SEARCH

SEARCH BY CITATION

Keywords:

  • Scleroderma;
  • Systemic sclerosis;
  • Pulmonary disease;
  • Ethnicity;
  • Race

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Objective

To determine the relative contributions of genetic, clinical, serologic, sociodemographic, and behavioral/psychological variables to early pulmonary involvement in the Genetics versus Environment in Scleroderma Outcome Study cohort.

Methods

At the baseline visit (V0), 203 patients with systemic sclerosis (SSc) were examined (104 whites, 39 African Americans, and 60 Hispanics). We obtained sociodemographic, behavioral/psychological (illness behavior, social support, learned helplessness, smoking, drinking), clinical, serologic (autoantibodies), and genetic (HLA class II and FBN1 genotypes) factors; pulmonary function test results; electrocardiograms; and chest radiographs. Data analysis included Fisher's exact test, chi-square test, Student's t-test, analysis of variance, and stepwise linear and logistic regression methods.

Results

Significant pulmonary involvement was seen in 25% of patients within 2.8 years of SSc diagnosis. At V0, pulmonary fibrosis was significantly higher in African Americans compared with whites or Hispanics. African Americans had significantly lower percent predicted forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV1) compared with whites and significantly lower percent predicted diffusing capacity for carbon monoxide (DLCO) compared with whites and Hispanics. Significant, independent associations impacting early pulmonary involvement included African American ethnicity, skin score, serum creatinine and creatine phosphokinase values, hypothyroidism, and cardiac involvement. Anticentromere antibody seropositivity was a significant, independent, protective factor for restrictive lung disease and FVC or DLCO values. African Americans had significantly increased frequencies of anti–topoisomerase I, fibrillarin, and RNP autoantibodies compared with whites. African Americans scored significantly lower on the Interpersonal Support Evaluation List and significantly higher on the Illness Behavior Questionnaire.

Conclusion

Early pulmonary involvement in SSc appears to be influenced by several factors delineated by ethnicity, including racial, socioeconomic, behavioral, and serologic determinants.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Pulmonary disease is the leading cause of death in patients with systemic sclerosis (SSc; scleroderma) (1, 2). Pulmonary manifestations of SSc include pulmonary arterial hypertension (PAH) and pulmonary fibrosis. The reported incidence of PAH is 15–35% in patients with limited cutaneous SSc, usually as isolated PAH, and 30% in patients with diffuse cutaneous SSc, frequently associated with pulmonary fibrosis (3). Most patients have some degree of pulmonary fibrosis; however, the clinical course varies from mild and asymptomatic to severely debilitating (4). Most studies suggest that treatment for SSc pulmonary parenchymal involvement is more effective when initiated during the early or inflammatory stage (i.e., active alveolitis) before irreversible scarring (fibrosis) has occurred. The recent development of effective regimens to treat fibrosing alveolitis in patients with SSc (5, 6) and PAH (7, 8) makes the identification of factors contributing to early lung involvement even more pressing.

Retrospective analyses of SSc populations assessing lung-associated and general mortality have identified baseline pulmonary abnormalities and several associated variables as risk factors for increased susceptibility to clinically significant and progressive pulmonary disease (9–11). Those studies have reported a 17–24% reduced survival rate due to pulmonary involvement. Other studies have suggested that not all pulmonary abnormalities may progress to clinically significant disease warranting therapy.

The etiologic or enhancing factors of pulmonary involvement in patients with SSc are incompletely understood. One previous study implicated genetic factors, i.e., HLA class II (12). Other studies have implicated immunologic factors for which certain autoantibodies such as anti–topoisomerase I (anti–topo I) may be markers (13). The few studies that address the impact of race or ethnicity on lung involvement in early SSc suggest a worse prognosis for nonwhite groups including African Americans, Japanese, and Choctaw Indians (9, 14, 15). However, ethnicity is not only defined by racial or genetic factors, but also by sociodemographic, behavioral, or cultural determinants particular to the group studied (16), and these factors could independently influence pulmonary disease in patients with SSc. Moreover, no comparative data exist for Hispanics.

This report examines the relative contributions of sociodemographic, behavioral, immunologic, and genetic variables to clinical, laboratory, and radiographic pulmonary abnormalities in a prospective multiethnic cohort of patients with SSc with <5 years disease duration at the baseline visit (V0). Early pulmonary abnormalities and ethnic factors seen in these patients with SSc at clinical presentation may contribute to improved prognostic models.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

The Genetics versus Environment in Scleroderma Outcome Study (GENISOS) project was designed to identify factors predictive of outcome in patients with SSc. It is a collaborative effort of the University of Texas Houston Health Science Center, the University of Texas Medical Branch at Galveston, and the University of Texas San Antonio Health Science Center. Institutional review board approval was obtained at each study site before enrollment, and all study participants gave written informed consent.

Patient selection.

Patient selection and recruitment is discussed in detail elsewhere (17). Patients were enrolled in the study if they 1) met the 1980 American College of Rheumatology (formerly the American Rheumatism Association) preliminary criteria for the classification of SSc (18), with disease duration of up to 5 years (from date of the first non-Raynaud's phenomenon manifestation of scleroderma); 2) defined their ethnicity; and 3) received medical care in the geographic catchment area of the participating center. A total of 203 patients were enrolled. Most patients were referred to the participating center for ongoing evaluation and treatment. Others were continuously followed by their referring rheumatologists and returned only for study visits. White (n = 104) and African American (n = 39) patients were those who identified themselves as white or African American. Hispanic (n = 60) patients were those of predominantly Mexican or Central American ancestry, a subset with more Native American than white European ancestry. In addition, 9 Asian patients and 1 Native American (Cherokee) patient were enrolled.

