DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

Efficacy and Safety of an Oral Ambroxol Spray in the Treatment of Acute Uncomplicated Sore Throat

Authors

C. de Mey1, J. Patel2, D. R. Lakha3, E. Richter4, S. Koelsch4

Affiliations

1

Key words ▶ sore throat ● ▶ ambroxol ● ▶ ambroxol spray ● ▶ local anaesthetic ●

Abstract

 ACPS – Applied Clinical Pharmacology Services, Mainz-Kastel, Germany  Boehringer Ingelheim (Pty) Ltd, Randburg, South Africa 3  1644 Starling Street, Lenasia South, Johannesburg, South Africa 4  Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim am Rhein, Germany



Objective:  Compare the efficacy and tolerability of oral spray formulations delivering 2.5, 5, and 10 mg ambroxol (AXS) per application (4 actuations/application) in relieving acute sore throat vs. spraying a matched placebo solution. Design:  Multi-centre, placebo-controlled, randomised, double-blind trial with up to 6 daily applications of the assigned medication for up to 3 days. Patients:  511 outpatients with acute sore throat were enrolled, 494 were treated. Treatments:  Up to 6 spray applications per day as needed for up to 3 days. Results:  All treatments led to a reduction in pain intensity (PI); the mean cumulative PIreductions over the first 2 h after the 1st dose

Introduction



received 15.12.2014 accepted 12.02.2015 Bibliography DOI http://dx.doi.org/ 10.1055/s-0035-1547229 Published online: 2015 Drug Res © Georg Thieme Verlag KG Stuttgart · New York ISSN 2194-9379 Correspondence PD Dr. med. C. de Mey ACPS – Applied Clinical ­Pharmacology Services Philippsring 11 D-55252 Mainz-Kastel Germany Tel.:  + 49/6134/180 467 Fax:  + 49/6134/180 468 [email protected]

Although self-limited, acute sore throat caused by viral pharyngitis is highly bothersome. Hence, offering prompt, efficient and safe pain relief must be a prime objective in managing sore throats. For the patients who seek medical advice this might help to avoid undue prescription of antibiotics. For the vast majority of the patients who do not seek such advice this is equally important since it reduces incapacitation. Ambroxol (trans-4-(2-Amino-3,5-dibrombenzyl­ amino)-cyclohexanol; CAS-Registry N° 1868391-5) is a well-known drug substance used systemically since many decades as secretolytic agent with muco-ciliary clearance facilitating properties. Topical ambroxol also has local anaesthetic properties [1, 2]. This led to the development and authorisation of ambroxol lozenges (Mucoangin® Boehringer Ingelheim) for pain relief in acute sore throat. To this purpose a series of randomised, placebo-controlled, double-blind clini-

(SPIDnorm(0–2)) were 24.7, 26.6, 26.0, and 32.2 % (SEM: 0.023) of the predose PI for treatment with placebo, and the 2.5, 5, and 10 mg AXS, respectively. These mean reductions were 2 (CI:  − 3.6; 7.5), 1.3 (CI:  − 4.3; 6.8), and 7.5 (CI: 2.0;13.1) percent points larger than for placebo. The 2.5 and 5 mg AXS were not distinguishable from placebo, but the 10 mg AXS was evidently superior. The numbers needed to treat (NNT) when comparing 10 mg AXS with placebo, were 9.5 and 8.8 for an average pain relief of 33 and 50 % of the maximum achievable effect over the first 2 h. Conclusions:  10 mg AXS showed a statistically significantly superior pain reduction relative to the placebo spray. Treatment with 10 mg AXS reaches an extent of pain relief that can be accepted to be clinically meaningful and was well tolerated.

cal trials had been performed that showed that sucking lozenges containing 20 mg ambroxol was efficacious and well tolerated in the treatment of acute uncomplicated sore throat of recent onset [3–5]. Spray formulations are generally well suited for the treatment of topical conditions of the oropharyngeal cavity. Spraying medication into the oro-pharyngeal cavity provides more direct access to the area of interest than sucking a lozenge. Accordingly, an ambroxol containing spray formulation was developed for the treatment of sore throat. The spray was developed with following goals in mind: to be well efficacious, safe and well tolerated with a convenient and sufficiently flexible dosage regimen; no to use excipients at levels that might result in excipient-related safety issues; and to keep the volume per actuation sufficiently low such that a swallowing reflex can be avoided that otherwise might have precluded the active ingredient to reside at the target site for a sufficiently long time. de Mey C et al. Efficacy and Safety of …  Drug Res

Downloaded by: University of Georgia Libraries. Copyrighted material.

2

This led to the development of test medications consisting of solutions with 3 different ambroxol concentrations (4.46 mg/ mL, 8.93  mg/mL, and 17.86  mg/mL ambroxol hydrochloride, respectively); with an actuation volume of 140 µl and 4 actuations per application, this corresponds with doses of 2.5, 5 and 10 mg ambroxol per application, respectively. Eucalyptus and menthol flavouring were used as further excipients. Control medications consisted of matched eucalyptus and menthol flavoured sprays that were different from the test medications only by the absence of ambroxol. The present publication reports the findings of a multi-centre, randomised, placebo-controlled, clinical trial with regard to the efficacy, safety and tolerability for pain relief in acute uncomplicated sore throat of recent onset using these test sprays relative to spraying a matched placebo of similar composition except for ambroxol. Additionally, in order to put the findings in perspective, a comparison is presented with a meta-analysis of the pooled efficacy data for the ambroxol lozenges.

Patients and Methods



The protocol, patient information and informed consent form of the study were reviewed and approved by an independent ethics committee. The study was planned, conducted, analysed and reported in accordance with the principles of Good Clinical Practice, the Declaration of Helsinki and the provisions for the orderly conduct of clinical trials in the country of conduct. The trial was registered at clinicaltrials.gov. The study was conducted from May 2011 to July 2011 in the Republic of South Africa in ambulatory fashion by primary care physicians who had been explicitly qualified with regard to the medical management of patients with acute sore throat and the orderly conduct of clinical trials.

