Risk factors of severe conditions in hospitalized children with adenovirus infection and chest CT features Authors: Shaofang Wang, Shichao Kang, Lili Guo, Shuchang Zhou, Yanjie Zhao, Huifen Shen, Shourui Jin, Hanxiong Guan, Liming Xia, Qiongjie Hu Published: 2024 ABSTRACT AbstractBackgroundChildren infected with human adenovirus (HAdV) were at particularly high risk of developing severe disease, but the risk factors of severe conditions are poorly understood.ObjectivesTo explore the risk factors for developing into severe conditions in pediatric patients with HAdV infection by analyzing baseline epidemiological data, clinical characteristics, and computed tomography (CT) imaging features.MethodsIn this retrospective study, 267 children with HAdV infection were included between October 2016 and September 2021 at Tongji Hospital, Wuhan, China. A descriptive analysis was conducted on the epidemiological and clinical data, as well as patient outcomes. CT manifestations were semi-quantitatively scored based on the presence of consolidation, ground-glass opacities, and emphysema in each lung lobe.ResultsOur analysis revealed that there was a statistically significant difference in the rate of abnormalities observed on chest CT imaging (P = 0.007) and the imaging characteristics of chest CTs (P = 0.002) when comparing severe and mild cases. We found that co-infection with two or more additional pathogens occurred more frequently in severe cases. Additionally, the proportion of lymphocytes in laboratory tests was significantly lower in patients with severe conditions. Furthermore, both the proportions and scores of consolidations were markedly higher in each lung lobe among the severe cases.ConclusionsOur findings may assist in identifying children hospitalized with HAdV who are at increased risk for severe conditions, thereby facilitating more aggressive treatment and care strategies.Keywords: Human adenovirus, Children, Risk factors, Computed tomography, Infection Introduction Human adenovirus (HAdV) poses a significant threat to public health, causing a range of illnesses in humans. Similar to influenza, the primary symptoms associated with HAdV infections include fever, cough, and sore throat. Additionally, HAdV can lead to lower respiratory tract infections, such as bronchitis, and is more likely to cause gastrointestinal symptoms like diarrhea, vomiting, and abdominal pain, particularly in children [1–3]. While most children with HAdV infections experience mild symptoms, have a favorable prognosis, and recover spontaneously within two weeks, there remains a subset whose condition deteriorates significantly, sometimes leading to death. This occurs despite advances in standardized treatment protocols and medical technology [4, 5]. Therefore, HAdV infections in children warrant ongoing attention. To date, there has been limited research focused on the risk of severe conditions resulting from HAdV infections in pediatric populations. Recent studies have concentrated on the epidemiology and molecular characterization of the virus in children to identify potential risk factors [3, 5, 6]. Notably, Rajkumar et al. [3] conducted an analysis identifying clinical features and risk factors associated with severe outcomes in pediatric HAdV infections, discovering that HAdV type 7, younger age (particularly those under 2 years), and the presence of significant comorbidities correlated with more severe outcomes necessitating invasive ventilation. However, few investigations have integrated radiological characteristics with clinical features to assess the risk of severe conditions in children with HAdV infections. Pediatric radiologic evaluations, increasingly indispensable in clinical practice, can be crucial for diagnosis and even lifesaving in some cases [7, 8]. Early imaging manifestations of HAdV pneumonia may not be apparent on chest X-rays; however, subtle alterations in the alveoli can lead to changes such as lung texture thickening, mild patchy infiltrates, or scattered fluffy nodular shadows [9]. As the disease progresses, these lesions may gradually coalesce or evolve into segmental and lobular consolidations, with complications like pleural effusion or pneumothorax potentially arising [10–12]. Recent data suggest that pediatric computed tomography (CT) can be performed at radiation doses comparable to those of chest radiography while maintaining excellent diagnostic performance [12, 13]. Chest CT provides detailed insights into the distribution and extent of pulmonary