The racial and ethnic identification of patients (white, African American, or Hispanic) were referred to as ethnic groups in this study. We only included patients with all 4 grandparents of the same ethnic group when analyzing data for the contribution of ethnicity to pulmonary involvement as discussed in a previous publication (17).

Clinical manifestation variables.

For each patient, a detailed clinical manifestation form (CMF) was completed, including demographic data; disease duration; disease manifestations; medication history; and clinical, laboratory, and radiologic parameters (17). The extent of skin involvement was measured by the modified Rodnan skin score (MRSS; range 0–51) (19–21).

Delineation of pulmonary involvement in patients with SSc.

Patients were identified as having SSc-related pulmonary involvement by abnormal pulmonary function test (PFT) parameters as noted below; by clinical findings; or, if established pulmonary disease, by radiographically evident pulmonary fibrosis on chest radiograph. Additionally, pulmonary involvement was considered present if the patient had changes of fibrosis or ground glass appearance by high-resolution computed tomography (CT), or inflammation by bronchoalveolar lavage (BAL; defined as >3% neutrophils and/or >2% eosinophils of the total white cell population according to American Thoracic Society guidelines). Pulmonary vascular involvement as PAH was noted if there had been an echocardiogram or right heart catheterization demonstrating elevated pulmonary artery pressure. High-resolution CT, BAL, echocardiogram, and right heart catheterizations were not included as a routine part of the GENISOS visit, but results were noted if obtained as part of usual clinical care. Clinical involvement was noted if the patient had dry bibasilar crackles or an increased pulmonic second sound (P2) relative to the aortic second sound (A2) on auscultation. The CMF also noted if the patient had a history of asthma or a diagnosis of chronic obstructive pulmonary disease. Gastrointestinal and cardiac manifestations derived from the study's clinical manifestations form are delineated in table legends, as appropriate.

Pulmonary function tests.

PFTs were performed at the 3 centers and results were interpreted by 1 faculty pulmonologist at each center. Variables were expressed as percent predicted forced vital capacity (FVC), percent predicted diffusing capacity in liters of carbon monoxide (DLCO), and forced expiratory volume in 1 second (FEV1). FVC and FEV1 were measured by spirometry using standardized methods. The percent predicted DLCO was obtained using the single breath technique. Predicted values for race, age, and sex were determined according to the American Thoracic Society recommendations (22). Restrictive lung disease (RLD) was diagnosed if the percent predicted FVC was <80%. Obstructive lung disease was diagnosed if the FEV1/FVC was <70% of that predicted. Abnormal diffusing capacity was defined by a percent predicted DLCO <80%.

Behavioral and psychological variables.

Three constructs (social support, coping behavior with illness, and helplessness) were evaluated. Social support was ascertained using the Interpersonal Support Evaluation List (ISEL), with higher scores indicating better social support (23). Coping with illness was ascertained with the Illness Behavior Questionnaire (IBQ), with higher scores indicating more inappropriate illness behaviors (24). Helplessness was measured with the Arthritis Helplessness Index in which higher scores indicate a higher degree of helplessness (25). The 36-item Short Form Health Survey (SF-36) was used to reflect the impact of disease on quality of life (26, 27).

Laboratory tests.

Routine laboratory tests obtained at baseline included a complete blood count with differential, serum creatinine, creatine phosphokinase (CPK), and urinalysis.

Autoantibody analyses.

Autoantibodies were determined at V0. Antinuclear antibodies and their various patterns, including anticentromere antibodies (ACAs) and antinucleolar antibodies, were determined by indirect immunofluorescence using HEp-2 cells as substrate (Autoantibodies Inc, Davis, CA or Inova Diagnostics, San Diego, CA) and were considered positive at a titer of ≥1:80 dilution. Anti-Ro/SSA, anti-La/SSB, anti-Sm, anti–U1 RNP, and anti–topo I were determined by immunodiffusion against calf thymus extract using commercial kits (Inova Diagnostics) in the University of Texas Houston rheumatology laboratories and were interpreted by 1 investigator (FCA). In addition, anti–RNA polymerase I/III autoantibodies, anti-Th/To, and anti–PM-Scl were determined by immunoprecipitation at the University of Pittsburgh (CAF-B) as previously described (28). Antibodies to anti–U3 RNP (fibrillarin) were identified by immunoprecipitation in the Advanced Diagnostics Laboratory at the University of Calgary (MF) (28).

Genetic analyses.

HLA class II alleles were determined by DNA oligotyping as previously described (29), with high-resolution DRB1 typing accomplished by direct sequencing. The 4 fibrillin 1 gene (FBN1) single-nucleotide polymorphisms were genotyped by direct sequencing using previously available primers (30). For sequence confirmation, the polymerase chain reaction products were analyzed on an ABI Prism 3100 automated DNA sequencer (Applied Biosystems, Foster City, CA).

Statistical analyses.

Unless otherwise stated, the cohorts used for comparison were whites (N = 104), African Americans (N = 39), and Hispanics (N = 60). The number of GENISOS patients not belonging to these ethnic groups (N = 10, with 9 Asian patients and 1 Native American patient) was too small for meaningful statistical analysis regarding ethnicity.

