Use of within-group designs to test anti-tussive drugs in conscious guinea-pigs

Abstract

Introduction: Cough is a common medical problem for which there are few effective drug treatments. A limited understanding of the mechanisms of induction and maintenance of cough and a paucity of suitable animal models frustrate drug discovery efforts to find novel anti-tussives. As in humans, guinea-pigs evoke a cough reflex upon exposure to tussive agents such as citric acid and capsaicin; both of which have been widely used to assess novel anti-tussive drugs. However, the potential for using within-group designs in drug development has received little attention and such designs may offer a way of assisting the drug discovery effort in the area of cough as well as other areas.

Methods: Cough can be monitored in conscious guinea-pigs by placing animals in a Perspex chamber, in which air is continually exchanged by use of negative pressure and drug delivery of aerosols to the chamber can be accurately timed. Cough in response to a tussive agent (e.g. 0.2-0.4M citric acid; 10-30 microM capsaicin) is detected by the simultaneous microphonic recording of audible signals characteristic of the cough response as well as by positive pressure changes in the chamber generated by a cough dependent rapid expiration of air from the lungs. Both the sound and pressure signals are recorded using an online analyzer, whilst the number of coughs can be analyzed off-line. The number of coughs over a 15 min period is used to quantitate tussive events.

Results: Reproducible cough can be detected in animals using cross-over designs that lend themselves to drug studies. Both the time and concentration dependence of anti-tussive drug action can be evaluated in the same animal. Furthermore, the effect of different anti-tussive drugs can be evaluated thereby reducing between group error and thereby improving the sensitivity of the test.

Discussion: Repeated measures design improves the precision with which to evaluate anti-tussive drugs in preclinical models and could be used to make the drug discovery process more efficient.