Reappraisal of Respiratory Abnormalities in Primary and Secondary Sjogren's Syndrome

Reappraisal of Respiratory Abnormalities in Primary and secondary Sjogren's Syndrome* A Controlled Study Matthildi P. Papathanasiou, M.D.;t Stavros H. Constantopoulos, M.D., F.C.C.P.;+ Constantinos Tsampoulas, M.D.;~ Alexandros A Drosos, M.D.;§ and Haralampos M. Moutsopoulos, M.D.II

In order to appraise the significance of respiratory abnormalities in primary and secondary Sjogren's syndrome (pSS and sSS), we evaluated 40 patients with pSS, 26 with sSS, 40 with rheumatoid arthritis (RA) but no SS, and 100 age- and sex-matched control subjects. The most common functional abnormality was diffuse interstitial lung disease (DILD) in patients with pSS (37.5 percent) and obstructive ventilatory defect in RA and sSS patients (40 and 19 percent, respectively). DILD was also present in the last two groups (Il.S percent in sSS and 27.5 percent in RA), while obstructive defect was rare in pSS (2.5 percent). Abnormalities suggesting small airways disease were present in all patient groups and also in the control group with similar frequency.

Patients with extraglandular pSS had most often DILD (52 percent). Patients with pSS and cryoglobulinemia had low C3 and C4 levels and decreased Dco, suggesting that interstitial lung disease may be a result of immune complex deposition. Clinical input of the functional abnormalities was minimal, expressed as dry cough and mild dyspnea. Pneumonia was not frequent, while pleurisy was present only in patients with sSS and RA. We suggest that, even though pulmonary abnormalities can frequently be detected with sensitive tests in patients with SS, they are not significant if compared with control subjects and are clinically negligible.

Sjogren's syndrome (88) is a chronic autoimmune disorder characterized by lymphocyte-mediated destruction of the exocrine glands, resulting in desiccation of conjunctival, oral, respiratory, gastrointestinal and genital mucosa. Pulmonary involvement in SS has attracted attention during the last years, but reports on its frequency and nature have varied among different studies. 1-7 This may be due to the different methods of evaluation applied by different investigators, and also to the mixed populations of both primary Sjogren's syndrome (pSS) and S8 associated with the presence of other autoimmune diseases (secondary SS).1,3.4 Absence of control populations in all the existing studies of pulmonary abnormalities in SS is a major shortcoming and may have caused overestimation of the results. Since clinical, serologic and genetic studies have all justified the distinction between primary and secondary 88 occurring in the presence of rheumatoid arthritis (RA), B.9 pulmonary function ought to be evaluated in the two groups separately. Mixed populations with

both pSS and sSS can carry along misleading conclusions, since pulmonary involvement may be attributed to the underlying autoimmune diseases, which are known to cause pulmonary impairment, and not to the presence of 88 itself. The purpose of this study is to reappraise pulmonary . function in patients with pSS compared with an ageand sex-matched control group and determine whether functional abnormalities, detected by sensitive methods, cause significant clinical respiratory incapacitation of these patients. Furthermore, the study addresses the question of different lung impairment in pSS and sSS.

*From the Departments of Medicine and Radiology, School of Medicine, University of Ioannina, Ioannina, Greece. tResearch Fellow, Derartment of Medicine. :J:Assistant Professor 0 Medicine, Department of Medicine. ~Lecturer in Radiology, Department of Radiology. §Attending rhYSiCian, Department of Medicine. [Professor 0 Medicine, Department of Medicine. Manuscript received December 27; revision accepted March 17. Reprint requests: Dr: Consta ntopoulos , Department of Medicine, University of 1oannina, loannina 45332, Greece