Comparisons of categorical variables among ethnic groups were performed using Fisher's exact test or chi-square test. For continuous variables, comparisons were made using an analysis of variance and Student's t-tests. Stepwise linear regression analysis was used to identify significant risk factors associated with FVC and DLCO at study entry, and stepwise logistic regression analysis was used to identify significant risk factors for RLD and radiographically evident pulmonary fibrosis. Significant values were set at P < 0.05.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Sociodemographic, clinical, and pulmonary parameters of patients with SSc by ethnic group.

The sociodemographic and behavioral influences and clinical involvement in the GENISOS cohort are summarized in Table 1. Mean disease duration at baseline for the entire cohort was 32.8 months. White patients had a significantly higher educational level (P < 0.0001), greater likelihood of being employed (P = 0.01), higher mean monthly income (P = 0.0001), and a greater proportion who consumed alcohol (P = 0.0007) than the nonwhite patients. A significantly smaller proportion of Hispanics smoked (5%) than did whites (26%) or African Americans (27%). White patients also had significantly higher ISEL scores for social support (P < 0.0001) and significantly lower IBQ scores for inappropriate coping behaviors (P = 0.006) than the nonwhite patients. White patients experienced significantly less overall gastrointestinal involvement in early disease (P = 0.029) compared with nonwhite patients (Table 1). There were no significant differences between whites, African Americans, or Hispanics in age, disease duration, MRSS, percentage of women, or percentage with diffuse cutaneous involvement, Raynaud's phenomenon, cardiac involvement, myositis, dysphagia, arthritis, tendon friction rubs, cancer, or diabetes mellitus. Overall, 58% of the cohort had diffuse cutaneous disease.

Table 1. Sociodemographic, behavioral, and clinical features of the Genetics versus Environment in Scleroderma Outcome Study cohort by ethnic group*
FactorWhite (n = 104)African American (n = 39)Hispanic (n = 60)P
  • *

    Values are the mean ± SD unless otherwise indicated. Comparison of groups was performed using analysis of variance. Data from 9 Asian patients with SSc are not included. SSc = systemic sclerosis; MRSS = modified Rodnan skin score; ISEL = Interpersonal Support Evaluation List; IBQ = Illness Behavior Questionnaire.

  • Compares white with nonwhite cohorts.

  • Including Medicare and Medicaid.

  • §

    Gastrointestinal involvement on the clinical manifestation form included symptoms of gastroesophageal reflux disease, dysphagia, diarrhea, constipation, episode(s) of severe abdominal pain requiring medical attention, pseudo-obstruction, malabsorption, unintended weight loss >11 pounds, documented esophageal dysmotility, upper or lower gastrointestinal involvement by diagnostic tests, or documented history of peptic ulceration.

Age, years51 ± 1249 ± 1348 ± 140.172
Disease duration, months29.8 ± 20.233.6 ± 19.935.2 ± 19.10.218
Diffuse cutaneous SSc, %5562620.617
MRSS15.1 ± 12.016.1 ± 14.015.6 ± 11.50.891
Education, no. of years13.6 ± 3.012.7 ± 2.510.8 ± 3.3<0.0001
Employed, %5034270.010
Income, monthly $US4,016 ± 3,4111,869 ± 1,3102,474 ± 2,4370.0001
Any insurance, %8264690.060
Tobacco smokers, %262750.002
Ethanol use, %5126250.0007
Social support (ISEL)8.5 ± 1.37.2 ± 1.87.8 ± 1.6<0.0001
Coping (IBQ score)9.8 ± 5.111.7 ± 6.212.7 ± 6.20.006
Gastrointestinal involvement, %§8497930.029

Autoantibody frequencies in patients with SSc by ethnic group.

Autoantibody frequencies among the 3 groups are summarized in Table 2. Significantly higher frequencies of anti-RNP antibodies were seen in African Americans compared with whites or Hispanics (21%, 6%, and 13%, respectively; P = 0.035). African Americans also had significantly higher frequencies of antifibrillarin antibodies compared with whites or Hispanics (23%, 8%, and 15%, respectively; P = 0.009). No significant differences were seen in other autoantibody frequencies among these groups.

Table 2. Autoantibody frequencies in Genetics versus Environment in Scleroderma Outcome Study patients by ethnic group*
FactorWhite (n = 104)African American (n = 39)Hispanic (n = 60)P
  • *

    Values are the percentage unless otherwise indicated. Data for 9 Asian patients with SSc are not included. ANA = antinuclear antibody; IIF = indirect immunofluorescence.

  • Based on the comparison of antibody distribution among whites, African Americans, and Hispanics.

ANA9197930.442
IIF pattern    
 Centromeric175120.185
 Speckled5367620.298
 Homogeneous0030.125
 Nucleolar2844370.186
 Cytoplasmic1713170.847
RNP621130.035
Fibrillarin (U3 RNP)823150.009
Topoisomerase I1421180.556
Sm0550.031
Ro3850.486
La1830.085
Th/To6070.295
RNA polymerase I2018280.379
RNA polymerase II6370.779
RNA polymerase III2021230.463

Patterns of early pulmonary involvement in patients with SSc by ethnic group.

Patterns of established pulmonary disease or clinical findings from the 3 ethnic groups at V0 are summarized in Table 3. The frequency of radiographically evident pulmonary fibrosis was significantly higher in African Americans compared with whites (46% versus 19%; P = 0.001) or Hispanics (46% versus 25%; P = 0.029). The cohort frequency of PAH (as reflected by elevated right ventricular systolic pressure on echocardiogram) was ≤10% overall but was seen in all groups at V0. However, the incidence of P2 louder than A2 on auscultation of heart sounds at study entry was significantly higher in African Americans compared with whites (28% versus 5%; P = 0.0003) and Hispanics (28% versus 8%; P = 0.009), possibly reflective of increased pulmonary pressures. The incidence of asthma was similar among the 3 cohorts, and most patients with asthma reported a long history of asthma, suggesting an idiopathic process unrelated to the reflux or aspiration that may be attributable to SSc-related gastroesophageal reflux disease (GERD).