abnormalities, including congenital lung lesions and foreign bodies [12, 14]. Approximately 14%-16% of children with HAdV pneumonia experience pulmonary sequelae, particularly those with severe forms of the disease [9, 15, 16]. Severe cases of HAdV pneumonia are prone to causing hemorrhagic pulmonary edema, evidenced by chest CT findings of bilateral multiple lobular and subsegmental consolidation along with ground-glass opacities (GGO) [9]. These lesions typically exhibit uniform enhancement and centripetal distribution. Collectively, analyzing CT characteristics may provide valuable insights into potential risk factors associated with poor prognosis in pediatric patients. This retrospective study aims to analyze CT imaging features and clinical data in children with HAdV infections, differentiating between severe and mild cases, to further explore the risk factors linked to severe conditions. We hypothesized that the enlargement of consolidations observed on high-resolution CT imaging, along with certain clinical features, would emerge as independent risk factors for severe conditions. Materials and methods Study design and participants In this retrospective study, we included pediatric patients who were hospitalized with HAdV infections from October 2016 to September 2021 at Tongji Hospital, Huazhong University of Science and Technology, in Wuhan, Hubei Province, China. Prior to the study, pediatricians informed the guardians about the potential benefits and risks associated with CT examinations for pediatric patients, and obtained their informed consent. This retrospective study was approved by the Ethics Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (Approval Number: TJ-IRB20180615). Due to the retrospective nature of the study, the Ethics Committee waived the requirement for individual informed consent. We obtained hospitalization data for the children through the Hospital Information System (HIS), while chest CT imaging data were collected from the Picture Archiving and Communication System (PACS). The patients underwent chest CT examinations within seven days before or after their admission. Additionally, we gathered data including gender, age, admission quarter, length of hospital stays (LOS), clinical symptoms (such as fever, cough, and sore throat), imaging modalities used, and laboratory indicators (such as white blood cell count and lymphocyte ratio). Mechanical ventilation was administered to children whose symptoms did not improve with supplemental oxygen provided via a face shield. Inclusion criteria 1) All children tested positive for HAdV infections and exhibited respiratory symptoms. The diagnoses of HAdV infections and any co-infecting pathogens were performed using the colloidal gold immunochromatographic assay (GICA) and indirect immunofluorescence assays (IFA). 2) Children aged 14 years or younger. 3) Complete clinical data were collected. 4) Chest CT examinations were conducted within 7 days before or after admission. Exclusion criteria 1) Patients with incomplete data. 2) Presence of significant symptoms unrelated to respiratory or gastrointestinal issues, such as keratoconjunctivitis, urinary tract infections, myocarditis, or encephalitis. 3) History of unresolved infectious diseases within one month prior to admission. 4) Poor quality of chest CT images that could not be adequately assessed due to substantial artifacts. A radiologist (SRJ, with 10 years of pediatric clinical experience) conducted patient evaluations. Ultimately, a total of 267 pediatric patients were included in this study. Of these, 23 children who finally received invasive mechanical ventilation were categorized into the severe group, while 244 children who did not require invasive mechanical ventilation were classified in the mild group. The detailed study flow is illustrated in Fig. 1 below. Study flow illustrating the enrollment of 267 children hospitalized with adenovirus infection Evaluation and analysis of chest CT Images All patients underwent a single non-contrast chest CT scan during an end-inspiratory breath-hold, utilizing various scanners and adhering to low-dose chest CT protocols (GE Healthcare, Philips, Toshiba, etc.). Two radiologists (KSC and QJH, with 5 and 10 years of radiological diagnosis experience, respectively) independently reviewed all CT images. Any discrepancies were resolved through discussion until a consensus was reached. The chest CT imaging changes in patients were assessed in accordance with the recommendations set forth in the