Inhibition of Cough-Reflex Sensitivity by Benzonatate and Guaifenesin In

Respiratory Medicine (2009) 103, 902e906

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/rmed

Inhibition of cough-reﬂex sensitivity by benzonatate and guaifenesin in acute viral cough

Peter V. Dicpinigaitis a,*, Yvonne E. Gayle a, Gail Solomon b, Richard D. Gilbert c a Albert Einstein College of Medicine and Monteﬁore Medical Center, 1825 Eastchester Road, Bronx, NY, 10461, USA b Reckitt Benckiser, Inc., 399 Interpace Parkway, Parsippany, NJ, 07054, USA c TKL Research, Inc., 365 West Passaic Street, Rochelle Park, NJ, 07662, USA

Received 8 September 2008; accepted 9 December 2008 Available online 1 January 2009

KEYWORDS Summary Cough; Acute cough due to viral upper respiratory tract infection (URI) is the most common form of Antitussive; cough and accounts for tremendous expenditure on prescription and non-prescription cough Guaifenesin; products worldwide. However, few agents have been shown in properly conducted clinical Benzonatate; trials to be effective for cough due to URI. The present study evaluated the effect of benzo- Capsaicin; natate 200 mg (B), guaifenesin 600 mg (G), their combination (B þ G), and placebo (P) on Common cold capsaicin-induced cough in 30 adult nonsmokers with acute URI. On 3 separate days within a 7-day period, 1 h after ingesting randomly assigned study drug in a double-blind fashion, subjects underwent capsaicin cough challenge testing, which involved inhalation of incre- mental doubling concentrations of capsaicin until the concentration of capsaicin inducing 5

or more coughs (C5) was attained. Each subject received 3 of 4 possible study drugs. G (p Z 0.01) but not B (p Z NS) inhibited cough-reﬂex sensitivity (log C5)relativetoP.The combination of B þ G suppressed capsaicin-induced cough to a greater degree than B alone (p < 0.001) or G alone (p Z 0.008). The mechanism by which the combination of B þ G causes a potentiation of antitussive effect remains to be elucidated. Our results suggest that B þ G may be an effective therapy for acute cough due to the common cold (URI). ª 2008 Elsevier Ltd. All rights reserved.

Introduction

Cough is the most common complaint for which patients in 1 * Correspondence to: Peter V. Dicpinigaitis, Einstein Division/ the United States seek medical attention. Acute cough Monteﬁore Medical Center, 1825 Eastchester Road, Bronx, NY due to viral upper respiratory tract infection (URI) is the 10461, USA. Tel.: þ1 718 904 2676; fax: þ1 718 904 2880. most common form of cough and results in tremendous E-mail address: [email protected] (P.V. Dicpinigaitis). ﬁnancial expenditure on prescription and non-prescription