Table 3. Pulmonary involvement in 3 ethnic groups with early systemic sclerosis*
FactorWhite (n = 104)African American (n = 39)Hispanic (n = 60)P
  • *

    Values are the percentage unless otherwise indicated.

  • By chi-square test for proportions comparing all 3 ethnic groups.

  • P = 0.001 for whites compared with African Americans; P = 0.029 for African Americans compared with Hispanics.

  • §

    Patients with P2 heart sounds louder than A2 heart sounds on cardiac auscultation.

  • P = 0.0003 for whites compared with African Americans; P = 0.009 for African Americans compared with Hispanics.

Pulmonary fibrosis1946250.005
Pulmonary hypertension21050.080
P2 > A2§52880.0002
Asthma105100.65

PFT parameters in patients with SSc by ethnic group.

Results of baseline pulmonary function tests by ethnicity are shown in Table 4. There were significant differences among the 3 groups in percent predicted FVC, DLCO, and FEV1. White patients had significantly higher percent predicted FVC (P = 0.002), percent predicted DLCO (P = 0.018), and FEV1 (P = 0.002) compared with African American patients. White patients had a significantly higher mean FEV1 (P = 0.0007) compared with Hispanics. The baseline PFT values of the Hispanics were more closely aligned with the African Americans except that Hispanics had significantly higher percent predicted DLCO values compared with African Americans (P = 0.004). The mean ± SD percent predicted DLCO of Hispanic patients was higher than that of white patients (76.8 ± 26.3 versus 72.2 ± 23.9, respectively) but did not reach statistical significance (P = 0.268).

Table 4. Pulmonary function test parameters in early systemic sclerosis by ethnicity*
ParameterWhite (n = 104)African American (n = 39)Hispanic (n = 60)PP value
White vs African AmericanWhite vs HispanicAfrican American vs HispanicWhite vs others
  • *

    Values are the mean ± SD unless otherwise indicated. FVC = forced vital capacity; DLCO = diffusing capacity for carbon monoxide; FEV1 = forced expiratory volume in 1 second.

  • Based on comparisons of pulmonary parameters between white and nonwhite cohorts obtained by Student's t-tests.

  • Based on comparisons among the 3 groups by analysis of variance tests.

Percent predicted FVC88.2 ± 19.975.0 ± 25.382.3 ± 24.60.0080.0020.0970.1700.007
Percent predicted DLCO, mmole × min−1 × kPa−172.2 ± 23.961.3 ± 19.676.8 ± 26.30.0130.0180.2680.0040.703
FEV12.38 ± 0.681.94 ± 0.642.08 ± 0.620.00060.00070.00070.2710.002

The PFT parameters were also assessed by smoking status. Among persons who never smoked, the mean ± SD percent predicted DLCO in African American patients was lower than in white patients (61 ± 21 versus 71 ± 24; P = 0.056) or Hispanic patients (61 ± 21 versus 76 ± 26; P < 0.011). Therefore, independent of smoking history, African Americans had a significantly lower mean percent predicted DLCO than Hispanics.

Of note, there were appreciable proportions of patients with PFT parameters <80% predicted at V0 in all groups. The frequency of percent predicted DLCO <80% was seen in 72 whites (71%), 29 African Americans (85%), and 34 Hispanics (63%). The frequency of percent predicted FVC <80% was seen in 30 whites (30%), 22 African Americans (59%), and 27 Hispanics (47%). The frequency of percent predicted FEV1 <80% was seen in 43 whites (43%), 22 African Americans (59%), and 21 Hispanics (36%).

Factors independently associated with pulmonary involvement in early SSc.

The factors that were independently and significantly associated with baseline pulmonary involvement by stepwise linear and logistic regression analyses in the GENISOS cohort are presented in Table 5. The independent factors that were significantly inversely associated with percent predicted DLCO values were a history of hypothyroidism, cardiac involvement, serum creatinine levels, and higher number of years of education. The factor that was significantly positively correlated with percent predicted DLCO values was the physical component score (PCS) of the SF-36 health form.

Table 5. Factors independently associated with pulmonary outcomes*
Dependent/independent variableParameter estimateStandard errorFPOR (95% CI)
  • *

    Stepwise linear regression analysis was used for the first 2 outcomes and stepwise logistic regression was used for the last 2 outcomes. OR = odds ratio; 95% CI = 95% confidence interval; DLCO = diffusing capacity for carbon monoxide; NA = not applicable; PCS = Physical Component Score derived from the SF-36 instrument; FVC = forced vital capacity; MRSS = modified Rodnan skin score; CPK = creatine phosphokinase.

  • R2 = 19.9%.

  • Diagnosed by elevated thyroid stimulating hormone.

  • §

    Patients were found to have evidence of cardiac involvement, including pericardial effusion; evidence of significant cardiac arrhythmia on electrocardiogram or Holter monitor; cardiomyopathy as defined by an ejection fraction <40% by echocardiogram or nuclear studies; evidence of pericarditis or pericardial effusion on electrocardiogram or echocardiogram, or a pericardial friction rub on auscultation; congestive heart failure by clinical diagnosis, the presence of an S3 sound on auscultation, or jugular venous distention; or clinical signs of right heart failure including P2 louder than A2 on auscultation or a right ventricular heave on palpation.

