Radiation Protection Dosimetry (2008), Vol. 132, No. 1, pp Advance Access publication 25 September 2008

Transcription

1 Radiation Protection Dosimetry (2008), Vol. 132, No. 1, pp Advance Access publication 25 September 2008 doi: /rpd/ncn247 CORRELATION OF PATIENT MAXIMUM SKIN DOSES IN CARDIAC PROCEDURES WITH VARIOUS DOSE INDICATORS J. Domienik 1, *, S. Papierz 1, J. Jankowski 1,3, J. Z. Peruga 2, A. Werduch 3 and W. Religa 4 1 Nofer Institute of Occupational Medicine in Lodz, Radiation Protection Department, Sw Teresy 8 St, Lodz, Poland 2 Medical University of Lodz, II Chair and Department of Cardiology, Kniaziewicza 1/5 St, Lodz, Poland 3 Institute of Physics, University of Lodz, Department of Nuclear Physics and Radiation Safety, Pomorska 149/153 St, Lodz, Poland 4 Ministry of Internal Affairs Hospital, Polnocna 42 St, Lodz, Poland Received April , revised August , accepted September In most countries of European Union, legislation requires the determination of the total skin dose received by patients during interventional procedures in order to prevent deterministic damages. Various dose indicators like dose area product (DAP), cumulative dose (CD) and entrance dose at the patient plane (EFD) are used for patient dosimetry purposes in clinical practice. This study aimed at relating those dose indicators with doses ascribed to the most irradiated areas of the patient skin usually expressed in terms of local maximal skin dose (MSD). The study was performed in two different facilities for two most common cardiac procedures coronary angiography (CA) and percutaneous coronary interventions (PCI). For CA procedures, the registered values of fluoroscopy time, total DAP and MSD were in the range ( ) min, (16 317) Gy cm 2 and ( ) mgy, respectively, and for interventions, accordingly ( ) min, (17 425) Gy cm 2, ( ) mgy. Moreover, for CA procedures, CD and EFD were in the ranges ( ) mgy and ( ) mgy and for PCI ( ) mgy and ( ) mgy, respectively. No general and satisfactory correlation was found for safe estimation of MSD. However, results show that the best dose indicator which might serve for rough, preliminary estimation is DAP value. In the study, the appropriate trigger levels were proposed for both facilities. INTRODUCTION Some radiation-induced skin injuries in patients may occur when doses received during cardiac intervention exceeded threshold limits. Thus in order to avoid serious injuries, the practical actions to control dose need to be implemented in clinical practice. The International Commission on Radiation Protection (ICRP) in Publication 85 recommends the evaluation of doses absorbed by patients in the area that receives maximum dose. The suggestion of ICRP is complied in the legislation of most countries of European Union, also in Poland (1 4). A few methods have been proposed to measure local maximum skin dose (MSD) from dose distribution. The most frequently discussed in scientific literature are large, self made, TL dosemeter (TLD) arrays, large-area slow films or self-developing Gafchromic films. Although, the accuracy of largefilm methods in determining local MSD might be good, at least theoretically, one has to take into account not only their suitability but also cost (which are rather considerable) and practicability (they are laborious) when planning to implement them in routine dosimetry. Due to the latter, the *Corresponding author: jdom@imp.lodz.pl measurements with films or TLD arrays are rather not likely to become preferable methods to monitor the local maximum dose in hospitals (5 7). The availability of direct estimation/evaluation of local maximum dose from beam monitoring quantities is desirable for prevention of deterministic injuries. Depending on the angiography unit, there are various beam monitoring quantities displayed by the system like dose area product or/and cumulative dose (CD). Sometimes additional ionising chamber is installed on an X-ray tube for measurements of beam air kerma at the reference point distance, specified by the operator. The correlation between the beam monitoring quantities and the local MSD will depend on the concentration of the dose which in turn depends on projection and collimator settings used by the operators as well as on the distance at which the beam monitoring quantity of interest is specified. Thus the relations between them defined in one hospital usually are not applicable in another one. In this study, we investigated the correlation between local MSD and beam monitoring quantities in two haemodynamic rooms in Poland. For local MSD measurements, the method of large area films (Kodak EDR2), used also in other studies (4,8 11), was applied. # The Author Published by Oxford University Press. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 PATIENT SKIN DOSE VERSUS OTHER DOSE INDICATORS MATERIALS AND METHODS The haemodynamic room I was equipped with GE Innova 2000 with digital flat panel imaging detector, frame acquisition rate 15 or 30 frames per second and four field sizes available, whereas in haemodynamic room II the measurements were performed with Philips Integris 3000 X-ray system (Philips Medical Systems) with a conventional image intensifier, frame acquisition rate 12.5 and 25 frames per second and three field sizes available. Both units have an undercouch X-ray tube configuration and work under automatic exposure control. The units performances were checked both according to the typical QA/QC tests described in Zoetelief et al. (12) as well as, for the purpose of presented study, according to the following method (13). The dose rates were measured, at the distance of 60 cm from focus to the entrance surface of the patient simulated with 20 cm of PMMA phantom and the image intensifier placed in the distance of 5 cm from the patient. The dose rates measured in room I were 13.2 mgy min 21 and 32.6 mgy min 21 for two modes low and normal, respectively; the one measured in room II for low fluoro mode was 26.8 mgy min 21. The entrance dose per image measured, for the same geometry and for routinely used acquisition rate in room I was equal to 116 mgy per image (for 15 frames s 21 ) and 153 mgy per image in room II (for 12.5 frames s 21 ). Furthermore, the system in room I is provided with a dose area product meter DIAMENTOR M4-KDK [PTW-FREIBURG] with additional function allowing to measure the beam air kerma, while the unit in room II was equipped with KAP-meter (Doseguard 100, RTI Electronics AB). The data were monitored for two cardiac procedures: diagnostic coronary angiography (CA) and percutaneous coronary intervention (PCI). The procedures were performed by eight specialists from haemodynamic room I and three specialists from haemodynamic room II. Altogether, in the survey, 162 cardiac procedures were randomly collected; in more detail: 106 procedures in room I (52 CA procedures and 54 PCI procedures) and 56 examinations in room II (29 CA procedures and 27 PCI procedures). The statistical data concerning gender, age, weight and height as well as dosimetric parameters like: fluoroscopy time (FT); total number of images; DAP, EFD (entrance dose in the patient plane) measured at the distance of 55 cm from the meter (internal flat ionisation chamber DIAMENTOR M4-KDK mounted directly to the light beam diaphragm housing), CD (calculated by the system on the basis of the algorithm using the physical exposure parameters, filtration and geometry of the X-ray beam) and MSD assessed with the use of Kodak EDR2 films were determined for the randomly selected patients. The Kodak EDR2 films were calibrated in the dose range from 100 to 1250 mgy. The simultaneously performed measurements of MSD with TLD arrays attached to KODAK films (7) show that the saturation is likely to appear for doses higher than 1.5 Gy. One of the main tasks of the study was to define the dose indicator that would approximate most closely patient MSD. In other words, the purpose was to allow the operator to monitor the MSD received by the patient and to assess the probability of deterministic effects. To this end, the statistical analysis for the correlation between dose indications like DAP, CD or EFD and actual MSD received by the patient was performed. RESULTS Table 1 supplies details of the min, max, first, second and third quartile calculated from the data collected in room I, concerning gender, age, weight and height. The distributions of the patient s age are given in Figure 1. The typical values of the following parameters: FT, the number of images acquired, MSD and beam monitoring quantities like DAP, CD and EFD are given in Table 2. The first, general and rather obvious, conclusion is that all investigated parameters take significantly higher values for more complex, therapeutic procedures; this is true for both rooms. However, significant differences are observed. In general, all parameters are higher in room I when the medians are compared. The effect of relatively poor dosimetric performance of the unit in room II (higher than in room I dose rates provided by AEC system) is reduced probably due to the operator s techniques employing shorter FT and lower number of images Table 1. Patient s age parameters for CA and PCI procedures in the room I. Procedure CA PCI Sex Male Female Male Female Number of procedures Min Max Mean Median Third quartile

