of seasonal rhinitis. J Laryngol Otol 1987;101:350-4. the North American studies and our data suggest 14 Juniper EF, Kline PA, Hargreave FE, Dolovich J. Comparison of beclomethasone dipropionate aqueous nasal spray, astemizole, and the combi- that intranasal corticosteroids are more cost effective nation in the prophylactic treatment of ragweed pollen-induced than oral antihistamines in the first line treatment of rhinoconjunctivitis. J Allergy Clin Immunol 1989;83:627-33.

1628 BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com Papers BMJ: first published as 10.1136/bmj.317.7173.1624 on 12 December 1998. Downloaded from 15 Robinson AC, Cherry JR, Daly S. Double-blind cross-over trial 39 Hilberg O. Effect of terfenadine and budesonide on nasal symptoms, comparing beclomethasone dipropionate and terfenadine in perennial olfaction, and nasal airways patency following allergen challenge. Allergy rhinitis. Clin Exp Allergy 1989;19:569-73. 1995;50:683-8. 16 Bunnag C, Jareoncharsri P, Wong EC. A double-blind comparison of 40 Meltzer EO. Nasal cytological changes following pharmacological nasal budesonide and oral astemizole for the treatment of perennial intervention. Allergy 1995;50(suppl 23):15-20. rhinitis. Allergy 1992;47:313-7. 41 Ramsdale EH, Kline PA. Comparison of an intranasal steroid 17 Darnell R, Pecoud A, Richards DH. A double-blind comparison of (budesonide) with terfenadine in the treatment of ragweed-induced fluticasone propionate aqueous nasal spray, terfenadine tablets and rhinoconjunctivitis. J Allergy Clin Immunol 1990;85:244. [Abstract.] placebo in the treatment of patients with seasonal allergic rhinitis to grass 42 Ramsdale EH, Kline PA. Comparison of once daily treatment with an pollen. Clin Exp Allergy 1994;24:1144-50. antihistamine (loratadine) and topical steroid (fluticasone propionate) 18 Simpson RJ. Budesonide and terfenadine, separately and in combination, during the ragweed pollen season. J Allergy Clin Immunol 1991;87:298. in the treatment of hay fever. Ann Allergy 1994;73:497-502. [Abstract.] 19 Van Bavel J, Findlay SR, Hampel FC Jr, Martin BG, Ratner P, Field E. 43 Stricker W, Klimas J, Mendelson L, Morris RJ, Reed CE, Shapiro G, et al. Intranasal fluticasone propionate is more effective than terfenadine Intranasal fluticasone propionate is more effective than astemizole for tablets for seasonal allergic rhinitis. Arch Intern Med 1994;154:2699-704. seasonal allergic rhinitis. Ann Allergy 1994;72:86. [Abstract.] 20 Schoenwetter W, Lim J. Comparison of intranasal triamcinolone 44 Nathan RA. Changing strategies in the treatment of allergic rhinitis. Ann acetonide with oral loratadine for the treatment of patients with seasonal Allergy Asthma Immunol 1996;77:255-9. allergic rhinitis. Clin Ther 1995;17:479-92. 45 Meltzer EO, Weiler JM, Widlitz MD. Comparative outdoor study of the 21 Bernstein DI, Creticos PS, Busse WW, Cohen R, Graft DF, Howland WC, efficacy, onset and duration of action, and safety of cetirizine, loratadine, et al. Comparison of triamcinolone acetonide nasal inhaler with astemi- and placebo for seasonal allergic rhinitis. J Allergy Clin Immunol zole in the treatment of ragweed-induced allergic rhinitis. J Allergy Clin 1996;97:617-26. Immunol 1996;97:749-55. 46 Munk ZM, LaForce C, Furst JA, Simpson B, Feiss G, Smith JA, et al. Effi- 22 Bronsky EA, Dockhorn RJ, Meltzer EO, Shapiro G, Boltansky H, LaForce cacy and safety of triamcinolone acetonide aqueous nasal spray in C, et al. Fluticasone propionate aqueous nasal spray compared with ter- patients with seasonal allergic rhinitis. Ann Allergy Asthma Immunol fenadine tablets in the treatment of seasonal allergic rhinitis. J Allergy Clin 1996;77:277-81. Immunol 1996;97:915-21. 