Study population

In order to be eligible, the patients had to provide witnessed and written informed consent before the start of any trial related procedure. Eligible patients were males or females of any race, at least 18 years of age, with acute uncomplicated sore throat of onset within the last 3 days before enrolment; baseline pain intensity had to be scored at least 6 points on an 11-point pain intensity scale (see below). The following conditions led to exclusion from enrolment: any sign of mouth-breathing due to nasal congestion; throat discomfort causing cough; active pulmonary disease such as bronchopneumonia; known allergy or hypersensitivity to ambroxol, sorbitol, or acetaminophen; drug dependence or alcohol abuse; use of any throat lozenge, throat spray, cough drop, mentholcontaining product, or any product with demulcent properties within 2 h before the intended first administration of the trial medication; use of any analgesic or anti-pyretic within 4 h before the intended first administration of the trial medication (24 h in the case of long-acting analgesics such as naproxen sodium); use of any 'cold medication' (e.  g., decongestant, antihistamine, expectorant, antitussive agent) within 8 h before the intended first administration of the trial medication (exception: secondgeneration antihistamines, which were not prohibited); use of any antibiotic within the last 24 h; more than incidental use of inhaled steroids or beta-agonists during the week prior to screening; use of any other investigational therapy within the de Mey C et al. Efficacy and Safety of …  Drug Res

DrugRes/2014-12-0925/6.3.2015/MPS

last month; pregnancy or lactation; presence of any medical or psychiatric condition which, in the opinion of the investigator, might have put the patient risk by participating in the trial and/ or might have a negative impact on compliance with the protocol and study directives. The sample size for the trial was estimated a priori based on 80 % power, a 2-sided α-error of 0.05, a common standard deviation σ of 0.3 and a mean treatment difference δ of 0.12 between treatments with ambroxol and placebo for the primary efficacy criterion. Assuming a drop-out rate of 10 % between screening and randomisation, a total of at least 440 patients was planned for enrolment with 100 patients assigned per treatment group.

Study medication

Computer-generated lists of random numbers were used for random allocation of individual supplies to trial participants. The randomisation list was generated using a validated system which involved a pseudo-random number generator so that the resulting treatment would be both reproducible and non-predictable. Matched supplies of active and control medication assigned equally to each of the 4 investigational treatments were used. Individualised medication boxes were packed and numbered for each participant according to the randomisation plan. At the site, each eligible patient was assigned the lowest available medication number at the time of randomisation and was only treated with the materials of the corresponding medication pack. Throughout the study until formal closure of the database, the randomisation codes were known only to the central site processing the medication supplies. Investigators, trial staff, outcome assessors, quality control monitors, data entry staff, data analysts, and trial participants were blinded with regard to treatment assignment. Investigational treatments consisted of a first application (4 actuations of active medication at different doses strengths or placebo) followed by pharmacodynamic pain evaluation over the first 3 h after dosing; during this period no further applications were made. Subsequently, up to 6 applications could be made as needed (prn - pro re nata) per day for a total of 3 treatment days. Placebo treatment consisted of the application of a matched eucalyptus- and menthol-containing ethanolic spray that differed from the test spray only by the absence of ambroxol spray. The placebo spray was to be applied in the same way as the active medication. Patients and investigators were blinded in this regard.

Study schedule and experimental conditions

All patients were investigated in a real-life primary care practice setting. After screening for eligibility, baseline pain was scored and pharyngeal redness (investigator assessment on thorough inspection - VRS: 0 = no redness; 1 = slightly red; 2 = markedly red; 3 = severe redness; 4 = severe inflammation) and difficulty swallowing (patients' assessment after drinking 100 mL noncarbonated water scoring difficulty of swallowing on an 11-point ordinal scale from 0 (not difficult) to 10 (very difficult) were assessed. Eligible patients were then assigned at random to their double-blind medication. A first application (4 attenuations) of the assigned medication was made and pain intensity (PI) was scored by means of an 11-point (0–10) numerical rating scale (NRS) 0.25, 0.5, 0.75, 1, 1.5, 2 and 3 h after the first application. During this time, the patients remained in the clinical practice. During this phase, the patients were not allowed to take other

Downloaded by: University of Georgia Libraries. Copyrighted material.

Original Article

DrugRes/2014-12-0925/6.3.2015/MPS

Efficacy and safety criteria Efficacy criteria

At the study clinic, changes in sore throat intensity after the first dose were investigated extensively up to 3 h after dosing. This is a pharmacodynamic evaluation of the extent and time course of sore throat relief after the first dose. Subsequent use was only restricted in terms of the total number of daily applications (up to 6 per day). This latter situation represents the likely way of pharmacotherapeutic use. Efficacy evaluations therefore addressed both the response in a pharmacodynamic approach (response to the first dose) and a pharmacotherapeutic way (symptomatic relief subsequent to the pharmacodynamic evaluation including relatively free use of the medication during the following days). Pain intensity was scored by the patient by means of an 11-point ordinal numerical rating scale (from 0 = not sore to 10 = very sore). For the pharmacodynamic response, pain intensities subsequent to dosing were expressed as arithmetic changes i. e., decreases in pain intensity (PID) from baseline intensity (negative changes reflecting a reduction in PI). The cumulative sum of the post-dosing PID over the first 2, 3 and 24 h after the first dose was weighted for the time elapsed since previous assessment by means of a rectangular area under the curve SPIDAUC (= Σ (Ti −  Ti–1) × PIDTi), in which Ti is the time of assessment in hours and PIDTi is the change in pain score at that time. Additionally, the SPIDAUC over the first 2, 3 and 24 h (τ) after the first dose was expressed as SPIDnorm(0–τ) = SPIDAUC/(τ * PIbaseline). The SPIDnorm(0–τ) thus represents the time-weighted average PID over the first τ hours after the first dose as a ratio of the baseline score; this equals the ratio of the achieved SPIDAUC relative to the maximum achievable AUC; hence, a SPIDnorm of  − 1.0, means that full pain relief (PI down to the zero score level) was achieved already at the first profiling time and was maintained up to last profiling time of the observation window. Similarly, any SPIDnorm represents the effect as a ratio of the maximum achievable effect.

The treatments were compared for SPIDnorm(0–2) as primary criterion; SPIDnorm(0–3) and SPIDnorm(0–24) and the time course of PID were evaluated as secondary efficacy endpoints along with the time to intake of a second dose, the number of applications per treatment day, the efficacy scores of the patient at 3 and 24 h after the first dose and at the end (in the evening) of each treatment day, redness of the pharyngeal mucosa at baseline and end-of-treatment, number of patients discontinued due to lack of efficacy and the overall assessment of efficacy by investigator and patient at the end-of-treatment.

