1 INTRODUCTION

Dermatomyositis (DM) is an autoimmune disease characterized by insidious or subacute onset with primary clinical manifestations of muscle weakness and skin rash. The diagnosis of DM currently depends on serological tests, imaging, electromyography, and clinical observations. As the disease progresses, it can gradually affect various organs, with interstitial lung disease (ILD) being the most common complication [1-4]. The incidence of DM ranges from 1 to 6 cases per 100 000 individual, with a female to male ratio of approximately 2:1 [5]. Notably, the overall incidence of ILD among patients with DM is 41%, and this figure rises to about 50% within the Asian population [4].

In its initial stages, dermatomyositis-associated interstitial lung disease (DM-ILD) can be highly occult and challenging to detect through conventional imaging and serological tests. When further complicated by interstitial pneumonia, reversing lung damage becomes increasingly difficult [6-8]. DM-ILD exhibits considerable heterogeneity, necessitating individualized treatment strategies based on the form of onset, rate of progression, extent of lesion involvement, and treatment response to treatment by other adverse factors. The management of DM primarily involves glucocorticoids, immunosuppressants, biological agents, and other pharmacological interventions [9, 10]. Timely intervention in DM-ILD is critical for improving patient outcomes, emphasizing the importance of early detection of ILD complications in DM patients.

Arterial blood gas analysis provides real-time insights into blood pH and oxygen levels [11]. In the presence of pulmonary interstitial lesions, alveolar gas exchange is compromised, leading to alterations in arterial oxygen partial pressure and pH levels [11]. Analysis of arterial blood gas is essential for assessing potential complications of ILD in DM patients. Additionally, serum ion levels serve as vital indicators of acid-base metabolism and overall metabolic function. Currently, hematological indicators specifically associated with DM-ILD are limited; however, serological and arterial blood testing offer the advantages of convenience and minimal invasiveness. This study aims to explore the potential roles of various biomarkers in monitoring interstitial pneumonia complications in DM by analyzing changes in blood gas indices, disease progression, complications, and other relevant factors.

2 MATERIALS AND METHODS

3 RESULTS

3.2 Comparison of indices between the control group, DM group, and DM-ILD group

Relative to the control group, the DM group exhibited increased levels of pH, Actual bicarbonate (AB), and BE, and decreased levels of K⁺, Na⁺, and Cl⁻, all with statistical significance (p < 0.05). Compared with the DM group, there was a decrease in pH and BE levels and an increase in Cl⁻ levels in DM-ILD group, also statistically significant (p < 0.05), as shown in Figure 1.

3.3 The relationship between ILD and disease course in DM

The course of illness in the DM-ILD subgroup was longer than in the general DM population (p = 0.013), as shown in Figure 2. The proportions of ILD among DM patients over time periods of 0–2, 3–4, 5–6, 7–8, and more than 8 years were 48.0%, 41.2%, 72.0%, 66.7%, and 88.9%, respectively, as shown in Table 2.

TABLE 2. Incidence of DM-ILD over different disease courses.

Disease course (yrs)	DM-ILD(n)/DM(n)[%]
0–2	12/25 (48.0)
3–4	7/17 (41.2)
5–6	18/25 (72.0)
7–8	8/12 (66.7)
>8	8/9 (88.9)

• Note: DM, dermatomyositis patients with this disease course; DM-ILD, dermatomyositis-associated interstitial lung diseases with this disease course.

3.5 Correlation analysis among indicators in patients with DM

Significant correlations were observed for the duration of interstitial pneumonia, and between gender/age, pH/interstitial pneumonia, age/PaO₂, gender/PaCO₂, pH/PaCO₂, PaCO₂/AB, all showing statistical significance (p < 0.05), as shown in Figure 3.

3.6 The value of age, pH, Cl⁻ and disease course in predicting DM-ILD

Differences in pH, BE, disease course and serum Cl⁻ levels were noted between the DM and DM-ILD groups. Therefore, these four indicators were included in a binary logistic regression model. To minimize the influence of uncontrollable patient factors, age and gender were also considered. It was determined that gender and BE did not reach statistical significance and were therefore excluded from further analysis. Age, disease course, pH, and Cl⁻ demonstrated statistical significance (p = 0.020, 0.006, 0.041, 0.016) as predictors of DM-ILD, as shown in Figure 4.

