|Year : 2019 | Volume
| Issue : 1 | Page : 28-33
Measurement of patient skin dose and establishment of local diagnostic reference levels for interventional cardiology procedures
Arti R Kulkarni1, Philomina Akhilesh2, Sunil Dutt Sharma3
1 Radiological Safety Division, Atomic Energy Regulatory Board; Homi Bhabha National Institute, Anushaktinagar, Mumbai, Maharashtra, India
2 Radiological Physics and Advisory Division, Bhabha Atomic Research Centre, Anushaktinagar, Mumbai, Maharashtra, India
3 Homi Bhabha National Institute; Radiological Physics and Advisory Division, Bhabha Atomic Research Centre, Anushaktinagar, Mumbai, Maharashtra, India
|Date of Submission||28-Feb-2019|
|Date of Decision||06-Mar-2019|
|Date of Acceptance||18-Mar-2019|
|Date of Web Publication||3-Jun-2019|
Sunil Dutt Sharma
Radiological Physics and Advisory Division, Bhabha Atomic Research Centre, CT and CRS Building, Anushaktinagar, Mumbai - 400 094, Maharashtra
Source of Support: None, Conflict of Interest: None
Use of diagnostic reference levels (DRLs) is an important tool for patient dose optimization in interventional radiology. DRLs are normally expressed in terms of kerma-area product (Pka). However, no direct method is available to estimate the probability of skin reactions using system displayed quantities. The aim of this work was to measure skin entrance doses (SEDs) in patients undergoing coronary angiography (CA) and percutaneous coronary interventions (PCI) using Gafchromic films and establish DRLs. The details of patients, exposure parameters, and dose quantities were recorded for 572 patients. Out of these, skin doses were measured for selected 64 patients using Gafchromic film. Measured SEDs are in the range of 48.2–740 mGy and 84–1242 mGy for CA and PCI, respectively. The data of 572 patients were analyzed, and 75th percentile of Pkawas calculated. DRLs for CA and PCI in terms of Pkaare found to be 34 and 134 Gy.cm2, respectively. DRLs for CA and PCI in terms of cumulative air kerma (Ka) at reference point are found to be 590 mGy and 1930 mGy, respectively. SED has a good correlation with Ka, however, it does not correlate well with Pka.
Keywords: Diagnostic reference level, dosimetry, Gafchromic film, interventional cardiology, skin dose
|How to cite this article:|
Kulkarni AR, Akhilesh P, Sharma SD. Measurement of patient skin dose and establishment of local diagnostic reference levels for interventional cardiology procedures. Radiat Prot Environ 2019;42:28-33
|How to cite this URL:|
Kulkarni AR, Akhilesh P, Sharma SD. Measurement of patient skin dose and establishment of local diagnostic reference levels for interventional cardiology procedures. Radiat Prot Environ [serial online] 2019 [cited 2019 Jun 20];42:28-33. Available from: http://www.rpe.org.in/text.asp?2019/42/1/28/259675
| Introduction|| |
The practice of interventional radiology (IR) in India began in the early 1970s, and the frequency of these procedures is rapidly increasing because it replaces most of the invasive procedures. IR has grown from around 2000 procedures in 1999 to over 50,000 procedures in 2017. IR procedures often involve high dose and hence the potential of these procedures to cause skin reactions in patients is of concern. As per the International Commission on Radiological Protection, diagnostic reference levels (DRLs) help in optimizing radiological protection in imaging procedures. It offers a method of discriminating unusually high or low patient dose for a particular procedure. As a step toward establishment of DRLs for IR practice in India, Atomic Energy Regulatory Board (AERB) has made mandatory that kerma-area product (Pka) meters shall be available with every IR equipment for monitoring and recording of patient doses during the procedure.
Several investigators have reported the use of Pka and cumulative air kerma (Ka) values of IR procedures for establishment of DRLs and estimation of skin doses.,,, However, as the system-displayed dose quantities do not provide any direct method for measuring skin doses, studies were also carried out using silver halide-based slow dosimetry film, XR-RV3 Gafchromic film, and thermoluminescent dosimeter (TLD) for skin dose measurements in IR procedures.
