Patient characteristics

We included 6868 inpatients who underwent plasma EBV-DNA, ALT, and ADA assays. The average age was 4.3 (± 3.0) years, with a male-to-female ratio of 1:4. The plasma EBV DNA positivity rate was 7.0% (483/6868); 1877 (27.3%) and 1667 (24.3%) children had elevated ADA and ALT levels, respectively.

The inpatients were divided into four categories based on their discharge diagnosis: inflammatory disease (n = 3030), autoimmune disease (n = 1006), malignant disease (n = 1061), and other diseases (n = 1065). Other diseases include anemia, hemophilia, jaundice, gastrointestinal malformations, inherited metabolic diseases, immunodeficiency diseases, facial paralysis, epilepsy, skin rashes, cardiomyopathy, and congenital heart disease.

General characteristics based on the ADA level

Overall, 4991 patients had a normal ADA level, and 1877 had an elevated level. Sex, age, the ALT level, plasma EBV DNA positivity rate, and the disease category significantly differed between the normal and elevated ADA groups (P < 0.05; Table 1). An elevated ADA level was most common in girls, those aged < 3 years, and those with inflammatory diseases, a high ALT level, and EBV positivity.

Table 1 Comparison of clinical characteristics in patients with normal and elevated ADAElevated ADA risk factors

Univariate regression analysis identified associations between an elevated ADA level and the ALT level, EBV infection, and the disease category (Table 2). Furthermore, multivariate regression analysis adjusted for sex and age indicated that the ALT level (adjusted OR [aOR] = 1.001, 95% CI: 1.001–1.002), EBV infection (aOR = 8.486, 95% CI: 6.753–10.663), inflammatory disease (aOR = 3.915, 95% CI: 3.198–4.794), autoimmune disease (aOR = 2.307, 95% CI: 1.823–2.920), and malignant disease (aOR = 1.381, 95% CI: 1.101–1.734) were risk factors for an elevated ADA level; of them, EBV infection was the strongest risk factor.

Table 2 Univariable and multivariable logistic regression models of factors associated with elevated ADA in hospitalized patientsThe ADA level in EBV-related diseases

Based on the discharge diagnosis, 483 children were plasma EBV-positive and diagnosed with a respiratory infection, IM, atypical EBV infection, malignant disease (e.g., hemophagocytic lymphohistiocytosis and tumors), and other diseases. Other diseases include Kawasaki disease, idiopathic thrombocytopenic purpura, epilepsy, stomatitis, and meningitis. Furthermore, 279 children had a normal ALT level, and 204 had an elevated level.

Of those with a normal ALT level, significantly more children with IM (97.3%) and atypical EBV infection (91.7%) had an elevated ADA level than those with a respiratory infection (45.8%), malignant disease (57.1%), or other diseases (51.4%) (P < 0.05; Table 3 and Fig. 1). Moreover, the ADA concentration was significantly higher in the IM and atypical EBV infection groups than in the other three groups (P < 0.05).

Table 3 The value of ADA in children with positive plasma EBV-DNAFig. 1

The value of ADA in patients with EBV-related diseases. *P < 0.05 (a versus b)

Finally, patients with IM, atypical EBV infection, and respiratory infection in the elevated ALT level group had higher ADA levels than those in the normal ALT level group (Fig. 2).

Fig. 2

ADA in normal and elevated ALT groups from patients with positive plasma EBV-DNA. A The value of ADA in normal and elevated ALT groups from patients with respiratory infection. B The value of ADA in normal and elevated ALT groups from patients with IM. C The value of ADA in normal and elevated ALT groups from patients with atypical EBV infection. D The value of ADA in normal and elevated ALT groups patients with malignant disease. E The value of ADA in normal and elevated ALT groups from patients with other diseases

Correlations between the ADA level and EB viral load in children with a normal ALT level

The EB viral load was significantly higher in the IM group than in the atypical EBV infection, respiratory infection, malignant disease, and other disease groups (Fig. 3). Furthermore, Spearman’s correlation analysis showed that the EB viral load positively correlated with the ADA level (r = 0.501, P < 0.05; Fig. 4).

Fig. 3

The viral loads of children with EBV-related diseases. * P < 0.05 (a versus b)

Fig. 4

Correlation between EB virus load and ADA in patients with normal ALT