ABSTRACT
BACKGROUND Asymptomatic SARS-CoV-2 infections are potential sources for transmission.
OBJECTIVES Here, we aimed to profile pre-symptomatic/asymptomatic SARS-CoV-2 infections in Mexico City, its associated comorbidities and outcomes.
METHODS Prospective observational study of SARS-CoV-2 cases assessed by contact tracing within the National Epidemiological Surveillance System in Mexico City. We classified subjects as cases with and without respiratory and non-respiratory symptoms (RS, NRS, respectively) assessed for SARS-CoV-2 using real time RT-PCR from nasopharyngeal swabs to evaluate clinical characteristics as well as features associated with severe COVID-19.
RESULTS Amongst 80735 cases with confirmed SARS-CoV-2 infection as of July 23th, 2020, we identified 9659 cases without RS (12.0%), 5605 had only NRS (6.9%) and 4054 who were initially asymptomatic (5.1%). Compared to asymptomatic cases, NRS cases and those with RS+NRS had higher rates of severe COVID-19 outcomes including hospitalization, ICU admission, and intubation (p<0.001). Mortality was higher for cases with RS+NRS (HR 7.58, 95%CI 4.56-12.58) or NRS (HR 3.13, 95%CI 1.80-5.45) compared to asymptomatic cases, adjusted for age, sex and comorbidities. Predictors for mortality in asymptomatic SARS-Co-V-2 infections, chronic kidney disease, previous exposure with suspected infection cases and older age were predictors for lethality.
CONCLUSIONS Definition of pre-symptomatic/asymptomatic cases has relevant implications for SARS-CoV-2 infection outcomes. Older age and comorbidity impact on the probability of developing severe complications for cases who were asymptomatic or had only NRS at evaluation and were sent for domiciliary treatment. Extending testing for detection of asymptomatic cases must be considered in Mexico to better understand the impact of the pandemic.
INTRODUCTION
Asymptomatic infections are a potential source of rapid SARS-CoV-2 spread [1]. Prevalence of pre-symptomatic/asymptomatic cases varies widely but represent a larger proportion of cases than previously anticipated [2]. Controversies surrounding asymptomatic SARS-CoV-2 infections have focused on the distinction between pre-symptomatic and asymptomatic infection and its definitions, which have focused on respiratory symptoms (RS), primarily cough, dyspnea and fever. Nevertheless, longitudinal data shows that only a fraction of initially asymptomatic infections develop subsequent symptoms [3–5]. In Mexico, SARS-CoV-2 testing focuses on sampling high-risk cases, which allowed characterization of increased susceptibility for severe COVID-19 attributable to cardio-metabolic comorbidities [6]; however, no data has been reported on asymptomatic SARS-CoV-2 infections. Lack of information regarding pre-symptomatic/asymptomatic cases in Mexico is concerning, particularly since many studies suggest that contact tracing of asymptomatic infections will be pivotal to contain further SARS-CoV-2 spread [7]. Here, we profile pre-symptomatic/asymptomatic cases of SARS-CoV-2 infections to assess outcomes depending on its definitions using cases detected in Mexico City.
METHODS
We analyzed data from the National Epidemiological Surveillance System in the Greater Mexico City area, an open-source dataset comprising daily updated cases tested using real-time RT-PCR to confirm SARS-CoV-2 using nasopharyngeal swabs according to the Berlin Protocol [8]. RS included fever, dyspnea, cough, polypnea or cyanosis at initial evaluation, whilst non-RS (NRS) included headache, myalgias, arthralgias, general malaise, abdominal pain, chest pain, conjunctivitis, irritability or vomiting. We defined cases with RS+NRS with ≥1 RS and any or none NRS. Cases with NRS had ≥1 NRS without RS, whilst asymptomatic cases had no symptoms. Pre-symptomatic cases were initially asymptomatic but eventually required hospital admission after follow-up. To contrast previous definitions of pre-symptomatic/asymptomatic cases based only on RS, we combined asymptomatic and NRS cases and referred to them as cases without RS (WoRS); we confirmed these definitions using correspondence analyses to cluster symptoms with comparable similarity (Supplementary Material). NRS and asymptomatic cases were assessed outside the established framework for suspected COVID-19 cases in Mexico by convenience sampling and contact tracing. We also estimated 7-day rolling average positivity rates (PR) for RS+NRS, NRS and asymptomatic/pre-symptomatic cases. Follow-up was conducted for all non-hospitalized cases up to 14 days; outcomes for all patients included hospital discharge due to clinical improvement or voluntary, ambulatory/treatment follow-up or death. Severe COVID-19 was defined as a composite of death, ICU admission and intubation. Complete description of statistical analyses and available variables are detailed in Supplementary Material.