47 Selner JC, Weber RW, Richmond GW, Stricker WE, Norton JD. Onset of 23 Brooks CD, Francom SF, Peel BG, Chene BL, Klott KA. Spectrum of sea- action of aqueous beclomethasone dipropionate nasal spray in seasonal sonal allergic rhinitis symptom relief with topical corticoid and oral anti- allergic rhinitis. Clin Ther 1995;17:1099-109. histamine given singly or in combination. Am J Rhinol 1996;10:193-9. 48 Rak S, Jacobson MR, Sudderick RM, Masuyama K, Juliusson S, Kay AB, 24 Géhanno P, Desfougeres JL. Fluticasone propionate aqueous nasal spray et al. Influence of prolonged treatment with topical corticosteroid compared with oral loratadine in patients with seasonal allergic rhinitis. (fluticasone propionate) on early and late phase nasal responses and cel- Allergy 1997;52:445-50. lular infiltration in the nasal mucosa after allergen challenge. Clin Exp 25 Vervloet D, Charpin D, Desfougeres JL. Intranasal fluticasone once daily Allergy 1994;24:930-9. compared with once daily cetirizine in the treatment of seasonal allergic 49 Gawchick SM, Saccar CL. The use of nasal corticosteroids in allergic rhinitis. Clin Drug Invest 1997;13:291-8. rhinitis. Pediatr Asthma Allergy Immunol 1995;9:25-38. 26 Frolund L. Efficacy of an oral antihistamine, loratadine, as compared with 50 Pipkorn U, Pukander J, Suonpaa J, Makinen J, Lindqvist N. Long-term a nasal steroid spray, beclomethasone dipropionate, in seasonal allergic safety of budesonide nasal aerosol: a 5.5-year follow-up study. Clin Allergy rhinitis. Clin Otolaryngol 1991;16:527-31. 1988;18:253-9. 27 Jordana G, Dolovich J, Briscoe MP, Day JH, Drouin MA, Gold M, et al. 51 Lipworth BJ, Seckl JR. Measures for detecting systemic bioactivity with Intranasal fluticasone propionate versus loratadine in the treatment of inhaled and intranasal corticosteroids. Thorax 1997;52:476-82. adolescent patients with seasonal allergic rhinitis. J Allergy Clin Immunol 52 Martinati LC, Sette L, Chiocca E, Zaninotto M, Plebani M, Boner AL. 1996;97:588-95. Effect of beclomethasone dipropionate nasal aerosol on serum markers 28 Sibbald B, Hilton S, D’Souza M. An open cross-over trial comparing two of bone metabolism in children with seasonal allergic rhinitis. Clin Exp doses of astemizole and beclomethasone dipropionate in the treatment Allergy 1993;23:986-91. of perennial rhinitis. Clin Allergy 1986;16:203-11. 53 Wolthers OD, Pederson S. Knemometric assessment of systemic activity 29 Andersson M, Lindqvist N, Svensson C, Ek L, Pipkorn U. Dry powder of once daily intranasal dry-powder budesonide in children. Allergy inhalation of budesonide in allergic rhinitis. Clin Otolaryngol 1994;49:96-9. 1993;18:30-3. 54 Brannan MD, Herron JM, Reidenberg P, Affrime MB. Lack of 30 Dickson DJ, Cruickshank JM. Comparison of flunisolide nasal spray and hypothalamic-pituitary-adrenal axis suppression with once-daily or terfenadine tablets in hay fever. Br J Clin Pract 1984;38:416-20. twice-daily beclomethasone dipropionate aqueous nasal spray adminis- 31 Backhouse CI, Finnamore VP, Gosden CW. Treatment of seasonal aller- tered to patients with allergic rhinitis. Clin Ther 1995;17:637-47. gic rhinitis with flunisolide and terfenadine. J Int Med Res 1986;14:35-41. 55 Simons FER, Simons KJ. The pharmacology and use of H1-receptor-

