Clinical characteristics of IIM patients

A total of 142 patients with IIM were recruited. In our study, 98 patients were new-onset and 44 patients were relapsed. They were hospitalized with clinical symptoms due to the activity or recurrence of the disease. The clinical characteristics and laboratory parameters of the patients were shown in Table 1. Of all the cases, 81 had DM, 14 had PM, 33 had ASS and 14 had IMNM. The age of the patients ranged from 19 to 82 years, with a median age of 55 years. The majority of patients were women (71.8%) and the median disease course was 5 months. We identified that 41 (28.9%) IIM patients had cardiac involvement. The types of cardiac involvement were detailed in supplementary Table 1. The distribution of antibody in patients with cardiac involvement was presented in Supplementary Fig. 1. We observed that patients with positive anti-signal recognition particle (SRP) autoantibody had the highest rate (50%) of cardiac involvement, followed by anti-glycyl (EJ) autoantibody (40%) and anti-histidyl (Jo-1) autoantibody (33.3%). We found that patients with cardiac involvement were older (61 years vs. 52 years, p < 0.001) and had higher levels of white blood cell (WBC) (10.6 *10^9/L vs. 8.1 *10^9/L, p = 0.014), C-reactive protein (CRP) (5 mg/L vs. 3 mg/L, p = 0.047), erythrocyte sedimentation rate (ESR) (12 mm/h vs. 7 mm/h, p = 0.044). As for cardiac biomarkers, CK-MB (5.4 ng/mL vs. 1.4 ng/mL, p < 0.001), hs-cTnI (37.6 pg/mL vs. 4.8 pg/mL, p < 0.001), Mb (73.6 ng/mL vs. 22.6 ng/mL, p = 0.001), NT-proBNP (419.7 pg/mL vs. 84.2 pg/mL, p < 0.001), lactate dehydrogenase (LDH) (334 IU/L vs. 240IU/L, p = 0.001) and aspartate aminotransferase (AST) (42 IU/L vs. 27 IU/L, p = 0.022) were significantly higher in patients with cardiac involvement than those without cardiac involvement. Furthermore, higher indicators of disease activity were observed in patients with cardiac involvement, including PhGA (4 vs. 3.2, p = 0.013), HAQ (0.5 vs. 0.3, p = 0.040), and MDAAT (3 vs. 2, p < 0.001).

Table 1 Demographic and clinical features in IIM patients The correlations between serum cardiac biomarkers and Disease activity in IIM patients

To further explore the relationship between serum cardiac biomarkers and disease activity, we performed the Spearman correlation test between the levels of CK, CK-MB, hs-cTnI, Mb, NT-proBNP and disease activity score (Table 2). All these biomarkers showed significant correlations with the disease activity. The hs-cTnI had significant correlations with PhGA (r = 0.403, p < 0.001), PGA (r = 0.296, p < 0.001), HAQ (r = 0.254, p < 0.001), MMT-8 (r = -0.187, p = 0.026) and MDAAT (r = 0.342, p < 0.001). The CK-MB showed significant correlations with PhGA (r = 0.431, p < 0.001), PGA (r = 0.299, p < 0.001), HAQ (r = 0.283, p = 0.001) and MMT-8 (r = -0.351, p < 0.001). The Mb was highly correlated with PhGA (r = 0.466, p < 0.001), PGA (r = 0.252, p = 0.002), HAQ (r = 0.270, p = 0.001) and MMT-8 (r = -0.303, p < 0.001). The serum CK levels was also correlated with PhGA (r = 0.311, p < 0.001), PGA (r = 0.203, p = 0.015), HAQ (r = 0.227, p = 0.007) and MMT-8 (r = -0.279, p = 0.001). The NT-proBNP showed significant correlations with PhGA (r = 0.307, p < 0.001), PGA (r = 0.227, p = 0.007) and MDAAT (r = 0.305, p < 0.001). In addition, we evaluated the serum cardiac biomarkers before and after treatment in Supplementary Fig. 2. We found that all these indicators significantly decreased after remission.

