Article

Pulmonary gas exchange impairment (GEI) is a common consequence of intrathoracic sarcoidosis presenting with important therapeutic and prognostic implications. Objective: To determine the role of clinical, radiographic and functional variables in predicting GEI during moderate exercise at the estimated lactate threshold (theta(L)) in patients with sarcoidosis. Methods: Fifty-four outpatients (29 females) with biopsy-proven sarcoidosis had clinical evaluation (baseline dyspnea index), lung function tests and an incremental cardiopulmonary exercise test with theta(L) estimation. On a separate day, patients underwent a constant work rate test at theta(L) with assessment of arterial blood gas tensions. Results: There was no evidence of GEI [DeltaP (A - a) O-2/VO2 >20 mm Hg.l.min(-1)] in patients with radiographic stages 0-1 (n = 13). In the remaining 41 patients, GEI was associated with more extensive radiographic involvement and reduced diffusing capacity of the lung for carbon monoxide (DLCO), forced expiratory volume in 1 s, total lung capacity and forced vital capacity (% predicted; p < 0.05); baseline dyspnea index and resting arterial blood gas tensions, in contrast, were not significantly related to GEI. DLCO correlated best with GEI. The negative predictive value of DLCO >70% predicted (absent-to-mild impairment) was 91.3% (sensitivity = 81.8%) and the positive predictive value of DLCO less than or equal to50% predicted (severe impairment) was 83.3% (specificity = 96.6%, likelihood ratio = 13.35). There was no improvement in diagnostic accuracy when other physiological tests were added to DLCO. These results were consistent with those found in a multiple logistic regression analysis with GEI as the dependent variable (p < 0.01). Conclusions: Conventional chest radiography and DLCO measurements suffice to estimate the individual risk of GEI at moderate exercise in patients with sarcoidosis.
Besnier-Boeck-Schaumann

Langue : ANGLAIS

Tiré à part : OUI