370

MATERIALS AND METHODS

Sixty-six patients with SS diagnosed and followed-up at the Rheumatology Department of Ioannina General Hospital were prospectively evaluated for respiratory involvement. Diagnosis ofSS was based on the presence of two out of three of the following findings: keratoconjunctivitis sicca (positive Schirmer's test <5 mm/5 min and/or slit lamp examination), xerostomia (parotid 80w rate <1 ml/5 min) and recurrent parotid gland enlargement. The diagnosis was confirmed by positive minor salivary gland lip biopsy results for an intense focal lymphocytic infiltrate with or without fibrosis (Tarpley's classification ~2 + ).10 Forty of these patients had pSS with no clinical or serologic evidence of another autoimmune disease, and 26 had sSS and definite or classic RA diagnosed according to the ARA's criteria." Forty patients with classic or definite RA without any clinical (xerophthalmia, xerostomia, recurrent parotid gland enlargement) or histologic evidence of SS (lip biopsy sampling revealed normal tissue classified as 0 according to Tarpley's classification)" also underwent evaluation of pulmonary Respiratory Abnormalities in SjOgren's Syndrome (Papathanas;ou et al)

Table I-Anthropometric and Clinical Characteristics Patients with Primary Sjogren's Syndrome (PSS), Secondary Sjogren's Syndrome (SSS), Rheumatoid Arthritis (RA) and Control Subjects

pSS sSS RA Control subjects

Age (mean±SD) (yrs) 53.05± 11.14 58.19 ± 10.315 5O.65± 14.375 52.31 ±9.76

of

Sex

Disease Duration (mean±SD) (yrs)

EaEgM* (% present)

40/0 24/2 34/6 91/9

6.7±6.5 4.48±5.21 8.24±7.73 none

52.5 27 27.5 none

*EaEgM = extraarticular, extraglandular manifestations function. Finally, a group of 100 age- and sex-matched individuals who volunteered for pulmonary function studies and chest roentgenographic examination was used for control. None of these subjects had recent or significant past medical history of respiratory illness. All patients and controls were lifelong non-smokers. Medical records were reviewed for the following information: patient's age and disease duration at the time pulmonary function tests were performed, presence of extraglandular manifestations in pSS (Raynauds phenomenon, splenomegaly, lymphadenopathy, intermittent dependent purpura, vasculitis, renal involvement, ie, glomerulonephritis or interstitial nephritis) and extraarticular manifestations in sSS and RA (subcutaneous nodules, episcleritis, pericarditis, splenomegaly, lymphadenopathy, Raynauds phenomenon and purpura). Serum values of C3 and C4, as well as the presence of different autoantibodies such as rheumatoid factor (RF), antibodies to extractable nuclear antigens (ENAs) (Ro-SSA, La-SSB or both) and cryoglobulins were recorded. All patients underwent clinical, roentgenologic and functional evaluation of the respiratory system. Clinical evaluation consisted of a respiratory questionnaire modified from that of the British Medical Research Council described elsewhere'" and physical examination. Standard posteroanterior and lateral chest roentgenographic examination of all patients and control subjects were arranged randomly and were evaluated by two of the authors for the presence of reticulonodular interstitital densities and other findings. Functional evaluation (Transferscreen 1110, Erich Jaeger, West Germany) included: spirometric tests of forced vital capacity (FVC), forced expiratory volume in one second (FEV J and FEV/FVC ratio, flow-volume curves with recording of the expiratory flow-volume curve, and maximum flow at 75, 50 and 25 percent of vital capacity (MEF~, MEF15O , MEF25) · 13 Total lung capacity (TLC) was measured with the helium dilution technique.f and carbon monoxide diffusing capacity (Dco) using the single breath method. I. Actual results of pulmonary function tests were expressed as percentage of predicted values (Knudson et al for spirometric tests and flow-volume curves," and Cotes for TLC and