Table 2 Correlation of serum cardiac biomarkers levels with disease activity in IIM patients The value of serum cardiac biomarkers in identifying cardiac involvement

To evaluate the ability of different serum biomarkers in distinguishing cardiac involvement, we performed the ROC analysis. We found that the hs-cTnI in the active period had the best predictive value (AUC = 0.848, 95%CI: 0.772,0.924; p < 0.001) (Fig. 1). The cut-off value at 13.8 pg/ml had a sensitivity of 80.5% and a specificity of 79.2%. Similarly, we calculated the AUC for the other four indicators including NT-proBNP (AUC = 0.835, 95%CI: 0.767,0.904; p < 0.001), CK-MB (AUC = 0.707, 95%CI: 0.616,0.797; p < 0.001), Mb (AUC = 0.683, 95%CI: 0.586,0.781; p = 0.001) and CK (AUC = 0.602, 95%CI: 0.493,0.710; p = 0.058). Although the markers were similarly useful in identifying cardiac involvement, they were still less efficient than hs-cTnI. Furthermore, we performed the logistic regression analysis to evaluate the potential risk factors of cardiac involvement in IIM patients (Table 3). Univariate logistic regression analysis showed that the age (> 49.5 years), the initial levels of WBC (> 11.4*10^9/L), CRP (> 11.5 mg/L), ESR (> 9.5 mm/h), CK-MB (> 2.1 ng/mL), hs-cTnI (> 13.8 pg/ml), Mb (> 38.3 ng/mL), NT-proBNP (> 247 pg/ml), LDH (> 405 IU/L) and AST (> 49.5 IU/L) were highly associated with cardiac involvement. Multivariate logistic regression analysis demonstrated that age (> 49.5 years) and the serum hs-cTnI (> 13.8 pg/ml) were independent risk factors of cardiac involvement with an odds ratio (OR) of 5.560 (95% CI: 1.264, 24.468; p = 0.023) and 7.810 (95% CI: 1.962, 31.097; p = 0.004), respectively.

Fig. 1

Receiver operating characteristic (ROC) curves of the predictive capacity of the serum cardiac biomarkers to distinguish cardiac involvement in IIM patients. The area under the curve (AUC) was 0.848 (95%CI: 0.772,0.924; p < 0.001) for hs-cTnI, 0.835 (95%CI: 0.767,0.904; p < 0.001) for NT-proBNP, 0.707 (95%CI: 0.616,0.797; p < 0.001) for CK-MB, 0.683 (95%CI: 0.586,0.781; p = 0.001) for Mb and 0.602 (95%CI: 0.493,0.710; p = 0.058) for CK. CK: creatine kinase; CK-MB: creatine kinase MB; hs-cTnI: high-sensitivity cardiac troponin I; Mb: myoglobin; NT-proBNP: N-terminal pro-B-type natriuretic peptide

Table 3 Logistic regression of risk factors for cardiac involvement in IIM patients The value of hs-cTnI in predicting the treatment and prognosis in IIM patients

To evaluate the effect of hs-cTnI on the treatment and prognosis, we divided the patients into two groups according to the levels of hs-cTnI (Table 4). Our results showed that patients with abnormal levels of hs-cTnI (> 30 pg/mL) were prescribed higher doses of glucocorticoids (p = 0.001). Besides, the beta blocker, angiotensin converting enzyme (ACE) inhibitor/angiotensin receptor blocker (ARB) (p = 0.002) and antiplatelet/anticoagulant treatment (p < 0.001) were more frequently used in patients with elevated hs-cTnI (> 30 pg/mL). During a median follow-up of 15 months, the overall survival tended to be lower in patients with elevated hs-cTnI (> 30 pg/mL) (p = 0.004) (Fig. 2A). Three deaths were recorded, and the causes of death were heart failure (n = 2) and respiratory failure (n = 1). In addition, an increased risk of cardiac involvement during follow-up was observed in the group with initial elevated levels of hs-cTnI (> 30 pg/mL) (p < 0.001) (Fig. 2B).

Table 4 The treatments and outcomes of IIM patients according to the levels of hs-cTnI. Fig. 2

The survival curve in different groups of patients according to the levels of hs-cTnI. The overall survival (A) and the incidence of cardiac involvement (B). hs-cTnI: high-sensitivity cardiac troponin I