Baseline characteristics of patients
We screened 745 patients using ICD-10 codes, of which 172 patients with SSc-ILD were enrolled. The overall prevalence of SSc-ILD among SSc patients was 23.1%, according to the patient recruitment chart. Table 1 presents the baseline patient characteristics. Females were dominant in both groups. Patients in the extensive group were younger (mean age ± SD, 49.3 ± 11.5) than those in the limited group (53.9 ± 12.5, p = 0.067). No significant differences in autoimmune markers, such as anti-centromere Ab, anti-RNA polymerase Ab, and anti-Scl-70 Ab, were found between the two groups. The extensive group showed frequent pulmonary hypertension at 43.5% compared to 16.7% in the limited group. The extensive group also had significantly higher ESR levels (61.3 ± 33.7 vs. 42.1 ± 26.0, p = 0.003) and higher mortality (32.6% vs. 10.0%, p = 0.011) compared to the limited group. The most common causes of death were pneumonia (n = 8) and malignancy (n = 6; two cases of mesothelioma and four cases of lung cancer), followed by pulmonary artery hypertension (n = 1), myocarditis (n = 1), sepsis (n = 1), renal crisis (n = 1), post-procedural (n = 1), and sudden cardiac death (n = 1).
Most patients were treated with immunosuppressive drugs such as steroids (76.4%), azathioprine (23.6%), cyclophosphamide (7.5%), mycophenolate mofetil (MMF, 7.5%), and methotrexate (MTX, 6.6%). D-penicillamine was used as adjunctive therapy (36.8%).
Table 1. Baseline characteristics of patients
Variable
|
Limited
(n=60)
|
Extensive
(n=46)
|
Total
(n=106)
|
P-value
|
Age, years
|
53.9±12.5
|
49.3±11.5
|
51.9±12.7
|
0.067
|
Sex, men
|
7 (11.7)
|
7 (15.2)
|
14 (13.2)
|
0.773
|
Smoking exposure, n (%)
|
|
|
|
0.813
|
Never
|
39 (88.6)
|
38 (88.4)
|
77 (72.6)
|
|
Former
|
1 (2.3)
|
3 (7.0)
|
4 (3.8)
|
|
Current
|
4 (9.1)
|
2 (4.7)
|
6 (5.7)
|
|
Pack-years
|
2.6±10.0
|
2.8±8.1
|
2.7±9.1
|
0.831
|
FVC % predicted
|
86±14
|
54±14
|
71.9±21.3
|
<0.001
|
FEV1 % predicted
|
96±17
|
62±16
|
81.3±23.9
|
<0.001
|
DLCO, % predicted
|
75±20
|
54±22
|
65.3±23.6
|
<0.001
|
Progression
|
8 (13.3)
|
7 (15.2)
|
15 (14.2)
|
0.999
|
Anti-centromere Ab, n (%)
|
4 (9.5)
|
2 (6.3)
|
6 (5.7)
|
0.693
|
Anti-RNA polymerase, n (%)
|
10 (23.8)
|
8 (24.2)
|
18 (17.0)
|
0.999
|
Anti-Scl 70, n (%)
|
35 (66.0)
|
25 (56.8)
|
60 (56.6)
|
0.404
|
Reflux/dysphagia symptoms, n (%)
|
23 (38.3)
|
23 (50.0)
|
46 (43.4)
|
0.243
|
Digital ulcers, n (%)
|
46 (76.7)
|
40 (87.0)
|
86 (81.1)
|
0.216
|
Synovitis, joint symptoms, n (%)
|
29 (48.3)
|
20 (43.5)
|
49 (46.2)
|
0.843
|
ILD related respiratory symptoms
|
42 (70.0)
|
39 (84.8)
|
81 (76.4)
|
0.585
|
ESR (mm/hr)
|
42.1±26.0
|
61.3±33.7
|
50.5±31.0
|
0.003
|
Pulmonary hypertension*
|
10 (16.7)
|
20 (43.5)
|
30 (28.3)
|
0.009
|
Immunosuppressant use, n (%)
|
|
|
|
|
Steroid
|
44 (73.3)
|
37 (80.4)
|
81 (76.4)
|
|
Azathioprine
|
12 (20.0)
|
13 (28.3)
|
25 (23.6)
|
|
Cyclophosphamide
|
1 (1.7)
|
7 (15.2)
|
8 (7.5)
|
|
Mycophenolate mofetil
|
2 (3.3)
|
3 (7.0)
|
5 (4.7)
|
|
Rituximab
|
0 (0.0)
|
0 (0.0)
|
0 (0.0)
|
|
Methotrexate
|
5 (8.3)
|
2 (4.3)
|
7 (6.6)
|
|
Antifibrotic drugs
|
|
|
|
|
D-penicillamine
|
25 (41.7)
|
14 (30.4)
|
39 (36.8)
|
|
Underlying malignancy†
|