  • R2 = 25.6%.

  • #

    In months from date of the first non-Raynaud's phenomenon manifestation of scleroderma.

Percent predicted DLCO     
 Hypothyroidism−12.45.235.670.019NA
 Cardiac§−10.84.416.000.016NA
 Serum creatinine (mg/dl)−4.251.468.430.004NA
 Education, no. years−1.90.5611.280.001NA
 PCS0.460.186.170.014NA
Percent predicted FVC     
 Anticentromere antibodies17.34.6913.520.0003NA
 Cardiac§−12.64.089.550.002NA
 MRSS−0.300.144.580.034NA
 CPK, IU−0.010.0055.280.023NA
 Employed7.013.474.090.045NA
Restrictive lung disease     
 Anticentromere antibodies−2.491.05NA0.0180.08 (0.01–0.65)
 Own car1.160.43NA0.0063.20 (1.39–7.37)
 Employed−1.230.41NA0.0030.29 (0.13–0.65)
Radiographic pulmonary fibrosis     
 African American ethnicity1.470.55NA0.0074.34 (1.48–12.72)
 Disease duration#0.040.01NA0.0011.04 (1.02–1.07)
 CPK, IU0.0040.002NA0.0131.004 (1.01–1.09)

The independent factors that were significantly inversely associated with percent predicted FVC values were the presence of cardiac involvement, MRSS, and CPK values. The independent factors that were significantly positively correlated with percent predicted FVC values were the presence of ACA and employment. The independent factors that were significantly inversely correlated with RLD at baseline were the presence of ACA and employment. Surprisingly, car ownership was positively associated with RLD.

The independent factors that were significantly associated with radiographically evident pulmonary fibrosis at study entry were African American ethnicity, disease duration, and CPK values. No portions of the IBQ were independently associated with early pulmonary involvement in these patients. Also, there were no significant associations between HLA class II alleles and pulmonary involvement, and these data failed to support the previously reported increased frequency of HLA–DR2 in patients with early-onset SSc pulmonary fibrosis (30).

Smoking and obstructive lung disease in early SSc.

The relationship of cigarette smoking to early restrictive and/or obstructive lung disease assessed by PFTs in the SSc GENISOS cohort is shown in Table 6. Among patients with PFT results and complete smoking histories available at V0, 155 were current nonsmokers and 41 were current tobacco smokers. There was a significantly higher percentage of patients with SSc with an obstructive lung pattern in the current smoking group compared with the current nonsmoking group (29% versus 12%; P = 0.008). Of potential predictive value, no significant difference in percent predicted DLCO <80% between nonsmokers and smokers was noted (25% versus 22%; P = 0.732). Further delineating the nonsmoking patient group into previous smokers and never smokers demonstrated a significant difference between the percentage of patients with an obstructive pattern (21% versus 7%; P = 0.03). In contrast, the use of tobacco products did not appear to influence either restrictive lung disease or isolated decrease in DLCO at this PFT evaluation. Obstructive lung disease was rare in patients with SSc who had never smoked (7%; data not shown); this finding confirmed a previous report of the effects of smoking on PFT parameters in early disease (10). The utility of the decreases in percent predicted FVC and DLCO as potential screening tools for PAH may hold true despite smoking history. A greater impact of tobacco use on respiratory pathology in patients with SSc may appear over time.

Table 6. Smoking and obstructive lung disease in early systemic sclerosis*
PFT profileNonsmokers (n = 155)Smokers (n = 41)P
  • *

    Values are the number (percentage) unless otherwise indicated. PFT = pulmonary function test; DLCO = diffusing capacity for carbon monoxide.

  • Patients currently smoking cigarettes.

  • Based on the comparison of nonsmokers with smokers.

  • §

    Percentage of patients with systemic sclerosis with obstructive lung disease (7%) is significantly lower in patients who never smoked than in patients who were past smokers (21%; P = 0.003).

Normal46 (30)11 (27)0.721
Obstructive19 (12)§12 (29)0.008
Restrictive45 (29)6 (15)0.062
Obstructive and restrictive7 (5)3 (7)0.691
Isolated decrease in DLCO38 (25)9 (22)0.732

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Diffuse cutaneous SSc accounted for 58% of all cases in the GENISOS cohort. This differs from the ratio of diffuse to limited disease typically reported in prevalence studies in which diffuse disease affects ∼40% of individuals (1). The GENISOS cohort was established to compare racial/ethnic differences early in the course of SSc. The ratio of diffuse to limited disease therefore reflects the patient population at our 3 sites, rather than the distribution of disease in the larger community. Because the proportion of patients with diffuse cutaneous SSc did not differ among ethnic groups and the disease duration also was not statistically significant, the comparisons for this study are considered valid.

The data from the GENISOS cohort demonstrate that racial, ethnic, and possibly other biologic factors are strongly implicated in pulmonary involvement in early SSc. Pulmonary involvement at V0 (mean disease duration 32.8 months), including radiographically evident pulmonary fibrosis and PAH, was noted in white, African American, and Hispanic patient groups. However, African Americans had significantly lower PFT parameters, higher proportions of patients with loud P2 auscultative sound, radiographically evident pulmonary fibrosis, and/or PAH at V0. African Americans had a significantly higher proportion of RNP, fibrillarin, and Sm autoantibodies. However, in our analysis, RNP and fibrillarin autoantibody seropositivity were not independently associated with early pulmonary fibrosis (17, 31–34), pulmonary hypertension (35), RLD, or lower FVC or DLCO values.

