Background/Purpose: Multisystem Inflammatory Syndrome in Children (MIS-C), also known as Pediatric Inflammatory Multisystem Syndrome temporally associated with COVID-19 (PIMS), is one of the most serious complications associated with COVID-19.The clinical features of MIS-C are not unique and are commonly seen in childhood febrile conditions. Thus, there is a need to identify those febrile children with MIS-C to enable early diagnosis and treatment. In response to the health care emergency, a multidisciplinary team from The Hospital for Sick Children (SickKids) of Toronto developed a preliminary screening pathway for the evaluation of possible MIS-C. The primary objective is to determine if the SickKids screening pathway is sensitive and specific to identify patients with MIS-C from other febrile children with suspected but not confirmed diagnosis of MIS-C. The secondary objective of this study is to determine how the ACR MIS-C algorithm performs compared to the SickKids screening pathway in differentiating children with MIS-C from the febrile controls.

Methods: Retrospective case-controlled, cross-sectional study including children who have been assessed at SickKids with suspected or confirmed MIS-C from March 2020 to March 2022. The MIS-C group included all children meeting the most permissive case definition as per international MIS-C and PIMS definitionsand adjudicated by our multi-disciplinary MIS-C working group; the febrile control group consisted of all patients with a history of three or more days of fever who were suspected of MIS-C and qualified for the SickKids MIS-C screening pathway, but did not fulfill criteria for MIS-C after adjudication by our multi-disciplinary group. SickKids and ACR pathways were retrospectively applied to both groups.The diagnosis result obtained using the pathways was compared with the final clinical diagnosis made using the WHO definition criteria, used as the gold standard. From the contingency table, sensitivity and specificity have been calculated along with a 95% confidence interval.

Results: 402 children (241 with MIS-C and 161 febrile controls) were included in the study. The median age was 4.18 years (IQR: 1.9 to 9.0) and 237 (60%) were male. For the SickKids screening pathway, the sensitivity was 62% (95%CI, 54.2% to 70.4%), and specificity was 91% (95%CI, 86.9% to 94.2%). The positive predictive value (PPV) was 79% and the negative predictive value (NPV) was 81%. Overall, the balanced accuracy was equal to 76%. The ACR screening pathway had 74% sensitivity (95%CI, 67.3% to 81.4%), and 99% specificity (95%CI, 98.1% to 100%), with PPV of 98% and NPV of 87%. The balanced accuracy was 87%.

Conclusion: The SickKids MIS-C screening pathway has a high specificity, but a low sensitivity and accuracy in capturing children with MIS-C at the onset of the disease. Overall, the ACR algorithm performs better at differentiating children with MIS-C from febrile controls.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/screening-multisystem-inflammatory-syndrome-in-children-accuracy-of-sickkids-screening-pathway-compared-to-acr-algorithm/