(over-the-counter) cough and cold products worldwide. Suggestive symptoms included acute onset of cough, fever, However, few of the currently available antitussive agents myalgias, rhinitis, nasal congestion, and sneezing. Symp- have been shown, in properly conducted clinical trials, to toms suggestive of a possible bacterial infection, such as be effective in suppressing cough due to URI.2,3 The sinus pain and purulent nasal discharge, excluded a subject narcotic opioids, such as codeine, have been demonstrated from enrollment. Other exclusion criteria included: history to be ineffective for acute cough associated with URI.4,5 If of asthma or other pulmonary disease; use of medications achieved, the antitussive effect of opioids often occurs at potentially affecting cough-reflex sensitivity (i.e., opiates the expense of troubling or intolerable side effects, such as and angiotensin converting-enzyme (ACE) inhibitors); and sedation or gastrointestinal disturbance.6 ingestion of the following types of medications within the Guaifenesin, the glyceryl ether of guaiacol (a constit- following time periods prior to study entry: short-acting uent of guaiac resin from the wood of Guajacum officinale antihistamines (48 h); short-acting pseudoephedrine (12 h); Linne´), is a component of numerous cough and cold prod- long-acting pseudoephedrine (24 h); non-steroid nasal ucts available worldwide. Guaifenesin is termed an sprays (12 h); cough suppressants and expectorants (24 h). expectorant since it is believed to alleviate cough discom- Female subjects were required to be using adequate fort by increasing sputum volume and decreasing its contraception. viscosity (hydration hypothesis), thereby promoting effec- 7 tive cough. Guaifenesin remains the only expectorant Study design approved by the United States Food and Drug Administra- tion (FDA).8 Despite its availability and common usage for This was a randomized, double-blind, placebo-controlled decades, few studies have evaluated the antitussive effect study of crossover design whose aim was to evaluate the of guaifenesin. A recently published trial demonstrated the effect of a combination of benzonatate and guaifenesin, ability of guaifenesin to inhibit cough-reflex sensitivity in relative to each individual drug and placebo, on cough- subjects with acute URI, thus suggesting an antitussive as reflex sensitivity to inhaled capsaicin in 30 otherwise well as expectorant action for the drug.9 healthy volunteers with acute viral URI. Upon enrollment, Benzonatate, a long chain polyglycol derivative chemi- subjects were randomized to receive 3 of 4 possible study cally related to procaine, is a peripherally-acting, oral drug combinations, one on each of 3 successive visits anesthetic agent whose mechanism of cough suppression is occurring on 3 separate days within a 7-day period. The 4 believed to involve inhibition of pulmonary stretch recep- possible study drug combinations included: benzonatate tors.10 Studies performed soon after its approval in the 200 mg (Inwood Laboratories, Commack, NY); guaifenesin United States a half century ago showed benzonatate able 600 mg (BI Chemicals, Petersburg, VA); benzonatate to inhibit experimentally-induced cough as well as to 200 mg þ guaifenesin 600 mg; and placebo, all prepared as suppress subjectively-measured pathological cough.11 No identically-appearing capsules. Subjects underwent capsa- further clinical trials of benzonatate have been published in icin cough challenge testing, to measure cough-reflex decades. sensitivity, 1 h after ingestion of study drug, administered Given the urgent need for better cough suppressant in a double-blind manner, on each of the 3 visits. The 1-h therapy, the goal of the present study was to evaluate, in pre-test interval was chosen to coincide with high plasma a prospective, randomized, double-blind, placebo- levels of each study drug. The dosages of study drugs were controlled manner, the antitussive effect of the combina- chosen to concur with current FDA-approved standards. tion of benzonatate and guaifenesin in subjects with acute The FDA regulates the use of capsaicin in clinical trials and URI. Cough-reflex sensitivity was measured by capsaicin required that no more than 3 successive doses (challenges) inhalation cough challenge. Capsaicin, the pungent extract be administered to a single subject in this study. Therefore, of red peppers, has been demonstrated in over two decades the 4 study treatments were investigated with a (4,3,N ) of clinical experience to induce cough in a safe, dose- balanced incomplete block design, in which each subject dependent and reproducible manner,12,13 and therefore has received 3 of the 4 treatments according to a pre- become an important tool for the assessment of the effect determined randomization scheme, over the course of 3 of a pharmacological intervention on cough-reflex sensi- study visits. This study was approved by the Institutional tivity.14 The most commonly used endpoints of capsaicin Review Board of Montefiore Medical Center, Bronx, New cough challenge are C and C , the concentrations of 2 5 York. All subjects provided written, informed consent. capsaicin inducing 2 or more, and 5 or more coughs, respectively. The C5 has been shown to be the most reproducible parameter in capsaicin cough-reflex sensi- Capsaicin cough challenge testing tivity measurement.12 Capsaicin was prepared and administered as previously described.12 Briefly, a stock solution (0.01 M) of capsaicin Methods (Formosa Laboratories, Taiwan, ROC) was prepared and subsequently diluted with physiological saline to yield serial Subjects doubling concentrations ranging from 0.49 mM to 1000 mM. Fresh dilutions were prepared on each day of testing. Subjects were otherwise healthy adult nonsmokers (life- Subjects inhaled single, vital-capacity breaths of time nonsmokers or ex-smokers having abstained from capsaicin aerosol administered via nebulizer (model 646; smoking >5 years) who developed symptoms consistent DeVilbiss Health Care Inc., Somerset, PA) controlled by with acute viral URI within 72 h of study enrollment. a dosimeter (KoKo DigiDoser; nSpire Health Inc., Louisville, 904 P.V. Dicpinigaitis et al.

