Comparison of four techniques to estimate radiation dose to skin during angiographic and interventional radiology procedures

PURPOSE: Four techniques used to estimate radiation risk were compared to determine whether commonly used dosimetry measurements permit reliable estimates of skin dose. Peak skin dose (PSD) is known to be the most reliable estimate of risk to skin. The purpose of this study is to determine peak skin dose with use of real-time software measurements and to correlate other measures of dose with PSD. MATERIALS AND METHODS: Two hundred twelve patients undergoing arch aortography and bilateral carotid arteriography (referred to as "carotid"), abdominal aortography and bilateral lower extremity runoff ("runoff"), or tunneled chest wall port placement ("port") were studied. Fluoroscopy time, dose-area product (DAP), and cumulative dose at the interventional reference point were recorded for all procedures; PSD was recorded for a subset of 105 procedures. The dose index, defined as the ratio between PSD and cumulative dose, was also determined. RESULTS: In general, correlation values for comparisons between fluoroscopy time and the other measures of dose (r = .29 to .78) were lower than values for comparisons among DAP, cumulative dose, and PSD (r = .52 to .94). For all procedures, pair-wise correlations between DAP, cumulative skin dose, and PSD were statistically significant (P < .01) The ratio between PSD and cumulative skin dose (dose index) was significantly different for ports versus other procedures (carotid, Z = 4.62, P < .001; runoff, Z = 4.52, P < .001), but carotid and runoff procedures did not differ significantly in this regard (Z = 0.746, P = .22). Within each individual procedure type, the range of values for the dose index varied 156.7-fold for carotid arteriography, 3.2-fold for chest ports, and 175-fold for aortography and runoff. CONCLUSION: Fluoroscopy time is a poor predictor of risk because it does not correlate well with PSD. Cumulative dose and DAP are not good analogues of PSD because of weak correlations for some procedures and because of wide variations in the dose index for all procedures.