Fleischer Society terminologies [1]. Consolidation was defined as a uniform increase in lung parenchyma density, characterized by obscured lung texture and identifiable air bronchograms. GGO was distinguished by areas of increased lung density with blurred margins, while the underlying lung texture remained visible. Emphysema was identified as localized regions of low density. CT findings were graded using a semi-quantitative analysis on a 5-scale to evaluate the extent of consolidation, GGO, and emphysema, in accordance with guidelines for the diagnosis and treatment of HAdV from previous studies [17, 18]. The scoring criteria were as follows: Score 0: no abnormal changes; Score 1: abnormal manifestation range ≤ 5%; Score 2: abnormal manifestation range > 5%, ≤ 25%; Score 3: abnormal manifestation range > 25%, ≤ 50%; Score 4: abnormal manifestation range > 50%, ≤ 75%; Score 5: abnormal manifestation range > 75%. The extent of abnormality was assessed independently for each zone of the left and right lungs, specifically: upper left, lower left, upper right, middle right, and lower right zones. The corresponding scores were classified into grades as follows: Grade 1 (Score 0), Grade 2 (Score 1), Grade 3 (Scores 2–3), and Grade 4 (Scores 4–5). Additionally, other manifestations such as bronchiectasis, mosaic sign, atelectasis, and extrapulmonary findings like pleural effusion and pneumothorax were also documented. Statistical analysis Statistical analysis was performed using SPSS version 22.0 (IBM Corp, New York, USA). Continuous variables were expressed as mean ± standard deviation (range). Differences between groups were assessed using the Mann–Whitney U test. Categorical variables were reported as absolute values (percentages), with differences between groups analyzed using Fisher's exact test. Multivariable logistic regression analysis was conducted to identify independent factors associated with severe conditions. All statistical tests were two-sided, and a p-value of < 0.05 was considered statistically significant. Results Epidemiological analysis A total of 267 patients with HAdV infection were enrolled in the study, comprising 165 male and 102 female children, as detailed in Table 1. The mean LOS was approximately 3 days for both the mild and severe groups (Mild: 3.19 ± 2.19 days; Severe: 2.78 ± 1.74 days; P > 0.05). Additionally, there were no significant differences in the quarterly distribution of admissions (Table 1). The duration of fever was 6.25 ± 8.80 days in the mild group and 7.44 ± 6.57 days in the severe group, with no statistically significant difference observed (P > 0.05). All children in the study were either cured or showed improvement and were discharged without any fatalities. Table 1Univariate analysis of clinical and imaging manifestations in two groups of patientsSevere groupMild groupPn (%)Gender >0.05 Male14 (60.87%)151(61.89%) Female9 (39.13)93 (38.11%)Co-infection kinds<0.01 ≥222(95.65%)164(67.21%) One1(4.35%)80(32.79%)Co-infection pathogens Aa1 (4.35%)80 (32.79%) Bb6 (26.09%)86 (35.25%) Cc5 (21.74)38 (15.57%) Dd11 (47.83%)34 (13.93%) Ee0 (0%)6 (2.46%)CT manifestations Normal<0.01 Yes3 (13.04%)102 (41.80) No20 (86.96%)142 (58.20%)Consolidation<0.001 Yes17 (73.91%)76 (31.15%) No6 (26.09%)168 (68.85%)GGO>0.05 Yes10 (43.48%)97 (39.75%) No13 (56.52%)147 (60.25%)Emphysema>0.05 Yes3 (13.04%)31 (12.70%) No20 (86.96%)213 (87.30%)Quarter>0.05 A (1-6)14 (60.87%)162 (66.39%) B (7-12)9 (39.13%)82 (33.61%)Mean±SDAges (y)2.78±1.743.19±2.19>0.05Fever duration(d)7.44±6.576.25±8.80>0.05Laboratory White cell (×109/L)11.49±8.5710.71±5.28>0.05 Lymphocytes proportion (%)27.40±17.1538.87±18.15<0.01Hospital LOS (d)13.87±6.926.92±3.38<0.001A, month1-6; B, month 7-12; LOS Length of stay, Aa Adenovirus infection, not combined with other pathogens; Bb, adenovirus combined with mycoplasma/chlamydia infections; Cc, adenovirus combined with other viral infections; Dd, adenovirus combined with mycoplasma/chlamydia infections, and viral infections; Ee, Others. A P-value of less than 0.05 was considered significant Univariate analysis of clinical and imaging manifestations in two groups of patients A, month1-6; B, month 7-12; LOS Length of stay, Aa Adenovirus infection, not combined with other pathogens; Bb, adenovirus combined with mycoplasma/chlamydia infections; Cc, adenovirus combined with other viral infections; Dd, adenovirus combined with mycoplasma/chlamydia infections, and viral infections; Ee, Others. A P-value of less than 0.05 was considered significant