Pulmonary function in Thai patients with systemic sclerosis; a single center 6-year retrospective study

Background

Systemic sclerosis (SSc) is an autoimmune disease that results from microvascular damage, dysregulation of innate and adaptive immunity, and widespread fibrosis that affects multiple organs. While skin fibrosis is a key feature in patients with SSc, the clinical prognosis is determined by the severity of internal organ involvement.¹ The prevalence of SSc ranged from 38 to 341 cases per million, and the 5- and 10-year survival rates following diagnosis are 75% and 63%, respectively.²

Major internal organ involvements in SSc include the pulmonary, cardiovascular, renal, and gastrointestinal systems. Notably, pulmonary involvement has emerged as a primary complication during the initial stages in Thai patients and is a significant contributor to mortality.^3–5 The clinical presentations of SSc patients with pulmonary involvement include dyspnea, non-productive cough, and fine crackles at the lung based on auscultation. The decline in forced vital capacity (FVC) was significantly higher in patients who had anti-topoisomerase autoantibody (anti-Scl70). In contrast, sex and age did not correlate with pulmonary function.³ In particular, both high and low BMI might influence lung function due to changing in chest wall elastance, potentially impacting disease outcome.⁶

This study aimed to demonstrate the characteristics of pulmonary function in Thai patients with SSc and to explore the potential association between pulmonary function, body mass index (BMI), and the presence of anti-Scl70. The findings of this study may hold significant value in shaping management guidelines and provide insights for future studies on pulmonary complications in Thai patients with SSc.

Methods

Results

From a total of 211 SSc patients, 128 who underwent PFT were enrolled. Spirometry results were available for 118 (92.2%) patients, and DLCO results were available for 108 (84.4%) patients. The mean (SD) age of the patients was 54.0 (12.5) years, with 102 (79.7%) being female. The median (IQR) BMI for all patients was 21.7 (19.6-25.5) kg/m². Regarding anti-Scl70, 14 patients (10.9%) were positive, 10 patients (7.8%) were negative, and the status was unreported for 104 patients (81.3%), as detailed in Table 1.¹⁰

The primary outcome of the study revealed mean (SD) % predicted of FEV1, FVC, and DLCO as 76.3 (16.3), 69.1 (15.8), and 75.5 (22.8), respectively. The mean (SD) FEV1/FVC ratio was 0.8 (0.1). Notably, a restrictive spirometry pattern was predominant among the majority of the patients, accounting for 78.8% (93/118) of the study population. Further details regarding pulmonary function are presented in Table 2.

In the secondary outcome analysis, following the categorization of patients based on the presence of anti-Scl70, no statistical differences were observed in FVC, FEV1, FEV1/FVC, or DLCO (Table 3). DLCO demonstrated a moderate positive linear correlation with FVC (r=0.50, p<0.001) and a moderate negative linear correlation with BMI (r=-0.36, p<0.001) (Figure 1). However, there was no correlation between FVC and BMI (r=-0.15, p-value=0.107).

Figure 1. The correlation between diffusing capacity for carbon monoxide (DLCO) and forced vital capacity (FVC) (left) and DLCO and body mass index (BMI) (right).

Patient characteristics, including age, sex, BMI, and the presence of anti-Scl70, did not demonstrate significant differences between those with and without a restrictive spirometry pattern. Notably, the majority of the patients with decreased DLCO exhibited a restrictive spirometry pattern (Table 4).

Discussion

In this study, the mean % predicted values of FVC and DLCO were 69.1 and 75.5, respectively. A substantial majority of the patients (78.8%) exhibited a restrictive spirometry pattern. These results are consistent with those of a prior study on ILD in Thai patients with SSc, where the mean % predicted FVC were 71.8 in diffuse cutaneous SSc (dcSSC) and 77.6 in limited cutaneous SSc (lcSSc). They also found that patients with dcSSc had a higher prevalence of ILD (78.7% vs. 45.8%, p=0.002).⁸ These findings emphasize the impact of pulmonary involvement on SSc.

