ABSTRACT
BACKGROUND: Nasopharyngeal (NP) specimen testing by reverse transcriptase polymerase chain reaction (RT-PCR) is the standard of care for detecting SARS-CoV-2. Data comparing the sensitivity and specificity of the NP specimen to the less invasive, mid-turbinate (MT) nasal specimen in children are limited. METHODS: Paired clinical NP and research MT specimens were collected from children <18 years with respiratory symptoms and tested by molecular assays to detect SARS-CoV-2 RNA. Sensitivity, specificity, and agreement (Cohen's kappa [κ]) were calculated for research MT specimens compared to the clinical NP specimens. RESULTS: Out of 907 children, 569 (62.7%) had parental consent and child assent when appropriate to participate and provided paired MT and NP specimens a median of 4 days after symptom onset (range 1-14 days). 16.5% (n = 94) of MT specimens were positive for SARS-CoV-2 compared with 20.0% (n = 114) of NP specimens. The sensitivity of research MT compared to clinical NP specimens was 82.5% (95% CI: 74.2%, 88.9%), specificity was 100.0% (95% CI: 99.2%, 100.0%), and overall agreement was 96.1% (κ = 0.87). The sensitivity of MT specimens decreased with time from 100% (95% CI: 59.0%, 100.0%) on day 1 of illness to 82.1% (95% CI: 73.8%, 88.7%) within 14 days of illness onset; sensitivity was generally >90% when specimens were collected within the first week of illness. CONCLUSION: MT specimens, particularly those collected within the first week of illness, have moderately reduced sensitivity and equivalent specificity to less-tolerated NP specimens in pediatric outpatients. MT specimen use in children may represent a viable alternative to NP specimen collection.