Hispanic patients were intermediate between white and African American patients in the proportion of those with pulmonary fibrosis, pulmonary hypertension, P2 louder than A2 on auscultation, percent predicted FVC, and FEV1, but the comparison between Hispanics and whites did not reach statistical significance for these features. African Americans had a higher frequency of gastrointestinal involvement (including GERD), possibly reflecting more severe disease. Lower percent predicted DLCO values have been associated with more advanced gastroesophageal reflux symptoms (36), and esophageal hypomotility has been associated with decreased FVC and DLCO values (37) presumably related to aspiration of gastric contents. African Americans also scored significantly lower than whites on the ISEL and significantly higher on the IBQ, neither of which had independent associations with pulmonary involvement. Again, Hispanic patients were intermediate between African American and white patients in ISEL score, proportion with either gastrointestinal involvement or pulmonary fibrosis, percent predicted FVC, and values for FEV1. African Americans are reported to have more diffuse cutaneous disease, an earlier age of onset, more pulmonary involvement, and lower frequency of ACA than their white counterparts (9, 10).

This is the first report to analyze pulmonary involvement in Hispanic patients with SSc. Hispanics were more closely aligned with African Americans in sociodemographic and behavioral parameters, with a similar incidence of overall gastrointestinal involvement and baseline PFT values, except that the Hispanic group had a significantly higher mean percent predicted DLCO compared with African Americans. However, Hispanics did not demonstrate the equivalent incidence of radiographically evident pulmonary fibrosis, a loud P2 auscultative sound, or PAH at V0, which was independent of smoking history. These data support the reported prognostic significance of the percent predicted DLCO values in early pulmonary disease for general mortality (10, 37, 38) and the biologic association of ethnicity to pulmonary involvement in patients with SSc (19, 32). Many of the psychological or behavioral determinants and questionnaire scores addressed in this study were similar for Hispanics and African Americans, suggesting their minor if any contribution to early lung disease.

In other chronic diseases, Hispanics share sociodemographic and behavioral characteristics of African Americans, but the milder clinical involvement in Hispanics is more aligned with white patients. This phenomenon has been termed the Hispanic paradox, although recent studies have critically refuted this concept (39–41). The ethnic contributions in African Americans for baseline PFT values, physical examination, and clinical measures related to pulmonary involvement are in agreement with studies reporting that more severe disease occurs in nonwhite cohorts (Japanese, Choctaw Indians, and African Americans).

The most significant, independent associations impacting early pulmonary involvement by radiographically evident pulmonary fibrosis, RLD, or percent predicted FVC or DLCO values included ethnic and clinical parameters: African American ethnicity, MRSS, serum creatinine and CPK values, history of hypothyroidism, and evidence of cardiac involvement. In our study and other studies, the skin score in SSc has been reported as being significantly and independently associated with percent predicted FVC and radiographically evident pulmonary fibrosis (42). Current employment status and the PCS from the SF-36 health form were significantly inversely associated with RLD and PFT parameters, likely reflecting the clinical condition. Additional disease involvement or protective influences based on ethnicity, or other variables with more robust associations may emerge over time (19, 32, 43).

The presence of ACA was a significant, independent protective factor for RLD and percent predicted FVC or DLCO values at study entry. A lower frequency of RLD, pulmonary fibrosis, and PAH among patients with SSc and ACA seropositivity has been reported (43). The protective influence of ACA has been further supported by a retrospective study, conducted at our institution (University of Texas Houston Health Science Center), of patients with SSc with longer disease duration, in which radiographic pulmonary fibrosis was seen in 3 (4%) of 68 patients with ACA compared with 125 (74%) of 178 patients without ACA (P < 0.001). Previous studies have reported a higher frequency of ACA in whites, although the significance of the abovementioned findings will need further study.

Previous studies looking at end-stage lung disease or early mortality proposed that low percent predicted FVC and DLCO values, sex, advanced age, proteinuria, and autoantibodies at the initial evaluation were significant prognostic markers (10, 38). However, we found a much higher frequency of percent predicted DLCO <80% (63–85% among groups) at V0 compared with previously published articles regarding early mortality, actual mortality from pulmonary disease, or development of PAH. The prognostic reliability of reduced percent predicted DLCO or FVC at baseline may also require additional or extra genetic factors to promote (or accurately reflect) profibrotic changes in the lung parenchyma or vasculature. It is unclear if the early reduced DLCO values are directly associated with decreased mortality due to progression of lung disease, or if additional clinical, sociodemographic, or behavioral factors will further influence the ongoing clinical activity and outcome. For example, evidence of some degree of interstitial fibrosis occurs in up to 90% of patients with SSc, results in RLD in 30–50% of patients, and progresses to significant lung damage in ∼15% of patients (44).

Our baseline evaluation of sociodemographic and behavioral/psychological differences in SSc ethnic groups did not elucidate the ultimate impact of these differences on clinical severity. For example, African Americans may have a greater incidence of pulmonary fibrosis reflecting the influence of shared genetic factors, but the extent to which the pulmonary fibrosis becomes clinically significant, or the time span to clinical significance may be influenced by behavior, such as smoking, personal activities, or clinical conditions such as GERD, relative to the other ethnicities.

The longitudinal study of this cohort may identify contributors that cumulatively lead to scleroderma lung disease or factors that distinguish stabilization from progression. This should allow for a more targeted approach and possibly a more judicious use of resources to identify patients at risk for pulmonary progression.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Dr. McNearney had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Reveille, Fischbach, Friedman, Tan, Ahn, Arnett, Mayes.