Safety criteria

Untoward changes reported spontaneously by the patient during the visit or recorded in the diary and related answers to nonleading questions by the physician were listed and analysed as adverse events. Additionally, at 3 and 24 h after the first dose, at the end (in the evening) of each treatment day and at the end-of-treatment, the patients scored their evaluation of treatment tolerability. At the end-of-treatment, the overall tolerability was also to be scored by the investigators.

Statistical analyses

Planned analysis:  For the responses to the first dose, the ­SPIDnorm (primary: SPIDnorm(0–2); secondary: SPIDnorm(0–3) and SPIDnorm(0–24)) were analysed by means of an analysis of covariance (ANCOVA) including treatment and centre as fixed effects and the pre-dose baseline PI value as a continuous covariate. Treatment differences were estimated by reference to the adjusted least square means and the corresponding 95 % confidence intervals (CI). All statistical testing was performed using a 2-sided, alpha = 0.05 level of significance. For the analysis of the time courses of the changes in pain intensity (PID) over the first 3 h after first dosing, restricted maximum likelihood based repeated measures approach was used using all available longitudinal post-dosing observations. The statistical model included the fixed, categorical effects of treatment, centre, time and treatment-by-time interaction, with baseline PI as continuous covariate. An unstructured covariance structure was used to model within-patient errors. A logistic regression model including terms for treatment and centre, with baseline PI as a covariate, was fitted to analyse the patient efficacy scores at 3 and 24 h after first dosing, at the end of each treatment day and at the end-of-treatment. The Kaplan-Meier estimator was used to contrast the treatments with regard to the 'time to second spray application'. The log rank test was used to evaluate differences between the treatment groups. The Cochran-Mantel-Haenszel test, stratified by the pre-dose baseline score, was used for the scores of the redness of the pharyngeal mucosa at end-of-treatment. Descriptive statistics by treatment group were used to summarise the number of spray applications per day and the number of patients who discontinued treatment due to lack of efficacy. Post-hoc analyses:  In order to put the data in perspective, the SPIDnorm(0–2) effects for the top spray dose (10 mg ambroxol per application) were compared with a meta-analysis of the pooled data from the efficacy studies with ambroxol lozenges containing 20 mg ambroxol. This pooled database consists of 7 multicentre, randomised, placebo-controlled parallel-group trials that were carried out with similar methodology; the database de Mey C et al. Efficacy and Safety of …  Drug Res

Downloaded by: University of Georgia Libraries. Copyrighted material.

medications, sucking lozenges, using toothpaste, mouthwash or breath spray, smoking, taking chewing gum or candies, eating and drinking. After completion of the measurement at 3 h after the first dose, the patients were allowed to leave the practice. A second dose of medication could be taken as needed, but not earlier than 3 h after the first; treatment (up to 6 applications as needed per day) was to be continued for up to 3 days. From 3 h after the first dose onwards, 325 mg p.o. paracetamol could be taken in the event of intolerable pain; concomitant treatment with other analgesics, anti-inflammatory agents, steroids (oral, inhaled, and topical), expectorants, antitussives was not allowed. During this ambulatory phase, PI-scores were recorded at 4, 6, 12, and 24 h after the first dosing. Efficacy was scored by the patients at 3 and 24 h after the first dosing and at the end (in the evening) of each treatment day. At the end of treatment, the patients returned to the study clinic and the diary records were collected and reviewed with regard to medication usage and compliance (diary records and return of medication containers), usage of other medications (diary records), and daily efficacy and tolerability scores; additionally at that time, difficulty of swallowing (see above) was scored and redness of the pharyngeal mucosa was evaluated (see above) for comparison with the baseline score, and the investigator made an overall assessment of efficacy and tolerability over the course of treatment.

Original Article

counts 816 and 841 evaluable patients treated with placebo and ambroxol lozenges 20 mg, respectively. 5 of the related studies and a meta-analysis of them have been published previously [5, 14]. The extended meta-analysis is being submitted for publication. With the data obtained from the spray study, also a post-hoc cumulative proportion of responders analysis (CPRA) was carried out as suggested by Farrar et al. [6, 7] for the evaluation of such pain relieving effects. Based on these CPRA, the Absolute Risk Difference (ARD) and Numbers Needed to Treat (NNT = 1/ ARD) were calculated for cut-offs of an average pain relief of 33 or 50 % over the first 2 (SPIDnorm(0–2)) and 3 h (SPIDnorm(0–3)) after a first dose.

Results



Demography, patient disposition and baseline features

A total of 511 patients were enrolled; 17 patients did not meet all eligibility criteria; 494 patients were randomised and treated. 11 patients discontinued from the trial prematurely, 6 due to adverse events, 3 due to lack of efficacy, one patient was lost to

follow-up, and a further patient discontinued prematurely for ▶  Fig. 1). other reasons ( ● Demography is detailed in ●  ▶  Table 1. 47 % of the treated patients were males, 53 % females. 53 % of the patients were black, 17.4 % Asian, and 27.9 % were white. The patients were 33.0 ± 12.6 years of age on average. The mean ( ± SD) duration since disease onset was 1.4 ± 0.8 days. On inspection of the throat at baseline, redness of the throat was slight in 14.6 %, marked in 54.5 %, and severe in 26.3 % of the treated patients; in 4.3 % there was severe inflammation. Pain intensity (PI) was to be scored on an 11-point ordinal numerical rating scale (from 0 = not sore to 10 = very sore); in order to be eligible the baseline score had to be at least 6. The distribution of the baseline PI is detailed in ●  ▶  Table 2. 6.9 % and 3.2 % had very high scores (either 9 or 10/11); for the remaining patients, scores of 6, 7, and 8 were about equally frequent. There was no relevant difference among the patient groups in this regard. In 17 patients (3.4 %), the use of co-medication (mostly paracetamol) was reported during the first 3 h after the first dose. After this trial phase, paracetamol was permitted and was used by 17.4 % of the patients, without relevant difference between the Fig. 1  Patient disposition.

Table 1  Patient demography.

