No abstract

We conducted a nationwide survey of multiple institutions and collected data of various interventional procedures in the field of cardiology. Included in the analysis were 126 institutions, 381 X-ray systems, and 805 protocols. The dose values were compared with the Japanese diagnostic reference levels (DRLs) 2015. Fluoroscopy time, air kerma at the patient entrance reference point (Ka, r), and air kerma-area product (PKA ) were analyzed for various interventional procedures in 5,734 cardiology patients. The fluoroscopic dose rate (FDR) for pulmonary vein isolation (PVI) was less than half that of the 75th percentile of the Japanese DRLs 2015. The 75th percentiles of fluoroscopy time, Ka, r, and PKA for the respective interventional procedures were as follows: 11.0 min, 735 mGy, and 64 Gy･cm 2 for diagnostic coronary angiography (CA); 13.2 min, 839 mGy, and 75 Gy･cm 2 for CA + left ventriculography; 34.4 min, 1,810 mGy, and 148 Gy･cm 2 for percutaneous coronary intervention (PCI) excluding chronic total occlusion; 80.1 min, 4,338 mGy, and 312 Gy･cm 2 for PCI for chronic total occlusion; 74.4 min, 833 mGy, and 90 Gy･cm 2 for PVI; and 34.0 min, 795 mGy, and 94 Gy･cm 2 for transcatheter aortic valve implantation, respectively. In assessing dose values in interventional radiology, the difficulty of the technique needs to be considered, and the DRL values for FDR, fluoroscopic time, Ka, r, and PKA for each interventional procedure are considered necessary when reassessing or updating DRLs.

In interventional radiology, dose estimation using the interventional reference point (IRP) is a practical method for obtaining the real-time skin dose of a patient. However, the IRP is defined in terms of adult cardiovascular radiology and is not suitable for dosimetry of the head. In the present study, we defined a new reference point (neuro-IRP) for neuro-interventional procedures. The neuro-IRP was located on the central ray of the X-ray beam, 9 cm from the isocenter, toward the focal spot. To verify whether the neuro-IRP was accurate in dose estimation, we compared calculated doses at the neuro-IRP and actual measured doses at the surface of the head phantom for various directions of the X-ray projection. The resulting calculated doses were fairly consistent with actual measured doses, with the error in this estimation within approximately 15%. These data suggest that dose estimation using the neuro-IRP for the head is valid.

We conducted a nationwide multicenter survey of various interventional radiology (IVR) procedures. Data were collected from 385 X-ray systems in 126 institutions, including 432 cine programs and 380 digital subtraction angiography (DSA) programs for diagnostic catheterization, percutaneous coronary intervention (PCI), ablation, transcatheter aortic valve implantation (TAVI), neurologic IVR, thorax IVR, abdominal IVR, and endovascular therapy (EVT). Fluoroscopic and cine dose rates were 10.1 mGy/min and 110.7 mGy/min, respectively, whereas for DSA programs, the median fluoroscopic and DSA dose rates were 8.0 mGy/min and 224.8 mGy/min, respectively. The DSA dose rate was more than twice the cine dose rate. The largest difference between dose rates was for diagnostic catheterization, which had a cine dose rate of 142.6 mGy/min and a fluoroscopic dose rate of 12.6 mGy/min (by a factor of 12.5), followed by EVT, which had a DSA dose rate of 216.0 mGy/min and a fluoroscopic dose rate of 7.7 mGy/min (by a factor of 29.6). The smallest difference between dose rates was for TAVI, which had a cine dose rate of 96.8 mGy/min and a fluoroscopic dose rate of 12.0 mGy/min (by a factor of 8.9), followed by neurologic IVR, which had a DSA dose rate of 227.9 mGy/min and a fluoroscopic dose rate of 9.6 mGy/min (by a factor of 22.6). Compared with the fluoroscopic dose rates, the cine dose rates were 9-13 times higher and the DSA dose rates were 22-30 times higher; the DSA dose rates were more than double the cine dose rates.