Acquisition of data. McNearney, Reveille, Fischbach, Friedman, Lisse, Goel, Zhou, Feghali-Bostwick, Fritzler, Mayes.

Analysis and interpretation of data. McNearney, Reveille, Friedman, Lisse, Goel, Tan, Zhou, Ahn, Fritzler, Arnett, Mayes.

Manuscript preparation. McNearney, Reveille, Fischbach, Friedman, Tan, Ahn, Feghali-Bostwick, Mayes.

Statistical analysis. McNearney, Tan, Zhou, Ahn.

Funding. Arnett.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

The authors would like to thank Evelyn Johnson and Charles Forbes for their assistance in data entry and database construct; Rudyard Lanete, Marilyn Berry, Heather Metts, Teddy Hunnicutt, Laura Clark, and Barbara Boyle for their roles as study coordinators; and Kim Jordan, Anthony Thomas, and Julio Charles for their assistance in the serologic and immunogenetic analyses.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES
  • 1
    Mayes MD. Scleroderma epidemiology [review]. Rheum Dis Clin North Am 2003; 29: 23954.
  • 2
    Altman RD, Medsger TA Jr, Bloch DA, Michel BA. Predictors of survival in systemic sclerosis (scleroderma). Arthritis Rheum 1991; 34: 40313.
  • 3
    Diot E, Giraudeau B, Diot P, Degenne D, Ritz L, Guilmot JL, et al. Is anti-topoisomerase I a serum marker of pulmonary involvement in systemic sclerosis? Chest 1999; 116: 71520.
  • 4
    Steen V, Medsger TA Jr. Predictors of isolated pulmonary hypertension in patients with systemic sclerosis and limited cutaneous involvement. Arthritis Rheum 2003; 48: 51622.
  • 5
    White B, Moore WC, Wigley FM, Xiao HQ, Wise RA. Cyclophosphamide is associated with pulmonary function and survival benefit in patients with scleroderma and alveolitis. Ann Intern Med 2000; 132: 94754.
  • 6
    Khanna D, Clements PJ, Furst DE, Chon Y, Elashoff R, Roth MD, et al, and the Scleroderma Lung Study Group. Correlation of the degree of dyspnea with health-related quality of life, functional abilities, and diffusing capacity for carbon monoxide in patients with systemic sclerosis and active alveolitis: results from the Scleroderma Lung Study. Arthritis Rheum 2005; 52: 592600.
  • 7
    Gildea TR, Arroliga AC, Minai OA. Treatments and strategies to optimize the comprehensive management of patients with pulmonary arterial hypertension. Cleve Clin J Med 2003; 70 Suppl 1: S1827.
  • 8
    Denton CP, Black CM. Pulmonary hypertension in systemic sclerosis [review]. Rheum Dis Clin North Am 2003; 29: 33549, vii.
  • 9
    Laing TJ, Gillespie BW, Toth MB, Mayes MD, Gallavan RH Jr, Burns CJ, et al. Racial differences in scleroderma among women in Michigan. Arthritis Rheum 1997; 40: 73442.
  • 10
    Morgan C, Knight C, Lunt M, Black CM, Silman AJ. Predictors of end stage lung disease in a cohort of patients with scleroderma. Ann Rheum Dis 2003; 62: 14650.
  • 11
    Steen VD, Medsger TA Jr. Severe organ involvement in systemic sclerosis with diffuse scleroderma. Arthritis Rheum 2000; 43: 243744.
  • 12
    Briggs DC, Vaughan RW, Welsh KI, Myers A, duBois RM, Black CM. Immunogenetic prediction of pulmonary fibrosis in systemic sclerosis. Lancet 1991; 338: 6612.
  • 13
    Ho KT, Reveille JD. The clinical relevance of autoantibodies in scleroderma. Arthritis Res Ther 2003; 5: 8093.
  • 14
    Greidinger EL, Flaherty KT, White B, Rosen A, Wigley FM, Wise RA. African-American race and antibodies to topoisomerase I are associated with increased severity of scleroderma lung disease. Chest 1998; 114: 8017.
  • 15
    Nishioka K, Katayama I, Kondo H, Shinkai H, Ueki H, Tamaki K, et al, and the Scleroderma Research Committee Japan. Epidemiological analysis of prognosis of 496 Japanese patients with progressive systemic sclerosis (SSc). J Dermatol 1996; 23: 67782.
  • 16
    Hutchinson J, Smith AD. Ethnicity. Oxford: Oxford University Press; 1996.
  • 17
    Reveille JD, Fischbach M, McNearney T, Friedman AW, Aguilar MB, Lisse J, et al. Systemic sclerosis in 3 US ethnic groups: a comparison of clinical, sociodemographic, serologic, and immunogenetic determinants. Semin Arthritis Rheum 2001; 30: 33246.
  • 18
    The Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980; 23: 58190.
  • 19
    Rodnan GP, Lipinski E, Luksick J. Skin thickness and collagen content in progressive systemic sclerosis and localized scleroderma. Arthritis Rheum 1979; 22: 13040.
  • 20
    Silman A, Harrison M, Brennan P, and the ad hoc International Group on the Assessment of Disease Outcome in Scleroderma. Is it possible to reduce observer variability in skin score assessment of scleroderma? J Rheumatol 1995; 22: 127780.
  • 21
    Pope JE, Bellamy N. Outcome measurement in scleroderma clinical trials [review]. Semin Arthritis Rheum 1993; 23: 2233.