CO). The nebulizer used in this study was modified in two with fixed terms for sequence, subject within sequence, ways. First, an inspiratory flow regulator valve (RIFR; nSpire period, and treatment. Treatments were compared pair- Health Inc.) was added, which limited inspiratory flow to wise using Fisher’s Least Significant differences. 0.5 L/s regardless of excessive inspiratory force, thereby guaranteeing a constant and reproducible inspiratory effort with each breath. Second, the straw and baffle assembly of Results the nebulizer was welded in place, thereby eliminating the variations in nebulizer output that occur when these Thirty subjects (19 female, 11 male; mean age 42.2 14.0 components are detached for washing and then reattached, [SD]) were enrolled and completed all 3 required study with resulting variable distances between the jet orifice visits. No significant adverse events were reported. Twenty- and straw. Given the nebulizer output of 1.007 mL/min, three subjects (77%) were lifetime nonsmokers; 7 subjects duration of aerosol delivery was programmed at 1.2 s to (23%) were distant ex-smokers (>5 years). The number of provide 0.02 mL aerosol per inhalation. Single breaths of subjects receiving each of the four possible study drugs capsaicin were given in ascending order, with inhalations of was: guaifenesin alone (n Z 22); benzonatate alone saline randomly interspersed to increase challenge blind- (n Z 23); guaifenesin þ benzonatate (n Z 23); placebo ness, until the concentration inducing 5 or more coughs (C5) (n Z 22). was attained. Inhalations were delivered at 1-min intervals. Characterizing the effect of study drug on cough-reflex Subjects were unaware that the end point of the study was sensitivity (C5), mean log C5 was 0.43 with benzonatate the number of coughs induced. alone, 0.49 with placebo, 0.55 with guaifenesin, and 0.70 with guaifenesin þ benzonatate. The overall difference Statistical analysis among the treatments was statistically significant (p < 0.001). In the pairwise comparisons, the log concen- A (4,3,N ) balanced incomplete block design was used, in tration of the guaifenesin/benzonatate combination was which each subject received 3 of the 4 possible treatments statistically significantly higher than either guaifenesin according to a predetermined randomization scheme, over alone (p Z 0.008) or benzonatate alone (p < 0.001). The log the course of 3 study visits. Subjects were assigned to one concentration was statistically significantly higher than of 16 possible sequences of 3 of the 4 treatments, with the placebo with both the guaifenesin/benzonatate combina- following constraints: each of the 4 treatments was to be tion (p < 0.001) and guaifenesin alone (p Z 0.010) [see received by the same number of subjects at any visit and Table 1]. each of the 16 sequences was to appear 2 or 3 times in the Effects of study drugs on cough-reflex sensitivity in subject cohort. terms of C2 very closely mirrored, but were not identical to, Based on the observed variability in capsaicin cough C5. In terms of mean log C2, the overall difference among challenge data previously reported by the authors,9 sample the treatments was statistically significant (p Z 0.007), as size estimates based on 80% power and 5% level of signifi- were the pairwise comparisons between guaifenesin and cance in a crossover design yielded n Z 16 to detect a 40% placebo (p Z 0.050); guaifenesin/benzonatate combination Z mean difference in log C5 before and after treatment. The and placebo (p 0.008); and, guaifenesin/benzonatate proposed sample size of 30 subjects in a (4,3,N ) balanced combination and benzonatate alone (p Z 0.004). Mean incomplete block design as described above provided log C2 data demonstrated an effect of guaifenesin alone approximately 20e24 observations per treatment. This compared to benzonatate alone (p Z 0.026) that was not level of statistical power was deemed acceptable for this evinced by log C5 data. However, mean log C2 values did not initial exploratory study. demonstrate a statistically significant difference between Log C5 and log C2 were analyzed using the analysis of guaifenesin alone and guaifenesin/benzonatate combina- variance model commonly applied to crossover designs, tion therapy.

