Evaluation of the negative predictive value of the SARS-CoV-2 PCR respiratory assays in asymptomatic children undergoing surgery

ABSTRACT

Background: Universal pre-operative screening with SARS-CoV-2 PCR has been adopted by institutions to mitigate risk to healthcare workers (HCW) during aerosol- generating procedures such as intubation. However, there remains uncertainty regarding rates of false negative results and optimal sampling type. The objective was to determine the reliability of single, pre-operative SARS-CoV-2 testing from the nasopharynx in children undergoing general anesthesia. Methods: Children < 18 years of age who underwent intubation for a procedure received pre-operative testing 24-48 hours prior with a nasopharyngeal (NP) swab or wash, in conjunction with intra-operative nasal wash (NW) and tracheal aspirate (TA) sampling. All paired samples underwent testing using the Simplexa DiaSorin platform or a modified Centers for Disease Control assay. Cohen's Kappa was used for interrater reliability of each sample result. McNemar's Test was used to compare result proportions by sample type. Positive and negative predictive values (PPV, NPV) were calculated based on the intraoperative NW as the reference standard. Analyses were conducted using SAS (v 9.4). Results: We collected full sample sets from 364 children from April 14 to May 15;66% of pre-operative samples were NP swabs. The median age was 6 years (IQR 2,13), 55% were male, 68% were white and 41% of children had a high-risk comorbidity. Most surgeries were conducted by general surgery (23%), followed by orthopedics (19%). Only 2.5% of children had respiratory symptoms, and 4.8% had a documented fever within a week of the procedure. SARS-CoV-2 positive samples occurred in 4/364 (1%) of pre-operative samples, 8/363 (2.2%) of intra-operative samples, and 8/348 (2.3%) of TA samples. The pre-operative test had 100% PPV and 99% NPV, and the TA had 100% PPV and 98.6% NPV (Table 1). There was very good agreement (Figure) between pre- and intraoperative upper respiratory sampling, with a Kappa of 0.66, (95% CI 0.35-0.97). There was no statistical difference in results by sample type. (Table Presented). Conclusion: There is a high PPV and NPV of pre-operative SARS-CoV-2 PCR testing among children undergoing anesthesia. These data can help inform guidelines regarding appropriate precautions for HCW performing high risk procedures in asymptomatic pediatric patients. (Table Presented).