Abstract
Introduction
Critically ill children with adenoviral pneumonia usually require intensive care treatment. In this study, the prevalence of viremia secondary to respiratory adenovirus infection and the host characteristics and inflammatory response of children with severe adenoviral pneumonia in the PICU were investigated.
Methods
We prospectively recruited children who were admitted to the PICU of a tertiary pediatric hospital due to severe adenoviral pneumonia from September 2022 to April 2025. Sputum and blood samples were tested for adenovirus DNA with polymerase chain reaction. The clinical characteristics, blood laboratory parameters, and outcomes of the patients were collected. Plasma cytokines were also detected by flow cytometry.
Results
Fifty-one patients with severe adenoviral pneumonia were enrolled, 20 of whom had viremia. The proportion of children with comorbidities in the viremia group and nonviremia group was 40.0% and 71.0%, respectively. The viral loads in sputum; CRP, PCT, D-dimer, AST, LDH, IL-6, IL-8, IL-10, and IFN-γ levels; and pSOFA, PSS and PRISM-Ⅲ scores were significantly greater in the viremia group than in the nonviremia group, whereas the IL-5 and albumin levels were markedly lower in the viremia group. The number of Natural killer cells in the viremia group was lower than that in the nonviremia group, but there was no statistical difference. Multivariate logistic regression found that sputum viral load was an independent risk factor associated with viremia. The areas under the receiver operating characteristic curves for AST, LDH, PCT, CRP, D-dimer, IL-6 and IFN-γ could predict viremia. The mortality rate of the viremia group was higher than that of the nonviremia group, while the difference was not significant.
Conclusions
The prevalence of viremia is nearly 40% in children with severe adenoviral pneumonia in the PICU. Compared with the nonviremia group, the viremia group had higher sputum viral loads, CRP, PCT, IL-6, IL-8, IL-10, and IFN-γ levels; and more serious disease severity.
Supplementary information
The online version contains supplementary material available at 10.1186/s12879-026-13314-6.
Keywords: Adenovirus, Viremia, Children, Viral load, Cytokines, Pneumonia
Introduction
Human adenovirus (HAdV) is a prevalent pathogen in children and can cause a series of diseases, such as upper respiratory tract infection, pneumonia, gastroenteritis, and central nervous system infection. In cases of severe HAdV infection, including pneumonia, myocarditis, encephalitis, etc., hospitalization is usually needed. Furthermore, children with severe adenoviral pneumonia are critically ill and often need intensive care in the PICU.
Etiological detection of respiratory tract secretions is the typical means of diagnosing HAdV infection in clinical settings, with pediatricians focusing mainly on respiratory infection and changes on pulmonary imaging. At the end of the 20th century, researchers reported that HadV could be detected by polymerase chain reaction (PCR) in the blood of immunodeficient patients, suggesting that HAdV infection can also lead to viremia [1]. PCR for the detection of adenovirus in blood is useful for the early diagnosis of disseminated HAdV infection.
Many studies have focused on HAdV viremia in patients undergoing hematopoietic stem cell transplantation and have shown that the presence of viremia is related to illness severity and prognosis in patients [2–5]. However, HAdV viremia can also occur in immunocompetent children [6, 7]. One study revealed that 42 of 111 immunocompetent children with HadV pneumonia had viremia [6]. In another study that included 113 immunocompetent children with HAdV respiratory tract infections, HAdV was detected in 28.3% of the patients’ blood [7].
At present, studies on viremia secondary to HAdV infection have focused mainly on the demographic and clinical characteristics of children in the pediatric general ward [6, 7]. Severe HAdV infections are sometimes fatal [8]. Furthermore, the mortality rate of children with severe HAdV pneumonia ranged from 6.9% to 28.8% [9–11].
Recently, the prevalence and clinical relevance of viremia secondary to respiratory virus infection have attracted much attention [12]. However, research on viremia in children with severe adenoviral pneumonia in the PICU is lacking. In this study, viremia was defined as the presence of adenovirus components detected by PCR in the blood. The main aim of this prospective case‒control study was to investigate the prevalence and clinical relevance of HAdV viremia in children with severe adenoviral pneumonia, as well as the associated inflammatory response. In addition, the predictive values of blood parameters for viremia were explored.
Methods
Study design and participants
The protocol for this prospective study was approved by the Institutional Ethics Review Board of Children’s Hospital, Chongqing Medical University (Approval Number: 2022–278). Children were consecutively recruited from the PICU between September 2022 and April 2025, and written consent was obtained from the participants and/or their parents prior to recruitment. Respiratory HAdV was detected by PCR within 24 h after admission to the PICU. The inclusion criteria for participants were as follows: (1) diagnosis of severe adenovirus pneumonia in accordance with the technical guidelines for prevention and control of human adenovirus respiratory infection (2019 edition) [13]; (2) PCR test with sputum and blood sample for adenovirus was performed and (3) aged ≥ 1 month and less than 18 years. The exclusion criteria were as follows: (1) more than 24 h between the extraction of respiratory tract specimens and blood samples for adenovirus detection by PCR; (2) death within 24 h after admission to the PICU; and (3) incomplete clinical data.
This trial was registered in the Chinese Clinical Trial Registry (Registration number: ChiCTR2200063040) on August 29, 2022. The study was conducted in accordance with the Declaration of Helsinki.
Study parameters
The demographic and clinical information collected included age, sex, glucose level, lactate level, INR, APTT, C-reactive protein (CRP) level, procalcitonin (PCT) level, white blood cell (WBC) count, hemoglobin level, platelet count, alanine aminotransferase (ALT) level, aspartate aminotransferase (AST) level, lactate dehydrogenase (LDH) level, pediatric Sequential Organ Failure Assessment (pSOFA) score, Phoenix Sepsis Score (PSS), PRISM-Ⅲ score, HAdV subtypes detected by next-generation sequencing (NGS), length of stay (LOS) in the PICU and hospital, need of mechanical ventilation, and mortality.