  • 22
    Watters LC, King TE, Schwarz MI, Waldron JA, Stanford RE, Cherniack RM. A clinical, radiographic, and physiologic scoring system for the longitudinal assessment of patients with idiopathic pulmonary fibrosis. Am Rev Respir Dis 1986; 133: 97103.
  • 23
    Cohen S, Mermelstein R, Kamarch T, Hoberman HM. Measuring the functional components of social support. In: SarasonIB, SarasonBR, editors. Social support: theory, research, applications. Boston: Martinus Nijhoff; 1985.
  • 24
    Pilowsky L, Spence ND. Illness behavior questionnaire (IBQ). In: CorcoranK, FischerJ, editors. Measures for clinical practice: a source book. New York: The Free Press; 1996. p. 182.
  • 25
    Engle EW, Callahan LF, Pincus T, Hochberg MC. Learned helplessness in systemic lupus erythematosus: analysis using the Rheumatology Attitudes Index. Arthritis Rheum 1990; 33: 2816.
  • 26
    Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992; 30: 47383.
  • 27
    Stewart AL, Hays RD, Ware JE Jr. The MOS short-form general health survey: reliability and validity in a patient population. Med Care 1988; 26: 72435.
  • 28
    Feghali-Bostwick C, Medsger TA Jr, Wright TM. Analysis of systemic sclerosis in twins reveals low concordance for disease and high concordance for the presence of antinuclear antibodies. Arthritis Rheum 2003; 48: 195663.
  • 29
    Reveille JD, Durban E, MacLeod-St Clair MJ, Goldstein R, Moreda R, Altman RD, et al. Association of amino acid sequences in the HLA-DQB1 first domain with antitopoisomerase I autoantibody response in scleroderma (progressive systemic sclerosis). J Clin Invest 1992; 90: 97380.
  • 30
    Tan FK, Wang N, Kuwana M, Chakraborty R, Bona CA, Milewicz DM, et al. Association of fibrillin 1 single-nucleotide polymorphism haplotypes with systemic sclerosis in Choctaw and Japanese populations. Arthritis Rheum 2001; 44: 893901.
  • 31
    Steen VD, Lanz JK Jr, Conte C, Owens GR, Medsger TA Jr. Therapy for severe interstitial lung disease in systemic sclerosis: a retrospective study. Arthritis Rheum 1994; 37: 12906.
  • 32
    Vlachoyiannopoulos PG, Dafni UG, Pakas I, Spyropoulou-Vlachou M, Stavropoulos-Giokas C, Moutsopoulos HM. Systemic scleroderma in Greece: low mortality and strong linkage with HLA-DRB1*1104 allele. Ann Rheum Dis 2000; 59: 35967.
  • 33
    Chang B, Wigley FM, White B, Wise RA. Scleroderma patients with combined pulmonary hypertension and interstitial lung disease. J Rheumatol 2003; 30: 2398405.
  • 34
    Sacks DG, Okano Y, Steen VD, Curtiss E, Shapiro LS, Medsger TA Jr. Isolated pulmonary hypertension in systemic sclerosis with diffuse cutaneous involvement: association with serum anti-U3RNP antibody. J Rheumatol 1996; 23: 63942.
  • 35
    Tormey VJ, Bunn CC, Denton CP, Black CM. Anti-fibrillarin antibodies in systemic sclerosis. Rheumatology (Oxford) 2001; 40: 115762.
  • 36
    Schachter LM, Dixon J, Pierce RJ, O'Brien P. Severe gastroesophageal reflux is associated with reduced carbon monoxide diffusing capacity. Chest 2003; 123: 19328.
  • 37
    Kinuya K, Nakajima K, Kinuya S, Michigishi T, Tonami N, Takehara K. Esophageal hypomotility in systemic sclerosis: close relationship with pulmonary involvement. Ann Nucl Med 2001; 15: 97101.
  • 38
    Scussel-Lonzetti L, Joyal F, Raynauld JP, Roussin A, Rich E, Goulet JR, et al. Predicting mortality in systemic sclerosis: analysis of a cohort of 309 French Canadian patients with emphasis on features at diagnosis as predictive factors for survival. Medicine (Baltimore) 2002; 81: 15467.
  • 39
    Patel KV, Eschbach K, Ray LA, Markides KS. Evaluation of mortality data for older Mexican Americans: implications for the Hispanic paradox. Am J Epidemiol 2004; 159: 70715.
  • 40
    Hunt KJ, Resendez RG, Williams K, Haffner SM, Stern MP, Hazuda HP. All-cause and cardiovascular mortality among Mexican-American and non-Hispanic White older participants in the San Antonio Heart Study: evidence against the “Hispanic paradox”. Am J Epidemiol 2003; 158: 104857.
  • 41
    Franzini L, Ribble JC, Keddie AM. Understanding the Hispanic paradox. Ethn Dis 2001; 11: 496518.
  • 42
    Morelli S, Barbieri C, Sgreccia A, Ferrante L, Pittoni V, Conti F, et al. Relationship between cutaneous and pulmonary involvement in systemic sclerosis. J Rheumatol 1997; 24: 815.
  • 43
    Ferri C, Valentini G, Cozzi F, Sebastiani M, Michelassi C, La Montagna G, et al. Systemic sclerosis: demographic, clinical, and serologic features and survival in 1012 Italian patients. Medicine (Baltimore) 2002; 81: 13953.
  • 44
    Atamas SP, White B. Cytokine regulation of pulmonary fibrosis in scleroderma [review]. Cytokine Growth Factor Rev 2003; 14: 53750.