Our study found a lower mean % predicted FVC (69.1%) compared to the previous large-scale study, The European Scleroderma Trials and Research group (EUSTAR) cohort, where the mean % predicted FVC in SSc patients were 90.3%. This difference may be attributed to the fact that the EUSTAR cohort assessed patients who presented within one year after the onset of Raynaud’s phenomenon, the most common initial presentation. In addition, a prior Thai study by Foocharoen et al.⁵ found that the shorter duration of pulmonary fibrosis after onset correlated with poorer outcome. This highlights the intricate nature of the association between disease duration and lung function in SSc. Another important contributing factor is the prevalence of dcSSc, which is a risk factor of pulmonary involvement, is higher in Thai (72.6%) compared to EUSTAR (20.6%) cohort.^3,5

Our study revealed no association between demographic data and restrictive spirometry results in patients with SSc. This suggests that the manifestation of a restrictive spirometry pattern in patients with SSc may be influenced by disease-specific factors rather than by general demographic characteristics. Similarly, the presence of anti-Scl70 did not significantly affect pulmonary function, although this finding diverges from previous studies suggesting a negative association with pulmonary outcome.^3,8 However, it is important to note that a majority of our patients in this study demonstrated a restrictive spirometry pattern and only a small number of our patients had documented records of the presence of anti-Scl70. These limitations may constrain our ability to thoroughly assess the impact of these factors on the pulmonary function.

DLCO and FVC are the two most frequently used PFT for assessing the pulmonary involvement of SSc.⁹ SSc usually undergo regular DLCO and FVC. Our findings revealed a moderate linear correlation between DLCO and FVC, and nearly all patients with decreased DLCO exhibited a restrictive spirometry pattern. This finding suggests that spirometry can serve as an effective screening test for pulmonary involvement in SSc and is a cost-effective option, particularly in resource-limited settings.

In this study, some essential medical history, particularly patient symptoms, previous treatments, disease duration, and co-existing organ involvement, were documented in diverse formats, posing challenges for analysis. Therefore, the earliest test conducted during the study period was employed to mitigate the impact of ongoing treatment and disease progression. However, this limits our ability to evaluate the relationship between pulmonary function and these factors. Further studies with pre-specified data collection of these factors may reveal new tools for the detection of pulmonary involvement.

The strengths of our study are its considerable number of participants in a data-scarce field and being the only study that aims to explore pulmonary function exclusively in Thai patients with SSc. The effect of selective bias in the study is modest because most of our patients with SSc were screened with spirometry annually, regardless of patient symptoms.

However, this study has some limitations. First, since SSc is a progressive disease, lack of disease duration data limited our ability to comprehensively assess disease severity and explore potential relationships between disease severity, anti-Scl-70 levels, and their combined impact on lung function in SSc patients. Second, the small number of documented anti-Scl70 decreased our power to examine its relationship with patient pulmonary function. Third, our study did not capture data on the type of systemic sclerosis in all participants. This omission limits our ability to assess the potential influence of SSc type on FVC and DLCO. Fourth, our study did not include pulmonary hypertension (PH) assessments, thus limiting our ability to conclusively determine the extent to which PH contributed to the observed DLCO values. Lastly, the 40% non-participation rate in spirometry testing (83/211) introduces potential selection bias, as participants who underwent testing may differ in disease severity, functional capacity, or willingness.

Conclusion

A restrictive spirometry pattern is common among Thai patients with SSc. However, the power of using demographic data and presence of anti-Scl70 to determine the probability of pulmonary complications remains limited. Further studies are required to evaluate anti-Scl70 data, SSc type, pulmonary hypertension assessment, including comprehensive data on disease duration to gain a more understanding of the disease course and its impact on SSc patients.

Ethics and consent

Ethical approval was obtained from the Human Research Ethics Committee of Thammasat University (Faculty of Medicine), Thailand (Project number MTU-EC-IM-1-177/65, Approval number 193/2022, Date of approval September 19, 2022), and the study was conducted according to the Declaration of Helsinki. The informed consent was waived in view of the retrospective nature of the study.