Calcitonin Gene–Related Peptide for Identifying Pediatric Bacterial Musculoskeletal Infections

Objectives: Bacterial musculoskeletal infections (MSKIs) can be challenging to diagnose. We compared the accuracy of calcitonin gene–related peptide (CGRP), a neuropeptide which is transcribed from the same gene as procalcitonin, to procalcitonin for the diagnosis of a MSKI in children. Methods: We conducted a prospective cohort study of patients 21 years old or younger who underwent evaluation for MSKIs at one of 3 emergency departments. Our primary outcome was a MSKI, defined as septic arthritis, osteomyelitis, or pyomyositis. We used a Spearman correlation coefficient to measure the association between serum CGRP and procalcitonin and compared the diagnostic accuracy using area under the receiver operating characteristic curve (AUC) analysis. Results: Of the 200 enrolled patients, 33 (17%) had a MSKI. Overall, median serum CGRP level did not differ between patients with and without a MSKI (13.5 pg/mL MSKI vs 10.9 pg/mL no MSKI; difference: 2.6, 95% CI: −0.6, 5.8), while PCT was higher in patients with a MSKI (0.12 ng/mL MSKIs vs 0.04 ng/mL no MSKI; difference: 0.08, 95% CI: 0.03 to 0.13). CGRP and PCT levels were not correlated (Spearman rank coefficient: −0.01, 95% CI: −0.15 to 0.13). CGRP had a lower AUC than procalcitonin [0.57, 95% CI: 0.47 to 0.66 CGRP vs 0.78, 95% CI: 0.69 to 0.87 PCT, P < 0.01]. Conclusions: Although biochemically related, CGRP was not correlated with procalcitonin in children undergoing evaluation for a MSKI. Our exploratory pilot highlights the ongoing need for novel biomarkers for the accurate and timely identification of children with a MSKI.