RESULTS
Study population
As of 23rd July 2020, a total of 212,708 subjects were tested for SARS-CoV-2 in Mexico City. From 80,735 cases with confirmed SARS-CoV-2 infection (Positivity rate 49.0%), 9,659 cases were WoRS (12.0%). Overall, 23,992 NRS cases had been evaluated, amongst whom 5,605 had confirmed SARS-CoV-2 infection (Overall Positivity rate: 26.5%); similarly, 21,628 initially asymptomatic cases were evaluated and 4,054 cases had confirmed SARS-CoV-2 infection (Positivity rate: 21.4%). Notably, 7-day PRs had been steadily decreasing for all patient categories (Supplementary Material). NRS cases were younger, predominantly female and, interestingly, had lower prevalence of NRS compared to cases with RS+NRS. Initially asymptomatic cases were younger compared to RS+NRS cases, with similar trends for lower rates of comorbidities compared to both RS+NRS and NRS cases (Figure 1, Table 1).
Descriptive characteristics of symptomatic and asymptomatic SARS-CoV-2 cases in Mexico City. Abbreviations: COPD: Chronic obstructive pulmonary disease; HIV/AIDS: Human immunodeficiency virus and/or acquired immunodeficiency syndrome; CKD: Chronic kidney disease, CDV: Cardiovascular disease; HCW: Healthcare worker
Histogram comparing case distribution according to age groups comparing cases with RS+NRS, NRS and asymptomatic SARS-CoV-2 infection, demonstrating the role of age in determining asymptomatic SARS-CoV-2 infections (A). We also show a comparison of the distribution of NRS related to SARS-CoV-2 infections in subjects with RS+NRS and NRS only. Abbreviations: RS: Respiratory symptoms; NRS: Non-respiratory Symptoms; Sudden onset: Patient report of sudden onset of symptomatology.
Profiling NRS cases with SARS-CoV-2 infection
Amongst NRS cases (n=5,605), 283 subjects were hospitalized (5.0%), 26 required mechanical ventilatory support (MVS) (0.5%), 27 required intensive care unit (ICU) admission (0.5%). Overall, 96 patients died (1.7%), of whom 22 were ambulatory and 74 hospitalized. Predictors increased risk for severe COVID-19 in NRS cases included vomiting, diabetes, and hypertension; decreased risk of severe COVID-19 was associated with rhinorrhea, previous exposure to suspected viral cases and healthcare workers. Independent predictors of mortality in NRS included older age, diabetes, chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD) and decreased risk associated with previous exposure to suspected viral cases (Supplementary Material). After follow-up, 481 patients were still under domiciliary surveillance, 109 remained hospitalized, 97 were discharged after recovery, 3 had voluntary discharge and 4,819 had completed follow-up with domiciliary surveillance without developing RS (86.0%).
Profiling pre-symptomatic/asymptomatic cases with SARS-CoV-2 infection
Amongst initially asymptomatic patients (n=4,054), 139 were hospitalized (3.4%), 9 required MVS (0.2%) and 29 required ICU admission (0.7%). We recorded 24 deaths (0.6%) amongst whom 8 were ambulatory and 16 were hospitalized (Table 1). Predictors for severe COVID-19 included younger age, cardiovascular disease, CKD, diabetes and a reduced risk associated to known previous exposure to suspected viral cases; predictors of lethality included older age and CKD. Currently, 444 patients were still under domiciliary surveillance, 84 remained hospitalized, 38 were discharged after recovery, 1 had a voluntary discharge, and 3,463 had completed follow-up without developing symptoms (85.4%).
Predictors and outcomes for RS+NRS, NRS and asymptomatic cases
Compared to asymptomatic, NRS cases were associated with younger age, reduced previous viral exposure, healthcare workers, diabetes, obesity, female sex, hypertension and smoking. Similarly, compared to WoRS, RS+NRS cases were associated with asthma, COPD, older age, male sex, diabetes, obesity, hypertension, reduced previous viral exposure (Figure 2). Compared to pre-symptomatic/asymptomatic cases, risk of severe COVID-19 was not significantly higher for NRS, but it was for RS+NRS (HR 3.91, 95%CI 2.95-5.18), adjusted for age, sex and comorbidities. However, NRS and RS+NRS had higher mortality compared to asymptomatic cases (Breslow-Cox p<0.001, Figure 2); adjusting for age, sex and comorbidities, NRS (HR 2.32, 95%CI 1.48-3.63) or RS+NRS cases (HR 6.22, 95%CI 4.17-9.29) retained higher mortality compared to asymptomatic cases.
