key: cord-1056229-jsjxdmi2 authors: Wang, Zijun; Zhou, Qi; Wang, Chenglin; Shi, Qianling; Lu, Shuya; Ma, Yanfang; Luo, Xufei; Xun, Yangqin; Li, Weiguo; Baskota, Muna; Yang, Yinmei; Zhai, Hui; Fukuoka, Toshio; Ahn, Hyeong Sik; Lee, Myeong Soo; Luo, Zhengxiu; Liu, Enmei; Chen, Yaolong title: Clinical Characteristics of Children with COVID-19: A Rapid Review and Meta-Analysis date: 2020-04-17 journal: nan DOI: 10.1101/2020.04.13.20064352 sha: bd028cdb7e70f740f9a1efe4b58255367814364b doc_id: 1056229 cord_uid: jsjxdmi2 Abstract Background: Most guidelines on COVID-19 published so far include recommendations for patients regardless of age. Clinicians need a more accurate understanding of the clinical characteristics of children with COVID-19. Methods: We searched studies reporting clinical characteristics in children with COVID-19 published until March 31, 2020. We screened the literature, extracted the data and evaluated the risk of bias and quality of evidence of the included studies. We combined some of the outcomes (symptoms) in a single-arm meta-analysis using a random-effects model. Results: Our search retrieved 49 studies, including 25 case reports, 23 case series and one cohort study, with a total of 1667 patients. Our meta-analysis showed that most children with COVID-19 have mild symptoms. Eighty-three percent of the children were within family clusters of cases, and 19% had no symptoms. At least 7% with digestive symptoms. The main symptoms of children were fever (48%, 95% confidence interval [CI]: 39%, 56%) and cough (39%, 95% CI: 30%, 48%). The lymphocyte count was below normal level in only 15% [95% CI: 8%, 22%] of children which is different from adult patients. 66% [95% CI: 55%, 77%] of children had abnormal findings in CT imaging. Conclusions: Most children with COVID-19 have only mild symptoms, and many children are asymptomatic. Fever and cough are the most common symptoms in children. Vomiting and diarrhea were not common in children. The lymphocyte count is usually within the normal range in children. Two reviewers (Z Wang and C Wang) selected the studies independently after first eliminating duplicates. The bibliographic software EndNote was used and any discrepancies were settled by discussion, consulting a third reviewer (Q Zhou) if necessary. Before the formal selection, the reviewers searched a random sample of 50 citations. The reviewers screened first all titles and abstracts with the pre-defined criteria, and categorized the articles into three (eligible, not eligible, and unclear) groups. In the second Step, full-texts of those potentially eligible or unclear studies were reviewed to identify the final inclusion. All the reasons for exclusion of ineligible studies were recorded, and the process of study selection was documented using a PRISMA flow diagram (14, 15) . Two reviewers (Q Shi and S Lu) extracted the data independently with a standardized data collection form, including:1) basic information (e.g. first author); 2) symptoms; 3) routine blood tests (e.g. leucocyte count); 4) blood biochemistry (e.g. alanine aminotransferase); 5) coagulation function (e.g. activated partial thromboplastin time); 6) imaging findings (e.g. abnormal imaging). For dichotomous outcomes, we abstracted the number of events and total participants per group. For continuous outcomes, we abstracted means, standard deviations (SD), and the number of total participants in per group. Outcomes with no events were reported, but these were excluded from the metaanalysis. If means and standard deviations (SD) were not reported, we calculated them from the reported indicators. (16) If data were missing or reported in an unusable way, we excluded the study from the meta-analysis and report the findings descriptively. All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint Two reviewers (Z Wang and C Wang) assess the risk of bias in each study independently. Discrepancies were settled by discussion, consulting a third reviewer (Q Zhou) if necessary. For randomized controlled trials (RCTs), we will assess the risk of bias independently using Cochrane risk-of-bias tool (17). It consists of seven domains, for each, we will grade as "Low", "Unclear", and "High". For nonrandomized controlled trials (nRCTs), ROBINS-I tool will be used(18). It consists of seven domains, for each, we will grade as "Low risk", "Moderate risk", "Serious risk", "Critical risk", and "No information". For case-control and cohort studies, the Newcastle-Ottawa Scale will be used (19, 20) . It consists of eight domains, for each, we will grade with stars. The more stars, the lower the risk of bias. For cross-sectional studies, we use a methodology evaluation tool recommended by Agency for Healthcare Research and Quality (AHRQ) (21). This tool assesses the quality of bias according