key: cord-0709050-kswmca36
authors: Kobayashi, Takaaki; Trannel, Alexandra; Holley, Stephanie A; Alsuhaibani, Mohammed; Abosi, Oluchi J; Jenn, Kyle E; Meacham, Holly; Sheeler, Lorinda L; Etienne, William; Dains, Angelique; Casado, Fernando; Kukla, Mary E; Ward, Emily; Ford, Bradley; Edmond, Michael B; Wellington, Melanie; Diekema, Daniel J; Salinas, Jorge L
title: COVID-19 Serial Testing among Hospitalized Patients in a Midwest Tertiary Medical Center, July–September 2020
date: 2020-10-26
journal: Clin Infect Dis
DOI: 10.1093/cid/ciaa1630
sha: 72a7c7bdab8f408a808be51c7bd80cbb29a25ab4
doc_id: 709050
cord_uid: kswmca36

We implemented serial COVID-19 testing for inpatients with a negative test on admission. The conversion rate (negative to positive) on repeat testing was one percent. We identified patients during their incubation period and hospital-onset cases, rapidly isolated them, and potentially reduced exposures. Serial testing and infectiousness determination were resource intensive.

A c c e p t e d M a n u s c r i p t 3

Coronavirus disease 2019 (COVID-19) is a multisystemic illness caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS CoV-2). Due to adverse patient outcomes and high costs associated with unrecognized COVID-19 transmission in hospitals, many centers have implemented admission COVID-19 testing protocols. However, COVID-19 testing on admission may miss cases if the patient is in the incubation period at the time of admission. Patients may also acquire COVID-19 during their hospitalization. Serial testing of hospitalized patients may be a plausible strategy to detect COVID-19 cases missed during admission screening. However, SARS CoV-2 testing can remain positive for months [1] [2] , and a positive test in an asymptomatic patient may represent remnant viral RNA from a past infection. Differentiating past from current infection is important because isolation and contact investigations may not be necessary for those with past infection.

Though the utility of COVID-19 admission testing has been assessed [3, 4] , limited data exists on the impact of serial testing for inpatients. We assessed the value of COVID-19 serial testing for hospitalized patients after a negative COVID-19 admission test.

The University of Iowa Hospitals & Clinics (UIHC) is an 811-bed academic medical center. In addition to admission COVID-19 testing, in July 2020 we implemented COVID-19 serial testing for inpatients every 5 days regardless of symptoms. Patients having surgical procedures also underwent pre-procedural testing if a COVID-19 test had not been obtained in the previous 48 hours. Patients with COVID-19 signs or symptoms were tested at the discretion of their treating provider. All admitted adults and children from July 7 to September 22, 2020 were included in this study. Testing was performed using the TaqPath In this paper we 1) assess the frequency of conversion from a negative admission COVID-19 test to a subsequent positive in repeat testing: serial, pre-procedural, or symptomatic testing during the same admission; 2) describe the clinical characteristics of patients found to be infected; 3) quantify exposure events; and 4) identify COVID-19 among contacts of infected persons. We obtained data from the electronic health record including age, sex, admission diagnosis, time from admission to positive repeat testing, symptoms, Mean RT-PCR cycle threshold (CT) values, SARS-CoV2 serum antibodies (IgG), and infectiousness as determined by the Program of Hospital Epidemiology. We used the Roche Diagnostics assay for total antibodies to SARS-CoV-2. All specimens positive by the Roche assay are tested by a separate DiaSorin SARS-CoV-2 IgG assay. An nasopharyngeal nucleic acid amplification test (NAAT) for COVID-19 was used for admission and repeat testing [5] . Tests performed 2 days before to one day after admission were considered the admission test. Information on exposure events for health care personnel (HCP), patients, and their follow up COVID-19 test results were obtained. were <30, the case was labeled as an acute infection; if repeat CTs were ≥30 or negative, the case was labeled as a past infection. Active infections remained on isolation precautions and contact tracing was performed. Cases who converted <14 days after admission were considered to have possibly been in the incubation period. Patients who converted ≥14 days after admission were considered hospital-associated. All HCP wore medical grade face masks and face shields for all patient care. HCPs wore N95 respirators and face shields during all aerosol generating procedures (AGPs) performed on patients known or suspected to have COVID-19. All inpatients were recommended to wear a mask in our hospital but adherence to recommendations was not assessed. Only one visitor was allowed per patient A c c e p t e d M a n u s c r i p t 5 for up to two hours per day. Face masks were required. If patients were identified as active infection, no visitors were allowed by our hospital policy. This study was approved by the Institutional Review Board of the University of Iowa.