3 J. DOMIENIK ET AL. Figure 1. The distribution of patient age for male and female patients. acquired. As a result, the MSD and DAP values are lower in room II when compared with room I. The present study allowed to determine the reference levels (RLs) defined as third quartile. The RLs for selected dose indicators are compared with those determined in the framework of SENTINEL project (14). The comparison is presented in Table 3. One notes, that the DAP reference levels are in our case higher for both diagnostic and therapeutic procedures than those recommended in European project. The discrepancies might be as much as by a factor 2.5 (Table 3, room I). On the other hand, the FT and numbers of images are lower in both rooms, except RLs for therapeutic procedures in room I. Taking into account the above, one would be tempted to conclude that the collimation technique presented by the operators is rather not satisfying. However, some care should be exercised because there is a number of other factors that could influence, when performing the procedure, the value of dose rates presented in Table 3. The main point is that these results were obtained with all factors set constant, including geometry. On the other hand, during the procedure with real patients the relevant parameters are being varied. According to Wagner et al. (15), the change in various parameters like kvp, fluoro and acquisition dose-rate settings, field of view, air gap between patient and image intensifier and source to skin distance may significantly influence the dose (even multiplicatively). From this point of view, one cannot ascribe any longer the difference in reported DAP values to collimation technique only. Nevertheless, optimisation of the unit performance would be also desirable. One of the main objectives of the paper was to verify whether there exist some beam monitoring quantity from those available in hemodynamic room which would allow the operator to control MSD, in a reliable way, in order to indicate the possible overexposure. To this end, the analyses of the correlations between the actual MSD and dose indicators (DAP, CD, EFD) were performed. Although, a similar analyses can be found in the literature (4,6,7,10,11), the need for further study is justified by the fact that, in general, the results depend on the facility where the measurements are performed. 20