Number of patients Sex [N ( %)] Male Female Race [N ( %)] Asian Black White Missing Age [years] *  Weight [kg] *  BMI [kg/m²] * 

Placebo

Ambroxol 2.5 mg

Ambroxol 5 mg

123 (100.0)

124 (100.0)

124 (100.0)

123 (100.0)

494 (100.0)

57 (46.3) 66 (53.7)

60 (48.4) 64 (51.6)

56 (45.2) 68 (54.8)

59 (48.0) 64 (52.0)

232 (47.0) 262 (53.0)

22 (17.9) 63 (51.2) 35 (28.5) 3 (2.4) 32.9 (13.8) 74.0 (16.6) 26.5 (6.2)

20 (16.1) 67 (54.0) 35 (28.2) 2 (1.6) 34.3 (13.3) 73.8 (18.8) 26.6 (6.9)

22 (17.7) 66 (53.2) 33 (26.6) 3 (2.4) 31.8 (11.9) 74.1 (17.1) 26.6 (6.0)

22 (17.9) 66 (53.7) 35 (28.5) 0 33.1 (11.3) 73.1 (17.5) 26.3 (6.2)

86 (17.4) 262 (53.0) 138 (27.9) 8 (1.6) 33.0 (12.6) 73.7 (17.5) 26.5 (6.3)

 * : mean ± SD

de Mey C et al. Efficacy and Safety of …  Drug Res

Ambroxol 10 mg

Total

Downloaded by: University of Georgia Libraries. Copyrighted material.

DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

Table 2  Baseline pain.

Number of patients [N ( %)] by baseline PI PI: 6 PI: 7 PI: 8 PI: 9 PI: 10 Missing Mean baseline PI (SD)

Ambroxol 2.5 mg

Ambroxol 10 mg

Total

123 (100.0)

124 (100.0)

124 (100.0)

123 (100.0)

494 (100.0)

32 (26.0) 44 (35.8) 32 (26.0) 10 (8.1) 5 (4.1) 0 7.28 (1.07)

27 (21.8) 47 (37.9) 39 (31.5) 8 (6.5) 3 (2.4) 0 7.30 (0.96)

38 (30.6) 40 (32.3) 37 (29.8) 7 (5.6) 1 (0.8) 1 (0.8) 7.13 (0.95)

35 (28.5) 38 (30.9) 34 (27.6) 9 (7.3) 7 (5.7) 0 7.31 (1.13)

132 (26.7) 169 (34.2) 142 (28.7) 34 (6.9) 16 (3.2) 1 (0.2) 7.26 (1.03)

treatment groups. The use of second-generation antihistamines was not prohibited; since such medications may cause dry mouth it is noteworthy that such medication was used only by 6 patients, one of whom as treated with placebo, but none was treated with the 10  mg ambroxol spray. 32 patients were excluded from the per-protocol dataset due to protocol deviations that were considered major (11, 8, 8, and 5 for treatment with placebo, 2.5, 5, and 10 mg ambroxol (per application), respectively); 17, 10, 7, and 1 deviation related to lack of medication compliance, incorrect diagnosis, use of prohibited co-medication, and missing diary, respectively. Patients were to take up to 6 spray applications per day as needed for up to 3 days. The amount of study medication used by each patient was determined by weighing the spray bottles before and after being used by the patient. Compliance (in percent) was calculated as the ratio of the estimated amount of medication used (measured weight difference) related to the use recorded in the patient diary (one spray = 0.56 g). Based on these rough estimates, two-thirds of the patients had a compliance in the range of 80–120 %. Deviations in compliance based on the weight of the spray bottle were not considered important protocol violations, which would have excluded patients from the per protocol set.

Efficacy

Ambroxol 5 mg

Administration of the first dose of the investigational treatments ▶  Fig. 2). From the resulted in a distinct relief of pain intensity ( ● time courses of the PID it is obvious that the treatments with 2.5 mg and 5 mg ambroxol per application did not differ from placebo, whereas the reduction in pain intensity was distinctly better with 10 mg ambroxol per application. This is confirmed by the least-square estimated treatment means and the estimated mean differences vs. placebo (plus 95 % CI) for the ­SPIDnorm ▶  Table 3). Over over the first 2, 3 and 24 h after the first dose ( ● the first 2 h after the first dose of the placebo spray, there was an average reduction in pain intensity of 24.7 % of the maximum achievable effect; over the first 2 h after the first dose of the 10 mg ambroxol spray, the average reduction was 32.2 % of the maximum achievable effect i. e., 7.5 % (CI: 2.0 to 13.1 %) larger as for placebo. The corresponding cumulative proportion of responders analysis (CPRA) plot is presented in ●  ▶  Fig. 3. This CPRA yields absolute risk differences (ARD) of 10.6 % and 11.4 % for an average pain relief of at least 33 or 50 % of the maximum achievable effect over the first 2 h; this corresponds with Numbers Needed to Treat (NNT) of 9.5 (95 % CI: 4.4 to ∞) and 8.8 (95 % CI: 4.8 to 52.3), respectively. The pain reduction after application of the 10 mg ambroxol spray was statistically significantly superior to placebo already at the first profiling time point 15 min after dosing (mean

Fig. 2  Time course of the mean ( ± SEM) change in pain intensity from pre-treatment baseline (BL) following the administration of the first spray application of placebo and the ambroxol sprays containing 2.5, 5, and 10 mg ambroxol per application (FAS: fully analysis dataset).

­ stimated difference in PID from placebo:  − 0.476; CI:  − 0.819 e to  − 0.132; p: 0.0068) lasting well up to 2 h after dosing (mean estimated difference in PID from placebo:  − 0.492; CI:  − 0.982 to  − 0.002; p: 0.0492); at 3 h after dosing, the distinction was less (mean estimated difference in PID from placebo:  − 0.346; CI:  − 0.873 to 0.182; p: 0.1978). The median time to perceptible pain relief after the first dose was 15 min for each treatment and there was no difference among the treatments in this regard. In the patients treated with the 10 mg ambroxol spray, the need for a second application tended to be less urgent (52.8 % and 65 % of patients taking the second application within 4 or 5 h, respectively) than in the patients treated with placebo (64.2 % and 80.5 % taking the second application within the first 4 or 5 h, respectively). The mean time ( ± SD) to 2nd application for the patients with a recorded time within the first 24 h was 4.91 ± 3.09 and 3.82 ± 1.35 for treatment with 10 mg ambroxol and placebo, respectively. Patients treated with the ambroxol containing sprays tended to use the medication less often. The mean ( ± SD) total number of applications over the course of the study was 8.0 ± 4.8 in the patients with the 10 mg ambroxol spray vs. 9.3 ± 5.2 in the patients treated with placebo. Although the patients' ratings of efficacy at the end of each treatment day tended to be higher when treated with the 10 mg ambroxol spray, there was no statistically significant difference from the placebo treatment in this regard.

de Mey C et al. Efficacy and Safety of …  Drug Res

Downloaded by: University of Georgia Libraries. Copyrighted material.