14 (21.7)
|
8 (24.2)
|
21 (19.8)
|
0.481
|
Follow up period (month)
|
100.0±44.7
|
86.0±53.4
|
92.8±49.3
|
N/A
|
Mortality
|
6 (10.0)
|
15 (32.6)
|
21 (19.8)
|
0.011
|
Abbreviations: ILD, interstitial lung disease; ESR, erythrocyte sedimentation rate; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 s; DLco, diffusing capacity of carbon monoxide. *Pulmonary hypertension was defined using traditional echocardiogram estimated right ventricular systolic pressure (RVSP) >40 mmHg.
†Malignancy:1) limited group: lung cancer (n=5), thyroid cancer (n=3), breast cancer (n=4), cervical cancer (n=2); 2) extensive group: lung cancer (n=3), thyroid cancer (n=1), breast cancer (n=1), ovarian cancer (n=1), colon cancer (n=1), ampulla of Vater cancer (n=1)
Table 2 details the clinical characteristics of the survivors and non-survivors. Those who did not survive were older than the survivors (mean age 56.9±13.4 vs. 50.7±12.3, p=0.049, non-survivors vs. survivors) and had a higher proportion of males (30.0% vs. 9.3%, p=0.024, non-survivors vs. survivors). FVC, forced expiratory volume in one second, and DLco were significantly lower in the non-survivor group; however, autoimmune antibodies and systemic symptoms were not statistically different between the groups. The ESR was higher in the non-survivor group, but the difference was not statistically significant. RVSP was significantly higher in the non-survivor group (32.3±16.8 vs. 44.7±25.3, p=0.005). The proportion of malignancy was higher in the non-survival group; however, this difference was not as statistically significant (38.1% vs. 16.5%, p=0.122) as the time to ILD development (4.3±5.5 vs. 3.2±3.9 years, p=0.760).
Table 2. Clinical characteristics between survivors and non-survivors in patients with SSc-ILD
Variable
|
Survivors
(n=85)
|
Non-survivors
(n=21)
|
P-value
|
Age, years
|
50.7±12.3
|
56.9±13.4
|
0.049
|
Sex, men
|
8 (9.4)
|
6 (28.6)
|
0.031
|
Smoking exposure, n (%)
|
|
|
0.449
|
Never
|
60
|
18
|
|
Former
|
3
|
2
|
|
Current
|
6
|
1
|
|
Pack-years
|
18.7±32.3
|
2.0±6.3
|
0.786
|
FVC % predicted
|
74.4±20.7
|
61.8±21.1
|
0.010
|
FEV1 % predicted
|
83.6±23.6
|
72.3±23.5
|
0.033
|
DLCO, % predicted
|
69.8±22.2
|
47.3±20.5
|
<0.001
|
GAP score
|
0.6±0.9
|
1.3±1.4
|
0.056
|
Anti-centromere Ab, n (%)
|
3 (3.5)
|
3 (14.3)
|
0.111
|
Anti-RNA polymerase, n (%)
|
16 (18.8)
|
2 (9.5)
|
0.337
|
Anti-Scl 70, n (%)
|
49 (57.6)
|
11 (52.4)
|
0.606
|
Reflux/dysphagia symptoms, n (%)
|
36 (42.4)
|
10 (47.6)
|
0.806
|
Digital ulcers, n(%)
|
72 (84.7)
|
14 (66.7)
|
0.069
|
ILD related respiratory symptoms
|
63 (74.1)
|
18 (85.7)
|
0.391
|
ESR (mm/hr),
|
48.0±30.4
|
61.6±31.8
|
0.090
|
RVSP (mmHg)
|
32.3±16.8
|
44.7±25.3
|
0.005
|
Underlying malignancy†
|
14 (16.5)
|
8 (38.1)
|
0.122
|
Extensive group
|
29 (34.1)
|
15 (71.4)
|
0.002
|
Follow up period (month)
|
96.4±47.0
|
80.4±56.3
|
N/A
|
Time to ILD development from first visit (year)
|
3.2±3.9
|
4.3±5.5
|
0.760
|
Abbreviations: ILD, interstitial lung disease; ESR, erythrocyte sedimentation rate; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 s; DLco, diffusing capacity of carbon monoxide; RVSP, right ventricular systolic pressure; GAP score, sex (G), age (A), physiology (P).
