Evaluating the Correlation Between Pediatric Exposure Rates and Common Body Size Surrogates in Fluoroscopy
Abstract
Purpose
To assess correlations between pediatric radiation exposure rates and commonly used body size surrogates, anteroposterior (AP) or lateral (LAT) thickness, weight, height, body mass index (BMI), body surface area (BSA), and age—for general fluoroscopy (GF) and fluoroscopically guided interventional (FGI) examinations, and to identify the most predictive surrogate for dose optimization across pediatric body sizes.
Methods
This retrospective study included 6,447 GF examinations from 4,452 pediatric patients and 2,968 cardiac FGI examinations from 1,471 patients, aged birth to 21 years. Reference air kerma (RAK) data from a dose monitoring database were calibrated to within ±5% accuracy. Exposure rate was defined as RAK divided by total fluoroscopy time. Body size surrogates (AP/LAT thickness, age, weight, height, BMI, and BSA) were extracted or calculated. Pearson correlation coefficients were computed between log-transformed exposure rates and each surrogate, with Steiger’s Z-test used to compare dependent correlations.
Results
For GF examinations, AP thickness correlated most strongly with exposure rate (r = 0.691), followed by BSA (r = 0.574) and weight (r = 0.568). BMI showed the weakest correlation (r = 0.446). For FGI examinations, weight demonstrated the highest correlation with exposure rate in both frontal (r = 0.689) and lateral (r = 0.794) planes, with BSA and LAT thickness performing similarly. All other correlations were significantly lower (p < 0.001). The superior performance of AP thickness in GF reflects its direct relationship to x-ray attenuation, whereas variable geometry and positioning during FGI favor weight as a more stable predictor.
Conclusion
AP thickness is the most predictive size surrogate for pediatric GF, while weight is most reliable for cardiac FGI. Modality-specific size metrics improve pediatric dose optimization and support consistent exposure management across patient sizes.