Which Groups of Children Are at More Risk of Fatality during COVID-19 Pandemic? A Case-Control Study in Yazd, Iran

Introduction The study aims to investigate the characteristics, comorbidities, laboratory findings, and clinical manifestations of under 18-year-old patients who died with the diagnosis of COVID-19 and determination of the most prevalent risk factors. Method This case-control study was performed at a referral hospital in Yazd from March 2020 to August 2021. All patients under 18 years who were diagnosed through real-time RT-PCR, chest computed tomography, and the World Health Organization definition were divided into deceased and survived groups. The characteristics (age and sex), disease severity, comorbidities, laboratory findings, and clinical manifestations of the two groups were compared and analyzed using SPSS, version 18 (SPSS Inc., Chicago, III., USA). Results A total of 24 patients in the deceased group and 167 patients in the survived group were compared. The highest mortality rate was observed in the age group of 1 month to 5 years, although no statistically significant relationship was found between age groups and the risk of mortality. Disease severity, dyspnea, low oxygen saturation on admission, length of hospital stays, and hospitalization history before the last admission were significantly correlated with mortality (P < 0.05). Lymphopenia increased the probability of mortality by more than two times (OR: 2.568; 95% CI (0.962–6.852)), but this was not the case for D-dimer and C-reactive protein. Furthermore, 27.5% of survived patients had normal chest CT scans, which was a statistically significant difference compared to the deceased patients (P: 0.031). Conclusion Based on the findings of this study, dyspnea, low oxygen saturation, and lymphopenia are critical indicators for identifying high-risk children with COVID-19 and triaging them for better care and treatment.


Introduction
Even though more than three years have passed since COVID-19 was declared a pandemic and resulted in the death of more than six million people worldwide, there are still many unanswered questions about this disease [1].Previous studies have revealed that children are less susceptible to COVID-19 and experience milder symptoms compared to adults.Furthermore, more recent investigations have indicated that the severity of the disease in children during the second and third waves of the pandemic is either comparable to or even lower than what was observed during the frst wave [2][3][4][5][6][7].
Te most frequently observed clinical symptoms in children with COVID-19 include fever, cough, and sore throat.As the disease progresses, there is a possibility of experiencing dyspnea, which can eventually lead to the development of acute respiratory distress syndrome (ARDS) or multiorgan failure [8].Additionally, lymphopenia, increased serum D-dimer, and procalcitonin levels are prevalent laboratory fndings.Chest computed tomography (CT) scans often reveal patchy lesions and ground-glass opacities [9,10].Certain factors, such as male gender, underlying diseases, age under one month, and symptoms of lower respiratory tract infection, increase the risk of fatal conditions and death [11,12].
Studies conducted in the United States of America and Europe have identifed a distinct medical condition known as pediatric infammatory multisystem syndrome temporally associated with SARS-CoV-2 (PIMS-TS) or multisystem infammatory syndrome in children (MIS-C).Tis condition has been found to be closely associated with the severity of COVID-19 [13,14].Te symptoms of PIMS-TS or MIS-C closely resemble those of Kawasaki disease (KD).Tey exhibit characteristics such as cytokine storms, macrophage activation, hyper-infammatory traits, and elevated levels of proinfammatory cytokines such as IL-1, IL-6, and TNF-α [15].Patients with MIS-C often present with gastrointestinal, mucocutaneous, and cardiovascular symptoms, and the condition is associated with higher disease severity [16,17].
Te progression of COVID-19 typically includes unpredictable events, beginning with an asymptomatic incubation period that can persist for several days, followed by a sudden onset of respiratory symptoms, which can eventually lead to severe stages requiring hospitalization in an intensive care unit and mechanical ventilation.Troughout infection, the body's immune and infammatory responses can be divided into three distinct phases: an initial innate immune response focused on the lungs, followed by a local/ systemic immune response, and potentially progressing to excessive infammatory responses that may trigger a cytokine storm and result in ARDS and multiple organ failure.Cytokine storm-like syndromes have the potential to induce a series of abnormalities and tissue damage, encompassing lymphocyte dysfunction, irregularities in granulocytes/ monocytes, damage to the vascular barrier, capillary impairment, difuse alveolar damage, multiorgan failure, and even fatal outcomes.In other words, the cytokine storm induced by the virus correlates with the severity of COVID-19 [8,15].On the other hand, research has indicated that patients with primary immune system defciencies, particularly those with agammaglobulinemia and the absence of B lymphocytes, tend to experience a milder course of COVID-19 with more favorable outcomes.However, in patients with common variable immunodefciency (CVID) and dysfunctional B lymphocytes, the disease often takes a severe form, necessitating ICU admission, mechanical ventilation, and treatment with IL-6-blocking drugs [18].
Te objective of this study was to explore the characteristics, comorbidities, laboratory fndings, and clinical manifestations of patients under 18 years old who died from COVID-19.Additionally, we aimed to identify the most prevalent risk factors associated with mortality in this population.