Deal.). The pattern of lung involvement was defined in a way similar to that used by Owens et al~ in a recent study of patients with progressive systemic sclerosis based on suggestions of the American Thoracic Society and others. In our study, we used MEFssinstead of FEF25-~ to evaluate small airways disease. Thus, we considered normal to be FVC~80 percent predicted, FEV/FVC~70 percent and MEF~~60 percent predicted. Restrictive was defined by FVC<80 percent predicted and FEV/FVC~70 percent, and obstructive as FEV/FVC<70 percent. Isolated Dco reduction was percent predicted, FEV/FVC~70 percent. indicated by FVC~80 MEF2S~60 percent predicted and Dco<80 percent predicted. percent predicted, Isolated MEF 25 reduction included FVC~80 FEV/FVC~70 percent, MEF25<60 percent predicted. We considered this evidence of small airways disease. Means and standard deviations of percent predicted values were calculated for MEF25' FEV/FVC, FVC and Dco in patients with pSS, sSS and RA and were compared to means of the control group values using the student's t-test. Patients with pSS were divided into two subgroups on the basis of presence or absence of other extraglandular manifestations and the above mentioned parameters were compared with controls for each subgroup. The same values were also compared between pSS and sSS and between sSS and RA. Clinical symptoms and patterns of lung impairment were compared by chi-square analysis between the two subgroups of pSS patients (glandular and extraglandular groups). Functional abnormalities were also correlated in all three groups with other extraglandular manifestations of SS and extraarticular manifestations ofRA, as well as presence of different autoantibodies, using the chi-square test. RESULTS

Anthropometric and clinical characteristics of patients and control subjects are presented in Table 1. Statistical analysis by chi-square did not reveal any significant difference. There was a predominance of women in all groups. It must be noted, however, that most of the patients belonging to either of the above groups who were excluded from this study due to a smoking habit were men. Analysis of the sera of all groups tested revealed that rheumatoid factor and cryoglobulins were present in all patient groups. Antibodies to Ro(SSA) and La(SSB) were found predominantly in patients with pSS, whereas antibodies to Ro(SSA) were detected in patients with either pSS or sSS and with RA without SSe Clinical and roentgenologic findings in the respiratory system of patients and control subjects are sum-

Table 2-Clinical and Roentgenologic Findings in Patients with Primary Sjogren's Syndrome (PSS), Secondary Sjogren's Syndrome (SSS), Rheumatoid Arthritis (RA) and Control Subjects pSS (n=40) Asymptomatic Dry cough Exertional dyspnea Xerotrachea Bibasilar rales Pleurisy Positive roentgenologic findings *pSS vs RA: X2 = 6.745, p
23 7 10 5 10 0 17

(57.5%) (17.5%) (25%) (12.5%) (25%) (42.5%)

sSS (n=26) 21 4 1 0 9 2 16

(80.8%) (15%) (3.8%) (34.5%) (7.7%) (61.5%):1:

RA (n=40) 31 (77.5%) 8 (20%) 3 (7.5%) 0

1 (2.5%)*. t 1 (2.5%) 11 (27.5%)

Control Subjects (n= 1(0) 100 (100%)

0 0 0 12 (12%) 0 10 (10%)

rsss vs RA: X = 10.27, p
2

CHEST I 90 I 3 I SEPTEMBER, 1986

371

Table 3-Functional Abnormalities ofPatients with Primary Sjogren's Syndrome ("SS), Secondary Sjogren's Syndrome (sSS), Rheumatoid Arthritis (RA) and Control Subjects Estimated by Reference to Predicted Values Primary SS Total (n=40) Normal

DILD·

Obstructive ~ MEF25t

15 15 1 9

(37.5%) (37.5%) (2.5%) (22.5%)

Glandular (n=19) 10 (52.5%) 4 (21%) 0 5 (26.5%)

Secondary SS (n=26)

Extraglandular (n=21) 5 11 1 4

(24%) (52%) (5%) (19%)

8 3 5 10

(31%) (11.8%) (19%) (38.5%)

RA (n=40) 12 7 16 5

(35%) (27.5%) (40%) (12.5%)

Control subjects (n= 1(0) 56 13 1 30

(56%) (13%) (1%) (30%)