Placebo

DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

Table 3  Estimated mean (SEM) pain reduction for each treatment and their differences vs. placebo.

N

estimated mean

Ambroxol dose

N

(SEM) (0–2) (0–3) (0–24) (0–2) (0–3) (0–24) (0–2) (0–3) (0–24)

123 123 123

 − 0.247 (0.023)  − 0.270 (0.024)  − 0.435 (0.028)

2.5 mg 2.5 mg 2.5 mg 5 mg 5 mg 5 mg 10 mg 10 mg 10 mg

124 124 124 123 123 124 123 123 124

Ambroxol minus PL estimated mean

estimated mean

(SEM)

(SEM)

 − 0.266 (0.023)  − 0.289 (0.024)  − 0.456 (0.028)  − 0.260 (0.023)  − 0.283 (0.024)  − 0.469 (0.029)  − 0.322 (0.023)  − 0.337 (0.024)  − 0.508 (0.028)

 − 0.020 (0.028)  − 0.019 (0.029)  − 0.022 (0.035)  − 0.013 (0.028)  − 0.014 (0.029)  − 0.035 (0.035)  − 0.075 (0.028)  − 0.067 (0.029)  − 0.074 (0.035)

95 % CI ( − 0.075;0.036) ( − 0.077; 0.038) ( − 0.090; 0.047) ( − 0.068; 0.043) ( − 0.071; 0.044)  − (0.104; 0.034) ( − 0.131; − 0.020) ( − 0.125; − 0.010) ( − 0.142;  − 0.005)

p-value 0.4847 0.5119 0.5344 0.6469 0.6455 0.3203 0.0079 0.0222 0.0365

trast of the treatments adjusted for redness at baseline confirmed the 10 mg ambroxol spray to be statistically significantly superior in this regard (p: 0.0420); no such distinction was seen for the treatments with the 2.5 mg (p: 0.3584) and 5 mg spray (p: 0.3910).

Safety and tolerability

The treatments were in general well tolerated. The safety database includes all treated patients and is the same as that used for the efficacy evaluation. Treatments were to last up to 3 days, but could be stopped earlier if there was no further need for treatment in the patient's opinion. On average, the treatment duration lasted 2.2 ± 0.8 days in each of the active treatment groups and 2.3 ± 0.8 days in the placebo group with a maximum of 3 days in all groups. Fig. 3  Cumulative proportion of responders analysis for treatment with the 10 mg ambroxol spray and placebo. The CPRA shows, for each treatment, the cumulative proportion of patients by level of average pain relief reached over the first 2 h after a first spray application (SPIDnorm(0–2)). In such CPRA graph, the x-axis represents cut-off points in percentage of change in pain intensity, and the y-axis shows the cumulative proportion of patients that have achieved that level of response or higher. The distance between the lines of 2 treatment groups is the absolute risk difference (ARD) between the 2 groups.

3 patients discontinued from the trial prematurely due to lack of efficacy: 2 patients in the placebo group and 1 patient in the ambroxol 2.5 mg group. No patients in the ambroxol 5 mg or 10 mg groups discontinued the trial medication due to lack of efficacy. At the end of treatment, the investigators and the patients were to score their overall evaluation of the treatment efficacy: for 22.8 %/17.9 % of the patients, the treatment’s efficacy was scored as excellent by the patient/investigator in the placebo group; in contrast, for 37.4 %/33.3 % of the patients, the treatment’s efficacy was scored as excellent by the patient/investigator in the 10 mg ambroxol spray group. Overall, treatment with the 10 mg ambroxol spray tended to be superior to placebo in this regard particularly in the evaluation by the patients (odds ratio: 1.745; CI: 1.067–2.852), but also, albeit less distinct for the evaluation by the investigators (odds ratio: 1.548; CI: 0.980–2.445); no such difference was seen for the lower strength ambroxol sprays. At the end of treatment, more patients showed "no redness" on inspection of the oro-pharynx when treated with 10  mg ambroxol spray (48.0 %) in contrast to placebo (35.8 %). The conde Mey C et al. Efficacy and Safety of …  Drug Res

Adverse events

Adverse events were reported in 8.9 %, 12.1 %, and 10.6 % of patients in the 2.5, 5, and 10 mg ambroxol treatment groups, respectively, and 9.8 % of patients in the placebo group (an overview is presented in ●  ▶  Table 4). There were no serious AE. One AE was reported as severe: one patient reported severe pharyngitis starting on the third day of treatment with placebo; the patient was discontinued due to this adverse event, was treated, and recovered. The event was not considered related to the investigational trial medication. 6 patients discontinued prematurely from the trial due to AE: one patient treated with placebo (severe pharyngitis considered not related); 2 patients treated with the 2.5 mg ambroxol spray (1 patient with mild cough and one patient with moderate aggravation of sore throat; both were considered not related); one patient treated with 5 mg ambroxol (mild dry throat and mild oral hypoaesthesia, both AEs considered related); 2 patients treated with 10 mg ambroxol (one patient due to moderate headache and one due to moderate aggravation of sore throat; both were considered not related to the investigational trial medication). The frequency of patients with adverse events by preferred term, which were reported in more than 1 % of patients in any treatment group is presented in ●  ▶  Table 5. The most frequently reported adverse event by MedDRA preferred term was headache, reported in 0.8 %, 3.2 %, and 2.4 % of patients in the 2.5, 5, and 10 mg ambroxol treatment groups, and 2.4 % of patients in the placebo group.

Downloaded by: University of Georgia Libraries. Copyrighted material.

Placebo (PL) SPIDnorm

DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

Table 4  Number and percentage of patients with any adverse event.