HAdV detection by PCR
Adenoviral DNA from respiratory tract specimens (sputum) and blood samples was obtained using a commercial quantitative fluorescence PCR kit (Sansure Biotech, Changsha, China) according to the manufacturer’s instructions. The limit of detection (LOD) for the assay was 200 copies/mL.
Cytokine measurement
Blood samples from patients and healthy controls were collected into tubes containing EDTA anticoagulant and centrifuged at 3000 rpm for 20 min to obtain plasma. The concentrations of plasma interleukin (IL)-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-17 A, interferon (IFN)-α, IFN-γ and tumor necrosis factor (TNF)-α were determined following the manufacturer’s instructions using a commercial human cytokine combined detection kit (Celgene Biotechnology, Jiangxi, China) and a BD FACSCanto flow cytometer (BD Bioscience, CA, USA).
Lymphocyte subset measurement
Two ml of blood sample collected and anticoagulated with heparin sodium. Twenty ul of corresponding fluorescent antibody was added to 50 ul of blood, mixed well with a shaker, avoided light at room temperature for 15 min, added 450 ul of FACS hemolysin, and kept it at room temperature for 15 min. Lymphocyte subset counts were detected using a BD FACSCanto™ II and analyzed with BD FACSCanto™ software version 3.1 (BD Bioscience, CA, USA).
Statistical analysis
Viral loads detected by PCR were log10 transformed. Blood viral loads below the LOD were assigned a viral load value of 1 copy/mL (0 log10 copies/mL). All data analyses were performed using IBM SPSS Statistics for Windows, version 23.0 (IBM Corp, Armonk NY, USA). Baseline characteristics and blood parameters were compared between the viremia group and the nonviremia group. Continuous variables are expressed as medians and interquartile ranges (IQRs) and were compared using the Mann‒Whitney U test. Categorical variables are expressed as percentages, and the chi-square test/Fisher’s test was employed to compare proportions. Correlation analyses were performed using the nonparametric Spearman rank correlation test, and a heatmap was constructed on https://www.chiplot.online/. Receiver operating characteristic (ROC) curves were constructed, and the predictive performance of the parameters for developing viremia was assessed. A P value < 0.05 was considered to indicate statistical significance.
Results
Demographic and clinical characteristics
A total of 51 individuals with severe adenoviral pneumonia were recruited from the PICU, 20 of whom had viremia. The demographic characteristics of the patients are listed in Table 1. The median age of all the patients was 18.00 months, the median weight was 10.00 kg, and 34 (66.7%) were male. There was no statistically significant difference in age, weight, sex and the time interval from onset to enrollment between the viremia group and the nonviremia group.
Table 1.
Demographics and clinical data
| All patients (n = 51) | Viremia (n = 20) | Nonviremia (n = 31) | P | |
|---|---|---|---|---|
| Age (months) | 18.00 (8.00, 65.00) | 12.50 (7.00, 46.75) | 22.00 (10.00, 75.00) | 0.051 |
| Weight (kg) | 10.00 (8.50, 18.00) | 9.25 (7.78, 17.00) | 11.30 (9.00, 18.00) | 0.199 |
| Male, n (%) | 34 (66.7%) | 11 (55.0%) | 23 (74.2%) | 0.156 |
| Sputum viral load (log10 copies/mL) | 6.64 (4.77, 8.15) | 7.92 (7.40, 8.59) | 5.00 (3.71, 7.69) | < 0.001 |
| The time interval from onset to enrollment (day) | 8.00 (5.00, 11.00) | 9.00 (7.25, 12.75) | 8.00 (4.00, 10.00) | 0.181 |
| WBC (×10^9/L) | 9.45 (5.52, 13.61) | 8.66 (5.44, 15.00) | 9.84 (6.08, 13.61) | 0.923 |
| Hb (g/L) | 101.00 (87.00, 113.00) | 96.00 (87.00, 105.75) | 105.00 (92.00, 115.00) | 0.076 |
| PLT (×10^9/L) | 267.00 (195.00, 398.00) | 313.50 (142.25, 402.50) | 261.00 (195.00, 377.00) | 0.750 |
| Neutrophil (×10^9/L) | 6.34 (4.14, 10.66) | 6.07 (3.95, 12.05) | 7.21 (4.20, 9.79) | 0.772 |
| Lymphocyte (×10^9/L) | 2.01 (1.11, 2.31) | 2.03 (1.41, 2.20) | 1.94 (0.96, 2.37) | 0.855 |
| CRP (mg/L) | 15.42 (1.80, 30.00) | 29.53 (8.98, 75.67) | 3.82 (0.80, 20.59) | 0.001 |
| PCT (ng/ml) | 0.49 (0.08, 1.92) | 1.79 (0.78, 8.78) | 0.20 (0.06, 0.61) | 0.000 |
| INR | 1.09 (0.95, 1.19) | 1.10 (0.94, 1.19) | 1.08 (0.99, 1.19) | 0.772 |
| APTT (s) | 30.60 (27.10, 34.80) | 30.80 (27.88, 42.33) | 29.30 (26.60, 33.10) | 0.243 |