Factors associated with presence of RS+NRS compared to WoRS (NRS + Asymptomatic), and NRS compared to asymptomatic cases (NRS) with SARS-CoV-2 infection (A). We also show the and comparison of 30-day mortality rates between RS+NRS cases, NRS and asymptomatic SARS-CoV-2 infections in Mexico City. Abbreviations: RS: Respiratory symptoms; NRS: Non-respiratory Symptoms; COPD: Chronic obstructive pulmonary disease; CKD: Chronic kidney disease; HCW: Health-care workers; Previous viral exposure: Prior contact with a patient with suspected viral infection.
DISCUSSION
Here, we present a comprehensive report of pre-symptomatic/asymptomatic COVID-19 cases in Mexico City. We show that the definition of asymptomatic SARS-CoV-2 infection should be extended beyond RS to identify subjects at lowest risk of subsequent complications for domiciliary treatment who might contribute to further SARS-CoV-2 spread. We also report a low prevalence of pre-symptomatic cases amongst those who were initially classified as asymptomatic but eventually were hospitalized or had outcomes related to severe COVID-19 [9]. This data confirms observations from other populations and highlights the relevance of identifying pre-symptomatic/asymptomatic cases, given conflicting evidence on viral shedding for pre-symptomatic/asymptomatic cases, and data suggesting evidence of subclinical lung abnormalities which must be assessed longitudinally to rule out long-term impacts of SARS-CoV-2 infection [10]. Notably, a consistent predictor of outcomes for NRS cases and pre-symptomatic/asymptomatic cases are previous exposure to suspected viral cases, highlighting the relevance of contact tracing and prompt case identification.
Studies assessing natural history of asymptomatic SARS-CoV-2 infections highlighted the role of age and comorbidity in the subsequent development of symptoms, which is consistent with our observations [11]. Domiciliary surveillance in Mexico City focused on outcomes but not individual symptom onset, which might limit our ability to identify all pre-symptomatic cases, particularly those who only develop mild symptoms. Regardless, only a small fraction of pre-symptomatic cases developed severe COVID-19 after being initially determined as asymptomatic, most of whom had underlying cardio-metabolic comorbidities or were susceptible for complications of COVID-19 attributable to unmasked cardiometabolic comorbidities. Asymptomatic SARS-CoV-2 infections have been postulated to elicit lower immune responses and diminished antibody production compared to symptomatic cases. Severe COVID-19 in pre-symptomatic cases might be attributable to a combination of decreased early immune response to SARS-CoV-2 infection and late enhanced pro-inflammatory responses in subjects with comorbidities [10,12]. This might also extend to subjects with NRS with COVID-19 whom have previously been reported to delay seeking medical attention and have prolonged viral shedding compared with subjects with RS [13]. Here, we assessed a thorough constellation of symptoms in one of the largest studies of pre-symptomatic/asymptomatic SARS-CoV-2 infections. Limitations include non-assessment of atypical SARS-CoV-2 symptoms which might occur amongst otherwise asymptomatic cases, particularly in older adults [14], and the lack of systematized sampling, which does not allow for an accurate estimation for the rates of asymptomatic cases in Mexico City. Our results highlight the need to systematize definition of asymptomatic cases and extend testing by contact tracing to detect asymptomatic SARS-CoV-2 infections as one of the mitigations strategies to reduce transmission of SARS-CoV-2 in Mexico City and Mexico in general [15].
Data Availability
All data sources and R code are available for reproducibility of results at https://github.com/oyaxbell/covid_asymptomatic_cdmx.
DATA AVAILABILITY
All data sources and R code are available for reproducibility of results at https://github.com/oyaxbell/covid_asymptomatic_cdmx.
AUTHOR CONTRIBUTIONS
Research idea and study design OYBC, NEAV, AVV, JPBL, CAFM, AMS; data acquisition: OYBC; data analysis/interpretation: OYBC, JPBL, NEAV, AVV, CAFM, AMS; statistical analysis: OYBC; manuscript drafting: OYBC, NEAV, AVV, JPBL, CAFM, AMS; supervision or mentorship: OYBC. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions about the accuracy or integrity of any portion of the work are appropriately investigated and resolved.
FUNDING
No funding was received.
CONFLICT OF INTEREST/FINANCIAL DISCLOSURE
Nothing to disclose.
ACKNOWLEDGMENTS
NEAV, JPBL, and AVV are enrolled at the PECEM program of the Faculty of Medicine at UNAM. JPBL and AVV are supported by CONACyT. The authors would like to acknowledge the invaluable work of all of Mexico’s healthcare community in managing the COVID-19 epidemic. Its participation in the COVID-19 surveillance program has made this work a reality, we are thankful for your effort.
Footnotes
CONFLICT OF INTERESTS: Nothing to disclose.
FUNDING: This research received no specific funding
REFERENCES
