JOURNAL CLUB: Can Coronal STIR Be Used as Screening for Acute Nontraumatic Hip Pain in Children?

Autor: Megan B. Marine, Boaz Karmazyn, Joseph T. Davis, Monica M Forbes-Amrhein, Ananth Kumar Ravi, Trenton D. Roth, Matthew R. Wanner
Rok vydání: 2017
Předmět:
Zdroj: American Journal of Roentgenology. 209:676-683
ISSN: 1546-3141
0361-803X
Popis: The objective of this study is to evaluate whether coronal STIR MRI can be used as a screening test for nontraumatic acute hip pain in children.From 2008 to 2012, we identified all patients younger than 18 years at our tertiary care facility who underwent pelvic MRI including coronal STIR for the following indications: acute hip pain, limping, or refusal to bear weight. Patients with a history of trauma were excluded. Each MR image was independently reviewed by four radiologists who were blinded to the clinical outcome. After first reviewing the coronal STIR images only, they then reviewed the full MRI studies in a random order different from that used for review of the coronal STIR images. The sensitivity and specificity of STIR-only images in identifying the presence of abnormality and specific diagnoses were calculated, with the full MRI study considered as the reference standard. Kappa values were calculated for STIR-only and full MRI studies.A total of 127 patients (67 female patients and 60 male patients; median age, 9 years; range, 5 months to 17 years) were identified. The most common abnormalities (calculated as the mean of frequency values noted by four readers) were hip effusion (52%; range, 46-58%), osteomyelitis (42%; range, 29-48%), and myositis (32%; range, 20-40%). For the detection of any abnormality, STIR-only images had a mean sensitivity of 95% and a mean specificity of 67%. For approximately one-third of STIR-only studies with true-positive results, additional abnormalities were found on full MRI studies.Coronal STIR imaging of the pelvis has high sensitivity (95%) in the detection of abnormalities associated with acute nontraumatic hip pain in children, but it often misses additional abnormalities.
Databáze: OpenAIRE