Systematically Mapping Clinical Features of Infections With Classical Endemic Human Coronaviruses

TO THE EDITOR—The classical endemic human coronaviruses including HKU1, OC43, NL63, and 229E are widely circulating among the global population. However, research on these coronaviruses has been largely neglected, until the recent outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that caused the coronavirus disease 2019 (COVID-19) pandemic. Emerging evidence indicates cross-reactive immunity toward SARS-CoV-2 by preexposure of endemic coronaviruses, which may induce possible protective effects but could also interfere with serological diagnosis of COVID-19 [1]. Therefore, several recent studies have investigated the prevalence of these endemic coronaviruses [2–4]. In line with these studies, Cummings et al [5] found that 8% of healthcare personnel were infected with endemic coronavirus, and they further investigated risk factors of the infection. However, little efforts have been dedicated to understand their clinical features. Such knowledge is essentially required for assessing the burden of endemic coronavirus infection per se, but also for better understanding the implications in the COVID-19 pandemic. In this study, we performed a systematic review and meta-analysis to comprehensively characterize the epidemiology and clinical features of endemic coronavirus infection (Supplementary Data). From 3225 searched records, we identified 157 eligible studies originated from 44 countries/territories across 6 continents. Among the reported 282 162 outpatient clinic or hospitalized patients with respiratory symptoms, OC43 was estimated as the most prevalent genotype (2.40% [95% confidence interval {CI}, 2.15%–2.66%]; I2 = 95%), followed by NL63 (1.60% [95% CI, 1.39%–1.82%]; I2 = 95%), HKU1 (1.04% [95% CI, .85%–1.25%]; I2 = 94%) and 229E (0.97% [95% CI, .82%–1.15%]; I2 = 95%) (Figure 1A and Supplementary Data). Interestingly, the young children group (aged 0–9 years) has the highest infection rate of 4.71% (95% CI, 3.89%–5.60%; I2 = 94%), compared with 0.94% and 2.35% in 2 older age groups (Figure 1B and Supplementary Data). This appears to be in line with the findings of Cummings et al [5] that younger healthcare personnel, those who saw pediatric patients, and those with household members under the age of 5 are at increased risk of infection. A, The estimated global prevalence of 4 endemic human coronaviruses including OC43, NL63, HKU1, and 229E. B, Forest plot of subgroups. (Note: country classification is based on development levels.) C, Forest plot of overall clinical characteristics in endemic coronavirus–confirmed patients. D, Forest plot of clinical characteristics in patients with monoinfection of an individual genotype of endemic coronavirus. Abbreviations: CI, confidence interval; NA, Not applicable. In total, 18 types of clinical symptoms and features have been identified among endemic coronavirus–confirmed patients by pooling all of the studies irrespective of the genotype or possible coinfection with multiple genotypes. Fever, cough, rhinorrhea, sore throat, and headache commonly occurred in more than half of the confirmed patients. Diarrhea was reported in about 5% of patients. Unexpectedly, >25% patients developed pneumonia, potentially resulting in serious complications, although the severity of pneumonia was not specified in these studies (Figure 1C). We further stratified patients with monoinfection of 229E, HKU1, NL63, or OC43, and characterized their clinical symptoms and features, respectively (Figure 1D). Strikingly, pneumonia was reported in >40% patients with NL63 or OC43 infection, although the included patient number is small (Supplementary Data). In summary, this study comprehensively characterized the clinical features of endemic coronavirus infection. Despite the intrinsic limitations of retrospective meta-analysis with potential selection and publication bias, our findings suggest that endemic coronavirus can impose substantial clinical burden in a subset of patients, which deserves more attention. Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.