Pe d i a t r i c I m a g i n g • O r i g i n a l R e s e a r c h Strauss et al. DRLs for Fluoroscopically Guided Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Pediatric Imaging Original Research
Estimates of Diagnostic Reference Levels for Pediatric Peripheral and Abdominal Fluoroscopically Guided Procedures Keith J. Strauss1 John M. Racadio1 Neil Johnson1 Manish Patel1 Rami A. Nachabe2 Strauss KJ, Racadio JM, Johnson N, Patel M, Nachabe RA Keywords: diagnostic reference level, pediatric, peripheral and abdominal angiography, radiation dose DOI:10.2214/AJR.14.13630 Received August 12, 2014; accepted after revision September 28, 2014. The Department of Radiology, Division of Interventional Radiology, Cincinnati Children’s Hospital Medical Center has a master research agreement with Philips Healthcare. K. J. Strauss has had his travel expenses paid to perform research and development activities for Philips Healthcare and has performed paid medical physics consulting services for Philips Healthcare. J. M. Racadio has had his travel expenses paid to participate in Phillips Healthcare– sponsored symposia. R. A. Nachabe is an employee of Philips Healthcare; however, the authors who are not Philips employees had full control of inclusion of any data and information that might present a conflict of interest for the author who is a Philips employee. Presented at the 2014 annual meeting of the Radiological Society of North America, Chicago, IL. 1 Department of Radiology, University of Cincinnati School of Medicine, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave, MLC 5031, Cincinnati, OH 45229-3026. Address correspondence to K. J. Strauss ([email protected]). 2 Interventional X-Ray Department, Philips Healthcare, Best, The Netherlands.
WEB This is a web exclusive article. AJR 2015; 204:W713–W719 0361–803X/15/2046–W713 © American Roentgen Ray Society
OBJECTIVE. The objective of our study was to survey radiation dose indexes of pediatric peripheral and abdominal fluoroscopically guided procedures from which estimates of diagnostic reference levels (DRLs) can be proposed for both a standard fluoroscope and a novel fluoroscope with advanced image processing and lower radiation dose rates. MATERIALS AND METHODS. Radiation dose structured reports were retrospectively collected for 408 clinical pediatric cases: Half of the procedures were performed with a standard imaging technology and half with a novel x-ray technology. Dose-area product (DAP), air Kerma (AK), fluoroscopy time, number of digital subtraction angiography images, and patient mass were collected to calculate and normalize radiation dose indexes for procedures completed with the standard and novel fluoroscopes. RESULTS. The study population was composed of 180 and 175 patients who underwent procedures with the standard and novel technology, respectively. The 21 different types of pediatric peripheral and abdominal interventional procedures produced 408 total studies. Median ages, mass and body mass index, fluoroscopy time per procedure, and total number of recorded images for the standard and novel technologies were not statistically different. The area of the x-ray beams was square at the level of the patient with a dimension of 10–13 cm. The dose reduction achieved with the novel fluoroscope ranged from 18% to 51% of the dose required with the standard fluoroscope. The median DAP and AK patient dose indexes were 0.38 Gy ⋅ cm2 and 4.00 mGy, respectively, for the novel fluoroscope. CONCLUSION. Estimates of dose indexes of pediatric peripheral and abdominal fluoroscopically guided, clinical procedures should assist in the development of DRLs to foster management of radiation doses of pediatric patients.