| Fib (g/L) | 2.66 (2.17, 3.37) | 2.80 (2.40, 3.48) | 2.51 (2.03, 3.26) | 0.091 |
| D-dimer (mg/L) | 0.76 (0.38, 1.96) | 1.67 (0.71, 7.90) | 0.45 (0.31, 1.11) | 0.001 |
| Albumin (g/L) | 37.00 (31.40, 39.20) | 31.55 (28.48, 38.08) | 38.20 (34.00, 39.70) | 0.009 |
| LDH (U/L) | 377.00 (290.00, 622.00) | 661.50 (448.75, 1413.25) | 339.00 (274.00, 384.00) | 0.000 |
| Cr (µmol/L) | 25.00 (19.00, 32.00) | 23.00 (18.00, 31.50) | 26.00 (21.00, 32.00) | 0.379 |
| Tin I (ug/L) | 0.010 (0.006, 0.050) | 0.020 (0.010, 0.078) | 0.008 (0.006, 0.030) | 0.066 |
| CK-MB (ug/L) | 1.88 (0.89, 4.00) | 2.245 (0.875, 4.035) | 1.75 (1.00, 4.00) | 0.772 |
| Glucose (mmol/L) | 6.80 (5.30, 9.30) | 6.85 (5.43, 9.53) | 6.50 (5.20, 9.30) | 0.623 |
| Lactate (mmol/L) | 1.00 (0.60, 1.60) | 1.00 (0.70, 2.55) | 0.90 (0.60, 1.50) | 0.250 |
| Total bilirubin (umol/L) | 4.80 (3.00, 9.00) | 3.90 (2.92, 8.08) | 6.50 (3.30, 11.40) | 0.390 |
| ALT (U/L) | 21.00 (15.00, 31.00) | 22.50 (15.50, 62.00) | 20.00 (15.00, 24.00) | 0.140 |
| AST (U/L) | 41.00 (32.00, 64.00) | 61.00 (40.50, 133.25) | 38.00 (27.00, 46.00) | 0.001 |
| PRISM-Ⅲ | 6.0 (5.0, 10) | 9.5 (6.0, 14.0) | 6.0 (3.0, 9.0) | 0.003 |
| Phoenix sepsis score | 3.00 (2.00, 4.00) | 3.00 (2.00, 5.00) | 2.00 (2.00, 3.00) | 0.042 |
| pSOFA | 3.00 (3.00, 5.00) | 4.00 (3.00, 6.00) | 3.00 (3.00, 4.00) | 0.043 |
| Coinfection, n (%) | 17 (33.3%) | 6 (30%) | 11 (35.5%) | 0.685 |
| MP, n (%) | 4 (7.8%) | 0 | 4(12.9%) | - |
| Other virus, n (%) | 8 (15.7%) | 5 (25%) | 3 (9.7%) | 0.142 |
| Bacterial, n (%) | 7 (13.7%) | 2 (10%) | 5 (16.1%) | 0.535 |
| Any comorbidity, n (%) | 30 (58.8%) | 8 (40.0%) | 22 (71.0%) | 0.028 |
| Airway malformation, n (%) | 7 (13.7%) | 1 (5.0%) | 6 (19.4%) | 0.146 |
| Congenital heart disease, n(%) | 6 (11.8%) | 3 (15.0%) | 3 (9.7%) | 0.332 |
| Central nervous system disease, n(%) | 6 (11.8%) | 2 (10%) | 4 (12.9%) | 0.832 |
| Immunocompromised, n (%) | 6 (11.8%) | 1(5%) | 5(16.1%) | 0.228 |
| Mechanical ventilation, n (%) | 38 (74.5%) | 17 (85%) | 21 (67.7%) | 0.167 |
| Glucocorticoid, n (%) | 36 (70.6%) | 14 (70.0%) | 22 (71.0%) | 0.941 |
| IVIG, n (%) | 36 (70.6%) | 15 (75.0%) | 21 (67.7%) | 0.579 |
| ECMO, n (%) | 3 (5.9%) | 3 (15%) | 0 (0) | - |
| LOS of PICU (day) | 10.00 (5.00, 19.00) | 9.50 (3.50, 18.00) | 11.00 (5.00, 21.00) | 0.406 |
| LOS of hospital (day) | 20.00 (13.00, 28.00) | 17.00 (12.00, 27.75) | 23.00 (14.00, 28.00) | 0.385 |
| Death, n (%) | 9 (17.6%) | 5 (25.0%) | 4 (12.9%) | 0.269 |
The data shown are the median (interquartile range)
As shown in Table 1, compared with that in the nonviremia group, the viral load in sputum was significantly greater in the viremia group. Moreover, compared with the nonviremia group, the viremia group had significantly higher CRP, PCT, D-dimer, AST and LDH levels but markedly lower levels of albumin. Moreover, compared with the nonviremia group, the viremia group had significantly higher pSOFA, PSS and PRISM-III scores. All patients did not receive antiviral treatment. Most of the children received immunomodulatory therapy. There was no statistically significant difference in the proportion of patients receiving intravenous glucocorticoid treatment or IVIG between the two groups.
There was no significant difference in the length of PICU stay or length of hospital stay between the two groups (Table 1). A total of 9 children died. The mortality rate of children in the viremia group was greater than that in the nonviremia group, but the difference was not statistically significant (Table 1). It is worth noting that all four deceased children in nonviremia group were accompanied by severe comorbidities, including acute leukemia, liver failure after liver transplantation, congenital heart disease, and severe bronchopulmonary dysplasia. However, among the 5 children with viremia who died, only two had severe comorbidities, namely malignant tumors and neuromuscular diseases. Three children with viremia received extracorporeal membrane oxygenation (ECMO) treatment, but two of them died.
The coinfection rates of mycoplasma, virus, and bacteria in the nonviremia group were 12.9%, 9.7% and 16.1%, respectively. There was no mycoplasma infection in the viremia group, and the coinfection rates of viruses and bacteria were 25% and 10%, respectively (Table 1).