3.7 ROC curve analysis of arterial blood gas pH and disease duration for predicting DM-ILD

The ROC curve analysis identified cutoff values for age, disease duration, pH, and Cl⁻ at 55.5 years, 5.5 years, 7.432, and 101.5 mmol/L, respectively, which provided the highest sensitivity and specificity for predicting DM-ILD. The combined assessment of these four variables showed statistical significance (p < 0.001). The sensitivity and specificity for predicting DM-ILD were 0.809 and 0.722, respectively, with an area under the curve of 0.830, as shown in Table 5 and Figure 5.

TABLE 5. Efficacy of arterial blood gas pH, disease course, serum Cl⁻, age, and their combined use in predicting DM with interstitial pneumonia.

Variable	AUC	Sensitivity	Specificity	Cutoff	p	95% CI for AUC
						Lower	Upper
Age	0.610	0.415	0.585	55.5 yrs	0.078	0.488	0.732
Disease course	0.656	0.472	0.528	5.5 yrs	0.013	0.543	0.769
pH	0.629	0.849	0.151	7.432	0.039	0.506	0.752
Cl−	0.648	0.642	0.639	101.5 mmol/L	0.018	0.530	0.766
Combined detection	0.809	0.830	0.722	—	<0.001	0.714	0.905

• Note: Combined detection: Age, disease course, pH, and Cl⁻ are combined.

4 DISCUSSION

DM is a complex disease with multiple etiological factors including immune dysfunction, genetic predispositions, environmental exposure, and infections [13, 14]. Epidemiological studies have shown a higher incidence of DM among women than men [5]. In this 3-year study with a sample size of 89 randomly selected cases, the proportion of male patients was significantly lower than that of female patients, which aligns with previous findings [5]. Patients with DM often exhibit abnormal immune functions, triggering inflammatory responses that lead to lung tissue damage through the infiltration of inflammatory cells. This process culminates in irreversible fibrosis within the pulmonary interstitium, progressing to dyspnea and other respiratory symptoms [8, 15]. The primary objective of this study was to explore the relationship between DM-ILD and both arterial blood gas indices and serum ion levels. It also aimed to analyze the timing of ILD onset and identify potential predictive indices for DM-ILD.

Our findings demonstrate that compared to the general population, individuals with DM exhibit higher pH and AB, along with reduced venous blood K⁺ levels. A decrease in K⁺ can lead to its extracellular migration, prompting an intracellular influx of H⁺ and Na⁺, which may result in a more alkaline blood pH and subsequent metabolic alkalosis. Our analysis revealed minimal differences in the PaCO₂ between the two groups, suggesting that the pH alteration in DM patients is not due to respiratory factors but rather to conditions predisposing them to metabolic complications [16, 17]. Further comparisons between DM patients with and without interstitial pneumonia showed that those without the complication had elevated levels of pH, AB, and BE, which is consistent with the metabolic alkalosis discussed earlier. Conversely, the pH values in DM-ILD patients were lower than those in DM patients without interstitial complications, indicating a potential occurrence of acidosis in the presence of interstitial pneumonia.

The lungs play a pivotal role in ventilation and gas exchange within the human body. In instances where lung interstitium is compromised, pulmonary hyperventilation typically leads to alterations in PaCO₂. Interestingly, our study noted that PaCO₂ levels remained stable, suggesting that the observed acidosis may be predominantly driven by metabolic factors rather than respiratory changes. DM is recognized as a systemic inflammatory disorder that affects multiple organs including the lungs, kidneys, and heart. This multifaceted involvement was underscored by variations in serum ion concentrations detected in our analysis. Notably, in the absence of overt respiratory symptoms, the body often initiates a range of compensatory responses [18]. The lungs are closely related to the regulation of blood oxygen concentration; as pulmonary disease progresses, a decline in blood oxygen levels is observed [7, 19, 20]. The heart and kidneys—organs characterized by high blood flow—work synergistically. The heart ensures continuous blood flow to the kidneys, which in turn regulate this flow, creating a complementary relationship [21]. As the burden on these organs increases, fluctuations in blood flow and resultant ionic imbalances occur leading to changes in blood pH levels. Despite the presence of lung lesions, their mild nature in this study meant that PaCO₂ levels were not significantly altered. The observed reduction in pH levels is likely attributable to compensatory mechanisms rather than direct respiratory alterations. This conclusion is supported by the concurrent reduction in AB, a positive shift in BE, and an increase in blood K⁺ levels noted during the study. In individuals with DM, the study revealed that metabolic alkalosis can develop even when the lungs are not affected. As the disease progresses and the partial pressure of oxygen declines, a reduction in metabolic bases occurs. The early onset of metabolic alkalosis before the development of interstitial pneumonitis facilitates a more effective monitoring of the lung's pathological state. Furthermore, this study identified a correlation between the duration of DM and the onset of ILD, with ILD presenting more frequently in patients who had experienced a longer duration of DM. These findings underscore the complex relationship between DM and ILD, reinforcing the interconnected nature of these conditions.