to 11 criteria. And each criterion is answered by "Yes", "No" or "unsure". For case reports and case series, we used a methodology evaluation tool recommended by National Institute for Health and Care Excellence (NICE) (22). The risk of bias is evaluated according to eight criteria. The results were summarized by scoring method, for the "Yes" items, the score was 1, and for the "no" items, the score was 0. The higher the total score, the lower the risk of bias. We summarized the results of the studies including less than 9 patients and did metaanalysis of included studies that have at least 9 patients. For dichotomous outcomes, we did a meta-analysis of proportions, reporting the effect size (ES) with 95% confidence intervals (CI). For continuous outcomes, we did a meta-analysis of continuous variable, calculating the effect size (ES) with 95% CI. We described the results of the studies with patients that below nine. As clinical and methodological heterogeneity in the study design, All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint characteristics of participants, interventions and outcome measures was expected, we used random-effects models (23). Two-sided P values < 0.05 were considered statistically significant. Heterogeneity was defined as P values <0.10 and I 2 >50%. All analyses were performed in STATA version 14. Two reviewers (Q Zhou and Y Xun) assessed the quality of main evidence independently using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool. We produced a "Summary of Findings" table using the GRADE pro software (24, 25). Direct evidence from RCTs is first set as high quality, and evidence from observational studies as low quality. Then initial quality can then be downgraded for five reasons (study limitations, consistency of effect, imprecision, indirectness, and publication bias) and upgraded for three reasons (large magnitude of effect, dose-response relation and plausible confounders or biases) (26-31). Finally, the quality of main evidence can be classified as high, moderate, low, or very low, which reflects the extent to what we can be confident that the effect estimates are correct. As COVID-19 is a public health emergency of international concern and the situation is evolving rapidly, our study was not registered in order to speed up the process. (32) Our initial search retrieved 774 records ( Figure 1) More information can be found in Table 2 and Supplementary Material 4. Seventeen case series reported the results of routine blood tests. The mean leucocyte count in children was 6 More information can be found in Table 3 and Supplementary Material 4. Children had on average milder symptoms, with many children having even no symptoms. Most children infected with COVID-19 were exposed through family clusters. About half of children present with fever or cough, and about one third of children with both fever and cough. Only a small minority of children had vomiting or diarrhea as initial symptoms. Leucocyte and lymphocyte counts are often in normal or above the normal range in children with COVID-19. Abnormalities in CT imaging were found in more than half of the children, the most common being ground-glass opacity in unilateral lung. The course of COVID-19 in children can be characterized by mild illness and no symptoms. According to a study by the Chinese CDC, as of February 11, 2020, 81% of all patients with COVID-19 showed only mild symptoms. (82) However, although the disease was less severe in the majority of adults (51%-74%) (83-85), 26%-32% of adults were still committed to ICU, and had often basic diseases such as hypertension or diabetes. (86) (87) (88) In contrast, we found only 3% report of children with severe illness. The Chinese CDC also pointed out that 889 (1%) patients with COVID-19 were asymptomatic (82); in our study, about 19% of children were completely asymptomatic, which is higher than the average level of the whole patient significantly. A study of asymptomatic infections with COVID-19 also showed that 29.2% of cases showed normal CT image and had no symptoms during hospitalization. What's more, these cases were younger (median age: 14.0 years; P=0.012) than the rest.