4 PATIENT SKIN DOSE VERSUS OTHER DOSE INDICATORS Table 2. Dosimetric parameters for monitored procedures. Procedure Parameter FT (min) Number of images DAP (Gy cm 2 ) EFD (mgy) CD (mgy) MSD (mgy) Room 1 Min CA Max Median Third quartile Min CA þ PCI Max Median Third quartile Room 2 Min CA Max Median Third quartile Min CA þ PCI Max Median Third quartile Procedure Sentinel Room 1 Room 2 Table 3. Reference levels for DAP, FT and number of images defined in room 1 and room 2. DAP (Gy cm 2 ) FT (mgy) Number of images Figures 2 7 show the comparisons between MSD and DAP, EFD, CD for CA and PCI procedures in room I. Some general remarks are in order. First, the correlation is much better for diagnostic procedures. This can be explained by the fact that the latter are simpler and standardised procedures contrary to the more complex and individual therapeutic ones. Secondly, in room I, the best correlation for CA procedures was found for DAP values. Third, even in the case of diagnostic procedures, the reliability of the estimation of MSD by dosimetric quantities can be put in doubt. In fact, if one relied on experimental points above the correlation line, one would underestimate the actual dose obtained by the patient. The rate of underestimation may be as much as one-third of the actual MSD (Figures 2 and 4). Taking into account the points (1017,1475) and (1768,1507) (Figure 4), we see that, in spite of a large difference in EFD, the corresponding MSD CD (mgy) Entrance surface air kerma rate CA Fluoroscopy low: 13.0 mgy min 21 ; Image acquisition: 100 mgy per image PCI CA Fluoroscopy low: 13.2 mgy min 21 ; Image acquisition: 117 mgy per image PCI CA Fluoroscopy low: 26.8 mgy min 21 ; Image acquisition: 153 mgy per image PCI values are roughly equal. This might suggest that the correlation coefficient R 2 ¼ 0.79 is not sufficient to provide a safe estimation of MSD. In addition, as far as Figures 2 and 3 are concerned, it is seen that the gauging of MSD with the help of DAP or CD values is not one-to-one, which is not comfortable from practical point of view. However, this might be improved by introducing the appropriate trigger level (10,11) ; here the trigger level for DAP readings indicating the MSD equal to 2 Gy and the threshold for deterministic injury (main erythema) (16) would be cgy (¼ 1 rad) cm 2. Looking at the Figures 5 7, representing the correlations for therapeutic procedures, one realises that the situation is even worse. The correlation coefficients are low and most of the points lie above the line which implies that in most cases the MSD are underestimated. Data collected in room 2 (Figures 8 and 9) illustrate low correlation between DAP value and MSD for CA (R 2 ¼ 0.61) and a relatively good 21

5 J. DOMIENIK ET AL. Figure 2. Correlation between MSD and DAP values for CA procedures in room 1. Figure 3. Correlation between MSD and CD values for CA procedures in room 2. Figure 5. Correlation between MSD and DAP values for CA þ PTCA procedures in room 1. Figure 6. Correlation between MSD and CD for CA þ PTCA procedures. Figure 4. Correlation between MSD and EFD for CA procedures in room Figure 7. Correlation between MSD and EFD for CA þ PTCA procedures in room 1.