To exclude the influence of coinfection factors, seventeen children with coinfections were removed, with 14 children in the viremia group and 20 children in the nonviremia group. Children in the viremia group had significantly higher sputum viral load and CRP, PCT, D-dimer, AST and LDH levels than children in the nonviremia group did (Supplementary Table 1). Moreover, children with viremia had significantly higher pSOFA, PSS and PRISM-III scores (Supplementary Table 1).
Among these 51 patients, 30 had concomitant diseases. The prevalence of comorbidities in the viremia group was significantly lower than that in the nonviremia group (Table 1). Six children were immunocompromised, namely, 2 leukemia patients, 2 primary immunodeficiency patients, 1 tumor patient, and 1 pulmonary hemosiderosis patient. In order to eliminate the impact of immunocompromise, these six patients were removed and statistical analysis was conducted on 45 immunocompetent children (Supplementary Table 2). Children with viremia still had significantly higher sputum viral load and CRP, PCT, D-dimer, AST and LDH levels than children without viremia (Supplementary Table 2). In addition, children with viremia also had significantly higher pSOFA, PSS and PRISM-III scores (Supplementary Table 2).
Comparison of inflammatory cytokines between the viremia group and the nonviremia group
As shown in Table 2, the cytokine concentrations of fifteen healthy children who comprised the healthy control group were examined. The levels of plasma IL-4, IL-6, IL-10, IFN-γ and IL-8 were significantly greater in the nonviremia group than in the healthy control group. However, compared with the healthy control group, the nonviremia group had significantly lower levels of plasma IL-17 A. Moreover, the levels of plasma IL-6, IL-10, IFN-γ and IL-8 were significantly greater in the viremia group than in the healthy control group.
Table 2.
Cytokine concentrations (pg/ml) in the healthy control group, the nonviremia group and the viremia group
| Cytokines | Controls (n = 15) | Viremia (n = 20) | Nonviremia (n = 31) | P ab | P ac | P bc |
|---|---|---|---|---|---|---|
| IL-2 | 0.47 (0.17, 0.90) | 0.77 (0.03, 1.44) | 0.91 (0.14, 1.17) | 0.542 | 0.185 | 0.498 |
| IL-4 | 0.62 (0.46, 1.05) | 0.90 (0.29, 1.53) | 1.56 (0.30, 1.94) | 0.499 | 0.048 | 0.209 |
| IL-6 | 1.59 (1.20, 2.28) | 160.18 (10.05, 446.40) | 9.46 (3.57, 33.88) | 0.000 | 0.000 | 0.008 |
| IL-10 | 3.32 (2.01, 3.87) | 19.40 (7.14, 61.76) | 6.34 (3.10, 11.16) | 0.000 | 0.005 | 0.012 |
| TNF-α | 1.16 (0.56, 1.33) | 1.35 (0.01, 2.25) | 1.33 (0.53, 1.85) | 0.610 | 0.348 | 0.692 |
| IFN-γ | 0.47 (0.28, 0.82) | 7.10 (1.41, 20.42) | 1.04 (0.71, 2.10) | 0.000 | 0.011 | 0.001 |
| IL-17 A | 6.84 (5.39, 9.59) | 3.92 (0, 11.88) | 4.34 (0.80, 8.47) | 0.268 | 0.037 | 0.922 |
| IL-1β | 0.73 (0.48, 1.77) | 0.80 (0, 2.43) | 1.41 (0.37, 2.30) | 0.730 | 0.716 | 0.364 |
| IL-5 | 0.21 (0.11, 0.59) | 0.18 (0, 0.39) | 0.46 (0.20, 0.76) | 0.240 | 0.174 | 0.019 |
| IL-12p70 | 1.88 (1.64, 2.22) | 1.39 (0.12, 2.12) | 1.84 (1.06, 2.69) | 0.214 | 0.907 | 0.164 |
| IFN-α | 1.27 (0.94, 1.63) | 3.50 (0.38, 55.90) | 1.51 (0.83, 4.06) | 0.347 | 0.535 | 0.369 |
| IL-8 | 7.92 (5.25, 9.21) | 78.04 (13.49, 137.04) | 15.59 (7.33, 30.74) | 0.000 | 0.003 | 0.011 |
The data shown are the median (interquartile range). Pab value: comparison of the controls to the viremia group; Pac value: comparison of the controls to the nonviremia group; Pbc value: comparison of the nonviremia group to the viremia group
The levels of plasma IL-6, IL-8, IL-10 and IFN-γ were significantly greater in the viremia group than in the nonviremia group (all P < 0.05), whereas the levels of plasma IL-5 were significantly lower (P < 0.05). The levels of plasma IL-1β, IL‐2, IL‐4, IL‐12p70, IL-17 A, IFN-α and TNF‐ɑ did not significantly differ between the two groups (Table 2).
To eliminate the influence of other pathogenic infections on cytokine expression, 17 children with coinfections were removed (Supplementary Table 3). Compared with children without viremia, children with viremia still had significantly higher levels of plasma IL-6, IL-10 and IFN-γ (all P < 0.01; Supplementary Table 3). However, there was no longer a statistically significant difference in the IL-5 and IL-8 levels between the children with viremia and those without viremia (Supplementary Table 3). This may be due to a decrease in sample size.
After removing 6 immunocompromised children, statistical analysis was conducted on 45 immunocompetent children. The levels of plasma IL-6, IL-8, IL-10 and IFN-γ were significantly increased in children with viremia (all P < 0.05), but the levels of plasma IL-5 was significantly decreased (P < 0.05) (Supplementary Table 4).