T
he Image Gently campaign urges careful management of the pediatric patient’s radiation dose from any imaging procedure that uses ionizing radiation, including complex angiographic studies [1]. Radiation management in the pediatric interventional fluoroscopic suite is a multifaceted process that needs to occur for each individual patient [2– 5]. The patient dose at the end of a clinical procedure can be estimated and compared with established standard-dose guidelines within the department as a function of the type of study and patient size. Estimated patient doses that exceed departmental standards should be investigated to ensure that appropriate steps to manage patient dose were followed throughout the procedure [6]. Departmental standard-dose guidelines are typically adapted from diagnostic reference levels (DRLs), which are based on na-
tional survey results of estimated patient doses during clinical studies [7]. Over the past decade, 13 published studies on the estimated doses during adult vascular angiographic procedures, the majority from neurovascular laboratories, were identified in the literature [8–20]. During the same time period, the estimated doses surveyed at seven pediatric cardiac catheterization departments were reported [21–27], and three different groups reported estimated patient doses during pediatric neurointerventions [28–30]. However, information on the estimated doses during pediatric peripheral and abdominal fluoroscopically guided studies from which appropriate DRLs could be established could not be identified in the published literature. We hypothesized that a novel image-processing technology capable of higher image quality at lower patient doses might reduce the required radiation dose during fluoro-
AJR:204, June 2015 W713
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Strauss et al. scopically guided studies of the trunk and limbs of pediatric patients ranging in age from neonates to 21 years. The purpose of this investigation was to survey the radiation dose levels during pediatric peripheral and abdominal fluoroscopically guided procedures at a tertiary care pediatric hospital. Data were collected for both a standard fluoroscopic unit (technology introduced in 2003 with typical patient dose rates) and a novel fluoroscopic unit (technology available in the United States in 2013) with advanced image processing and lower patient radiation dose rates. The surveyed radiation dose levels should assist in the development of DRLs, which foster management of radiation dose rates of pediatric patients during interventional fluoroscopic procedures. Materials and Methods Study Patients and Data Collection The institutional review board of Cincinnati Children’s Hospital Medical Center approved the collection of clinical data for this retrospective study. Patient informed consent was waived; however, patient confidentiality was secured according to HIPAA guidelines. Dose indexes from 204 interventional radiology (IR) procedures performed between September 2013 and January 2014 using a standard technology system were collected. All procedures performed using the department’s two standard fluoroscopes during this 5-month period except neurointerventional procedures were included. Dose indexes from an equal number of IR procedures (excluding neurointerventions) performed between December 2013 and March 2014 using a novel technology were collected. The novel technology fluoroscope (commissioned in December 2013) replaced one of the department’s standard technology units. The department’s three pediatric IR operators performed the 408 procedures. Radiation dose structured reports (RDSRs) generated by the fluoroscopes were collected via a networked third-party radiation dose database server (PEMNET, Clinical Microsystems). The RDSRs contain radiation dose values such as air Kerma (AK) at the interventional reference point, dose-area product (DAP), fluoroscopy time, and total number of images created for each radiation acquisition (e.g., fluoroscopy, digital subtraction angiography [DSA], rotational acquisitions, and cone beam CT acquisition). In addition, the patient medical record number, patient demographics (e.g., age, sex, mass, height, date of birth), date of the procedure, and study type were available. The accuracy of the DAP and AK values, which are indexes of estimates of the patient’s radiation
W714
dose, that are displayed on the fluoroscopes was verified with a calibrated ionization chamber and electrometer by the department’s staff medical physicist. A standard calibration protocol developed by the American Association of Physicists in Medicine [31] to measure the discrepancy between the displayed values on the fluoroscope and the calibrated ionization chamber positioned at the interventional reference point was used by the medical physicist. The measured correction factors, 1.03 and 1.05 for the standard and novel units, respectively, were applied to all dose indexes displayed on the units and were recorded in each unit’s RDSRs.
Imaging Technology This study was completed using two C-arm imaging systems: a novel advanced image-processing and dose reduction technology (AlluraClarity, Philips Healthcare), which was approved by the U.S. Food and Drug Administration for sale in the United States in 2013, and a standard technology (Allura Xper, Philips Healthcare), which has been available in the United States since 2003. The novel system uses real-time image noise reduction algorithms driven by parallel processing hardware to improve image quality in real time. This improved image quality allows the manufacturer to reduce patient entrance dose with changes to configuration settings [10, 32]. The added beam filter, reduced pulse width, smaller focal spot size, managed x-ray tube voltage and tube current, and
reduced detector AK differ appropriately as a function of the body region to be imaged and the physical size of the pediatric patient. Therefore, the configuration of the acquisition parameters on the novel system reduces patient entrance dose while maintaining image quality at levels similar to the standard technology [31].
Data Analysis Procedural dose indexes, expressed as both AK in units of mGy at the interventional reference point and DAP in units of Gy · cm2, were obtained from the RDSR. Indicators of the complexity of an interventional fluoroscopic procedure are total fluoroscopy time and total number of recorded DSA images [33]. To compress the distribution of AK and DAP due to the varying complexities of the 408 procedures, fluoroscopy and fluorography radiation doses of individual cases were normalized by the fluoroscopy time and the number of recorded DSA images for each procedure, respectively, to provide four rates: DAP/min and AK/min for fluoroscopy; and DAP/number of DSA images and AK/number of DSA images for fluorography. To further reduce the distribution of the dose rate indexes due to varying thicknesses of the body part imaged, the rates calculated in the previous paragraph were normalized by the patient’s mass [27] to provide DAP/min ⋅ kg, AK/min ⋅ kg, DAP/number of DSA images · kg, and AK/number of DSA images ⋅ kg.