Study Design and Data
Collection.Tis case-control study enrolled all patients under 18 years old who were admitted to a referral hospital in Yazd and diagnosed with COVID-19 between March 2020 and August 2021.Te study utilized a census sampling method, which did not exclude any patients from the research.All data were extracted from the patients' archived medical records at the hospital.Te patients were categorized into two distinct groups: the deceased group, consisting of patients who died during their hospitalization, and the survived group, comprising those who were discharged.Patient categorization as confrmed, probable, and suspected cases was determined according to the criteria defned by the World Health Organization (WHO) [19].
We conducted a comparative analysis of various factors, including demographic variables (age and sex), disease severity, underlying medical conditions, clinical presentations, laboratory results, hospitalization duration, pediatric intensive care unit (PICU) admission, reverse transcription polymerase chain reaction (RT-PCR) test outcomes, and spiral chest CT scan fndings (pattern and severity score) between the two study groups.Disease severity was categorized into three distinct levels: nonsevere, severe, and critical, in accordance with WHO guidelines [20].According to the World Health Organization (WHO) criteria, patients were categorized into three groups: confrmed (positive PCR regardless of symptoms), probable (positive lung CT fndings in patients with negative or unknown PCR results), and suspected (children with a history of respiratory symptoms, with or without fever, and a documented close contact with a probable or confrmed case of COVID-19).
To diagnose confrmed cases of COVID-19, we employed the Novel Coronavirus (2019-nCoV) Nucleic Acid Diagnostic Kit (PCR-Fluorescence Probing) developed by the Sansure Biotech Company.Tis kit is used for the qualitative detection of the ORF1ab and N genes of SARS-CoV-2 RNA, and a cycle threshold (Ct) ≤ 40 was considered positive.All the results were interpreted by a trained technician based on positive and negative controls.Furthermore, the test exhibited a sensitivity of 95%.
A D-dimer level above 200 mg/dl and a C-reactive protein (CRP) level above 10 mg/dl were considered positive.Te defnition of lymphopenia difered in diferent age groups of children, and we assessed it individually.

Statistical Analysis.
Te data were analyzed using descriptive and dispersion statistics to report the results.Te chi-square test and Fisher exact test were employed to compare the frequency of explanatory variables between the survived and deceased patients.Logistic regression was 2 Canadian Journal of Infectious Diseases and Medical Microbiology conducted to estimate the risk of death among deceased patients based on the groups of explanatory variables, with comparison to the survived children.Te statistical software SPSS Statistics for Windows, version 18 (SPSS Inc., Chicago, III., USA) was used for data analysis.A signifcance level of 0.05 was employed for all tests, and all statistical analyses were conducted with a two-tailed approach.

Demographic Characteristics and Disease Severity.
Out of these patients, 133 (69.6%) were confrmed cases, 26 (13.6%) were considered probable cases, and 32 (16.8%) were suspected cases.Te deceased and survived groups included 24 and 167 patients, respectively.Te PCR test was positive in 62.5% of the deceased group and 70.7% of the survived group (P � 0.416).Te demographic characteristics and disease severity of the 191 patients under 18 years old are presented in Table 1.

Underlying Disease.
In the group of deceased patients, it was found that 15 individuals (62.5%) had preexisting health conditions, with malignancy and neurologic disease being the most frequently observed underlying diseases, respectively (Table 2).

