RT-PCR yields high false-negative rates in early SARS-CoV-2 infection
Reverse transcriptase polymerase chain reaction, or RT-PCR, yielded a false-negative rate as low as 20% and as high as 100% among patients who were tested for infection with SARS-CoV-2, according to a literature review and pooled analysis published in Annals of Internal Medicine.
“The most important thing to take away from our study is that if clinicians have a very high suspicion that someone is infected — they had a very high risk exposure, or if they have symptoms that are very consistent with COVID-19 — then it's important to treat them as positive even if you have a negative test result,” Lauren Kucirka, MD, PhD, study co-author and gynecology and obstetrics resident physician at the Johns Hopkins Bloomberg School of Medicine, told Healio Primary Care.
The researchers analyzed seven previously published studies that assessed RT-PCR performance by time since SARS-CoV-2 exposure or symptom onset via upper respiratory tract samples. Only 12 patients in the studies were from the United States, according to Kucirka.
Kucirka and colleagues found that over the 4 days of infection before the typical time of symptom onset, the probability of a false-negative result in an infected person decreased from 100% (95% CI, 100%-100%) on day 1 to 67% (95% CI, 27%-94%) on day 4. The median false-negative rate further decreased to 38% (95% CI, 18%-65%) on day 5 and 20% (95% CI, 12%-30%) on day 8 (3 days after symptom onset). The rate then increased from 21% (95% CI, 13%-31%) on day 9 to 66% (95% CI, 54% -77%) on day 21. The high false-negative rates were likely due to the variability in the individual amount of viral shedding and sample collection techniques, according to researchers.
Despite the findings, Kucirka encouraged clinicians to still use the RT-PCR.
“The RT-PCR is incredibly important,” she said. “Testing and tracing strategies are key to efforts to controlling COVID-19. Physicians need to be aware of the RT-PCR’s limitations and take that into account when designing and implementing testing policies.”
Kucirka advised physicians to consider waiting 1 to 3 days after symptom onset to “minimize the probability of a false-negative result.” She also recommended that physicians make a COVID-19 diagnosis based on a combination of the patient’s test result, epidemiologic situation and “constellation of clinical symptoms — such as ground glass opacities appearing on a patient’s CT scan and oxygen saturation levels plummeting — that physicians have reported seeing in patients with COVID-19.”