·DILD = diffuse interstitial lung disease; t = isolated decrease of MEF25<60 percent

marized in Table 2. All control subjects and the majority of patients were free of respiratory symptoms. However, some patients in all of the three groups complained of dry cough and/or exertional dyspnea. Five patients with pSS (12.5 percent) suffered from xerotrachea" (dry cough which varied from a foreign body sensation in the pharynx to exhaustive dry cough ending in vomiting). This symptom was never reported alone by any patient with sSS. On physical examination, ten (25 percent) pSS patients, nine (34.5 percent) sSS patients and one (2.5 percent) RA patient, as well as 12 (12 percent) control subjects, were found to have fine bibasilar rales audible over the lower lung fields. Finally, two patients with sSS (7.7 percent) and one with RA (2.5 percent) had clinical and radiographic evidence of pleural effusion. Pleurisy was neither present nor reported in the past medical history of any patient with pSS. Chest x-ray evaluation was nondiscriminatory in any of the groups tested. A diffuse reticulonodular pattern usually most prominent in the lower lung fields was identified in 42.5 percent of pSS, 61.5 percent of sSS, and 27.5 percent of RA patients, and 10 percent of control subjects. One patient with RA who had worked as a miner for 25 years had chest x-ray evidence compatible with Caplan's syndrome, with signs of pneumonoconiosis and five nodules. The results of pulmonary function tests are pre-

sented in Table 3. They were completely normal in 15 (37.5 percent) patients with pSS, eight (31 percent) patients with sSS, 12 (35 percent) patients with RA and only 56 (56 percent) of control subjects. The nature of functional pulmonary abnormality was mixed in all groups evaluated. The predominant disorder was diffuse interstitial lung disease in pSS patients, reaching very high frequency (52 percent) in the extraglandular group. In addition, an obstructive ventilatory pattern was present in patients with sSS and RA. Finally, isolated decrease in MEF25 was frequently observed in all patient groups, as well as in control subjects (Table 3). Mean values of MEF25' FEV/FVC, FVC and Dco for patients and control subjects are compared in Table 4. Primary SS patients and control subjects differ in FVC, which was significantly lower in pSS patients. When parameters for each of the two subgroups ofpSS (ie, with and without other extraglandular manifestations) were compared separately with control subjects, Dco and FVC of patients with extraglandular involvement were also found to be significantly lower. Patients with sSS had MEF25 and FEV/FVC values significantly lower than normal control subjects. FVC and Dco did not differ significantly. In patients with RA, similar findings were observed; MEF25 and FEV/FVC were significantly lower than in control subjects, whereas Dco was not.

Table 4-Mean Values ± Standard DeviationB ofMEF JIS' FEV/FVC, TLC, FVC, and Dco in Patient Groups and Control Subjects

pSS patients Glandular Extragland ular sSS patients RA patients Control subjects pSS vs control pSS glandular vs control pSS extraglandular vs control sSS vs control RA vs control pSS vs SSS sSS vs RA

MEF25

FEV/FVC

T1£

FVC

01£0

61.67±25.6 66.9±24.6 6O.16±24.8 46.15:t 24.95 5O.925±27 63.5:t 23.95 0.389 (NS) 0.555 (NS) 0.564 (NS) 3.185* 2.569+ 2.44+ 0.735 (NS)

83.1 ±8.05 83.4±8.9 82.85±7.4 79.3±7.14 78.4±7.9 83.4::t5.03 0.16 (NS) o (NS) 0.325 (NS) 2.755* 3.71t 2.01+ 0.479 (NS)

94.95::t 13.3

92.47:t 12.9 93.47± 15 9O.8±8.9 95.8:t 15.5 93.37± 18.3 100.42:t 14.27 3.188· 1.86 (NS) 4.0t 1.375 (NS) 2.19* 0.9 (NS) 0.577 (NS)

91.23 ± 18.86 96.4± 18.2 85.0± 17.26 91.9:t 18.09 87.3± 18.7 93.55:t 15.43 0.677 (NS) 0.636 (NS) 2.01+ 0.424 (NS) 1.817 0.413 (NS) 0.980 (NS)