Signs and Symptoms.
In our study, fever emerged as the most prevalent symptom in both the deceased and survival groups in on-admission and preadmission time.Dyspnea was more common in the deceased group with a signifcant diference (Table 3).

Laboratory Findings.
A comparison of laboratory fndings on the frst day of admission was conducted for both groups, and no signifcant diferences were observed (Table 4).Lymphopenia was observed in 47.4% of patients in the deceased group and 26.4% of patients in the survived group (P = 0.054) (OR � 2.568, 95% CI (0.962-6.852)).
D-dimer and C-reactive protein (CRP) levels were measured in two groups, revealing no signifcant diference between them.

Chest CT Scan Pattern and Severity.
In our study, we evaluated the pattern and severity score of chest spiral CT scans in 82 patients.Te results showed that ground-glass opacity and consolidation were the predominant fndings, as summarized in Table 5.

Other Findings.
A signifcant diference in the length of hospital stay was observed between the two groups.Specifcally, the percentage of patients who were hospitalized for less than 7 days was 78.4% in the group that survived, while it was 50% in the deceased group (P = 0.011).
As expected, the number of patients admitted to the pediatric intensive care unit (PICU) was signifcantly higher in the deceased group compared to the survived group (P < 0.001) (OR � 37.342, 95% CI: (8.399-166.030)).

Statistically Signifcant Findings.
In this study, we conducted a comparison between the characteristics, comorbidities, laboratory fndings, and clinical manifestations of 24 patients under 18 years old who died and 167 patients who survived COVID-19.Our fndings revealed several signifcant factors that increased the risk of death.Tese factors included disease severity, presence of dyspnea, lymphopenia, hospital stay of more than 7 days, previous hospitalization before admission to our center, and low oxygen saturation levels.Additionally, we observed that the survived group had a higher prevalence of normal chest CT scan fndings.In our study, consistent with other similar studies, ground glass opacity (GGO) was found to be the most common fnding in lung CT scans of pediatric COVID-19 cases.However, no signifcant correlation was found between GGO and mortality.Instead, our fndings highlight the importance of considering other indicators, such as oxygen saturation and dyspnea upon admission, for a more comprehensive assessment of outcomes.It is possible that a larger study with a greater sample size could potentially yield diferent or contrasting results.

Disease Severity Was Highly Correlated, but Age and
Gender Were Not.As expected, the severity of the disease showed a signifcant correlation with the risk of mortality in the children we studied.A recent study conducted in Iran, involving 47 hospitalized COVID-19 patients, found that symptoms such as dyspnea, vomiting, and chest pain were signifcantly associated with disease severity.Some of the fndings of their study are consistent with the results of our study [22].
Furthermore, our study identifed a higher frequency of deaths within the 1 month to 5 years age group compared to other age groups.However, it is important to note that no statistically signifcant relationship was observed between any specifc age group and the risk of mortality.However, some studies suggest a higher frequency of deaths among younger age groups compared to older ones [23,24].

Canadian Journal of Infectious Diseases and Medical Microbiology
Additionally, a systematic review of 83 studies did not fnd any association between gender and death in children.However, they did identify an increased risk of death in patients aged 1 to 4 years and infants [25].Consistent with other studies, our study also found that male patients were the majority in the deceased group [11,26].
Furthermore, our study revealed a signifcant correlation between previous hospitalization due to COVID-19 and increased mortality (P < 0.001).Te reason behind this fnding can be attributed to the fact that our center is specifcally designated as a referral hospital for COVID-19 cases in our region, and critically ill patients who need special care and hospitalization in the PICU were referred to this center.

