To characterize the intrathoracic thermal events that occur during breathing in humans, we developed a flexible probe (OD 1.4 mm) containing multiple thermistors evenly spaced over 30.2 cm, that could be inserted into the tracheobronchial tree with a fiberoptic bronchoscope. With this device we simultaneously recorded the airstream temperature at six points from the trachea to beyond the subsegmental bronchi in six normal subjects while they breathed ambient and frigid air at multiple levels of ventilation (VE). During quiet breathing of room air the average temperature ranged from 32.0 +/- 0.05 degrees C in the upper trachea to 35.5 +/- 0.3 degrees C in the subsegmental bronchi. As ventilation was increased, the temperature along the airways progressively decreased, and at a VE of 100+ 1/min the temperature at the above two sites fell to 29.2 +/- 0.5 and 33.9 +/- 0.8 degrees C, respectively. Interval points were intermediate between these extremes. With cold air, the changes were considerably more profound. During quiet breathing, local temperatures approximated those recorded in the maximum VE room-air trial, and at maximum VE, the temperatures in the proximal and distal airways were 20.5 +/- 0.6 and 31.6 +/- 1.2 degrees C, respectively. During expiration, the temperature along the airways progressively decreased as the air flowed from the periphery of the lung to the mouth: the more the cooling during inspiration, the lower the temperature during expiration. These data demonstrate that in the course of conditioning inspired air the intrathoracic and intrapulmonic airways undergo profound thermal changes that extend well into the periphery of the lung.

The increased minute ventilation (VE) associated with exercise produces similar degrees of airway cooling in normal and asthmatic subjects, but only those with asthma develop postexertional bronchoconstriction in response to this stimulus. We have found that when normal subjects breathing subfreezing air perform isocapnic hyperventilation to levels exceeding those associated with even exhausting exercise, 1-s forced expiratory volumes and maximum midexpiratory flow rates fall significantly. When tests more sensitive in detecting bronchoconstriction are employed, changes are seen at lower levels of hyperventilation that simulate the VE associated with moderately heavy work loads. We conclude that normal subjects respond to airway cooling, but are much less sensitive than those with asthma.