Cellular immunity between the viremia group and the nonviremia group
Thirty-eight children underwent cellular immune testing, with 17 in the viremia group and 21 in the nonviremia group. However, there was no significant difference in CD3 + count, CD3 + CD4+ count, CD3 + CD4+CD8 + count, CD3 + CD8+ count, CD19 + count, CD45 + count and CD4/CD8 ratio between the viremia group and the nonviremia group (Supplementary Table 5). The number of Natural killer cells in the viremia group was lower than that in the nonviremia group, but there was still no statistical difference.
Independent risk factors for viremia
Univariate and multivariate logistic regression analyses were used to explore which factors were independently associated with viremia secondary to HAdV infection (Table 3). After controlling for other variables, sputum viral load was the only independent risk factor for viremia (adjusted OR 1.656; 95% confidence interval [CI] 1.040–2.637; P = 0.034).
Table 3.
Univariate and multivariate logistic regression were used to identify risk factors associated with viremia
| Univariate logistic regression Unadjusted OR (95% CI) |
P | Multivariate logistic regression Adjusted OR (95% CI) |
P | |
|---|---|---|---|---|
|
Sputum viral load (log10 copies/mL) |
2.008 (1.333–3.026) | 0.001 | 1.656 (1.040–2.637) | 0.034 |
| PCT (ng/ml) | 1.361 (1.010–1.834) | 0.043 | 1.094 (0.917–1.304) | 0.320 |
| IFN-γ (pg/ml) | 1.255 (1.054–1.496) | 0.011 | 1.153 (0.965–1.377) | 0.116 |
| pSOFA | 1.348 (1.004–1.811) | 0.047 | 1.195 (0.790–1.808) | 0.399 |
Correlations between baseline blood indices and viral load
The heatmap shows the correlations among the baseline blood indices and viral load (Fig. 1). CRP (r = 0.31, P < 0.05), PCT (r = 0.45, P < 0.001), AST (r = 0.34, P < 0.05), LDH (r = 0.37, P < 0.01), IL-6 (r = 0.28, P < 0.05), IL-8 (r = 0.33, P < 0.05), IL-10 (r = 0.36, P < 0.01) and IFN-γ (r = 0.37, P < 0.01) levels were positively correlated with sputum viral loads in children with severe adenoviral pneumonia. Moreover, CRP (r = 0.43, P < 0.01), PCT (r = 0.56, P < 0.001), AST (r = 0.47, P < 0.001), LDH (r = 0.55, P < 0.001), IL-6 (r = 0.40, P < 0.01), IL-8 (r = 0.37, P < 0.01), IL-10 (r = 0.42, P < 0.01) and IFN-γ (r = 0.51, P < 0.001) levels were positively correlated with blood viral loads.
Fig. 1.
A correlation analysis was performed to explore the relationships between blood indices and viral load. * P < 0.05, ** P < 0.01, *** P < 0.001
ROC curve analyses of the occurrence of viremia
An area under the curve (AUC) analysis was performed to predict viremia. CRP, PCT, D-dimer, LDH, IL-6, IFN-γ and AST levels could predict viremia (Table 4). The AUC for CRP was 0.776 (95% CI: 0.641 to 0.910; P = 0.001), that for PCT was 0.836 (95% CI: 0.721 to 0.951; P < 0.001), that for D-dimer was 0.788 (95% CI: 0.664 to 0.913; P = 0.001), that for LDH was 0.834 (95% CI: 0.701 to 0.968; P < 0.001), that for IL-6 was 0.715 (95% CI: 0.569 to 0.861; P < 0.01), that for IFN-γ was 0.769 (95% CI: 0.632 to 0.907; P = 0.001), and that for AST was 0.766 (95% CI: 0.627 to 0.906; P = 0.001) (Table 4).
Table 4.
AUCs determined from the ROC curves for the predictors of viremia
| AUC (95% CI) | P | |
|---|---|---|
| CRP | 0.776 (0.641–0.910) | 0.001 |
| PCT | 0.836 (0.721–0.951) | < 0.001 |
| D-dimer | 0.788 (0.664–0.913) | 0.001 |
| LDH | 0.834 (0.701–0.968) | < 0.001 |
| IL-6 | 0.715 (0.569–0.861) | 0.008 |
| IFN-γ | 0.769 (0.632–0.907) | 0.001 |
| AST | 0.766 (0.627–0.906) | 0.001 |
Serotypes of HAdV
NGS of sputum samples from 10 children demonstrated the serotypes of HAdV. Among the 9 children with viremia, there were a total of 6 cases of HAdV-7, 1 case of HAdV-5, 1 case of HAdV-21, and 1 case of HAdV-55. One children without viremia was infected with HAdV-7. Three out of these ten children died, with two serotypes being HAdV-7 and one serotype being HAdV-55, all of whom were viremia patients.
Discussion
This study revealed that viremia may play an important role in the pathogenesis of severe HAdV infection in critically ill children. Approximately 40% of children with severe adenoviral pneumonia had viremia in the PICU. Compared with children without viremia, children with viremia had markedly higher sputum viral loads; higher LDH, AST and D-dimer levels; more severe inflammatory reactions; and higher disease severity scores. Compared with children with nonviremia, children with viremia had significantly increased levels of plasma IL-6, IL-8, IL-10 and IFN-γ but decreased levels of plasma IL-5. After excluding children with coinfections, children with viremia still had significantly higher CRP, PCT, LDH, AST, D-dimer, IL-6, IL-10 and IFN-γ levels than children without viremia did. Multivariate logistic regression revealed that sputum viral load was independently associated with viremia. AUC analysis demonstrated that the levels of CRP, PCT, D-dimer, LDH, IL-6, IFN-γ and AST were able to predict viremia in children with HAdV infection. Furthermore, the CRP, PCT, D-dimer, AST, LDH, IL-6, IL-8, IL-10 and IFN-γ levels were positively correlated with the blood viral load.