Number ( %) of patients – Total Patients with any adverse event Patients with severe adverse events Patients with adverse drug reactions Patients discontinued due to AE Patients with serious adverse events

Placebo

Ambroxol 2.5 mg

Ambroxol 5 mg

Ambroxol 10 mg

123 (100.0) 12 (9.8) 1 (0.8) 3 (2.4) 1 (0.8) 0

124 (100.0) 11 (8.9) 0 3 (2.4) 2 (1.6) 0

124 (100.0) 15 (12.1) 0 7 (5.6) 1 (0.8) 0

123 (100.0) 13 (10.6) 0 5 (4.1) 2 (1.6) 0

Table 5  Number and percentage of patients by preferred AE-term for events reported in at least 1 % of any treatment group. 123 (100.0) 12 (9.8) 3 (2.4) 2 (1.6) 2 (1.6) 0 0 0 0 2 (1.6) 0

124 (100.0) 11 (8.9) 1 (0.8) 1 (0.8) 1 (0.8) 0 2 (1.6) 2 (1.6) 2 (1.6) 0 0

Overall tolerability scores

Statistical analysis of patient assessment of tolerability at 3 h and 24 h after the first spray application and at the end of each treatment day as well as the final global assessment of tolerability by the investigator and the patient showed no relevant treatment differences between ambroxol and placebo.

Discussion



Efficacy

Although self-limited, sore throat is highly bothersome. Hence, offering prompt, efficient and safe pain relief must be a prime objective in managing sore throats. Several simple remedies are generally advocated: getting plenty of rest (either in or out of bed), drinking plenty of water to prevent dehydration, gargling with warm salty water, drinking warm liquids (tea or broth) or cool liquids, eating gelatine desserts or flavoured ice, suing a cool mist vaporiser to relive throat dryness, etc. In cases that are refractory to such remedies, medicinal treatment may be quite useful. Ambroxol is a local anaesthetic agent [1, 2] with well-documented secretolytic and secretomotoric actions [8, 9] and ancillary anti-inflammatory [10] and antioxidant properties [11–13]. Lozenges containing 20 mg ambroxol (Mucoangin® Lozenges) have been developed and were authorised for pain relief in uncomplicated acute sore throat [5]. Presently, a randomised, placebo-controlled, double-blind clinical trial is reported that investigated the efficacy, safety and tolerability for pain relief in acute sore throat by means of another form of access to the oro-pharyngeal cavity i. e., a spray formulation. Test medications consisted of solutions of 3 different ambroxol concentrations (4.46  mg/mL, 8.93  mg/mL, and 17.86  mg/mL ambroxol HCl, respectively); with an actuation volume of 140 µl and 4 actuations per application, this corresponds with doses of 2.5, 5 and 10  mg ambroxol per application, respectively.

124 (100.0) 15 (12.1) 4 (3.2) 3 (2.4) 1 (0.8) 0 1 (0.8) 0 0 0 2 (1.6)

123 (100.0) 13 (10.6) 3 (2.4) 3 (2.4) 0 3 (2.4) 0 1 (0.8) 1 (0.8) 0 0

Ambroxol is poorly soluble and the development of higherstrength aqueous solutions would have required increasing the concentration of surfactants, e. g., polysorbate and co-solvents like ethanol. Such change, particularly with regard to ethanol, could result in excipient-related risk and inconvenience. Accordingly, the development of a spray with an application dose higher than 10 mg was not pursued. The test solutions also contain eucalyptus and menthol as further excipients; 10 % ethanol is added to facilitate the dissolution of such aetheric flavouring oils. The control medication ("placebo") was a matched eucalyptus and menthol containing spray that was different from the test medications only by the absence of ambroxol. The spray device is similar to other such pump devices for topical administration of medication in the oro-pharyngeal cavity. With such device medication is targeted directly to the site of therapeutic action: with the proposed spray angle of 60 ° and a distance of about 5 cm from the nozzle to the target region, a round area of about 6 cm in diameter is covered by the spray readily. Droplets are well distributed in the oro-pharyngeal cavity without reaching the lower respiratory tract to a relevant extent. Although obviously taking reference upon the efficacy of lozenges containing ambroxol (see also below), such oral spray product has unique features: sucking a lozenge or spraying directly to the target yield essentially different rates of delivery to the site of therapeutic action and a different spread throughout the oro-pharyngeal cavity; also, sucking a lozenge actively and intentionally amplifies the salivary flow; this dilutes and washes the active ingredient off the site of action; accordingly, the residence time of active ingredient at the site of action might be shorter and its concentration lower than with the spray. A direct comparison between the 2 formulations might have been interesting, but is hardly feasible since this would have required a double-dummy approach to secure objectivity. Considering that the "placebo" treatments (sucking a minty lozenge or spraying an eucalyptus and menthol flavoured spray) cannot be considered inactive, such comparison would have been seriously confounded. de Mey C et al. Efficacy and Safety of …  Drug Res

Downloaded by: University of Georgia Libraries. Copyrighted material.

Number ( %) of patients – Total Total with any adverse event Headache Pharyngitis Nausea Dry mouth Cough Dypepsia Condition aggravated Upper respiratory tract infection Hypoaestheisa oral

Original Article

DrugRes/2014-12-0925/6.3.2015/MPS

Accordingly, we first focused on evidencing whether and to which extent such spray formulations might be effective and well tolerated in the treatment of acute sore throat. To this purpose, similar trial methods were used as in the extensive clinical trial database already available on the ambroxol containing lozenges in this indication [5]: outpatients with acute uncomplicated sore throat were assigned at random to double-blind treatment with either placebo or one of the ambroxol sprays (delivering 2.5, 5, or 10 mg ambroxol per application) in parallelgroup fashion. The reduction in pain intensity (evaluated on 11-point ordinary numerical rating scale) over the first 3 h after a first dose of the assigned medication was evaluated at the study clinic; subsequently the patients left the clinic and they recorded efficacy and tolerability scores daily in their diary. A second dose of the medication could not be taken earlier than 3 h after the first; during the ambulatory up to 6 doses (each time 4 actuations) could be applied daily as needed for up to 3 days. A close-out visit at the study clinic took place at the end of treatment. Application of the control ("placebo") spray was associated with a distinct reduction in pain intensity; application of the ambroxol sprays that delivered doses of 2.5 and 5 mg ambroxol per application were not distinguishable from placebo for any efficacy criterion. In contrast, the drop in pain intensity with the 10 mg ambroxol spray was statistically significantly larger than for the control treatment already at the first profiling 15 min after dosing and this difference remained distinct well up to 2 h after dosing. Over the first 2 h after the first dose of the placebo spray, there was an average reduction in pain intensity of 24.7 % of the maximum achievable effect (SPIDnorm(0–2)); over the first 2 h after the first dose of the 10 mg ambroxol spray, the average reduction was 32.2 % of the maximum achievable effect i. e., 7.5 % (95 % CI: 2.0 to 13.1 %) larger. Over the first 3 h (SPIDnorm(0–3)) the difference between the treatment groups was 6.7 % (95 % CI: 1.0 to 12.5 %). In terms of the SPIDnorm(0–2) the effect for the spray containing 10 mg ambroxol is smaller, but well in the range of the effects observed with the lozenges containing 20  mg ambroxol: ­reference is taken upon the pooled database that consists of 7 multi-centre, randomised, placebo-controlled, double-blind, parallel-group trials that were carried out with similar methodology comparing the effects of sucking a lozenge containing de Mey C et al. Efficacy and Safety of …  Drug Res