6 PATIENT SKIN DOSE VERSUS OTHER DOSE INDICATORS Figure 8. Correlation between MSD and DAP for CA procedures in room 2. Figure 9. Correlation between MSD and DAP for CA þ PTCA procedures in room 2. correlation for PCI procedures (R 2 ¼ 0.82), which is quite opposite to the relations in room I. This confirms the fact that for every facility the individual study of correlations between beam monitoring quantities provided by the system and MSD is needed. The appropriate DAP-reading trigger level for therapeutic procedures in room II for the above proposed MSD would be 346 Gy cm 2. DISCUSSION The need for optimisation of unit performance and staff training in radiation protection, including the dose-reduction techniques, were ascertained. The other results of our paper seem to support the point of view presented by a number of authors like Vano et al. (4), van de Putte (6) and Morell et al. (9), who state in their papers that the DAP cannot predict the MSD in a reliable way. In our study, we extended the analyses to other dosimetric quantities beyond DAP-like CD or EFD showing that neither of them is satisfactory as the estimation of MSD on the level which could be accepted from the point of view of patient safety; there are too many factors which can influence the relations between this MSD and beam monitoring quantities making them rather complicated. However, the better correlation of DAP and MSD obtained in our study and, for example, in Trianni et al. (10) justify the statement that DAP can still serve, when dealt with sufficient care, as rough, preliminary, estimation of the risk experienced by the patient. The results allowed to propose trigger level for DAP readings for CA and PCI, respectively, in room I and room II. It should be stressed once more, as it was mentioned in the previous section, that the study of correlations between the system beam monitoring quantities and MSD should be performed individually in the case of every facility. REFERENCES 1. Vano, E., Arranz, L., Sastre, J. M., Moro, C., Ledo, A., Garate, M. T. and Minguez, I. Dosimetric and radiation protection considerations based on same cases of patient skin injuries in interventional cardiology. Br. J. Radiol. 71, (1998). 2. International Commission on Radiological Protection. Avoidance of radiation injuries from medical interventional procedures. ICRP Publication 85. Ann. ICRP 30(2) (2000). 3. Koenig, T. R., Wolff, D., Mettler, F. A. and Wagner, L. K. Skin injuries from fluoroscopically guided procedures. Am. J. Radiol. 177, 3 11 (2001). 4. Vano, E., Prieto, C., Fernandez, J. M., Gonzalez, L., Sabate, M. and Galvan, C. Skin dose and dose-area product values in patients undergoing intracoronary brachytherapy. Br. J. Radiol.76, (2003). 5. Waite, J. C. and Fitzgerald, M. An assessment of methods for monitoring entrance surface dose in fluoroscopically guided interventional procedures. Radiat. Prot. Dosim. 94(1/2), (2001). 6. van de Putte, S., Verhaegen, F., Taeymans, Y. and Thierens, H. Correlation of patient skin doses in cardiac interventional radiology with dose-area product. Br. J. Radiol. 73, (2000). 7. Study on local maximum skin doses to patients undergoing cardiac procedures in dedicated Polish center (submitted for publication). 8. Guibelalde, E., Vano, E., Gonzalez, L., Prieto, C., Fernandez, J. M. and Ten, J. I. Practical aspect for the evaluation of skin doses in interventional cardiology using a new slow film. Br. J. Radiol. 76, (2003). 9. Morell, R. E. and Rogers, A. T. Kodak EDR2 film for patient skin dose assessment in cardiac catheterization procedures. Br. J. Radiol. 79(943), (2006). 23

7 10. Trianni, A., Chizzola, G., Toh, H., Quai, E., Cragnolini, E., Bernardi, G., Proclemer, A. and Padovani, R. Patient skin dosimetry in haemodynamic and electrophysiology interventional cardiology. Radiat. Prot. Dosim. 117(1 3), (2005). 11. Karambatsakidou, A., Tornvall, P., Saleh, N., Chouliaras, T. L, öfberg, P.-O. and Fransson, A. Skin dose alarm levels in cardiac angiography procedures: is a single DAP value sufficient? Br. J. Radiol. 78, (2005). 12. Zoetelief, J., Schultz, F. W., Kottou, S., Gray, L., O Connor, U., Salat, D., Kepler, K., Kaplanis, P., Jankowski, J. and Schreiner, A. et al. Quality control measurements for fluoroscopy systems in eight countries participating in the Sentinel EU coordination action. Radiat. Prot. Dosim. 129, (2008). J. DOMIENIK ET AL. 13. Padovani, R., Trianni, A., Bokou, C., Bosmans, H., Jankowski, J., Kottou, S., Kepler, K., Malone, J., Tsapaki, V. and Salat, D. et al. Survey on performance assessment of cardiac angiography systems. Radiat. Prot. Dosim. 129, (2008). 14. Padovani, R., Vano, E., Trianni, A., Bokou, C., Bosmans, H., Bor, D., Jankowski, J., Torbica, P., Kepler, K. and Dowling, A. et al. Reference levels at European level for cardiac interventional procedures. Radiat. Prot. Dosim. 129, (2008). 15. Wagner, L., Archer, B. and Cohen, A. Management of patient skin dose in fluoroscopically guided interventional procedures. J. Vasc. Interv. Radiol. 11, (2000). 16. Wagner, L., Ejfel, P. and Geise, R. Potential biological effects following high x-ray dose interventional procedures. J. Vasc. Interv. Radiol. 5(1), (1994). 24