There are many limitations in using Pka values for skin dose estimation as it is highly area dependent, however, it can be used for estimation of effective dose.Use of TLD is a laborious method, positioning is difficult, and some of the high-dose points may be missed if not appropriately placed. Furthermore, many of the TLDs are not tissue equivalent and may need various correction factors. Comparatively, film dosimetry is considered to overcome these limitations, and Gafchromic film has the advantage that the readout is directly related to the radiation that enters locally on the skin, includes backscatter and is independent of the beam projection angle. Further, the Gafchromic film can be handled in normal room lights, self-developing, and responds nearly immediately to radiation exposure. The aim of the present study was to measure skin entrance doses (SEDs) for coronary angiography (CA) and percutaneous coronary intervention (PCI) procedures using Gafchromic XR-RV3 films and analyze the system-displayed dose quantities Pka and cumulative Ka at interventional reference point (IRP) to establish local DRLs for these interventional procedures.
| Materials and Methods|| |
Patient data and local diagnostic reference level
The most commonly performed CA and PCI procedures were selected for the present study. IR equipment used for patient data collection were Artis Zee and Axiom Artis (Siemens Ltd., Germany). The equipment performance (kilovoltage accuracy, beam current linearity, output consistency, spatial resolution, tabletop dose rate, tube housing leakage, and accuracy and reproducibility of Kerma area product meter) were tested and found within tolerance limits as per the AERB quality assurance protocol for X-ray equipment.
The details of patient (age, gender), procedure (CA/PCI), exposure parameters (average kVp, fluoroscopy time [min], frame rate [frames/s], number of acquisitions), and system displayed dose quantities (Pka[μGym2], and Ka[mGy]) were recorded. The data of 572 patients (374 CA and 198 PCI) were recorded and analyzed to calculate 75th percentile values of Pka and Ka. Usually, the 75th percentile of dose quantities is designated as DRLs.
Measurement of skin entrance dose using Gafchromic film
Of the 572 patients, skin dose measurements were carried out for selected 64 cases (39 cases of CA and 25 cases of PCI) using Gafchromic XR-RV3 film. Both male and female patients were included in this study. The complexity index and variation of dose with the weight of the patients were not considered as most of the selected procedures are standard, and patient weights were within the range of 65–80 kg.
Before its use for patient skin dose measurement, the dose-response calibration curve for the Gafchromic XR-RV3 film was generated for 80 kV with 3.5 mm of Al filtration in a diagnostic X-ray beam (Polydoros, Siemens Ltd., Germany). Calibrated solid-state Xi detector (Raysafe AB, Sweden) was used for measuring the output of this X-ray machine and verifying the accuracy of the operating potential. The calibration of Xi detector is traceable to Physikalisch-Technische Bundesanstalt, Germany, with calibration uncertainty of 2%. The XR-RV3 film was cut into pieces of 6 cm × 6 cm prior to experiment with extra care to protect it from any physical damage and arranged at the center of the polymethyl methacrylate phantom (30 cm × 30 × cm × 30 cm) for simulating the actual exposure conditions of the patient. Field size of 10 cm × 10 cm was used during irradiation of the film samples. The film was calibrated for dose range of 15–2000 mGy. The films were digitized using Epson Expression 1000XL scanner, and reflective density to Ka calibration curve was plotted.
For skin dose measurement, the film samples of sufficiently larger size (30 cm × 20 cm) were used to cover the overlapping of anteroposterior, posteroanterior, and oblique orientations and placed on couch below the patient as per the calibration setup. The contribution from lateral field could not be covered with this setup. However, lateral field projection is rarely used in cardiac procedures and hence its exclusion will not affect the measured dose significantly. Film darkening includes backscatter, beam orientations, and field nonuniformities. These films were scanned after 24 h as per calibration conditions. The reflective density of the film was then measured using Epson Expression 1000XL scanner, and Ka(mGy) was calculated using the calibration curve. For conversion of entrance surface air kerma (Ke) to absorbed dose to the skin, published correction factor of 1.06 can be used. However, this correction factor has not been used in the present study.
| Results|| |
[Figure 1] shows the dose-response curve of Gafchromic XR-RV3 film used for skin dose measurement. This curve shows that the reflective density of the film increases nonlinearly with the dose necessitating the use of a mathematical fit for estimating the dose from the recorded reflective density.
|Figure 1: Dose-response calibration curve of Gafchromic XR-RV3 film for 80 kV X-ray|
Click here to view
The age range of the patients on which skin dose measurements were carried out was 28–86 years. The kV P values recorded were within 76–92 kVp. Frame rate of 15 frames/s was used. [Figure 2] shows the samples of film exposed for skin dose measurement during CA and PCI procedures. The density distribution of the films indicates that the area of maximum dose is located in the primary beam.
|Figure 2: Sample of the film exposed for skin dose measurement during (a) coronary angiography and (b) percutaneous coronary intervention procedures|
Click here to view
The statistics of recorded data and measured peak skin dose in CA procedures are given in [Table 1]. Data of 374 patients undergoing CA procedures were recorded from different hospitals and mean, median, standard deviation, and 75th percentile were evaluated. It can be observed from [Table 1] that skin dose for CA procedures ranges from 48.2 to 740 mGy.
