Customary 13CO2 breath tests--and also 15N urine tests--always start with an oral administration of a test substrate. The test person swallows a stable isotope labelled diagnostic agent. This technique has been used to study several pathophysiological changes in gastrointestinal organs. However, to study pathophysiological changes of the bronchial and lung epithelium, the inhalative administration of a stable isotope labelled agent appeared more suitable to us. [1-13C]Hexadecanol and [1-13C]glucose were chosen. Inhaled [1-13C]hexadecanol did not yield 13CO2 in the exhaled air, but [1-13C]glucose did. To study the practicability of the [1-13C]glucose method and the reproducibility of the results, 18 inhalation tests were performed with healthy subjects. In 6 self-tests, the optimum inhalative dose of [13C]glucose was determined to be 205 mg. Using the APS aerosol provocation system with the nebulizer 'Medic Aid' (Erich Jaeger Würzburg), a 25% aqueous solution was inhaled. Then, breath samples were collected at 15 min. intervals and analysed for 13CO2. 75-120 min after the end of inhalation a well-reproducible maximum delta13C value of 6%o over baseline (DOB) was detected for 12 healthy probands. Speculating that the pulmonary resorption of the [13C]glucose is the rate-limiting step of elimination, decompensations in the epithelium ought to be reflected in changed [1-13C]glucose resorption rates and changed 13CO2 output. Therefore, we speculate that the inhalation of suitable 13C-labelled substrates will pave the way for a new group of 13CO2 breath tests aiding investigations of specific pathophysiological changes in the pulmonary tract, such as inflammations of certain sections and decompensations of cell functions.