Table 1 Effect of study drugs on cough-reﬂex sensitivity to inhaled capsaicin. Guaifenesin Benzonatate Guaifenesin þ benzonatate Placebo p valuea (n Z 22) (n Z 23) (n Z 23) (n Z 22) b log C5 Mean 0.545 0.430 0.704 0.491 <0.001 S.D. 0.389 0.502 0.467 0.399 Median 0.60 0.60 0.60 0.60 Range 0.30e1.20 0.30e1.50 0.30e1.50 0.30e1.20 p value vs. placebo 0.010 0.104 <0.001 p value vs. 0.008 <0.001 Guaifenesin þ benzonatate p value vs. benzonatate 0.298 a The differences in log-transformed concentrations between treatments were analyzed using an analysis of variance with factors for sequence, subject within sequence, period, and treatment. b C5 Z concentration of capsaicin inducing 5 or more coughs. Inhibition of cough-reﬂex sensitivity by benzonatate and guaifenesin in acute viral cough 905

Carryover effects were tested using the analysis of evaluation in clinical trials to assess its effect in acute as variance with terms for sequence, subject within sequence, well as other types of troublesome cough. carryover, and treatment. In the analysis of C5, the p value for the effect of carryover was 0.3739 and, in the analysis Conflict of interest statement of C2, the p value was 0.4280. Thus, carryover effects were not statistically significant. A similar analysis was per- PVD received an unrestricted, investigator-initiated formed to test the effect of visit number (i.e., time) and research grant to perform this work. PVD is currently was also found not to be significant. serving as a consultant to Reckitt Benckiser on a project unrelated to the content of this manuscript. YEG has no Discussion conflicts related to this manuscript. GS is an employee of Reckitt Benckiser. RDG is a biostatistician whose company, TKL Research, Inc., is engaged on an as-needed basis by the We have demonstrated the ability of guaifenesin, and the study sponsor. combination of guaifenesin and benzonatate, to inhibit cough-reflex sensitivity in subjects with acute URI, whose cough receptors are presumed to be transiently hypersen- Role of funding source sitive.15,16 While the inhibitory effect, if any, of benzonatate alone on the capsaicin-induced cough reflex did not This study was supported by an unrestricted, investigator- achieve statistical significance in this study, guaifenesin initiated grant from Reckitt Benckiser. The study sponsor alone produced significantly greater cough-reflex inhibition had no role in study design; collection, analysis and inter- than placebo. In addition, the results of the study showed pretation of data; in the writing of the manuscript; or in the a statistically significant increase in cough-reflex suppres- decision to submit the manuscript for publication. sion when guaifenesin and benzonatate were used in combination compared to either agent alone and to References placebo. These data suggest that benzonatate acts to potentiate the antitussive efficacy of guaifenesin through 1. Burt CW, Schappert SM. Ambulatory care visits to physician a mechanism as yet unknown. The antitussive effect of offices, hospital outpatient departments, and emergency guaifenesin using the capsaicin challenge model in subjects departments: United States, 1999e2000. Vital Health Stat 9 with acute URI has been observed in an earlier trial, but its 2004;13(157):1e70. potentiation by benzonatate has not been previously 2. Schroeder K, Fahey T. Systematic review of randomised reported. controlled trials of over-the-counter medicines for acute cough Acute viral cough is often difficult to treat. Indeed, two in adults. Br Med J 2002;324:329e31. prospective placebo-controlled studies have shown the 3. Bolser DC. Cough suppressant and pharmacologic protussive narcotic opioid codeine to be clinically ineffective in therapy: ACCP evidence-based clinical practice