32 Benincasa C, Lloyd RS. Evaluation of fluticasone propionate aqueous antagonist drugs. N Engl J Med 1994;330:1663-70. http://www.bmj.com/ nasal spray taken alone and in combination with cetirizine in the prophy- 56 Kozma CM, Schulz RM, Sclar DA, Kral KM, Mackowiak JI. A comparison lactic treatment of seasonal allergic rhinitis. Drug Invest 1994;8:225-33. of costs and efficacy of intranasal fluticasone and terfenadine tablets for 33 Drouin MA, Yang WH, Horak F, van de Heyning PH, Kunkel GH, seasonal allergic rhinitis. Clin Ther 1996;18:334-46. Backhouse CI, et al. Adding loratadine to topical nasal steroid therapy 57 Berka C, Chin W. Fluticasone propionate intranasal steroid: cost- improves moderately severe seasonal allergic rhinoconjunctivitis. Adv effectiveness vs antihistamines. J Allergy Clin Immunol 1994;93:165. Ther 1995;12:340-9. [Abstract.] 34 Davies RJ, Lund VJ, Harten-Ash VJ. The effect of intranasal azelastine and (Accepted 15 September 1998) beclomethasone on the symptoms and signs of nasal allergy in patients with perennial allergic rhinitis. Rhinology 1993;31:159-64. 35 Pelucchi A, Chiapparino A, Mastropasqua B, Marazzini L, Hernandez A, Foresi A. Effect of intranasal azelastine and beclomethasone dipropion- on 24 September 2021 by guest. Protected copyright. ate on nasal symptoms, nasal cytology, and bronchial responsiveness to Correction methacholine in allergic rhinitis in response to grass pollens. J Allergy Clin Immunol 1995;95:515-23. Effect of government recommendations on methadone prescribing 36 Harding SM, Heath H. Intranasal steroid aerosol in perennial rhinitis; in south east England: comparison of 1995 and 1997 surveys comparison with an antihistamine compound. Clin Allergy 1976;6:369-72. An error occurred in this paper by John Strang and Janie 37 Lau SK, Wei WI, van Hasselt CA, Sham CL, Woo J, Choa D, et al. A clinical comparison of budesonide nasal aerosol, terfenadine and a combina- Sheridan (28 November, pp 1489-90). The end of the last tion therapy of budesonide and oxymetazoline in adult patients with sentence in the first paragraph should have read: “. . . and perennial rhinitis. Asian Pac J Allergy Immunol 1990;8:109-15. (c) the optimal dosage for methadone maintenance 38 Negrini AC, Troise C, Voltolini S, Horak F, Bachert C, Janssens M. Oral antihistamine/decongestant treatment compared with intranasal cortico- treatment was probably between 50 mg and 100 mg [not steroids in seasonal allergic rhinitis. Clin Exp Allergy 1995;25:60-5. 100 g] daily.”

Fifty years ago The new NHS: Moscow calling

Moscow has discovered that all is not well with our Health commercial chemists, resulting in the neglect of the sufferings of Service, and the other day took the trouble to tell us so in plain the people. We hasten to print this analysis of events lest Moscow English. A broadcast asserted that the Health Bill came to accuse us again of misrepresentation. Eighteen months ago the nothing because “the Right Labourites surrendered to the Meditsinsky Rabotnik (“Medical Worker”) reminded us that “from capitalist Medical Association without even a fight” (Listener, Nov time to time one finds in the Journal lines of which it is difficult to 18). The Labour Government was blamed for not having decide which is the greatest in them—malicious slander or nationalized the pharmaceutical industry; quackery still exists and timeless ignorance.” doctors have become “merchants” in association with the (BMJ 1948;ii:208(suppl))

BMJ VOLUME 317 12 DECEMBER 1998 www.bmj.com 1629