Although the primary infection route is the respiratory tract [14], adenovirus may damage the alveolar and vascular endothelium through inflammatory reactions, activation of the coagulation system or direct invasion, causing viremia and organ damage [7]. Studies on respiratory tract infection caused by HAdV have revealed that the incidence of viremia ranges from 16% to 100% [6, 7, 15–19]. Viral load in respiratory tract secretions is a predictor of disease severity in children with HAdV pneumonia [20]. We found that compared with children without viremia, children with viremia had a higher sputum viral load. This study revealed a positive correlation between the sputum viral load and the blood viral load. We also found that children with viremia had significantly higher pSOFA, PSS and PRISM-III scores. Therefore, viremia can be used as a marker of disease severity, which may contribute to the early hierarchical management of children and the formulation of treatment plans.
D-dimer is a fibrin degradation product. An increase in D-dimer level is a marker of a severe condition caused by HAdV [7, 14, 21]. One study indicated that compared with children with common HAdV pneumonia, children with severe HAdV pneumonia had significantly higher D-dimer levels [14]. D-dimer is an independent risk factor for plastic bronchiolitis in children with HAdV pneumonia [21]. Viremia is related to the activation of the coagulation system [7]. Chen et al. [7] reported that patients with viremia had significantly higher D-dimer levels than patients without viremia did. In this study, children with viremia had significantly higher concentrations of D-dimer than children without viremia did. Moreover, D-dimer level was able to predict HAdV viremia.
In clinical practice, CRP and PCT are generally considered inflammatory markers. Studies have demonstrated that CRP and PCT levels are markedly elevated in patients with severe HAdV pneumonia [22, 23]. In cases of HAdV infection, children with viremia had significantly higher PCT levels than those without viremia did [7]. Our study also revealed that compared with children without viremia, children with viremia had significantly higher levels of CRP and PCT. Furthermore, the results of the ROC curve analyses indicated that both the PCT level and the CRP level had acceptable discriminatory ability for detecting viremia. We found that both CRP level and PCT level were positively correlated with viral load. Some researchers have suggested that alterations in PCT and CRP levels should be considered in the medical management of HAdV infection [24]. We also argued that the levels of CRP and PCT could reflect the degree of inflammatory response and the severity of disease in children with severe adenoviral pneumonia.
Liver function damage caused by HAdV infection has received increased attention. Liver injury is related to prolonged hospitalization among patients with HAdV infections [25]. AST is more commonly increased than ALT is in children with HAdV infection [7, 14, 25, 26]. One study revealed that compared with patients without viremia, patients with viremia have higher levels of AST and ALT [7]. Our study also revealed that AST levels were significantly higher in children with viremia than in children without viremia, but the difference in ALT levels was not statistically significant between the two groups. AST level was positively correlated with the viral load in blood and was able to predict viremia. Moreover, we found that albumin levels were markedly decreased in children with viremia compared with those in children without viremia.
HAdV infection can lead to a “cytokine storm” in immunocompetent patients. One study indicated that the expression of IL-10, M-CSF, MIG, HGF, IL-1β, IFN-γ and IL-2Ra was closely associated with disease severity and could predict disease progression in patients with HAdV infection [27]. Studies have demonstrated that HAdV infection can alter the expression of cytokines [25, 28, 29]. Biserni et al. [28] reported that in immunocompetent children with HAdV upper respiratory tract infections, blood IL-6 and IL-8 levels significantly increase during the acute phase. The viral loads of patients with respiratory tract HAdV infections is strongly correlated with the level of IFN-γ, and severe liver damage caused by HAdV infection might be correlated with increased IFN-γ levels [25]. Fu et al. [29] reported that TNF-α, IL-1β, IFN-γ and IL-6 levels significantly increased after HAdV 7 infection. Oda et al. [30] reported that compared with that in children with RSV infection, the serum level of IL-6 in children with HAdV infection significantly increased on the fourth day.
However, the expression of plasma inflammatory cytokines in children with viremia secondary to respiratory HAdV infection is still unclear. This study revealed that compared with children without viremia and healthy controls, children with viremia had significantly higher IL-6, IL-8, IL-10 and IFN-r levels. The levels of plasma IL-6, IL-8, IL-10, and IFN-r were positively correlated with viral loads. In addition, in contrast to a previous study [19], we reported that the level of IL-5 in the viremia group was significantly lower than that in the nonviremia group. Whether removing immunocompromised children or coinfected children, viremia caused similar inflammatory cytokine changes. We argued that the presence of viremia was associated with obvious alterations in the host inflammatory response in children with HAdV infection. Using specific methods to reduce the inflammatory response may alleviate the condition of children with severe adenoviral pneumonia.
However, this study found no significant difference in the number of lymphocyte subsets between the viremia group and the nonviremia group, which may be due to the small sample size.
LDH is a critical metabolic enzyme in human cells, serving as the last step of aerobic glycolysis by reversibly catalyzing the conversion of pyruvate to lactate. LDH is very sensitive to tissue damage and infection. LDH was an independent risk factor for plastic bronchiolitis in children with HAdV pneumonia [21]. LDH levels increase markedly in severe HAdV infections and are linked to disease severity and patient outcome [14, 31]. One study demonstrated that in cases of HAdV infection, children with viremia had significantly higher LDH levels than children without viremia did [7]. LDH levels were significantly higher in children with viremia than in children without viremia. ROC curve analysis revealed that LDH levels could predict viremia. Moreover, LDH was positively correlated with viral load. Therefore, LDH is an important indicator for determining the severity of disease and the risk of viremia.