guidelines. e patients with acute cough due to URI.4,5 Recently published Chest 2006;129:238S 49S. cough management guidelines by the American College of 4. Eccles R, Morris S, Jawad M. Lack of effect of codeine in the treatment of cough associated with acute upper respiratory Chest Physicians (ACCP) did not recommend central cough tract infection. J Clin Pharm Ther 1992;17:175e80. suppressants for cough due to URI, and furthermore, did not 5. Freestone C, Eccles R. Assessment of the antitussive efficacy of recommend over-the-counter combination cold medica- codeine in cough associated with common cold. J Pharm tions (with the exception of an older generation antihista- Pharmacol 1997;49:1045e9. mine-decongestant) for acute cough due to the common 6. Chung KF. Currently available cough suppressants for chronic cold, until randomized, controlled trials demonstrate their cough. Lung 2008;186(Suppl. 1):S82e7. efficacy.3 7. Ayres PJW. Experimental and clinical methodologies in efficacy Cough-reflex sensitivity during acute viral URI appears of expectorants using the example of guaifenesin. In: Loew D, difficult to suppress in the laboratory. The authors, using Rietbrock N, editors. Phytopharmaka III. Darmstadt, Germany: e a PubMed search of the English language literature limited Steinkopff Verlag; 1997. p. 151 60. 8. United States Dept. of Health and Human Services. Expecto- to human trials, were unable to find a single study rant drug products for over-the-counter human use: final demonstrating the ability of codeine or dextromethorphan, monograph. Washington, DC: Federal Register 54; 1989. p. probably the two most commonly used antitussive agents, 8494e509. to inhibit capsaicin-induced cough in subjects with URI. The 9. Dicpinigaitis PV, Gayle YE. Effect of guaifenesin on cough reflex present study confirms the results of an earlier trial that sensitivity. Chest 2003;124:2178e81. showed a single, 400-mg dose of guaifenesin able to inhibit 10. Banner AS. Pharmacologic treatment of cough. In: Leff AR, capsaicin-induced cough in subjects with the common editor. Pulmonary and critical care pharmacology and thera- cold.9 Thus, guaifenesin appears to have a dual mechanism peutics. New York: McGraw-Hill; 2002. p. 673e9. of action: as an antitussive, as well as an expectorant, for 11. Ziment I. Agents that affect mucus and cough. In: Witek TJ, which it is approved by the FDA. A recent study has also Schacter EN, editors. Pharmacology and therapeutics in respiratory care. Philadelphia: WB Saunders; 1994. p. demonstrated the ability of the anticholinergic bronchodi- 239e57. lator tiotropium to inhibit cough-reflex sensitivity to 12. Dicpinigaitis PV. Short and long-term reproducibility of capsa- 17 inhaled capsaicin in subjects with URI. icin cough challenge testing. Pulm Pharmacol Ther 2003;16: Given the need for better cough suppressant therapies 61e5. and the results of the present exploratory study, the 13. Dicpinigaitis PV, Alva RV. Safety of capsaicin cough challenge combination of guaifenesin and benzonatate merits further testing. Chest 2005;128:196e202. 906 P.V. Dicpinigaitis et al.

14. Morice AH, Fontana GA, Belvisi MG, et al. European Respiratory 16. O’Connell F, Thomas VE, Studham JM, Pride NB, Fuller RW. Society guidelines on the assessment of cough. Eur Respir J Capsaicin cough sensitivity increases during upper respiratory 2007;29:1256e76. tract infection. Respir Med 1996;90:279e86. 15. Empey DW, Laitinen LA, Jacobs L, Gold WM, Nadel JA. Mecha- 17. Dicpinigaitis PV, Spinner L, Santhyadka G, Negassa A. Effect of nism of bronchial hyperreactivity in normal subjects after upper tiotropium on cough reﬂex sensitivity in acute viral cough. respiratory tract infection. Am Rev Respir Dis 1976;113:131e9. Lung 2008;186:369e74.