Laboratory studies included complete blood cell count, platelet count, Westergrens erythrocyte sedimentation rate (ESR), rheumatoid factor by latex fixation, and antinuclear antibody. Chest roentgenograms were performed in all patients and were rated for the presence or absence of the following characteristics: hyperinflation; increased interstitial markings; pleural abnormalities; or localized scarring or atelectasis. Chest roentgenograms were classified as normal if there was no abnormality other than localized disease. read more
All patients completed a standardized respiratory questionnaire detailing a history of respiratory symptoms or past history of pulmonary disease.
All patients had B cell alloantigenic typing for HLA DRwl-11, DQw 1-3, DRw52, and DRw53 using a two-color fluorescent microcytotoxicity test. B-cell antisera standardized against those of the Ninth International Histocompatibility Workshop were used to define HLA antigens. HLA-A, HLA-B, and HLA-C antigens were not studied. One of the 48 subjects was untypeable at the DQw locus.
Pulmonary Function Testing
Forced expiratory spirometry was performed on all patients with a Stead-Wells spirometer (WC. Collins) according to standardized techniques. Pulmonary volumes were measured by helium dilution according to standardized techniques. Diffusing capacity for carbon monoxide (D) was determined for all patients using the singlebreath technique, except that no correction was made for oxygen absorption during the breath-hold. Reference values based on height, age, and sex for pulmonary volumes and D were taken from the literature.
Group comparisons were performed by Students test for continuous variables and by Fishers exact test for dichotomous variables. In order to determine the predictive value of possible risk factors for abnormality of pulmonary function, we used a stepwise multiple regression technique. Independent variables included the following: DQwl/DQw3; DQwl/DQw2; DQw2/DQw3; DQwl; DQw2; DQw3; DR4; Sjogren’s syndrome, smoking status, sex, race, duration of disease, level of ESR and titer of rheumatoid factor. Dependent variables included the forced expiratory volume in one second (FEVJ, forced vital capacity (FVC), and total lung capacity (TLC); residual volume (RV) and D expressed as percent of predicted; and the FEV^FVC ratio expressed as a percentage (FEVj/FVC’^). Independent variables were entered into the model when the partial F value exceeded 4.0 and were removed from the model when the partial F value was less than 4.0. Analysis of variance was used to determine the independent and interactive effect of each of the three DQw subtypes. Statistical significance was inferred when p<0.05.