Introduction 1 When in 2004 it was decided to provide the radiologists with an additional dosimeter, the AMC had already substantial data of doses monitored on both sides of the lead apron. These data were collected at the Department of Cardiology, where additional dosimeters were introduced in 1998. Until 2004 the data collected at the Department of Cardiology established no relation between the doses measured above and those measured under the lead apron. This was thought to be caused by the fact that the two dosimeters were often transposed, that the dosimeters were not always returned at the same time, and that unmistakable instruction regarding the position of the two personal dosimeters was missing. To rule out the influence of these factors on the dose measured, it was decided to develop a special holder for the two dosimeters at the time the radiologists were provided with additional dosimeters. Moreover, it was decided to replace all personal dosimeters in the study on the same day by one person, to preclude the influence of differences in measuring periods and problems with transposed dosimeters. 2 The aim of the present study was to determine whether a relation exists between the doses measured above and those measured under the lead aprons of radiologists performing interventional procedures and whether two personal dosimeters provide additional information compared to a single dosimeter. Materials and Methods 1 2 Fig. 1 The specially designed holder fixed to the lead apron breast-high  Fig. 2 The specially designed holder with the two personal dosimeters  The doses were measured during 39 periods of 4 weeks in 3 consecutive years. At the start of the study six of the radiologists were provided with two personal dosimeters, while throughout the study another two radiologists were provided with two dosimeters. Three radiologists left the AMC before the end of the study. p 3 At the Department of Radiology the interventional procedures were performed with two Philips Integris Allura machines (Philips Medical Systems, Best, the Netherlands) with the same potential range of 50−133 kVp. During all interventional procedures the radiologists used wrap-around lead aprons (skirts and vests; Scanflex Medical AB, Täby, Sweden) with an overall lead equivalence of 0.25 mm at 100 kV. The wrap-around apron has an overlap due to which the actual thickness of the protective apron in the front is 0.5 mm lead equivalent. The doses were monitored above and under a layer of 0.25 mm lead equivalence. p Results During the study 392 dosimeters were provided to eight radiologists. All dosimeters were returned to the NRG. The doses were read and reported in 13 4-weekly dose reports. In total 196 doses were recorded above the lead apron, ranging from <0.01 to 16.78 mSv. Under the lead apron the doses ranged from <0.01 to 0.83 mSv. Five readings were excluded from the study, as the doses measured above and under the lead apron were reported to be <0.01 mSv. In 27 cases the doses measured under the lead apron were <0.01 mSv. These doses were treated as 0 mSv in the statistical analysis. 1 2 3 4 p  p p  p  p  p  Table 1 The number of measurements, mean dose, standard deviation, and percentiles (25th, 50th [median], and 75th) of the doses measured above the lead aprons of radiologists 1 to 8 Radiologist no. N Mean (mSv) SD 25th (mSv) 50th (mSv) 75th (mSv) Minimum (mSv) Maximum (mSv) 1 4 3.41 1.22 2.16 3.58 4.48 1.88 4.58 2 34 3.67 3.21 1.32 3.15 5.26 0.03 16.78 3 9 2.33 1.24 1.38 2.17 3.59 0.28 4.00 4 9 3.65 2.61 1.76 3.56 4.72 0.35 9.24 a 38 1.79 1.09 0.88 1.70 2.29 0.14 5.40 b 39 3.68 2.37 2.25 3.26 4.18 0.13 11.52 b 21 3.66 1.71 2.49 3.08 4.57 0.66 6.94 b 37 4.75 2.31 3.10 4.87 6.13 1.05 13.23 Total 191 3.44 2.41 1.74 3.05 4.61 0.03 16.78 Note p Table 2 The number of measurements, mean dose, standard deviation, and percentiles (25th, 50th [median], and 75th) of the doses measured under the lead aprons of radiologists 1 to 8 Radiologist no. N Mean (mSv) SD 25th (mSv) 50th (mSv) 75th (mSv) Minimum (mSv) Maximum (mSv) 1 4 0.18 0.09 0.10 0.17 0.27 0.09 0.29 2 34 0.13 0.15 0.03 0.11 0.17 0.00 0.83 3 9 0.06 0.06 0.01 0.07 0.11 0.00 0.15 4 9 0.14 0.09 0.06 0.15 0.21 0.00 0.24 a 38 0.07 0.09 0.00 0.06 0.09 0.00 0.49 6 39 0.12 0.08 0.06 0.11 0.19 0.00 0.37 a 21 0.20 0.12 0.12 0.18 0.27 0.02 0.50 8 37 0.13 0.12 0.05 0.12 0.18 0.00 0.66 Total 191 0.12 0.11 0.05 0.1 0.17 0.00 0.83 a p  Fig. 3 Boxplot of the doses (mSv) measured above the lead apron. The black line in the red box marks the median, the box demarks the 25th and 75th percentiles, and the whiskers mark the minimum and maximum observed values that are not statistical outliers. Extreme values are marked with an asterisk. The extreme values are more than three box lengths from the upper edge of the box. The box length is the interquartile range  Fig. 4 Boxplot of the doses (mSv) measured under the lead apron. The