The Accuracy of Prevalent Vertebral Fracture Detection in Children Using Targeted Case-Finding Approaches.

Area of Science:

• Pediatric Endocrinology
• Radiology
• Orthopedics

Background:

• Pediatric vertebral fractures (VF) are concerning due to cumulative radiation exposure from standard spine radiographs.
• Identifying children at risk for prevalent VF (PVF) is crucial for appropriate diagnostic imaging.
• Existing conditions like leukemia, rheumatic disorders, and nephrotic syndrome increase VF risk in children.

Purpose of the Study:

• To evaluate the accuracy of back pain and lumbar spine bone mineral density (LS BMD) in identifying prevalent vertebral fractures (PVF) in children.
• To develop and compare case-finding paradigms using these clinical predictors for PVF detection.
• To guide judicious use of spine radiography in pediatric populations at risk for VF.

Main Methods:

• Studied 400 children at risk for PVF, assessing back pain via patient report and LS BMD using dual-energy X-ray absorptiometry.
• Quantified PVF on spine radiographs using the modified Genant semiquantitative method.
• Analyzed data using logistic regression to determine the predictive value of LS BMD and back pain, developing four case-finding paradigms.

Main Results:

• Prevalent VF (PVF) were found in 11.0% of the study cohort (44 out of 400 patients).
• Low LS BMD (Z-score < -1.6) and back pain alone showed similar sensitivity (55%, 52%) and specificity (78%, 81%).
• A combined approach of low LS BMD or back pain yielded the highest sensitivity (75%) for PVF detection, with high negative predictive value (95%).

Conclusions:

• Lumbar spine bone mineral density and back pain history are valuable tools for identifying children with the highest risk of prevalent vertebral fractures.
• The developed case-finding paradigms allow for more judicious application of spine radiography, minimizing unnecessary radiation exposure.
• The choice of paradigm should consider the expected PVF rate and the clinical setting to optimize diagnostic accuracy and resource utilization.