Conversion factors to estimate effective doses from kerma area product in interventional cardiology. Impact of added filtration

Highlights

• New X-ray systems with “patient dose reduction techniques” use high copper filtration for cine acquisitions.

• Monte Carlo software with data from each radiation event (using RDSR) allow refining the calculation of effective doses.

• Conversion factors for the new dose reduction systems, increased by 38% in comparison to the standard X-ray systems.

• The impact of filtration should be considered to estimate effective and population doses from interventional cardiology.

Abstract

There is a large variation in the factors used to estimate effective doses from kerma area product (KAP) for interventional cardiology. These factors are required to estimate population doses. This paper presents the results for this conversion factor for cardiac procedures using tissue weighting factors of ICRP-103 and the impact of the added copper filtration in the X-ray beam.

The data from 925 cardiac procedures and 75,347 radiation events were collected from two angiography laboratories using the DICOM Radiation Dose Structured Reports (RDSR). Effective doses were calculated with Monte Carlo software and the dosimetric, technical and geometrical information included in the RDSR.

In one laboratory, with an X-ray system without Cu filtration for the cine runs, a factor of 0.21 ± 0.05 mSv/(Gy·cm²) was obtained. In other laboratory, incorporating a patient dose reduction technique, and 0.4 mm of Cu filtration for cine runs, the conversion factor was 0.29 ± 0.05 mSv/(Gy·cm²). The analysis of the radiation events for the different Cu filtrations (0.0; 0.1; 0.4 and 0.9 mm) resulted in conversion factors of: 0.16; 0.27; 0.34 and 0.40 mSv/(Gy·cm²) respectively.

The estimation of effective and population doses from KAP should take into account the Cu filtration in the X-ray beam. For the X-ray system with patient dose reduction technique, using 0.4 mm Cu for cine runs, the global conversion factor increased by 38%, from 0.21 to 0.29 mSv/(Gy·cm²) in comparison to the standard X-ray system with a protocol that did not include copper filtration for cine acquisitions.

Introduction

The X-ray systems used for interventional procedures are able to give patient exposure indexes during and at the end of the procedures. In the most recent models, the X-ray systems are able to transfer this dosimetric and other technical and geometrical information to the DICOM radiation dose structured reports (RDSR).

The dosimetric data, kerma area product –KAP– and air kerma –AK– at the patient entrance reference point [1], are used for optimisation of the radiation protection of patients. Now, with the information contained in the RDSR (kV, filtration, C-arm angulation, radiation field size, etc) the optimisation actions may be improved. For specific clinical indications, their median values should be periodically compared with the diagnostic reference levels (DRLs) [2].

Population doses are estimated using the frequency of the procedures and the quantity effective dose. The Monte Carlo software [3] allows organ and effective doses to be calculated from the primary dosimetric quantities reported by the X-ray systems, from all the information on the X-ray beam quality (kV and filtration) and from other geometrical and technical factors.

According to the recommendations of the International Commission on Radiological Protection (ICRP), effective doses should not be used to estimate individual radiation risks to patients but may be useful to compare the relative risks when different imaging procedures with different modalities (e.g. X-ray catheterization, computed tomography and nuclear medicine) are performed on the same patient [4], [5].

The new European Directive on Basic Safety Standards [6] requires patient dose audits and the use of DRLs for optimisation, and in addition, this regulation requires that “population doses derived from medical exposures be estimated for radiodiagnostic and interventional radiology”. Such a requirement makes it necessary to estimate effective doses and the frequency of the procedures.

The UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation) collects and periodically publishes the worldwide data on population doses arising from medical exposures. The last global report for the data corresponding to the period 1997–2007 was published in 2008 [7] and the next report is expected in 2020.

UNSCEAR recognises that there are some limitations to the survey data as the majority of responses received come from the relatively more developed countries. Principally because of their numerous significant benefits, interventional radiology and cardiology procedures have experienced a dramatic increase in frequency in recent years. The UNSCEAR report of 2008 stated that the average conversion factor to deduce the effective dose from KAP readings for cardiac interventional procedures was 0.17 mSv/(Gy.cm²) [7].

In 2015, UNSCEAR published in view of the next report, a “User Manual for the Global Survey on Medical Exposures” [8] in which it is indicated that “effective dose estimation is usually performed using simpler methods on the basis of easily measurable dose metrics and the application of conversions coefficients that have been determined through scientific research”.

The factors recommended in this manual for the estimation of effective doses in cardiac interventional procedures were: 0.12 mSv/(Gy.cm²) for diagnostic coronary angiography and a wide range of factors from 0.20 to 0.28 for percutaneous transluminal coronary angioplasty (PTCA) [8].

In Europe, all the interventional procedures are estimated to contribute 3.5–14%, to the collective dose and the PTCAs, 0.5–3.6% [9]. In the latest European report, published in 2015 [10], the contribution of interventional radiology to the total collective dose in medical imaging in Europe was estimated 8–9% with a mean value in frequency of 0.6%. For coronary angiography and coronary angioplasty, in many countries, the effective dose is being estimated by multiplying the KAP by 0.2 mSv/(Gy.cm²), a factor recommended by the European Guidelines for adult patients in IC (interventional cardiology) [9].

In Spain, the fraction of the collective dose per million inhabitants derived from IC was 34 man-Sv (1600 man-Sv in Spain with 46.5 million inhabitants). The frequency of the IC procedures in Spain is 0.66% from the total medical procedures with X-rays, and IC represents 4% of the X-ray collective dose. These estimations result from the information on frequencies of examinations sourced from the Spanish Society of Cardiology and the patient dose values obtained by the national programme on patient dose DOCCACI [11].

The patient dose management systems already introduced in many hospitals allow to collect the dosimetric data, together with other technical details and geometrical data automatically for each radiation event (fluoroscopy and cine runs) of the interventional procedures. Consequently, a more detailed Monte Carlo calculation can be made to estimate organ and effective doses for a large number of patients.

Another importance issue is that some new X-ray systems dedicated to interventional procedures are implementing patient dose reduction techniques with different X-ray beam qualities (higher Cu filtration) and with advanced post-processing of the images, which may lead to a substantial variation in the conversion factor to estimate effective doses.

It seems appropriate to revisit and update these conversion factors.

The aims of this paper are:

• To calculate the conversion factors (using all the information contained in the RDSR for each of the radiation events), to estimate effective doses from KAP values, using a large number of cardiac procedures and Monte Carlo software.

• To evaluate the impact of the added copper filtration in the X-ray beams on these conversion factors, and to give practical advice for the evaluation of the collective doses derived from IC.