94.6± 13 95.3± 13.88 97.04:t 12.8 101.08 ± 14.4 98.92± 11.84 1.613 (NS) 1.345 (NS) 1.06 (NS) 0.677 (NS) 0.816 (NS) 0.631 (NS) 1.17 (NS)

.p <0.01; tp <0.001; +p <0.05 372

Respiratory Abnormalities In SjOgren's Syndrome (Papathanasiou et 8/)

Mean values of patients with sSS were compared with mean values of pSS and RA patients. In the first case, only MEF 25 and FEV/FVC were significantly lower in patients with sSS. In the second case, no significant differences were found in any functional parameter between patients with sSS and RA. Finally, a comparison of clinical symptoms and patterns of lung impairment in pSS patients with glandular involvement vs pSS patients with extraglandular involvement showed that the presence of xerotrachea and interstitiallung disease were significantly different. Namely, five out of 19 patients with glandular pSS had xerotrachea compared to 0 out of21 patients with extraglandular pSS (X 2 = 4.14, p<0.05) and only three out of 19 patients of the first group had diffuse interstitial lung disease, compared to 12 out of 21 patients with extraglandular involvement (X 2 = 5.62, p<0.05). Correlation of respiratory involvement with clinical and serologic parameters revealed that there was no correlation of pulmonary involvement in general and in the presence of extraglandular or extraarticular involvement, rheumatoid factor (RF) and antibodies to extractable nuclear antigens (ENAs). A comparison of pulmonary abnormalities and C3, C4 levels in pSS patients with or without cryoglobulinemia revealed that, in patients with cryoglobulinemia, the mean values of Dco and C3, C4 were significantly lower (Table 5). Furthermore, C4 levels were significantly lower in pSS patients with diminished Dco compared to levels in pSS patients with normal Dco, but C3 levels were not. DISCUSSION

In an attempt to further understand pulmonary involvement in SS, we have undertaken the present study of pulmonary function in non-smoking patients with pSS and sSS separately, and compared their results with that of a group of unmatched non-smoking control subjects. Never before has a control population been included in the evaluation of pulmonary involvement in SS patients. In addition, a group of patients with RA, with no signs of SS and normal minor labial salivary gland biopsy results, was used to compare

pulmonary function of these patients and patients with the same disease accompanied by sSS in an attempt to assess whether the presence of sSS influences pulmonary function. Our results indicate that pulmonary involvement is quite frequently detected by pulmonary function tests in all three groups of patients with pSS, sSS and RA, and also in control subjects when very sensitive parameters like MEF 25 and Dco are used. Diffuse interstitial lung disease was the predominant disorder in patients with pSS, especially those with extraglandular manifestations. This is in agreement with Strimlan et al' and Oxholm et al," who also reported diffuse interstitial lung disease as the major abnormality in pSS. Oxholm et al2 also indicated that interstitial disease is more common when there is evidence of other extraglandular involvement. Extraglandular involvement in sS has been proven to be associated with a higher incidence of autoantibodies and immune complexes, and lower complement levels. I6-18 This finding, as well as the correlation of diminished Dco with cryoglobulinemia and low C4levels in our patients, suggests that immune complex deposition may playa role in the pathogenesis of interstitial lung disease in patients with pSS. Patients with sSS had an additional defect: an obstructive ventilatory pattern which was not different from that of patients with RA only. Obstructive defect has been reported as the most common finding in controlled series of patients with RA.18 •19 Begin et al20 have also documented, both functionally and histologically, chronic obstructive airways disease in six life-long nonsmokers with RA, five of whom also had evidence of sSS. Impairment of small airways' function, which is known to be a result of either an obstructive or an interstitial process," was similar in patients with pSS and sSS as judged by impaired MEF25 and compatible flow/volume curves. However, the high percentage of isolated MEF2.5 reduction in control subjects (30 percent) implies that this test, though quite sensitive, is not specific and one should be cautious when interpreting its significance. Abnormal results of Dco were

