Since the SARS-CoV-2 outbreak, Emergency Departments (EDs) have been through continuous
reorganization in order to deal with both COVID-19 and ordinary patients. Nevertheless,
previous evidence reports hospitals to be an important source of contagion during
epidemic [1,2], making it essential to control the infection risk in healthcare settings.
For this reason, and given the reported low sensitivity (63%) of a single test ,
several hospitals decided to require two negative nasopharyngeal swabs before admitting
patients to non-COVID wards as has been reported in previous studies regarding MERS-CoV
and SARS-CoV-2 . However, such a strategy involves keeping patients in an isolated
area awaiting admission, thus increasing boarding and overcrowding. The aim of our
study was to analyse the diagnostic yield of a second swab in patients with or without
symptoms of COVID-19.
We performed a cross-sectional study enrolling all adult patients (i.e. ≥18 years)
admitted to the ED of Ospedale Maggiore Policlinico of Milan from March 1 to April
15, 2020 and then hospitalized. Ethical approval for this study was obtained from
Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico ethics committee. As suggested
by international guidelines [5,6], patients with COVID-19-typical symptoms (fever,
dyspnea, cough, sore throat, loss of taste and smell) or close contact with laboratory-confirmed
cases were addressed to the COVID-suspected path. Patients without these features
were addressed to the non-COVID path.
Epidemiological, clinical and laboratory characteristics of the two groups were obtained
from electronic medical records.
Nasopharyngeal swab specimens were collected following a standardized procedure 
and RT-PCR was performed on the specimen to detect SARS-CoV-2 RNA. The specimens were
processed using GeneFinder COVID-19 Plus RealAmp Kit.
In case of a positive test, we admitted patients to a COVID ward; otherwise, we performed
a second swab after 24 h.
We evaluated the diagnostic yield of the second swab as the proportion of patients
with a second positive test on the total number of patients that underwent the second
swab and calculated the number needed to diagnose as the ratio between the number
of the second swabs performed and the positive ones.
We expressed data as proportions and 95% confidence intervals (CIs) and performed
all analyses with Microsoft Excel (Microsoft Corporation, Redmond, WA, USA).
During the study period, 2721 patients presented to the ED and 835 met the inclusion
criteria. Among these, we admitted 630 patients to the COVID-suspected path and 205
to the non-COVID path.
In the non-COVID population, we performed a swab in 122 (60%) patients and six (5%;
95% CI: 1%, 9%) patients were positive. Among the 116 negative patients, 61 underwent
a second swab within 24 h and 60 patients (98%; 95% CI: 95%, 100%) were negative,
while one (2%; 95% CI: 0%, 5%) was positive, so that we had to perform 50 tests to
detect one positive patient.
In the COVID-suspected population, we tested 579 patients (92%) for SARS-CoV-2 infection:
410 (71%; 95% CI: 67%, 74%) were positive, while 169 (29%; 95% CI: 26%, 33%) were
negative. We performed a second swab in 24 h on 89 negative patients among whom 79
patients (89%; 95% CI: 82%, 95%) tested negative and 10 (11%; 95% CI: 5%, 18%) tested
positive. Therefore, the number of tests needed to detect one positive patient was
nine. All these data are graphically reported in Figure 1
Patients selected for the study and the results of nasopharyngeal swabs.
Our results show that when using swabs in series in patients with low probability
of SARS-CoV-2 infection, the second swab has a low diagnostic yield (2%), while when
using the second nasal swab as a diagnostic test in patients with a high probability
of infection, the diagnostic yield is higher (11%) and may justify performing a second
swab in clinical practice.
Moreover, considering the overall low sensitivity of the nasopharyngeal swab ,
performing a second swab test does not completely remove the risk of hospitalizing
infected patients in non-COVID wards while it increases boarding and overcrowding.
In addition, testing patients with two swabs results is an increased financial burden
and could lead to difficulties in providing swabs in a context of limited supply of
Therefore, we believe that healthcare settings can not rely on the swab test to rule
out SARS-CoV-2 infection. Instead, given the high risk of in-hospital contagion, risk
control strategies based on personal protective equipment use and preventive isolation
for every patient – even when tested negative – is strongly recommended.
All authors contributed to the design of the study, data collection, analysis and
interpretation, revised the article critically and approved the final version.
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Declaration of Competing Interest