Respiratory muscle function and hypoxic ventilatory control in patients with type I diabetes

The interaction among pulmonary mechanics, respiratory muscle performance, and ventilatory control in subjects with insulin-dependent diabetes mellitus has so far received little attention. We therefore decided to assess the role of central factors and peripheral factors on the ventilatory response to a hypoxic stimulus in type I diabetic patients.

Eight patients in stable condition aged 19 to 48 years old, with insulin-dependent diabetes mellitus (duration of the disease, 36 to 240 months) and no history of smoking, cardiopulmonary involvement, or autonomic neuropathy; and an age- and gender-matched control group.

In each patient, we measured the following: pulmonary volumes; diffusing capacity of the lung for carbon monoxide (D(LCO)); time and volume components of ventilation (tidal volume [V(T)] and respiratory frequency); static compliance (Clstat) and dynamic compliance (Cldyn); swings in pleural pressure (Pes) and gastric pressure (Pg); and transdiaphragmatic pressure (Pdi), obtained by subtracting Pes from Pg. Maximal inspiratory Pes and Pdi during a maximal sniff maneuver were also measured. Swings in Pes and Pdi during V(T) as a percentage of Pes and Pdi during the maximal sniff maneuver [Pessw(%Pessn) and Pdisw(%Pdisn), respectively] were both considered as a measure of central respiratory output, and the Pessw(%Pessn)/V(T) ratio was considered as an index of neuroventilatory dissociation (NVD) of the inspiratory pump. Subjects were studied at baseline and during hypoxic rebreathing.

Pulmonary volumes and D(LCO) were normal or slightly reduced. A lower Cldyn, higher central respiratory output, and NVD were found. During hypoxic rebreathing, patients had lower V(T), similar central respiratory output, and greater NVD per unit change in arterial oxygen saturation compared with values in control subjects. An increase in dynamic elastance, computed as 1/Cldyn, during hypoxia was found in patients, but not in normal subjects, and was directly related to concurrent changes in NVD.

We have shown that the assessment of a normal Clstat and normal routine parameters of airway obstruction does not permit the definite exclusion of the role of peripheral airway involvement in insulin-dependent diabetes mellitus. Peripheral airway involvement is likely to influence indices of hypoxic ventilator) drive by modulating a normal central motor output into a rapid and shallow pattern of ventilatory response.