In the case of adenovirus infection, a high serum viral load is an independent risk factor for mortality [19], and persistent viremia may predict mortality [15, 16]. In this study, the mortality rate of children with viremia was higher than that of children without viremia, although the difference was not significant, this may be attributed to the small sample size. We speculate that viremia may lead to an excessive inflammatory response, tissue damage and coagulation activation, which are linked to poor outcomes.
Compared with the nonviremia group, the viremia group did have significantly greater PRISM-III, PSS and pSOFA scores, indicating a more severe condition. The mortality rate and mechanical ventilation demand in the viremia group were higher than those in the nonviremia group, but there was no statistical difference, which may be due to the small sample size included in the study. In addition, the impact of comorbidities on mortality must also be taken into account. The deceased children without viremia all had severe comorbidities, which were the important factors related to death. On the contrary, only two deceased children in the viremia group had severe comorbidities.
Viremia also did not prolong the PICU length of stay and hospital length of stay. This may be related to a higher percentage of comorbidities in the nonviremia group. The percentage of children with any comorbidity in the nonviremia group was 71%, while in the viremia group it was only 40%.
In PICU, clinicians were more inclined to perform NGS testing on severely ill children to screen for pathogens. The three deceased children were infected with HAdV-7 and HAdV-55. Therefore, clinicians should pay attention to adenovirus infections of these two serotypes.
There are several limitations in this study. First, the sample size was relatively small, a priori sample size calculation was not performed, and recruitment was based on consecutive admissions. The small sample size may affect statistical validity, especially for multivariate and subgroup analyses. Second, most of the children’s adenovirus serotypes had not been tested. Moreover, the analyses linking the serotypes of HAdV to hospital stay, morbidity, or mortality were not able to performed. Third, in vitro experiments, such as cell specific cytokine release, were not conducted. Fourth, not all children underwent cellular immune testing. Fifth, the prevalence of comorbidities and airway malformations was high. Of course, this may also accurately reflected the status of severe adenovirus infection in children in PICU. However, it must be noted that severe comorbidities may have an impact on mortality rate, length of hospital stay, etc. At present, we cannot completely rule out the influence of comorbidities. Sixth, respiratory tract or blood samples for viral detection were not obtained at a fixed time during the course of HAdV infection because the time at which each child entered the PICU after illness could not be unified. Finally, because of the heterogeneity of clinical conditions, the treatment for each child was not unified. Despite these limitations, the findings of this study are exploratory and hypothesis-generating. A multicenter prospective study with a large sample is expected to explore the prevalence, inflammatory response and outcome of adenovirus viremia in critically ill children.
Conclusions
The prevalence of viremia is nearly 40% in children with severe adenoviral pneumonia who require treatment in the PICU. Children with viremia have higher sputum viral loads and higher levels of CRP, PCT, IL-6, IL-8, IL-10 and IFN-γ than children without viremia do. Children with viremia have more serious disease severity. The mortality rate of children with viremia is higher than that of those without viremia, although there is no statistical difference. Viremia may be an important pathophysiological event underlying the pathogenesis of severe adenovirus infection.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We would like to acknowledge the patients who participated in this study.
Abbreviations
- ALT
Alanine aminotransferase
- AST
Aspartate aminotransferase
- AUC
Area under curve
- CRP
C reactive protein
- CI
Confidence interval
- HAdV
Human adenovirus
- IFN
Interferon
- IL
Interleukin
- IQR
Interquartile range
- IVIG
Intravenous immunoglobulin
- LOD
Limit of detection
- LDH
Lactate dehydrogenase
- LOS
Length of stay
- MP
Mycoplasma pneumoniae
- PCT
Procalcitonin
- PCR
Polymerase chain reaction
- PICU
Pediatrics intensive care unit
- pSOFA
Paediatric sequential organ failure assessment
- ROC
Receiver operating characteristic
- TNF
Tumor necrosis factor
- WBC
White blood cell
Author contributions
YF designed and supervised the study and revised the paper. SL and NZ equally contributed to experiment, data analysis and writing of the manuscript. SL, NZ, SC, JH, YR, CX and FZ participated in sample acquisition and data collection. DP, CL and HD participated in statistical analysis. All authors reviewed and approved the final version of the manuscript.
Funding
Project Supported by Scientific and Technological Research Program of Chongqing Municipal Education Commission(Grant No. KJZD-K202200402).
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
This study was approved by the Institutional Ethics Review Board of the Chongqing Medical University Children’s Hospital (Approval Number: 2022 − 278) and conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from the patients and/or their parents.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Siwei Lu and Nang Zang are co-first authors and contributed equally to this work.