20 mg ambroxol (N: 841) with sucking a matched placebo ­lozenge (N: 816). A post-hoc meta-analysis of these data for the pain relieving effect over the first 2 h after a first lozenge dose (SPIDnorm(0–2)) yields an average pain relieving effect of the ambroxol lozenges that was 11 % (95 % CI: 9 to 14 %) larger than ▶  Fig. 4). for sucking a control lozenge ( ● There is no ruling on the meaningfulness of such effects: one may argue that the effects of the placebo treatment are already quite large and that the extra benefit from the use of active medication is quite small in comparison. At least 2 aspects need to be taken into account when assessing the observed effects in terms of relevance. Firstly, sore throat is a self-limiting condition, which heals spontaneously in most patients over the course of 2–3 days; this is well reflected by the changes from pre-dose baseline observed in the patients treated with placebo. Secondly, spraying an eucalyptus and menthol tasting placebo solution into the oro-pharynx is not an "inactive" treatment. In spite of these obvious constraints there is interest in "measuring" meaningfulness. In their critique of an earlier publication that presented the outcome of the clinical trials with the ambroxol lozenges [5], Chenot et al. suggested a probability of 59 % that a patient treated with ambroxol lozenges achieves a greater or faster pain reduction within 3 h than when treated with the control [14]; this value corresponds with the so-called probability index ("relative effect") advocated by Kieser et al. for sample size estimates of studies aiming at demonstrating both a statistically significant and clinically relevant effect [15]. A better approach might be to identify, which change in pain intensity is likely to be clinically relevant and then to compare the treatments with regard to their likelihood of yielding such relevant improvement. To this purpose, firstly reference was taken on Farrar et al. who introduced a cut-off for "clinically important changes" in acute pain relief of 33 % from baseline [6, 16]. Also, using a 0–10 numeric scale, Cepeda et al. identified changes of 20 %, 35 %, and 45 % as cut-offs for what was to be perceived by patients with moderate postoperative pain as "minimal improvement", "much improvement" and "very much improvement" from pretreatment baseline, respectively [17]. These cut-offs are in agreement with those reported in further trials that are pertinent to the validation and understanding of pain intensity scores by means of numerical or verbal rating scales in pain trials in emergency

Downloaded by: University of Georgia Libraries. Copyrighted material.

Fig. 4  Forest plot of the SPIDnorm over the first 2 h after the first lozenge comparing lozenges containing 20 mg ambroxol with matched placebo. Metaanalysis of 7 randomised, placebo-controlled, double-blind, parallel-group clinical trials; this analysis is an extension of a previously published pooled analysis of 5 trials carried out with ambroxol containing lozenges under similar methodological conditions [5, 14].

DrugRes/2014-12-0925/6.3.2015/MPS

Original Article

care (see for instance Bijur et al., 2011 [18]; Kendrick and Strout, 2005 [19]; Farrar, 2010 [7]). Secondly, a cumulative proportion of responders analysis (CPRA) was performed as suggested by Farrar et al. [6, 7]; from this CPRA ▶  Fig. 3), the Absolute Risk Difference (ARD) and Numbers ( ● Needed to Treat (NNT = 1/ARD) were calculated for Minimum Efficacy cut-offs selected on the basis of these considerations. This CPRA yields absolute risk differences (ARD) of 10.6 % and 11.4 % for an average pain relief of at least 33 or 50 % of the maximum achievable effect over the first 2 h; this corresponds with Numbers Needed to Treat (NNT) of 9.5 (95 % CI: 4.4 to ∞) and 8.8 (95 % CI: 4.8–52.3), respectively. Accordingly, although spraying a placebo solution into the oro-pharyngeal cavity led to a distinct reduction in pain intensity, treatment with the 10 mg ambroxol spray is more likely to reach an extent of pain relief that can be accepted to be clinically meaningful. This benefit was most distinct after the first dose, but was confirmed also in the further course of treatment. During the subsequent ambulatory phase, treatment with the 10 mg ambroxol spray was confirmed to be superior to placebo since the need for a second application tended to be less urgent and the patients tended to use less prn-medication. Although the patients' ratings of efficacy at the end of each treatment day tended to be higher when treated with the 10 mg ambroxol spray, there was no statistically significant difference from the placebo treatment in this regard. At the end of treatment, for 22.8 %/17.9 % of the patients, the treatment’s efficacy was scored as excellent by the patient/investigator in the placebo group; in contrast, for 37.4 %/33.3 % of the patients, the treatment’s efficacy was scored as excellent by the patient/investigator in the 10 mg ambroxol spray group. Also, at the end of treatment, more patients showed "no redness" on inspection of the throat when treated with 10 mg ambroxol spray (48.0 %) in contrast to placebo (35.8 %).

­ vidence of superiority. Although spraying a placebo solution e into the or-pharyngeal cavity led to a distinct reduction in pain intensity, treatment with the 10 mg ambroxol spray was shown to be more likely to reach an extent of pain relief that can be accepted to be clinically meaningful. Additionally, treatment ­ with the ambroxol sprays proved to be very well tolerated.

Safety and tolerability

The South African Ambroxol Study Principal Investigators: Dr Deepak Lakha; Dr Hemant Makan; Dr Jaco Jurgens; Dr Rohit Dulabh; Dr Emlyn Janari; Dr Essack Mitha; Dr Phillip Nel; Dr Zubar Vawda; Dr Uttam Govind; Dr Visvakuren Naidoo; Dr Pradipkumar Patel; Dr Stephen Schmidt; Dr Julian Trokis; Dr Shaunagh Emanuel; Dr Johan Lombaard.