(89) Of children with COVID-19, 83% had other family members infected. The majority of asymptomatic children in family clusters were confirmed after a positive nucleic acid test, which was conducted because of the close contact to infected family members. It seems that family cluster in children were more likely to be tested than adults. So, we suspect whether asymptomatic children are really asymptomatic or are still in the incubation period. Another explanation could be that only symptomatic cases transmit; So, adults without symptoms are seldom diagnosed, while children without symptoms (who had not that many any other contacts during the holiday season than their family members) get diagnosed anyway because of the obvious exposure. The diagnosis of suspected cases in children needs comprehensive consideration. Fever and cough were the main symptoms in patients with COVID-19, which is reported by the most of current guidelines and recommendations. (12, 90-93) Although fever and cough All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint also the first two symptom of children. When compared with adults, fever and cough occurred only in 48% and 39% of children, respectively. The rates of fever and cough in adults are up to 98%(88) and 87% (84) , which indicates that fever and cough in children are not as common as in adults. Compared with children infected with SARS, MERS and other viral diseases (94, 95) , there are no specific symptoms in children with COVID-19 that could help to diagnose the disease accurately. Therefore, detection methods are particularly important for the diagnosis of COVID-19. Chinese National Health Commission also pointed out that fever and/or respiratory symptoms, imaging features indicating of pneumonia, leukocyte and lymphocyte counts characteristics in the early stage, and epidemiological history should be comprehensively used to determine suspected cases. After that, RT-PCR, sequencing or specific antibody were used to make a definite diagnosis.(12) Attention should be paid to the children with COVID-19 who start with gastrointestinal symptoms. Although gastrointestinal symptoms such as nausea, vomiting and diarrhea are less common in children with COVID-19, recent studies have SARS-CoV-2 in the feces of patients (96, 97) and a study showed that some children persistently tested positive on rectal swabs even after nasopharyngeal testing was negative, raising the possibility of fecal-oral transmission. (98) According to one study, diarrhea was the first symptom in three out of 31 children. (88) Moreover, three of the eight children with severe cases of COVID-19 that reported symptom clearly had gastrointestinal symptoms, one of them started with gastrointestinal symptoms, without any obvious respiratory tract infection in the early stage (37). In addition, comorbidity cannot be ignored either. Our results showed that half of these eight children with severe cases had other diseases, including two with intussusception and one of them was dead (47). Similarly, a study of All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint children with MERS also suggest that serious illness can occur in children with underlying disease. (99) Although there is no evidence that gastrointestinal symptoms and comorbidity in children are related to the severity of the disease, clinicians should pay attention to the gastrointestinal symptoms and comorbidity in the process of diagnosing children with COVID-19 and apply real-time monitoring and protection. Abnormal CT imaging was less common in children with COVID-19 than adults, but the imaging findings were similar in children and adults. Unilateral pneumonia is common in children with COVID-19, and the main change in imaging is ground-glass opacity. However, bilateral pneumonia is more common in adults, and the main change in imaging is also ground-glass opacity. (86, 87) One guideline (93) pointed out that there were multiple small patch shadows and interstitial changes in the early stages of the disease in adults, especially in the extrapulmonary zone in chest imaging. Furthermore, multiple ground-glass opacity or infiltrative shadows may develop in both lungs. In severe cases, pulmonary consolidation may occur, and pleural effusion is rare. An analysis of CT features in children with COVID-19 showed that in 15 cases, inflammatory infiltration was found in the chest CT imaging during initial diagnosis and reexamination. Most inflammatory infiltrations were manifested as small nodular ground-glass opacities, and multiple lobe segments were less involved. Multiple lobe segments were involved in only one case, and the imaging changes were not typical in the advanced stage as well (39). These findings suggest that pulmonary inflammation in children is mild and localized. The results of laboratory tests of children with COVID-19 were more often within the normal range than those of adults. The leukocyte count of children with COVID-19 was usually normal or below the normal range. The lymphocyte count was generally normal or above the normal range, and only 15% of cases were below the normal range. While for adults, the leukocyte count was usually normal or above the normal range, and the lymphocyte count were mostly below the normal range (35%-63%). (86, 87) It can be seen that there were significant differences in routine blood values between adults and children. However, some published COVID-19 guidelines and consensus for children consider a reduction in lymphocyte count as one of the factors for diagnosing suspected cases (90) (91) (92) . Our study demonstrates that guidelines for children should not be formulated in full accordance with adult standards, otherwise true cases may be missed. Most of the existing systematic reviews on characteristics of patients with COVID-19 are based on adult patients or patients regardless of age.