By grouping patients with DM according to age and gender, it was concluded that alterations in pH did not correlate with these factors. This observation reinforces the hypothesis of the present study that pH variations in patients with DM are predominantly linked to interstitial pneumonia rather than age or gender. Upon analyzing data—including blood gas pH, disease duration, serum Cl⁻ levels, age, and gender—through a binary logistic model, it was determined that age, disease duration, pH, and serum Cl⁻ were predictors of interstitial pneumonia development. The predictive odds ratio for disease duration was 1.37, indicating that with each additional year of disease duration, the likelihood of developing interstitial pneumonia was 1.37 times that of the previous year. Further analysis revealed that monitoring these four indices significantly enhances the early detection of interstitial pneumonia. ROC curves demonstrated that the combined assessment substantially outperformed individual tests in predicting interstitial pneumonia, achieving a prediction accuracy of 80.9%, with notable improvements in both sensitivity and specificity. These findings underscore the importance of regular monitoring of arterial blood gas pH and serum Cl⁻ levels to effectively preempt interstitial pneumonia as DM progresses and patient age. Specifically, the predicted threshold for the course of the disease was found to be 5.5 years. In Table 2, ILD incidence remained below 50% for patients whose disease duration was under 5 years but increased notably between five and six years. These results suggest that ILD should be monitored closely when the disease duration reaches 5–6 years. Although arterial blood gas analysis is routinely performed and is minimally invasive, it involves slightly more trauma than venous blood collection. Based on these findings, we recommend that patients with dermatomyositis under 55 years of age undergo annual monitoring after five years from disease onset, while those over 55 years should undergo annual blood gas analyses to detect potential complications early.

The generalizability of these findings is subject to certain limitations. Notably, although this study investigated the relationship between patients with DM and arterial blood gas and ions, it is important to acknowledge that due to treatment factors, the impact of patient medication cannot be disregarded. Nonetheless, the findings of this study may serve as a reference for evaluating complications of DM during the treatment period. Given that dermatomyositis is a clinically rare disease, the sample size in this study was necessarily small. Future research will aim to enlarge the study cohort to facilitate comprehensive follow-up, including monitoring patient survival, dynamic observation of patient conditions, and enhancing service provision in clinical settings.

In conclusion, the analysis of arterial blood gas and serum Cl⁻ seems to play a more significant indicative role in the development of DM-ILD. Furthermore, the combined monitoring of arterial blood gas pH, disease duration, patient age, and serum Cl⁻ could prove to be an effective method for predicting DM-ILD, offering substantial potential clinical value.

AUTHOR CONTRIBUTIONS

Xu Zhang: Conceptualization (equal); data curation (equal); formal analysis (equal); resources (equal); validation (equal); writing – original draft (equal). Xuemei Wei: Conceptualization (equal); resources (equal); writing – original draft (equal). Xiaojuan Luan: Data curation (equal); resources (equal). Xiujuan Li: Data curation (equal); resources (equal). Jin Dong: Investigation (equal); validation (equal); writing – review & editing (equal). Jingzhu Nan: Formal analysis (equal); supervision (equal). Yanhong Gao: Writing – review & editing (equal).

CONFLICT OF INTEREST STATEMENT

All authors declare no conflicts of interest.

ETHICS STATEMENT

The studies involving human participants were reviewed and approved by the First Medical Center of Chinese PLA General Hospital, and the approved number was S2021-087-01.

INFORMED CONSENT

Not applicable.