We tested 4,580 patients on admission. A total of 4,438 (96.9%) had negative results. Of those, 1,950 (42.6%) had at least one repeat test during their admission (Appendix 1). Overall compliance with serial testing was 96%. We identified 19 patients (1.0%) who converted from negative to positive during their admission. 

We implemented a serial testing strategy for inpatients with a negative COVID-19 admission screening test. The conversion rate from negative to positive was one percent.

Nine patients (47%) were infectious. The serial testing strategy helped us identify seven infectious COVID-19 patients, most of them asymptomatic. We detected COVID-19 cases sooner and potentially prevented further in-hospital exposure events. However, serial testing and infectiousness determination were time and resource intensive.

Implementing serial testing of inpatients for COVID-19 was feasible. While the conversion rate from negative to positive was relatively low (1.0%), this strategy identified asymptomatic patients who developed COVID-19 during their hospitalization and potentially prevented exposure events. Some studies have assessed the impact of repeating tests in symptomatic patients or those undergoing a surgical procedure. They found that 1-3% were positive on retesting after an initial negative result. [6] [7] [8] . However, their approach was different from ours because they repeated tests in patients with a high suspicion of COVID-19 or in whom they suspected a false negative result. Studies focusing only on admission screening or pre-procedural testing could have missed patients in their incubation period or hospital-onset COVID-19 patients.

This strategy helped us identify patients that became infectious during their hospital stay. However, nearly 50% were likely past infections and therefore not infectious. These findings highlight the limitations of using SARS-CoV2 NAAT in asymptomatic individuals for screening purposes. Patients may continue having a positive NAAT even months after an acute infection. Some authors postulate that different CT limits may be needed or alternative A c c e p t e d M a n u s c r i p t 7 testing methods be used for public health screening efforts [2] [9] . We were able to assess case infectiousness by serial testing and discontinue isolation precautions in those with prior infections. This approach helped better utilize scarce resources (e.g., private rooms, personal protective equipment) and facilitated medical care for patients (e.g., allowing visitors, avoidance of delays for certain procedures).

COVID-19 infection can be asymptomatic in 30-40% of patients. [10, 11] Because of the long incubation period (up to 14 days) and the possibility of hospital onset, a negative admission test may not guarantee absence of risk during the hospitalization. Furthermore, detecting asymptomatic or presymptomatic cases early can avoid outbreaks in healthcare facilities. Patients who have recently converted are more likely to have higher viral loads and may be more infectious. Symptomatic patients may have other diagnoses (e.g., chronic cardiac or pulmonary disease) that may make it difficult for providers to suspect COVID-19.

Because serial testing is costly and time intensive, implementing it for all inpatients may not be cost-effective in facilities with a low incidence of COVID-19. Some institutions may want to consider it, especially if they have semi-private rooms or lack respirator availability for all aerosol generating procedures.

This study has limitations. It was conducted at a single center with a relative low COVID-19 incidence and the results may not be generalizable. Interpretation of infectiousness using CT values is not yet standardized. CT values vary widely between assays and gene targets and may not translate numerically to CT values obtained from other testing. We could not confirm if exposed persons who subsequently tested positive acquired COVID-19 in the hospital or in the community. However, we present one of the first experiences of COVID-19 serial testing and a framework for infectiousness interpretation using CT values and serologic status. In conclusion, we demonstrated that a serial testing strategy for inpatients could help detect COVID-19 cases. These cases could have been in the incubation period on admission, or healthcare-associated infections. CT value kinetics enabled us to assess case infectiousness and discontinue isolation precautions in those unlikely to be infectious.

Because serial testing and infectiousness determination were time and resource intensive, screening strategies should balance diagnostic and resource stewardship with patient and health care professionals safety. M a n u s c r i p t 

Temporal dynamics in viral shedding and transmissibility of COVID-19

Duration of SARS-CoV-2 Infectivity: When is it Safe to Discontinue Isolation? Clinical Infectious Diseases

Universal screening for the SARS-CoV-2 virus on hospital admission in an area with low COVID-19 prevalence

RE: Universal SARS-CoV-2 testing on admission to the labor and delivery unit: Low prevalence among asymptomatic obstetric patients

CDC's Diagnostic Test for COVID-19 Only and Supplies

Utility of Repeat Testing for COVID-19: Lab Stewardship When the Stakes are High

Retesting for severe acute respiratory coronavirus virus 2 (SARS-CoV-2): Patterns of testing from a large US healthcare system

Utility of retesting for diagnosis of SARS-CoV-2/COVID-19 in hospitalized patients: Impact of the interval between tests

Rethinking Covid-19 Test Sensitivity -A Strategy for Containment

Suppression of a SARS-CoV-2 outbreak in the Italian municipality of Vo'

Prevalence of Asymptomatic SARS-CoV-2 Infection : A Narrative Review

M a n u s c r i p t