Dyspnea and Low Oxygen Saturation Raised the Risk of Death, and Fever Was the Most Common.
Our study confrmed that dyspnea on admission time was strongly correlated with mortality, consistent with a study conducted in Vietnam [27].Fever was the most prevalent symptom observed in both the deceased and survived groups, followed by vomiting and dyspnea in the deceased group, and cough in the survived group, aligning with previous research [28][29][30][31][32][33][34].Notably, despite the high frequency of fever and vomiting among the deceased group, there was no statistically signifcant increase in the risk of death, which corresponds with fndings from a study conducted in Brazil [35].Several studies have    Our study showed that none of the children who died due to COVID-19 had a normal CT scan, which could be due to the impact of lung involvement on the patient's outcome.Te most common pattern seen in both groups was ground glass opacity.Tese fndings were consistent with a systematic review that included 3557 children with COVID-19 [37].Another metaanalysis by Nino et al. also reported ground glass opacity as the most common pattern in children, which aligns with our study [38].Consistent with our study, an abnormal chest CT scan in a study conducted in Istanbul showed a statistically signifcant relationship with the severity of the disease of COVID-19, which was associated with a relative increase in the risk of mortality [39].

Lymphopenia Was the Only Laboratory Parameter with a Statistically
Signifcant Relationship with Mortality.Tis study revealed that lymphopenia was the only laboratory parameter that showed a statistically signifcant correlation with mortality in COVID-19 patients.Tis could be attributed to the uncontrolled progression of the pathogen in critically ill patients, which is often accompanied by hypercytokinemia [40].As lymphocytes play a crucial role against viruses, lymphopenia can create a vicious cycle that ultimately exacerbates the severity of the disease.A separate cohort study conducted in Korea also indicated that lymphopenia, as well as its severity, can serve as reliable predictors of poor clinical outcomes, including mortality, the need for intensive care, and oxygen requirement [41].
Furthermore, a review study suggested that lymphopenia could be a strong predictive factor for the severe and progressive course of COVID-19 [42].
A systematic review published in October 2020 revealed that hypoalbuminemia, lymphopenia, leukocytosis, elevated levels of interleukin 6, and prolonged PT time were found to be associated with mortality in children [43].

No Statistically Signifcant Relationship Found in Un-
derlying Disease.In this study, malignancy was the most common underlying disease in died and survived groups.However, no signifcant correlation was observed between the presence of underlying disease and the risk of mortality.Some studies showed that underlying respiratory and neurologic diseases were the most common underlying conditions in children who died from COVID-19 [44,45].

Conclusion
Based on the fndings of this study, dyspnea, low oxygen saturation, and lymphopenia emerged as pivotal markers for identifying high-risk children.Tese critical factors can serve as valuable tools for efcient triage and ensuring enhanced care and treatment for these individuals.Tese factors can be used for triaging purposes to ensure better care and treatment for these individuals.Also, it is necessary for the treatment staf and children's caregivers to receive more training about these indicators.

Data Availability
Te registered data utilized to generate the fndings of this study are subject to restrictions imposed by the Ethics Committee of Shahid Sadoughi University of Medical Sciences and Health Services, aiming to safeguard patient privacy.However, researchers who satisfy the criteria for accessing confdential data can make a data request to Mehran Karimi at drmehrankarimi@yahoo.com.

Conflicts of Interest
Te authors declare that there are no conficts of interest in the publication of this paper.

Table 1 :
Demographic characteristics and disease severity of children with COVID-19.

Table 2 :
Underlying diseases in survived and deceased groups of children with COVID-19.
Te odds of death in children with underlying diseases was approximately 2.23 times higher than that in children without underlying diseases.

Table 3 :
Signs and symptoms in preadmission and on-admission time in deceased and survived children with COVID-19.

Table 4 :
Laboratory fndings of deceased and survived children with COVID-19.Laboratory fndings include sodium, potassium, platelets count, weight blood cell (WBC) count, calcium, phosphorus, prothrombin time (PT), partial thromboplastin time (PTT), alanine transaminase (ALT), aspartate aminotransferase (AST), magnesium, erythrocyte sedimentation rate (ESR), creatine phosphokinase (CPK), albumin, blood glucose, and lactate dehydrogenase (LDH) were assessed into two groups at the frst day of admission and number of patients who had pathological laboratory marker as upper and lower limits of normal range were recorded. *

Table 5 :
6attern and severity scores of spiral chest CT scan imaging in pediatric patients with COVID-19.Normal chest CT scan was not observed in any of the deceased children, In addition, the predominant fndings of chest CT scans in both deceased and survived patients were ground glass opacity and consolidation.6CanadianJournal of Infectious Diseases and Medical Microbiology