NEW YORK — Nucleic acid amplification tests (NAATs) for SARS-CoV-2 are sensitive enough that repeat testing of patients in hospitals is unnecessary, according to a new study appearing in JAMA Network Open this week.
NAATs are considered the gold standard for detecting SARS-CoV-2 but obtaining a positive result on an infected individual is contingent on the presence of a sufficient amount of virus at the anatomical sampling site. Meanwhile, changes in viral shedding have been associated with a decrease in test sensitivity and an increase in false-positives within days of symptom onset.
Aiming to better understand the clinical performance of NAATs over a period of time and in association symptom onset and disease prevalence, a team led by investigators from the University of British Columbia examined data obtained from 47 hospitals across Canada that are participating in the Canadian COVID-19 Rapid Response Emergency Department Network, which comprises a registry that retrospectively enrolled patients presenting to emergency departments (EDs) with suspected COVID-19.
The study included nearly 133,000 patients, evenly divided by sex and with a median age of 57 years, who were tested for SARS-CoV-2 with a NAAT within 24 hours of their arrival at the ED. Most NAATs performed used nasopharyngeal swab samples.
Of the patients examined, about 96,000 reported an onset of symptoms within two weeks. Approximately 17,000 tested positive during their hospital visit, with 538 initially having a negative test result. Among those with the initial negative test result, 434 — nearly 81 percent — tested positive within 14 days and thus were considered to have had at least one false-negative test result.
Overall, the study's authors found that the sensitivity of SARS-CoV-2 NAATs was 96.2 percent among all of the tests performed. The sensitivity of testing by symptom duration ranged from a high of 97.7 percent on the second day of symptoms to a low of 90.4 percent on day 11 of symptoms. The sensitivity of testing among patients who reported COVID-19 symptoms was 97.1 percent compared with 87.6 percent among those without symptoms.
The diagnostic yield of nucleic acid amplification testing was 12.0 percent, the researchers found. The lowest diagnostic yield — 8.1 percent — was observed among patients presenting within the first 24 hours of symptom onset, while the highest diagnostic yield — 20.0 percent —was observed in patients tested on day 10 after symptom onset.
In terms of initial false-negative test result, the team found that they were more likely to occur in older patients, those with chronic kidney disease or psychiatric conditions, and those reporting abdominal pain. Patients experiencing symptoms typically associated with COVID-19, such as cough, fever, and muscle aches, were less likely to have a false-negative result.
"Our results indicate high sensitivity of the test, with little variation during the first 14 days of symptoms," according to the study's authors. However, the diagnostic yield varied by symptom duration, peaking for patients presenting at around day 10 of symptoms. This, they added, suggests that those who present to the ED with prolonged respiratory symptoms are more likely to receive a COVID-19 diagnosis.
The study's findings, the scientists wrote, "indicate high sensitivity for infections prompting emergency department visits and hospitalizations, indicating little value for repeated testing unless the clinical suspicion for disease remains very high."
The researchers noted that their study does have certain limitations, including the lack of data around specimen quality and the inability to control for new COVID-19 exposures over time. Still, "we believe our study provides a robust clinical estimation of NAAT sensitivity and can guide emergency department physicians and admitting teams in their interpretation of diagnostic test results at the time patients present."