AMMONIA COUGH ELICITED THROUGH A TRACHEAL SIDE TUBE IN UNANESTHETIZED DOGS. COMPARATIVE ANTITUSSIVE BIOASSAY OF FOUR MORPHINE DERIVATIVES AND METHADONE IN TERMS OF AMMONIA THRESHOLDS

¹ Department of Pharmacology, Parke, Davis & Co., Detroit 32, Michigan

Small amounts of a crude mixture of ammonia with air and water vapor injected directly through a chronic tracheal side tube (procedure A) or of pure ammonia injected into a constant stream of humidified and warmed air flowing through the side tube (procedure B) elicited realistic, vibrant coughs or paroxysms of coughing in unanesthetized dogs, with retching when larger amounts were used. Proportions of administrations eliciting a critical cough, expressed as probits, were related linearly to log ml. ammonia, with ED₅₀'s of 0.9-2.4 ml. and standard deviation factors (f) of 1.24 to 2.18. Lower values of f were associated with procedure B and use of a near minimal cough than with procedure A and use of more vigorous cough as criterion of response.

In only one of 14 experiments was there evidence of drift in sensitivity associated with administration of ammonia 100 times during 60 or 100 minutes.

Dogs prepared with polyethylene tracheal tubes with inner and outer tracheal flanges wired together through the tracheal wall remained technically useful for periods of 4 days to over 20 months, average over 5 months.

In dogs not previously subjected to numerous narcotic treatments, brief series of ammonia stimulations before and after administration of 2 to 5 mgm. codeine P0₄ per kgm. body weight sufficed to exhibit significant reduction in the proportion of stimuli followed by cough. Such quantal procedure suggests itself for preliminary examination for antitussive action.

For more definitively quantitative study, tussal thresholds to ammonia were determined. In statistically designed and evaluated experiments ammonia threshold determinations were made before and repeatedly after subcutaneous administration of normal saline or various doses of various drugs.

Log posttreatment thresholds were found usefully related to log pretreatment thresholds by analysis of covariance. Treatments were balanced among dogs so that demonstrably important differences among dogs could be taken into account in the analysis.

Log thresholds at the time effects were on the average maximal for respective treatments (peak effects), or the average of successive posttreatment log thresholds (an approximate function of total effect) did not deviate significantly from parallel linear relationships with log dose for codeine. PO₄ (1.0-5.7 mgm./kgm.), morphine. SO₄ (0.15-0.42), dihydrocodeinone. HCl (0.20-0.54) and methadone HC1 (0.20-0.28). Lambda varied between 0.20 and 0.40. In terms of peak effects dihydrocodeinone. HCl was 12 (95 per cent limits, 7-31), morphine. SO₄ 14 (8-33), and methadone 9 (3.9-15) times as potent as codeine.PO₄. In terms of the two-hour average effects the corresponding results were 16 (9.3-62), 17 (10-54), and 10.(5.8-16).

At a codeine-like dose level ethylmorphine.HCl yielded a moderate elevation of threshold (50-60 per cent) that did not grade upward with dosage.