†Malignancy:1) Survivors: lung cancer (n=3), thyroid cancer (n=3), breast cancer (n=5), cervical cancer (n=2), colon cancer (n=1); 2) Non-survivors: lung cancer (n=5), thyroid cancer (n=1), ovarian cancer (n=1), ampulla of Vater cancer (n=1)
Table 3 shows the results of the logistic regression analysis to determine the risk factors for mortality. In the univariate analysis, the factors related to death were age (odds ratio [OR] 1.043, p=0.043), male sex (OR 4.179, p=0.020), baseline FVC (OR 0.970, p=0.017), and high RVSP (OR 1.028, p=0.043).
The multivariate analysis determined that advanced age, lower FVC, and underlying malignancy were significantly associated with mortality.
Table 3. Multivariable logistic regression models for the risk of mortality in patients with SSc-ILD
|
Univariate
|
Multivariate
|
Variables
|
OR (95% CI)
|
P-value
|
OR (95% CI)
|
P-value
|
Age, years
|
1.043 (1.001-1.086)
|
0.043
|
1.073 (1.009-1.140)
|
0.024
|
Sex, men
|
4.179 (1.256-13.897)
|
0.020
|
2.779 (0.592-13.047)
|
0.195
|
FVC % predicted
|
0.970 (0.946-0.995)
|
0.017
|
0.951 (0.910-0.993)
|
0.022
|
Anti-centromere Ab
|
4.231 (0.765-23.407)
|
0.098
|
|
|
Anti-Scl 70 Ab
|
0.622 (0.226-1.713)
|
0.358
|
|
|
ESR
|
1.016 (0.999-1.032)
|
0.062
|
|
|
RVSP (mmHg)
|
1.028 (1.001-1.056)
|
0.043
|
1.012 (0.978-1.047)
|
0.500
|
Underlying malignancy
|
2.549 (0.870-7.464)
|
0.088
|
7.531 (1.066-53.177)
|
0.043
|
Abbreviations: FVC, forced vital capacity; ESR, erythrocyte sedimentation rate; RVSP, right ventricular systolic pressure.
Table 4 shows the disease course according to the initial disease stage. Approximately 20% of patients in the limited group and 15.2% in the extensive group showed a progressive disease course. FVC declines were similar in the limited and extensive groups, at 15–20% in the first year; however, approximately 8–10% of patients showed progressive lung function decline regardless of the initial disease extent.
Table 4. Disease course according to initial disease stage
Variable
|
First 12-month period with FVC decline
|
After 2nd year period with FVC decline
|
Overall FVC decline
|
Progression
|
Stable
|
Progression
|
Stable
|
Extensive (n=46)
|
7 (15.2)
|
39 (84.8)
|
5 (10.7)
|
41 (89.1)
|
Limited (n=60)
|
12 (20.0)
|
48 (80.0)
|
5 (8.3)
|
55 (91.7)
|
Abbreviation: FVC=forced vital capacity
Kaplan–Meier survival curves (Figure 2) estimated all-cause mortality between the limited and extensive groups. The extensive group showed significantly higher mortality during the follow-up period (log-rank test, p=0.017).