black line in the red box marks the median, the box demarks the 25th and 75th percentiles, and the whiskers mark the minimum and maximum observed values that are not statistical outliers. Extreme values are marked by an asterisk. The extreme values are more than three box lengths from the upper edge of the box. The box length is the interquartile range  5 p R 2 Y X Y X p  6 Fig. 5 X Y  Fig. 6 Distribution of the residuals (mean = 0, standard deviation = 1)  p R 2 Discussion The dose measurements above and under the lead apron established that the occupational exposure of the radiologists performing interventional procedures in the AMC is comparable. There was only one exception: radiologist 5. As radiologist 5 performed fewer procedures than his colleagues, his exposure was expected to be lower. It was, however, unknown what reduction in doses was to be expected, as data on the number and the length of procedures were not collected for the study. Despite this, it is likely that the lower exposure of radiologist 5 was caused by a lower workload, as the doses above and under the lead apron decreased proportionally compared to the other radiologists’: 46% and 50%. 4 5 6 7 In the studies by Williams et al. and Nicklason et al. the doses under the lead apron were also measured. Williams et al. found that the average monthly dose under the lead apron ranged from 0 to 0.48 mSv. As for the results above the lead apron, Williams et al. reported that the highest average dose was caused by the differences in the nature of work of one of the radiologists. The average monthly dose under the lead apron of all other radiologists (the radiologist with the highest dose was excluded) was 0.08 mSv. In the present study the doses under the lead apron varied between 0.06 and 0.2 mSv in a 4-week period. The average dose, 0.12 mSv, was higher than the average dose reported by Williams et al. The differences between the two studies are possibly caused by the differences in thicknesses of the lead aprons between them. Williams et al. reported that the doses were measured under lead aprons of 0.35 and 0.5 mm, while in the present study doses were measured under a layer of 0.25 mm. Niklason et al. measured doses under lead aprons of 0.25 and 0.5 mm. As for the doses measured above the lead apron, the doses under the lead apron were annual doses. The average annual dose under the lead apron reported by Niklason et al. was 0.88 mSv, while in the present study the average annual dose was 1.6 mSv (median, 1.3 mSv). As for the results reported by Williams et al., the differences in doses were probably caused by the variation in thickness of the lead aprons worn by the radiologists. In the study reported by Williams et al., the radiologists ensured that the personal dosimeters were worn in all cases. However, the personal dosimeters under the lead aprons were worn not only during interventional procedures but also during other radiology procedures. The authors described that the dedication of the radiologists to wearing additional dosimeters under the lead apron was probably less consistent than for the other dosimeters. In the present study the dedication of the radiologists was not tested, as the dosimeters were simultaneously worn in the special holder. Moreover, the dosimeters in this study were worn exclusively during interventional procedures. The duration of the measurements in the study by Niklason et al. was approximately 2 months, while in the present study the doses of the majority of the radiologists were measured during 3 consecutive years. Y X 8 9 5 10 Y X X 11 12 In the present study the doses under the lead apron were determined almost at the same place as the doses above the lead apron. This was possible as a special holder was used. Before the special holder was introduced it was not possible to derive a relation between the doses above and those under the lead apron. Conclusion The aim of the present study was to determine whether a relation exists between the doses measured above the lead apron and those measured under the lead apron of radiologists performing interventional procedures and whether two personal dosimeters provide additional information compared to a single dosimeter. It might be concluded that a linear relation between the measurements above and those under the lead apron is proved in this study. With this, there is no evidence for a more accurate estimation of the effective dose when an additional dosimeter is used. As the results of this study were derived from measurements by radiologists performing interventional procedures, it is not known whether a factor of 0.036 is applicable for other specialists performing interventional procedures, such as interventional cardiologists and electrophysiologists. For this reason, more investigation is needed to derive correction factors for other physicians performing interventional procedures.