The relationship between single-breath diffusion capacity of the lung for nitric oxide and carbon monoxide during various exercise intensities.

STUDY OBJECTIVES To determine the relationship between single-breath diffusion capacity of the lung for nitric oxide (DLNO) and single-breath diffusion capacity of the lung for carbon monoxide (DLCO), and to determine the single-breath DLNO/DLCO ratios during rest and at several exercise intensities using a commercial lung diffusion system that uses electrochemical cells to analyze gases. SETTING AND PARTICIPANTS Eight healthy men (age, 27 +/- 5 years; weight, 83.0 +/- 11.8 kg; height, 180.4 +/- 9.5 cm; maximal oxygen uptake [VO(2)max], 47.6 +/- 10.2 mL/kg/min [mean +/- SD]) performed single-breath DLNO measurements (inspired nitric oxide concentration, 66.5 +/- 10.6 ppm) and carbon monoxide (0.30%) randomized on different days at rest and at various exercise intensities (40%, 75%, and 90% of VO(2)max reserve [VO(2)R]) on a electrically braked load simulator. The DLCO measured on day 1 was compared to the DLCO measured during the DLNO method from another day. RESULTS The relationship between DLNO and DLCO was linear (DLNO = 4.47 x DLCO; r(2) = 0.91; standard error of the estimate = 0.04; p < 0.05). DLNO was 4.52 +/- 0.24 times greater than DLCO, independent of exercise intensity. DLNO increased from 210.3 +/- 18.2 mL/min/mm Hg at rest to 284.2 +/- 38.6 mL/min/mm Hg at 90% VO(2)R (oxygen uptake = 42.6 +/- 9.8 mL/kg/min; 284.2 +/- 31.6 W; p < 0.05). Stepwise regression demonstrated that DLNO is predicted by alveolar volume (VA) [in liters] and workload (watts) such that DLNO = 13.4 x VA + 0.23 x workload + 107.7 (r(2) = 0.90; SEE = 17.5; p < 0.05). CONCLUSION (1) Single-breath DLNO and DLCO increase linearly with increasing workload; (2) the single-breath DLNO/DLCO ratios are independent of exercise intensity, suggesting that using either nitric oxide or carbon monoxide as transfer gases are valid in the study of lung diffusion during any level of exercise; and (3) DLNO is mainly predicted by VA and workload.