The manuscript reports data from a randomised, placebo-controlled, double-blind clinical trial organised and funded by the Boehringer Ingelheim GmbH, Division Medicine, Ingelheim am Rhein, Germany. There are no competing interests to declare. Conception and design of the study was developed by the sponsor under the responsibility of S. Koelsch and E. Richter. Orderly implementation of the study was managed by J. Patel in close collaboration with D. Lakha as principal and coordinating investigator. Statistical analysis was the responsibility of E. Richter. Overall evaluation, interpretation of the data and preparation of the manuscript was the joint effort of all listed authors as coordinated by C. de Mey. All authors reviewed the manuscript critically and approved the final version to be published. C. de Mey was remunerated by the sponsor for analysing the data. D. Lakha was remunerated by the sponsor for his services as principal and coordinating investigator. S. Koelsch, E. Richter, and J. Patel are employees of the sponsor who provided important contributions to the design and interpretation of the trial. At no time the sponsor took undue influence on the data presented here and the conclusions drawn from them.

Acknowledgment



References

Conclusion



Accordingly, it can be concluded that spray solutions delivering 2.5 mg or 5 mg ambroxol have a pain relieving effect that is not distinguishable from that of spraying a matched placebo solution without ambroxol. In contrast, a spray solution delivering 10  mg ambroxol per application (4 actuations/application) showed a statistically significantly superior pain relieving effect. The superiority is well detectable already 15 min after dosing and lasts well up to 2 h after a first dose. The effect is most ­distinct after the first spray application; over the first 2 h after dosing, relative to placebo, the effect size for the 10 mg ambroxol spray was less than the pooled effect size for lozenges containing 20  mg ambroxol relative to sucking a placebo lozenge ­(SPIDnorm(0–2) difference of 11 % points), but well in the range of effect sizes across the several lozenges trials. During the further course of ambulatory treatment using the spray as needed with up to 6 daily doses (each time 4 actuations) for up to 3 days, the treatment with the 10  mg ambroxol spray shows further

1 Tyrakowski T, Greczko I, Sedlaczek A et al. Electrophysiological investigation of the effects of Ambroxol on the transepithelial Na +  ion transport pathway in airways. Pol J Pharmacol 1998; 50: 31–38 2 Weiser T, Wilson N. Inhibition of tetrodotoxin (TTX)-resistant and TTXsensitive neuronal Na( + ) channels by the secretolytic ambroxol. Mol Pharmacol 2002; 62: 433–438 3 Schutz A, Grund HJ, Pschorn U et al. Local anaesthetic properties of ambroxol hydrochloride lozenges in view of sore throat – Clinical proof of concept. Arzneim.-Forsch/Drug Res 2002; 52: 194–199 4 Fischer J, Pschorn U, Vix JM et al. Efficacy and tolerability of ambroxol hydrohloride lozenges in sore throat – Randomised, double-blind, placebo-controlled trials regarding the local anaesthetic properties. Arzneim.-Forsch/Drug Res 2002; 52: 256–263 5 de Mey C, Peil H, Kölsch S et al. Efficacy and safety of ambroxol lozenges in the treatment of acute uncomplicated sore throat. EBM-based clinical documentation. Arzneim.-Forsch/Drug Res 2008; 58: 557–568 6 Farrar JT, Berlin JA, Strom BL. Clinically important changes in acute pain outcome measures: a validation study. J Pain Symptom Manage 2003; 25: 406–411 7 Farrar JT. Advances in clinical research methodology for pain clinical trials. Nat Med 2010; 16: 1284–1293 8 Disse BG. The pharmacology of Ambroxol - review and new results. Eur J Respir Dis Suppl 1987; 153: 255–262

de Mey C et al. Efficacy and Safety of …  Drug Res

Downloaded by: University of Georgia Libraries. Copyrighted material.

The treatments were in general equally well tolerated. The AE were mostly mild-to-moderate and regressed readily without sequels. Rarely, the investigational treatments had to be discontinued due to AE. In this regard, there was no noteworthy difference between the treatments with the ambroxol containing ambroxol sprays and the placebo spray.

Author Contributions

Original Article

14 Chenot JF, Weber P, Friede T. Efficacy of Ambroxol lozenges for pharyngitis: a meta-analysis. BMC Fam Pract 2014; 15: 45 15 Kieser M, Friede T, Gondan M. Assessment of statistical significance and clinical relevance. Stat Med 2013; 32: 1707–1719 16 Farrar JT, Young JP, LaMoreaux L et al. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001; 94: 149–158 17 Cepeda MS, Africano JM, Polo R et al. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003; 105: 151–157 18 Bijur PE, Chang AK, Esses D et al. Problems with measurement of the minimum clinically significant difference in acute pain in elders. Acad Emerg Med 2011; 18: 135–139 19 Kendrick DB, Strout TD. The minimum clinically significant difference in patient-assigned numeric scores for pain. Am J Emerg Med 2005; 23: 828–832

Downloaded by: University of Georgia Libraries. Copyrighted material.

9 Disse BG, Ziegler HW. Pharmacodynamic mechanism and therapeutic activity of Ambroxol in animal experiments. Respiration 1987; 51 (Suppl 1): 15–22 10 Pfeifer S, Zissel G, Kienast K et al. Reduction of cytokine release of blood and broncheoalveolar mononuclear cells by Ambroxol. Eur J Med Res 1997; 2: 129–132 11 Gillissen A, Bartling A, Schoen S et al. Antioxidant function of Ambroxol in mononuclear and polymorphnuclear cells in vitro. Lung 1997; 175: 235–242 12 Felix K, Pairet M, Zimmermann R. The antioxidative activity of the mucoregulatory agents: Ambroxol, bromhexine and N-acetyl-Lcysteine. A pulse radiolysis study. Life Sci 1996; 59: 1141–1147 13 Lapenna D, de Gioia S, Ciofani G et al. Ambroxol is a scavenger of hypochlorous acid and monochloramine. Pharmacology 1994; 49: 132–135

DrugRes/2014-12-0925/6.3.2015/MPS

de Mey C et al. Efficacy and Safety of …  Drug Res

Efficacy and Safety of an Oral Ambroxol Spray in the Treatment of Acute Uncomplicated Sore Throat.

Compare the efficacy and tolerability of oral spray formulations delivering 2.5, 5, and 10 mg ambroxol (AXS) per application (4 actuations/application...
529KB Sizes 0 Downloads 15 Views