|Table 1: Statistics of data collected and measured peak skin doses for coronary angiography procedures|
Click here to view
The statistics of recorded data and measured peak skin dose in PCI procedures are given in [Table 2]. It can be observed from [Table 2] that skin dose for PCI procedures ranges from 84 to 1242 mGy.
|Table 2: Statistics of data collected and measured peak skin doses for percutaneous coronary interventions|
Click here to view
[Table 3] presents the local DRLs for CA and PCI procedures in terms of Pka and cumulative Ka. As indicated earlier, the DRL is 75th percentile of the data recorded for a given procedure.
|Table 3: Local diagnostic reference levels for coronary angiography and percutaneous coronary intervention procedures|
Click here to view
[Table 4] presents comparison of DRL values reported by several investigators. The DRL values estimated in this study is also shown in this table. A close observation of the data indicates that DRL values for CA and PCI procedures range from 21 to 102 and 50 to 193 Gy.cm2, respectively. The DRL values estimated in this study lies well within the DRL reported in the literature.
|Table 4: Comparison of diagnostic reference levels for coronary angiography and percutaneous coronary intervention procedures expressed in kerma-area product values (Gy.cm2)|
Click here to view
The correlation between Ka and peak skin entrance dose (PSED) is shown in [Figure 3] for CA and [Figure 4] for PCI. The correlation of Pka with PSED is shown in [Figure 5] for CA and [Figure 6] for PCI. [Figure 3] and [Figure 4] shows that the regression coefficients (R2) for correlation between Ka and PSED are 0.8363 and 0.8337, respectively, which indicates good correlation between these two quantities and hence Ka can be used for estimation of skin reaction probability in CA and PCI procedures. [Figure 5] and [Figure 6] reveals that the correlation of measured values of PSED and recorded values of Pka appeared to be average, and the observed values of R2 are 0.6 and 0.74 for CA and PCI, respectively. This implies that the Pka values although a useful quantity for estimation of effective dose, but not a reliable indicator of probable skin reactions.
|Figure 3: Correlation between cumulative air kerma and measured peak skin entrance dose using XR-RV3 film for coronary angiography procedures|
Click here to view
|Figure 4: Correlation between cumulative air kerma and measured peak skin entrance dose using XR-RV3 film for percutaneous coronary intervention procedures|
Click here to view
|Figure 5: Correlation between kerma area product (KAP) and measured peak skin entrance dose using XR-RV3 film for coronary angiography procedures|
Click here to view
|Figure 6: Correlation between kerma area product (KAP) and measured peak skin entrance dose using XR-RV3 film for percutaneous coronary intervention procedures|
Click here to view
| Discussion|| |
The variations in the radiation doses incurred during PCI are observed to be larger than the variation of doses incurred in the CA procedures. The reasons for variation in the doses are due to clinical complexity involved in the procedure, number of stents implanted, and skill of the medical practitioner. The data presented in this study were collected from a few hospitals in Mumbai. The present data can be considered as a representation of the Mumbai region, and the local DRLs presented in this work can be implemented in the city. However, median of these data may not be the representative of the practices in the country. There is a need of conducting country-wide survey for establishing national DRLs.
During the course of the present study, it was noted that the important parameters affecting patient dose used for establishment of DRLs are: proper nomenclature and categorization of procedure, identifying clinical complexity of procedure, grouping the data based on body mass index or patient weight, accuracy and consistency of dose display quantities, and importantly skill of medical practitioner. The entire cardiology team needs to be involved in the generation of data, analysis, review, and possible improvements in the procedure. The uncertainty because of kerma area product meter performance can be reduced by periodic calibration. In the present study, the observed variation was 20%. When setting local DRLs, each institute should reduce this uncertainty by correcting the values with appropriate calibration coefficient.
The DRL derived in this study for CA is comparable to the values reported by the earlier study in India and other countries, however, the DRLs derived in this study for PCI appears to be higher than the values reported from other countries. This may be because the patient doses in IR procedures have not been studied comprehensively and guidance for clinical practitioner is not available. CA being simpler procedure and standardized, patient doses are more or less optimized and hence DRLs are comparable to reported values. Introduction of flat panel technology and dose reduction features in the recent years also have affected the patient doses.
Pka values can be used to estimate the effective dose for adult patients using conversion factor as 1 Gy.cm2 (Pka) yields 0.18 mSv effective dose. It shows very poor correlation with the peak skin dose as the large dose received in small area and small dose spread over the large area of the skin will show the similar values. However, skin reactions will be highly dependent on the spread of the dose in the skin area.
Another quantity displayed on the system during IR procedure is Ka at IRP. Here, the reference point may lie inside the patient, outside the patient, or nearly on the skin of the patient depending on the site of procedure and thickness of the patient. This point dose calculation does not take into account the angulations used during the procedure. However, in the absence of any direct method for skin dose measurement, the quantity Ka is used as surrogate for estimation of SED. As the IRP lies in air, to convert the values to SED, backscatter factor (BSF) is required to be used.
