Discussion
During the first five months of the COVID-19 epidemic in Vietnam, we characterized all upper respiratory specimens received by NIHE coming from cities and provinces in the north of Vietnam. Just over one percent of all samples received yielded positive results by real-time RT-PCR and yet, by the end of May 2020, fewer than 400 cases had been identified in total in Vietnam, with no deaths and with no community transmission after 14 April 2020.
Rapid scale-up and decentralization of testing was a key component to Vietnam’s response to minimize entry and transmission of SARS-CoV-2. Another component of Vietnam’s response was its implementation of centralized quarantine for all persons entering the country from affected countries since early February 2020 (4). Travelers in quarantine were required to provide upper respiratory specimens for testing upon arrival and before the end of the 14-day quarantine period. Among these travelers in quarantine centers, we identified 89 laboratory-confirmed cases, 78 (88%) of which were positive on their first sampling and 11 during quarantine. Lastly, quarantining and testing of first- and second-degree exposed community members was implemented to stop further transmission of virus from primary cases who had not been detected on entry and subsequent cases. These data suggest that centralized quarantine coupled with testing upon arrival and quarantine based on exposure rather than symptoms is an effective measure for preventing exposure of communities and transmission of SARS-CoV-2 imported from other countries.
Although viral culture remains the gold standard for confirmation of viral infection, because of shorter turn-around time and higher sensitivity, detection of SARS-CoV-2 by real-time RT-PCR is the accepted gold standard for detecting SARS-CoV-2 for purposes of isolation and contact tracing. Semi-quantitation of viral nucleic acids using Ct value can be used to select samples for virus isolation (2,10–12). We observed a strong correlation between Ct values and cell culture positivity rate (table 4). This suggests viral load data may be used as a rough proxy for the infectivity of infected patients.
There is concern about prolonged viral nucleic acid detection in samples from patients recovered from COVID-19. The large majority of these samples, both in the literature and in our collection, have high Ct values and so far culture attempts have not been successful. This observation supports the hypothesis that prolonged shedding or re-positivity of samples is not associated with continued replication, but rather an indicator of removal of damaged lung tissue containing intact stretches of viral RNA by coughing or ciliary transport.
Among these 158 confirmed COVID-19 cases, we also identified seven individuals with positive real-time RT-PCR results after two consecutive negative results within 15 days (Table 3). We were unable to culture virus from these specimens, all of which had Ct values greater than 33, suggesting that these cases represent viral remnants rather than infectious virus. These findings are consistent with findings from South Korea and China (13–15). Positive real-time RT-PCR results can be confusing for patients and hospital staff who understandably wish to prevent continued transmission, either among patients and healthcare workers or among the general community. Our findings should provide reassurance that patients with positive real-time RT-PCR results with Ct values >30 more than 10 days after onset or first positive result and after having had a negative result are at extremely low risk of transmission. These findings also support a strategy of testing based on signs of clinical recovery, rather than a “test-of-cure” strategy.
Our study had several limitations. First, the specimens we received were collected as part of the national strategy for prevention and control of COVID-19 without accompanying systematic clinical metadata and we were thus unable to stratify between asymptomatic, mild and severe cases. Second, we were not able to systematically assess the possible duration of viral shedding because most of our cases were detected upon arrival, through contact tracing, and by the quarantine process. Thus, sampling times were determined by disease control staff in the field, rather than in the context of a rigorously designed study. Third, the source of specimens for viral isolation was upper respiratory tract specimens only. We did not receive any sputum or tracheal aspirate fluids, which may have different characteristics in terms of Ct values or culturable virus.
In summary, we describe here the virologic and epidemiology characteristics of cases of laboratory-confirmed COVID-19 in northern Vietnam from two clusters of cases during the first months of the pandemic. We determined that most cases that will be laboratory-confirmed are confirmed within the first few samplings. We also determined that most cases that are positive very late in their clinical course are extremely unlikely to represent active infection but, rather, remnants of viral RNA.