We have recently analyzed our data collected over a period of the last 10 years dealing with patients with hyposmia treated with theophylline.

We initially analyzed results in 230 patients tested by 8 different physicians and psychophysicists using the three stimuli, staircase, forced choice, sniff technique previously demonstrated to be effective in measuring smell function in both normal volunteers and in patients with various forms of smell loss (hyposmia). Using this technique detection thresholds (DT) for 4 odorants, pyridine (dead fish odor), nitrobenzene (almond odor), thiophene (gasoline-like odor) and amyl acetate (banana odor), recognition thresholds (RT) for these 4 odorants and magnitude extinction (ME) values for these 4 odorants were measured.

Patients were tested at their initial visit to The Taste and Smell Clinic in Washington, DC. This group was found to have normal zinc metabolism and no other metabolic cause for their smell loss. They were then placed in a controlled single blind treatment protocol and given an oral delayed release theophylline preparation at doses of 200 mg, 400 mg and 600 mg at intervals of 2-4 months for each oral dose. All patients started treatment on 200 mg of theophylline. If patients improved their smell function subjectively and by the testing noted above they remained on the 200 mg dose and were reevaluated at 4-6 month intervals. If they did not improve on 200 mg the drug dose was increased to 400 mg and a similar protocol was used. If improvement did not occur on 400 mg the drug dose was increased to 600 mg.

Data were analyzed in 12 different categories for each drug dose. Analysis of these data indicated that DTs and RTs for patients improved significantly for all odors presented at each drug dose (p values varied from <0.05 < 0.001). Analysis of ME data indicated that while values improved under each treatment condition significance was achieved only for nitrobenzene estimation at a dose of 200 mg. (p <0.05).

Results were then stratified with respect to degree of patient hyposmia. Patients with the most severe smell loss, Type I hyposmia, improved much less than those with a lesser degree of smell loss, Type II hyposmia, at each level of drug treatment.

These results indicate that patients with smell loss (without abnormalities of zinc or other metabolic processes) improved their smell function after treatment with theophylline. These results also indicate that patients with lesser degrees of smell loss (Type II hyposmia) improved much more than did patients with greater degrees of smell loss (Type I hyposmia), indicating the importance of quantitative measurements of smell loss by a standard clinical technique to determine treatment efficacy.

These studies reflect the efforts of our summer associate, Judy Winglee, of Tufts University who has worked diligently to process this enormous and complex data set.