Imaging evaluation On chest CT images, the incidence of positive abnormalities was significantly higher in the severe group (86.96%) compared to the mild group (58.20%) (P < 0.01, Table 1). The details of the CT manifestations for both groups are summarized in Table 2. Notable differences in abnormal features were observed between the two groups. In the severe group, consolidation (35%, n = 8) and consolidation combined with GGO (35%, n = 8) accounted for 70% (16/23) of all findings; nearly 75% (6/8) of these consolidations exhibited bilateral involvement (Fig. 2). For the mild group, the most frequent CT manifestations were simple GGO (18.9%, n = 46) and consolidation combined with GGO (16.4%, n = 40), whereas the proportion of simple consolidation was lower (10.2%, n = 25) (Fig. 3). Table 2Manifestations of chest CT images in Severe and Mild groupsChest CTSevere, total N=23 (n, %)Mild, total N=244(n, %)Abnormal Findings20 (86.96%)142 (58.20%)Consolidation8 (34.78%)25 (10.25%)GGO1 (4.35%)46 (18.85%)Emphysema2(8.70%)14(5.74%)Consolidation & GGO8(34.78%)40(16.39%)Consolidation & Emphysema0(0%)6(2.46%)GGO & Emphysema0(0%)6(2.46%)Consolidation & GGO & Emphysema1(4.35%)5(2.05%)χ223.057P<0.01A P-value of less than 0.05 was considered significant Manifestations of chest CT images in Severe and Mild groups A P-value of less than 0.05 was considered significant CT scans of children hospitalized with adenovirus infection. The CT findings illustrate extensive consolidation in both lungs of a 3-year-old child with severe illness who required mechanical ventilation (A&B). In contrast, scattered ground-glass opacities (GGO) and localized emphysema are visible on the CT images of a 1-month-old child with a mild condition (C&D). Black arrow indicates consolidation; red arrow indicates GGO; yellow arrow indicates emphysema Consolidation scores were used to evaluate the extent of abnormalities in each zone of the left and right lungs separately. Note: *P < 0.05, **P < 0.01, ***P < 0.001, indicating comparisons between the severe and mild groups. U-Left refers to the upper left lobe; L-Left refers to the lower left lobe; U-Right refers to the upper right lobe; M-Right refers to the middle right lobe; L-Right refers to the lower right lobe. A P value of < 0.05 was considered statistically significant In the semi-quantitative assessment of CT images, scores for consolidation, GGO, and emphysema were calculated individually. Only the distribution of consolidation showed a significant difference (Fig. 3), with the severe group presenting higher scores compared to the mild group. The scores for GGO and emphysema did not reveal significant differences between the two groups (data not shown). To further investigate factors related to disease progression, we conducted a univariate analysis of clinical and imaging data. The results indicated that a lower proportion of lymphocytes in laboratory findings significantly correlated with the occurrence of severe conditions. Additionally, children with severe conditions were more likely to have two or more (≥ 2) concurrent infections, and the presence of consolidation on chest CT was closely associated with the progression of severe conditions, with a significant difference observed between the two groups. Detailed findings are presented in Table 1. Based on the multivariate analysis of clinical and imaging manifestations, lower lymphocyte proportions, the presence of two or more co-infections, and higher scores for consolidation on chest CT were identified as independent risk factors for severe conditions associated with HAdV infections, with odds ratios (OR) of 0.955, 1.361, and 9.339, respectively (Fig. 4). Nomograms illustrating the risk factors based on the results of multivariate analysis of clinical and imaging manifestations in the two patient groups. Note: *P < 0.05, ***P < 0.001, indicating comparisons between the severe and mild groups Discussion Firstly, this retrospective cohort study indicates that age is not be correlated with severe conditions in children with HAdV infections. Moreover, we identified co-infection with two or more pathogens, a lower lymphocyte ratio, and a greater degree of consolidation on chest CT images as independent risk factors for the progression of HAdV infections to severe conditions. Our findings suggest that CT imaging features may serve as potential indicators of disease severity at the time of examination. The primary objective of our study was to predict the progression of adenovirus infection in