SED = Ka× BSF (1)
Where BSF = 1.07. As the films were calibrated with phantom, BSF is already incorporated.
It appears that although these are very common procedures all over the world, many of the procedures are not expected to result in the skin injury. Further for optimization of the radiation dose, every institute needs to establish the local DRLs for common procedures and review the radiation dose by comparing among various practitioners for improving safety culture. For interventional procedures, it is important to understand that the reference levels are never meant for individual patients and to be applied with flexibility to allow higher exposures if these are indicated by clinical judgment.
The reference levels suggested in this study as the 75th percentile of Pka values can be used by the institute for initial comparison of patient dose for optimization and can review the same after analysis of sufficient number of procedures. It is seen that the median values of Pka, i.e., 21 Gy-cm2 for CA and 111 Gy-cm2 for PCI are below the respective reference levels which implies the acceptability of the procedures in the selected institutions.
| Conclusions|| |
Patient dose data were collected from different hospitals to establish local DRL for CA and PCI procedures, and skin dose measurements were also carried out using Gafchromic XR-RV3 film. Sample study of the SED measurements shows that the maximum localized skin dose received in CA and PCI procedures are 740 mGy and 1242 mGy, respectively, which are well within the threshold of skin injury and also within the internationally published values in most of the developed countries. The measured PSED shows very poor correlation with fluoroscopy time (FT). Cumulative Ka can be used as close surrogate of SED and are required to be monitored for avoiding probable skin injuries in the complex IR procedures, specifically when the same beam angle is being used for longer time. Till date, there are no DRLs established for these procedures in India, and this study could be used as an interim yardstick for other cardiac interventional laboratories in India until a large national study could be performed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Indian Society for Vascular and Interventional Radiology. Available from: http://www.isvirindia.org
. [Last accessed on 2019 Jan 20].
Vañó E, Miller DL, Martin CJ, Rehani MM, Kang K, Rosenstein M, et al.
ICRP publication 135: Diagnostic reference levels in medical imaging. Ann ICRP 2017;46:1-44.
AERB Safety Code No. Radiation Safety in Manufacture, Supply and Use of Medical Diagnostic x-ray Equipment. AERB/RF-MED/SC-3 (Rev. 2); 2016.
Bogaert E, Bacher K, Lemmens K, Carlier M, Desmet W, De Wagter X, et al.
Alarge-scale multicentre study of patient skin doses in interventional cardiology: Dose-area product action levels and dose reference levels. Br J Radiol 2009;82:303-12.
Maghbool M, Hosseini MA. An assessment of radiation exposure doses in patients undergoing primary percutaneous coronary intervention based on angiographic records. Radiation Protection Dosimetry, ncy180, https://doi.org/10.1093/rpd/ncy180.
van de Putte S, Verhaegen F, Taeymans Y, Thierens H. Correlation of patient skin doses in cardiac interventional radiology with dose-area product. Br J Radiol 2000;73:504-13.
Kwon D, Little MP, Miller DL. Reference air kerma and kerma-area product as estimators of peak skin dose for fluoroscopically guided interventions. Med Phys 2011;38:4196-204.
Vañó E, Guibelalde E, Fernández JM, González L, Ten JI. Patient dosimetry in interventional radiology using slow films. Br J Radiol 1997;70:195-200.
Uniyal SC, Chaturvedi V, Sharma SD, Rawat A. Patient dosimetry during interventional cardiac procedures in a dedicated catheterization laboratory. Radiat Prot Dosimetry 2017;175:201-8.
Siiskonen T, Ciraj-Bjelac O, Dabin J, Diklic A, Domienik-Andrzejewska J, Farah J, et al.
Establishing the European diagnostic reference levels for interventional cardiology. Phys Med 2018;54:42-8.
McCabe BP, Speidel MA, Pike TL, Van Lysel MS. Calibration of GafChromic XR-RV3 radiochromic film for skin dose measurement using standardized X-ray spectra and a commercial flatbed scanner. Med Phys 2011;38:1919-30.
International Atomic Energy Agency. Establishing Guidance Levels in X-ray Guided Medical Interventional Procedures: A Pilot Study. Safety Report Series 59. 17: International Atomic Energy Agency; Vienna, 2009.
Brindhaban A. Radiation dose to patients in coronary interventional procedures: A survey. Radiat Prot Dosimetry 2018; doi: 10.1093/rpd/ncy179. [Epub ahead of print]
Pantos I, Patatoukas G, Katritsis DG, Efstathopoulos E. Patient radiation doses in interventional cardiology procedures. Curr Cardiol Rev 2009;5:1-11.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3], [Table 4]