Table 5-Comparison of Pulmonary Abnormalities and C3, C4 Levels in Primary Sjogren's Syndrome (PSS) Patients with or

without Cryoglobulinemia

Primary Sjogren's Syndrome (pSS) Patients

MEF2.~

FEV/FVC TLC

FVC

Dco

C3 levels C4 levels

Cryoglobulins Present

Cryoglobulins Absent

74.64±23.97 84.73±5.8 93.54± 14.9 92.8± 12.2 79.545 ± 15.782 80.27±24.37 23.08± 16.6

59.14 ± 25.28 82.38±8.8 94.57± 13.35 93.37 ± 12.76 95.76 ± 17.35 108.25 ± 31.32 39.96± 13.8

Statistical Analysis p

1.705 0.938 0.196 2.666 2.87 3.67

CHEST / 90 / 3 / SEPTEMBER, 1986

NS NS NS NS

<0.05 <0.01 <0.001

373

also detected in our control group, although less frequently (13 percent). These findings have also been described by other investigators who have studied respiratory involvement in autoimmune rheumatic diseases and have used control groups. Thus, Hyland et al23 found the mean values of Dco and MEF25 to be 74±17 percent and 58±32 percent of predicted, respectively, in their control group. This may be attributed to the difference in populations selected to provide predicted normal values for pulmonary function tests, and to the populations that can be reasonably selected for a control group matched to a group of patients with autoimmune rheumatic diseases. The individuals comprising our control group were comparable to our patients not only in terms of age and sex, but also occupation, cultural and social background and available medical care. The only difference was the absence of an autoimmune disease. Comparison with such a control group is more realistic and contributory in appraising the pulmonary involvement in these diseases, rather than comparison with an ideal group of normal subjects who have been selected according to very strict criteria to provide predicted values on PFT Certainly, none of the individuals chosen to participate in our study as a control subject suffered from any pulmonary disease, had any symptoms related to the respiratory system or any significant past medical history. Finally, as indicated by the high percentage (10 percent) of positive roentgenographic examinations in our control group, roentgenologic evaluation of interstitial lung disease is highly subjective and nonspecific. This has been previously described." These positive roentgenographic films were unrelated to the finding of impaired Doo. Our study reveals that pulmonary involvement, although frequent, is rarely clinically significant in SSe It also clarifies that differences exist between patients with pSS and sSS, since pulmonary involvement in patients with sSS and RA is most probably the result of RA itself. These differences in pulmonary involvement further justify the distinction of pSS and sSS. ACKNOWLEDGMENTS: We are grateful to Drs. A. G. Tzioufas and M. Manoussakis for performing immunologic studies, and to Ms. E. E. Papanikolaou for excellent secretarial assistance.

5 6 7

8

9 10 11 12

13 14 15 16 17 18

19 20 21

REFERENCES Strimlan VC, Rosenow EC III, Divertie MB, Harrison EG Jr. Pulmonary manifestations of Sjogren's syndrome. Chest 1976; 70:354-61 2 Oxholm ~ Bundgaard A, Birk Madsen E, Manthorpe R, Vejl~ Rasmussen E Pulmonary function in patients with primary Sjogren's syndrome. Rheumatol Intern 1982; 2:179-81 3 Newball HH, Brahim SA. Chronic obstructive airway disease in patients with Sjogren's syndrome. Am Rev Respir Dis 1977; 115:295-304 4 Fairfax AJ, Halsam PL, Pavia D, Sheahan NF, Bateman JRM,