pediatric patients by analyzing baseline CT signs collected around the time of admission. The results demonstrated that the established CT score, along with the presence of abnormal findings, were independent predictors of subsequent progression to severe pneumonia in children infected with adenovirus. To our knowledge, there is a scarcity of studies assessing the risk of severe conditions in children with HAdV infections, and no previous research has effectively combined radiological characteristics with clinical features to explore these risk factors in this population. Previous studies have reported that severe conditions are more likely to occur in children younger than 2 years compared to those with mild cases [19]. However, our findings indicate that age is not a risk factor for developing severe cases of HAdV infection. This discrepancy may be attributed to factors such as the targeted population and sampling period. Our cohort was collected from a more comprehensive hospital over a span of more than five years, contrasting with the limitations of previous studies. Additionally, the inclusion criteria for our study required a CT examination, which younger children are less likely to undergo. Although the majority of HAdV-infected children in our cohort were male, we found no significant gender differences between the mild and severe groups, consistent with earlier literature [19]. Co-infection with other respiratory viruses is frequently reported in cases of HAdV infections [6, 20–22]. In our study, co-infection with two or more additional pathogens was identified in 166 out of 267 HAdV-positive cases. We observed that the frequency of co-infection in children with severe conditions was significantly higher than in those with mild cases. Notably, approximately 95.7% of severely infected children presented with co-infections, and 56.5% had two or more other pathogens. Mycoplasma, Chlamydia, and various viral infections were the most common co-infecting pathogens. For instance, we identified approximately 86 cases in the mild group and 6 cases in the severe group that were co-infected with adenovirus along with Mycoplasma or Chlamydia. However, due to the limited number of severe cases, we did not perform statistical analyses to compare the specific types of co-infecting pathogens between the two groups. One child was even found to be infected with four different pathogens: Mycoplasma, Chlamydia, Epstein-Barr virus, and Influenza B virus. These results align with previous findings indicating that interactions among multiple pathogens can exacerbate pediatric lung disease. Consequently, co-infection is often associated with more severe clinical symptoms, potentially leading to a heightened risk of severe infection [22]. Furthermore, we observed that the proportion of lymphocytes in children with severe HAdV infections is lower than that in those with mild cases. This suggests that lymphocyte levels may reflect disease progression, a finding consistent with other reports [22]. Further multivariate analysis of clinical and CT imaging data indicated that lymphocyte proportion could be a protective factor against the progression of HAdV infections to severe conditions. Research suggests that the Th1 cell response and cellular immunity are primary mechanisms for controlling viral infections [23]. Several studies have identified a correlation between Coronavirus disease-2019 (COVID-19) severity and lymphopenia, with levels of lymphopenia serving as a predictor of disease severity [24, 25]. It has been hypothesized that factors such as cytokine storms, T cell exhaustion induced by COVID-19, and Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) virus-T cell interactions contribute to changes in lymphocyte proportions in severely affected COVID-19 patients. We propose that the mechanism underlying lymphopenia in HAdV infections may be similar to that observed in COVID-19 patients; however, further exploration is warranted. Investigations have indicated that typical chest CT findings in pediatric adenovirus pneumonia include consolidation and GGO, predominantly exhibiting a centripetal distribution. While HAdV infections are often mild to moderate, severe cases can occur in some children [26]. In the present study, we found a correlation between CT imaging features and outcomes of HAdV infections. Compared to mild cases, CT images of severe HAdV infections showed a higher frequency and proportion of consolidations, resulting in significantly elevated consolidation