Universal PCR screening for coronavirus disease 2019 in asymptomatic patients on admission.

To the Editor,

Coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is highly infectious during the pre-symptomatic period in patients. The nosocomial transmission of COVID-19 to health-care workers and other patients can have serious impacts on hospital operation, including the suspension of new admissions and the closing of hospital wards. Patients should be screened for likely COVID-19 symptoms, such as fever, cough, sore throat, dyspnoea, or the loss of smell or taste, before hospital entry. So far, universal PCR screening has been adopted for specific situations, such as for obstetric patients and health-care workers. Although performing specific PCR screening for patients with COVID-19 symptoms is essential, the Centers for Disease Control and Prevention in the USA has recommended the administration of universal PCR screening for all pre-admission patients, depending on testing capacity and disease prevalence [1]. However, this testing strategy may return negative results for patients who are tested during the incubation period, resulting in false-negative results for patients who later become infectious [1]. To address this possibility, the Public Health England guidelines recommend screening all patients on admission and re-screening all inpatients again after 5–7 days [2]. Whether PCR testing should be performed for all hospital admission patients in Japan, even those who are asymptomatic, remains unclear, and available data regarding the effectiveness of this strategy have been limited.

At Tokyo Medical University Hospital (TMUH, Shinjuku, Tokyo), universal PCR screening was implemented for all patients on admission, even those without symptoms, to identify all individuals with COVID-19 at the time of admission. The initial expected sensitivity and specificity of the PCR test (cobas, Roche Molecular Systems, Pleasanton, CA, USA) used at TMUH were 95.2% and 99.3%, respectively [3]. Here, we summarize the results of performing universal PCR testing on nasopharyngeal specimens obtained from all inpatient admissions for 16 weeks, from 11 May 2020 to 30 September 2020. Universal PCR testing was performed for 6224 patients on admission (Table 1 ). No individuals had positive results from May 2020 to July 2020, and positive COVID-19 results were limited to two individuals (2/6224, 0.03%), who had scheduled elective operations during August 2020 and September 2020. The patients with positive COVID-19 results were asymptomatic at the time of PCR testing and did not develop any symptoms before discharge. Both patients reported no overseas travel history for the previous month. No other patients became symptomatic and contagious while hospitalized during the study period.

Table 1

The number of COVID-19 PCR tests performed for inpatient admissions each month, from May 2020 to August 2020

	No. of COVID-19 PCR tests	No. of positive results	Positive rate	Total no. of COVID-19 patients in Tokyo Metropolitan areaa
11–31 May	558	0	0%	244
1–30 June	1222	0	0%	944
1–31 July	1494	0	0%	6466
1–31 August	1496	1	0.07%	8126
1–30 September	1454	1	0.07%	4921
Total period	6224	2	0.03%	20 738

Number based on officially released data from the Tokyo Metropolitan Government.

The results of this study indicated a low positive COVID-19 rate among asymptomatic patients who come to the hospital for scheduled elective procedures, similar to the results of the national COVID-19 antibody survey. We recently faced a second COVID-19 wave during this period, resulting in a total number of 20 738 cases in the Tokyo Metropolitan area, which has a population of 13 999 568 people. The cumulative incidence rates per 100 000 population during the period from 1 May 2020 to 30 September 2020 were 153.0 (cumulative incidence per month; minimum 6.8, maximum 58.0) for Tokyo and 54.6 (cumulative incidence per month; minimum 1.4, maximum 25.4) for all of Japan, which indicates low levels of community transmission. The positive rate for COVID-19 antibody testing in June 2020 in Tokyo was officially reported at 0.10% [4]. We must, therefore, consider whether the effectiveness of the universal PCR mitigation strategies that were implemented at the onset of the pandemic contributed to the prevention of nosocomial infections in the low levels of community transmission observed in Japan. Our results during this study period indicated that universal PCR testing might be overly labour-intensive and not cost-effective, even though the universal screening protocol detecting two positive cases, who remained asymptomatic after the positive diagnosis. Whether these asymptomatic individuals posed an infectious threat to health-care workers or other patients in this study also remains unclear, although a previous report has indicated that asymptomatic individuals might contribute to further spread. Four individual studies have found that the rate of asymptomatic infection ranged from 0% to 2.2%, so the WHO has suggested that the proportion of truly asymptomatic individuals who transmit the virus to others remains unknown [5].

Distinguishing whether patients are truly asymptomatic or are only pre-symptomatic and will eventually develop symptoms is extremely challenging. Appropriate hand hygiene procedures, universal masking, and environmental surface cleaning should be more strongly emphasized as realistic strategies to avoid infection. COVID-19 PCR testing is thought to be ineffective during the virus incubation period, so PCR testing before admission alone is not sufficient to prevent nosocomial transmission risks.

Our study period coincided with a period of low overall community transmission in Japan; for this reason, additional studies remain necessary to determine whether the strategy of universal testing on hospital admission may be more useful in other countries, both in Asia and globally, that are characterized by higher levels of community transmission.

Transparency declaration

The authors declare that they have no competing interests. The study received no external funding.