(100-105) Two of these compared the differences between children and adult patients. (100, 101) Only one review reported the clinical characteristics of children with COVID-19 at present.(106) All reviews that considered children had similar outcomes: for example, children had milder symptoms than adults, some children had no symptoms, and lymphopenia in children did not occur as often as in adults. But our rapid review has included 49 studies of children patients, which is more than most of these reviews together. In contrast to another systematic review of children with COVID-19 (106), we included studies not only China but also from other countries such as Singapore, which were published in Chinese and English. Moreover, we conducted a meta-analysis whereas the previous study only did a systematic literature review. This rapid review has several strengths. First, although this is not the first systemic review about the clinical characteristics of children with COVID-19, but is to our knowledge the first to combine the results with meta-analysis and GRADE evaluation of the quality of main evidence, which is of great importance for clinicians to diagnose and treat children All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04.13.20064352 doi: medRxiv preprint rapidly. Second, our study points out the loopholes in some current guidance documents that suggest the diagnosis of suspected cases -also in children -based on the lymphocyte count. Third, as a rapid review, this study summarizes the latest published information on clinical cases, which provides relatively high-quality evidence for the formulation of clinical practice guidelines in the rapidly evolving public health emergency situation and helps policy-makers to make evidence-based decisions quickly (107). Our study has also some limitations. First, due to the rapid fermentation of the public health emergency and new cases emerging continuously, the findings of this review may get outdates relatively soon. Second, cannot be sure if some cases were included in multiple studies. Third, at present, there is no unified definition for clinical classification of the severity of COVID-19, so we had to combine light, mild and moderate disease into one category (mild), while severe and critical cases were both considered as severe cases. The researchers should aim to conduct more targeted studies on COVID-19 in specific subpopulations. Policy makers should develop accurate guidelines for both children and adults. Clinical practitioners should pay attention on the specific characteristics of different patient populations to improve the accuracy of diagnosis and treatment. Children with COVID-19 are more common to have only mild symptoms, and many children are even completely asymptomatic. Fever and cough are the main symptoms of COVID-19 in both children. Vomiting and diarrhea occurring less frequently in children. Ground-glass opacity is the most common CT imaging of children. Whereas adults tend to have elevated lymphocyte count at the beginning of the disease, in children the lymphocytes were usually within the normal range. As the characteristics of COVID-19 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint differ between adults and children in multiple ways, specific criteria for the diagnosis and treatment of COVID-19 in children are urgently needed. We thank Janne Estill, Institute of Global Health of University of Geneva for providing guidance and comments for our review. We thank all the authors for their wonderful collaboration. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04.13.20064352 doi: medRxiv preprint Contents Supplementary Material 1-Search Strategy ............................................................... 22 Supplementary Material 2-Risk of bias ...................................................................... .................................................................... 28 Supplementary Material 3-GRADE evidence profile ................................................ 29 Supplementary Material 4-Case report summary ..................................................... 32 Table A. Symptom ............................................................................................. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is Web of Science #1 TOPIC: "Epidemiological Studies" #2 TOPIC: "Epidemiologic* Study" #3 TOPIC: "Epidemiological characteristics" #4 TOPIC: "Clinical features" #5 TOPIC: "Clinical characteristics" #6 TOPIC: "Clinical Presentation" #7 #1-6/OR #8 TOPIC: "COVID-19" #9 TOPIC: "SARS-COV-2" #10 TOPIC: "Novel coronavirus" #11 TOPIC: "2019-novel coronavirus" #12 TOPIC: "coronavirus disease-19" #13 TOPIC: "coronavirus disease 2019" #14 TOPIC: "COVID19" #15 TOPIC: "Novel CoV" #16 TOPIC: "2019-nCoV" #17 TOPIC: "2019-CoV" #18 TOPIC: "Wuhan-Cov" #19 TOPIC: "Wuhan Coronavirus" #20 TOPIC: "Wuhan seafood market pneumonia virus" #21 #8-20/ OR #22 #7 AND #21 Cochrane #1 "COVID-19":ti,ab,kw #2 "SARS-COV-2":ti,ab,kw #3 "Novel coronavirus":ti,ab,kw #4 "2019-novel coronavirus" :ti,ab,kw #5 "Novel CoV" :ti,ab,kw #6 "2019-nCoV" :ti,ab,kw All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04.13.20064352 doi: medRxiv preprint #7 "coronavirus disease-19" :ti,ab,kw #8 "coronavirus disease 2019" :ti,ab,kw #9 "COVID19" :ti,ab,kw #10 "Wuhan-Cov" :ti,ab,kw #11 "Wuhan Coronavirus" :ti,ab,kw #12 "Wuhan seafood market pneumonia virus" :ti,ab,kw #13 #1-12/OR #14 "Epidemiological Studies":ti,ab,kw #15 "Epidemiologic* Study":ti,ab,kw #16 "Epidemiological characteristics":ti,ab,kw #17 "Clinical features":ti,ab,kw #18 "Clinical characteristics":ti,ab,kw #19 "Clinical Presentation":ti,ab,kw #20 #14-19/ OR #21 #13 AND #20 the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04.13.20064352 doi: medRxiv preprint the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13 the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.04. 13.20064352 doi: medRxiv preprint World Health Organization; c2020 A Novel Coronavirus from Patients with Pneumonia in China Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia A familial cluster of infection associated with the 2019 novel coronavirus indicating potential person-to-person transmission during the incubation period Outbreak of pneumonia of unknown etiology in Wuhan, China: The mystery and the miracle Potential for global spread of a novel coronavirus from China Chest CT Findings in Patients with Corona Virus Disease 2019 and its Relationship with Clinical Features Diagnosis, treatment, and prevention of 2019 novel coronavirus infection in children: experts' consensus statement Diagnosis and treatment of 2019 novel coronavirus infection in children: a pressing issue The novel coronavirus pneumonia diagnosis and treatment plan (trial version 7) [Internet]. National Health Commission of the People's Republic of China; c2020 Global surveillance for COVID-19 disease caused by human infection with the 2019 novel coronavirus World Health Organization; c2020 The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration The methodological and reporting quality of systematic reviews from China and the USA are similar Comparison of the Clinical Characteristics Between RNA Positive and Negative Patients Clinically Diagnosed With 2019 Novel Coronavirus Pneumonia Analysis of Clinical Features of 29 Patients With 2019 Novel Coronavirus Pneumonia Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in Nanjing Recommendation for the diagnosis and treatment of novel coronavirus infection in children in Hubei (Trial version 1) Diagnosis and treatment recommendation for pediatric coronavirus disease-19 Diagnosis and treatment of the 2019 novel coronavirus infection in children. (Trial version 2 of Children's Hospital of Chongqing Medical University The programme for prevention and control of SARS-CoV-2 infection in neonates in Hunan Provence Clinical analysis of pediatric SARS cases in Beijing Middle East respiratory syndrome coronavirus disease in children Epidemiologic Features and Clinical Course of Patients Infected With SARS-CoV-2 in Singapore Surface Environmental, and Personal Protective Equipment Contamination by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) From a Symptomatic Patient Characteristics of pediatric SARS-CoV-2 infection and potential evidence for persistent fecal viral shedding Middle East Respiratory Syndrome Coronavirus Disease in Children Clinical, laboratory and imaging features of COVID-19: A systematic review and meta-analysis COVID-19: Children Comparison with Adults Based on the Latest Data COVID-19 patients' clinical characteristics, discharge rate, and fatality rate of meta-analysis Clinical Characteristics of 2019 Novel Infected Coronavirus Pneumonia：A Systemic Review and Meta-analysis Clinical characteristics of hospitalized patients with SARS-CoV-2 infection: A single arm meta-analysis downgrade one level: The risk of bias is high due to the limitations of study design. 2. downgrade one level: Heterogeneity of data synthesis results, I 2 > 50% downgrade one level: Sample size is less than optimal information sample (OIS) upgrade two levels: Large magnitude of effect Outcome (number of studies)