scores across all lobes. The presence of consolidation indicates ineffective gas exchange in the lung tissue, and an increase in consolidation area represents a reduction in effective ventilation [27]. Therefore, when large areas of consolidation occur in one or multiple lobes, patients are at risk for hypoxemia and respiratory distress. On the other hand, GGO is a nonspecific imaging finding characterized by a reduction in lung parenchyma volume without complete destruction of alveolar tissue. Unlike consolidation areas, regions exhibiting GGO retain some degree of ventilation function [27]. This pathologic characteristic may explain our specific finding that GGO did not emerge as a statistical risk factor for severe HAdV infections. This study has several limitations. First, it was a retrospective analysis conducted at a single center, which may affect generalizability. Second, the total number of patients, particularly in the severe cases category, was small, limiting the reliability of our findings. Additional large-scale studies are needed to enhance our understanding of the risk factors associated with severe conditions, especially those related to intubation and severe pneumonia. Furthermore, we did not perform detailed serotype identification of adenoviruses to assess the impact of serotype variations on HAdV infections. In addition, the long-term effects of adenovirus infection on pediatric patients in adulthood, such as pulmonary kinetic energy and exercise tolerance, remain unclear and require extended follow-up studies. Finally, the influence of several other confounding factors was not fully considered in this study. In summary, the present study evaluated the CT imaging features and clinical characteristics of children with HAdV infections, including those with both severe and mild conditions. Our findings can aid in identifying hospitalized children with HAdV infections who are at an increased risk of developing severe complications, thereby prompting the need for more aggressive treatment and care. Acknowledgements We would like to express our gratitude to everyone who took part in this project. We especially thank Suping Chen from CT research center, GE Healthcare for the technical guidance and language modification. Authors’ declaration I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part. All the authors listed have approved the manuscript that is enclosed. Abbreviations Human adenovirus Computed tomography Ground-glass opacity Hospital Information System Picture archiving and communication system Length of hospital stays Colloidal gold immunochromatographic assay Indirect immunofluorescence assay Odds ratio Coronavirus disease-2019 Severe acute respiratory syndrome coronavirus-2 Authors’ contributions Both Shaofang Wang and Shichao Kang contributed equally to the concept design, data collection, image analysis and the paper writing. The corresponding authorship is shared by Qiongjie Hu and Liming Xia. All authors (*Shaofang Wang, *Shichao Kang, Lili Guo, Shuchang Zhou, Yanjie Zhao, Huifen Shen, Shourui Jin, Hanxiong Guan, #Liming Xia, #Qiongjie Hu) reviewed the manuscript. Funding This work was supported by the National Natural Science Foundation of China (Grant No.82101277 and 82001785), and by the Grant from the Natural Science Foundation of Hubei Province (2021CFB076). Data availability All the data and materials can be made available from the Corresponding Author on reasonable requests. Declarations Ethics approval and consent to participate This retrospective study was approved by the Ethics committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, PR China (TJ-IRB20180615). We confirmed that all methods were carried out in accordance with relevant guidelines and regulations. Consent for publication Not applicable Patient's Informed consent was waived by the Ethics committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, PR China, because of the retrospective nature of the study. In this study, the pediatricians had informed the guardians of the potential for benefit or harm of CT exams on pediatric patients and obtained their informed consent. 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. Shaofang Wang and Shichao Kang contributed equally to this work. Contributor Information Liming Xia, Email: xialiming2017@outlook.com. Qiongjie Hu, Email: qjhu@outlook.com. References Associated Data Data Availability Statement All the data and materials can be made available from the Corresponding Author on reasonable requests.