References
- 1.Munoz FM, Piedra PA, Demmler GJ. Disseminated adenovirus disease in immunocompromised and immunocompetent children. Clin Infect Dis. 1998;27(5):1194–200. [DOI] [PubMed] [Google Scholar]
- 2.Chandorkar A, Anderson AD, Morris MI, et al. Viral kinetics and outcomes of adenovirus viremia following allogeneic hematopoietic cell transplantation. Clin Transpl. 2021;35(12):e14481. [DOI] [PubMed] [Google Scholar]
- 3.Balletto E, Ponzano M, Raiola AM, et al. Adenovirus infection in adult patients undergoing allogeneic hematopoietic stem cell transplant: incidence, clinical management, and outcome. Transpl Infect Dis. 2024;26(2):e14215. [DOI] [PubMed] [Google Scholar]
- 4.Lee YJ, Huang YT, Kim SJ, et al. Adenovirus viremia in adult CD34(+) selected hematopoietic cell transplant recipients: low incidence and high clinical impact. Biol Blood Marrow Transpl. 2016;22(1):174–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Omar H, Yun Z, Lewensohn-Fuchs I, et al. Poor outcome of adenovirus infections in adult hematopoietic stem cell transplant patients with sustained adenovirus viremia. Transpl Infect Dis. 2010;12(5):465–9. [DOI] [PubMed] [Google Scholar]
- 6.Zhang R, Wang H, Tian S, et al. Adenovirus viremia may predict adenovirus pneumonia severity in immunocompetent children. BMC Infect Dis. 2021;21(1):213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chen Q, Liu J, Liang W, et al. Clinical features, replication competence, and innate immune responses of human adenovirus type 7 infection. J Infect Dis. 2021;223(8):1390–99. [DOI] [PubMed] [Google Scholar]
- 8.van der Zalm MM, Sam-Agudu NA, Verhagen LM. Respiratory adenovirus infections in children: a focus on Africa. Curr Opin Pediatr. 2024;36(3):342–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Xu XH, Fan HF, Shi TT, et al. Analysis of mortality risk factors in children with severe adenovirus pneumonia: A single-center retrospective study. Pediatr Neonatol. 2023;64(3):280–7. [DOI] [PubMed] [Google Scholar]
- 10.Shi J, Zhou Y, Wang F, et al. A case series of children with adenovirus pneumonia: three-year experiences in a tertiary PICU. BMC Pediatr. 2020;20(1):375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Peng H, Chen F, Zuo Y, et al. Mortality reduction in pediatric patients with severe fatal human adenoviral pneumonia treated with high titer neutralizing antibodies (NAbs) plasma: a retrospective cohort study. BMC Pediatr. 2022;22(1):151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hagman K, Postigo T, Diez-Castro D, et al. Prevalence and clinical relevance of viraemia in viral respiratory tract infections: a systematic review. Lancet Microbe. 2025;6(2):100967. [DOI] [PubMed] [Google Scholar]
- 13.Expert Writing Group of Technical Guidelines for Prevention and Control of Human Adenovirus Respiratory Infection. Technical guidelines for prevention and control of human adenovirus respiratory infection (2019 edition). Zhonghua Yu Fang Yi Xue Za Zhi. 2019;53(11):1088–93. [In Chinese]. [DOI] [PubMed] [Google Scholar]
- 14.Liang Y, Wu J, Chen G, et al. Risk factor analysis and prediction model construction for severe adenovirus pneumonia in children. Ital J Pediatr. 2024;50(1):210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sun B, He H, Wang Z, et al. Emergent severe acute respiratory distress syndrome caused by adenovirus type 55 in immunocompetent adults in 2013: a prospective observational study. Crit Care. 2014;18:456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gu L, Qu J, Sun B. Sustained viremia and high viral load in respiratory tract secretions are predictors for death in immunocompetent adults with adenovirus pneumonia. PLoS ONE. 2016;11:e0160777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Schjelderup Nilsen HJ, Nordbø SA, Krokstad S, et al. Human adenovirus in nasopharyngeal and blood samples from children with and without respiratory tract infections. J Clin Virol. 2019;111:19–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hijano DR, Ferrolino JA, Hidinger J, et al. Clinical correlation of adenoviral load in the respiratory tract measured by digital PCR in immunocompromised children. Open Forum Infect Dis. 2023;10:ofad030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wei J, Zang N, Zhang J, et al. Genome and proteomic analysis of risk factors for fatal outcome in children with severe community acquired pneumonia caused by human adenovirus 7. J Med Virol. 2023;95:e29182. [DOI] [PubMed] [Google Scholar]
- 20.Xie L, Zhang B, Zhou J, et al. Human adenovirus load in respiratory tract secretions are predictors for disease severity in children with human adenovirus pneumonia. Virol J. 2018;15(1):123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wang Y, Jiang Y, Xia R, et al. Clinical features and risk factors of adenovirus-related plastic bronchitis in children. Ital J Pediatr. 2025;51(1):128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zhong H, Dong X. Analysis of clinical characteristics and risk factors of severe adenovirus pneumonia in children. Front Pediatr. 2021;9:566797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dai H, Xi H, Huang L, et al. Molecular epidemiology and clinical features analysis of respiratory adenovirus infections reveals correlations between genotype, inflammatory biomarkers, and disease severity. Biomed Res Int. 2020;2020:4357910. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gómez de Oña C, Alvarez-Argüelles ME, Rojo-Alba S, et al. Alterations in biochemical markers in adenovirus infection. Transl Pediatr. 2021;10(5):1248–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tang S, Qin R, Zhang D, et al. Liver injury and prolonged hospitalization as indicators of severity in patients with adenovirus infections. BMC Infect Dis. 2024;24(1):430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Săndulescu O, Streinu-Cercel A, Miron VD, et al. Liver transaminases in pediatric adenovirus infection-A five-year study in two major reference centers from romania. Microorganisms. 2023;11(2):302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Li J, Wei J, Xu Z, et al. Cytokine/chemokine expression is closely associated disease severity of human adenovirus infections in immunocompetent adults and predicts disease progression. Front Immunol. 2021;12:691879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Biserni GB, Dondi A, Masetti R, et al. Immune response against adenovirus in acute upper respiratory tract infections in immunocompetent children. Vaccines (Basel). 2020;8(4):602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Fu Y, Tang Z, Ye Z, et al. Human adenovirus type 7 infection causes a more severe disease than type 3. BMC Infect Dis. 2019;19(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Oda K, Yamamoto Y. Serum interferon-gamma, interleukin-4, and interleukin-6 in infants with adenovirus and respiratory syncytial virus infection. Pediatr Int. 2008;50(1):92–4. [DOI] [PubMed] [Google Scholar]
- 31.Lin F, Zhou Q, Li W, et al. A prediction model for acute respiratory distress syndrome in immunocompetent adults with adenovirus-associated Pneumonia: a multicenter retrospective analysis. BMC Pulm Med. 2023;23(1):431. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.