374

22 23

24

Agnew JE, et al. Pulmonary disorders associated with Sjogren's syndrome. Q J Med 1981; 199:279-95 Constantopoulos SH, Drosos AA, Maddison PJ, Moutsopoulos HM. Xerotrachea and interstitial lung disease in primary Sjogren's syndrome. Respiration 1984; 46:310-14 Constantopoulos SH, Papadimitriou CS, Moutsopoulos HM. Respiratory manifestations in primary Sjogren's syndrome: a clinical, functional and histologic study. Chest 1985; 88:226-29 Vitali C, TavoniA, Viegi G, Begliomini E, Agnesi A, Bombardieri S. Lung involvement in Sjogren's syndrome: a comparison between patients with primary and with secondary syndrome. Ann Rheum Dis 1985; 44:455-61 Moutsopoulos HM, Webber BL, Vlagopoulos T~ Chused TM, Decker JL. Differences of the clinical manifestations of sicca syndrome in the presence and absence of rheumatoid arthritis. Am J Med 1979; 66:733-36 Moutsopoulos HM, Chused TM, Mann DL, Klippel JH, Fauci AS, Frank MM, et al. Sjogren's syndrome (sicca syndrome): current issues. Ann Intern Med 1980; 92:212-26 Tarpley TM, Anderson LG, White GL. Minor salivary gland involvement in Sjogren's syndrome. Oral Surg 1974; 37:64-74 Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar R. 1959 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 1958; 9:175-76 Constantopoulos SH, Acritidis NK, Dalavanga YA, Moutsopoulos HM. Respiratory involvement in autoimmune rheumatic diseases is frequent and easy to detect. Clin Exp Rheumatol 1983; 1:233-36 Knudson RJ, Statin RC, Lebowitz MD, Burrows B. The maximal expiratory flow-volume curve: normal standards, variability and effects of age. Am Rev Respir Dis 1976; 113:587-600 Cotes LE. Lung function: assessment and application in medicine, 3rd ed. London: Blackwell Scientific Publications, 1975. Owens GR, Fino GJ, Herbert DL, Steen VD, Medsger TA Jr, Pennock BE, et al. Pulmonary function in progressive systemic sclerosis. Chest 1983; 84:546-50 Bombardieri S, Paoletti ~ Ferri C, Di Mun no 0, Fornai E, Giuntini G. Lung involvement in essential mixed cryoglobulinemia. Am J Med 1979; 66:748-56 Alexander EL, Hirsch TJ, Arnett FC, Provost n: Stevens MD. Ro(SSA) and La(SSB) antibodies in the clinical spectrum of Sjogren's syndrome. J Rheumatol1982; 6:184-95 Tzioufas AG, Manoussakis MN, Costello R, Silis M, Papadopoulos NM, Moutsopoulos HM. Cryoglobulinemia in autoimmune rheumatic diseases: Evidence of circulating monoclonal cryoglobulins in patients with primary Sjogren's syndrome. Arthritis Rheum (in press) Collins RL, Turner RA, Johnson AM, Whittley NO, McLean RL. Obstructive pulmonary disease in rheumatoid arthritis. Ann Rheum Dis 1976; 19:623-28 Geddes DM, Webley M, Emerson PA. Airways obstruction in rheumatoid arthritis. Ann Rheum Dis 1979; 38:222-25 Begin R, Masse S, Cantin A, Menard H, Bureau M. Airway disease in a subset of non-smoking rheumatoid patients: characterization of the disease and evidence for an autoimmune pathogenesis. Am J Med 1982; 72:743-50 Fulmer JD, Roberts WC. Commentary: small airways and interstitial pulmonary disease. Chest 1980; 77:470-72 Hyland RH, Gordon DA, Roderk I, Davis GM, Russell ML, Hutcheon MA, et aI. A systematic controlled study of pulmonary abnormalities in rheumatoid arthritis. J Rheumatol 1983; 10:395-405 Epler GR, McLoud TC, Gaensler EA, Mikus J~ Carcington GB. Normal chest roentgenograms in chronic diffuse infiltrative lung disease. N Engl J Med 1978; 298:943-49

Respiratory Abnormalitiesin SjOgren's Syndrome (Papathanasiou at aJ)