TABLE 1: Patients’ Characteristics and Baseline Standard X-Ray Technology (n = 204 Procedures)
Novel X-Ray Technology (n = 204 Procedures)
p
180
175
NA
12.6 [5.5–17.2]a
11.6 [3.4–15.9]a
0.07, 0.17, 0.39b
90:114
113:91
0.03
Mass (kg)
39.3 [19.2–60.6]a
35.9 [15.1–57.9]a
0.27, 0.58, 0.39b
Body mass indexc
18.2 [15.6–22.0]a
18.3 [16.4–22.3]a
0.21, 0.14, 0.10b
Characteristics No. of patients Age (y) Male-female ratio
No. of examinations per patient (no. of patients)
0.69
1
163
155
2
11
15
3
5
3
4
1
1
5
0
0
6
0
1
37
34
No. of procedures performed in 31 patients who were treated with both technologies
0.79
Note—NA = not applicable. aValues are median [first quartile–third quartile]. bThe p values are listed for median, variance, and distribution, respectively. cWeight in kilograms divided by the square of height in meters.
AJR:204, June 2015
DRLs for Fluoroscopically Guided Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
TABLE 2: Procedure Types Standard X-Ray Technology (n = 204 Procedures)
Novel X-Ray Technology (n = 204 Procedures)
Angiography or venography
6
5
Pyelography
1
4
Central line injection
6
2
Biliary or percutaneous transhepatic cholangiography
2
3
Biopsy
2
2
Procedure
Chest tube placement
11
8
Embolization
1
1
Direct sclerosing
27
18
Drain placement
8
9
ERCP
2
10
Esophageal dilatation
25
15
Fistulography
5
14
Gastrojejunostomy
20
18
Nasogastric or nasojejunal tube placement
17
12
Peripherally inserted central catheter
7
5
Steroid injection
17
23
Lymphangiography
1
1
Vascular access
15
13
Arthrography
14
24
Lumbar Puncture
11
11
Nephrostomy
6
6
release R2011a), and a p value of < 0.05 was considered statistically significant.
Note—Data are number of procedures.
The following equations allow the estimate of total procedure (TP), fluoroscopy (FS), or fluorography (FG) DAP or AK for an individual patient from the normalized estimates of DAP and AK. In each of the following three equations, substitute either DAP or AK for X: X / (min ⋅ kg) × FT × PM = FS X
X / (DSA ⋅ kg) × DSA × PM = FG X TP X = FS X + FG X
(1), (2), (3),
where min ⋅ kg indicates units of the estimated fluorography DRL, FT is fluoroscopy time in minutes, PM is patient mass in kilograms, DSA ⋅ kg are units of the estimated fluoroscopy DRL, and DSA is the number of DSA images. Finally, division of DAP by AK estimates the area of the x-ray beam at the isocenter of the fluoroscope.
Statistical Analysis The Pearson chi-square test was applied to categoric variables such as sex, clinical study type, number of procedures per operator, and patient inter- and
intramodality replicates. Two-sample Mann-Whitney U, Levene’s, and Kolmogorov-Smirnov tests were applied to compare medians, variances, and distributions. The variables such as age, mass, body mass index (BMI [defined as weight in kilograms divided by the square of height in meters]), procedure dose, fluoroscopy time, number of acquired frames, fluoroscopy dose rate, and acquisition dose rate were investigated with the aforementioned three tests. Given the nongaussian distribution of the collected data, variables are reported as medians along with the first and third quartiles. Statistical calculations were performed using statistics software (Matlab, MathWorks, version 7.12,
Results The age, mass, and BMI of patients did not show any statistical difference in median, variance, and distribution for both technologies (Table 1). The fraction of males in the standard group is smaller (44%) than that in the novel group (55%). The total number of patients treated with the standard technology and the novel technology showed no statistical difference (p = 0.69) in intramodality replicates (Table 1). Thirty-one patients were treated with both technologies (37 and 34 procedures with the standard and novel technology, respectively) showing no statistical difference in intermodality number of repeated patients (p = 0.79) (Table 1). Twenty-one different types of interventional procedures were performed (Table 2). Despite the variety of clinical procedures, the proportion of clinical study types performed with both technologies was not statistically significant (p = 0.22). Table 3 summarizes the distribution of the number of procedures among the three operators. This table also lists the distribution of the procedure AK per operator. There is no statistical difference between the distribution of the procedures among the operators (p = 0.06). A significant procedure dose reduction (p < 0.05) between the two technologies was observed for each of the three operators. No statistical differences in the median, variance, and distribution were observed in fluoroscopy time and number of acquired DSA images for both technologies (Table 4). DSA acquisitions were performed in 35 and 31 procedures with the standard and novel technology, respectively. Thus, no statistical difference exists in the number of procedures with DSA in both groups (p = 0.69). This statistical analysis validates the comparison of the two types of fluoroscopes. Table 4 lists the median, first quartile, and third quartile of the surveyed dose indexes
TABLE 3: Distribution of Procedure Air Kerma (AK) Values Among Operators Standard X-Ray Technology (n = 204 Procedures) Operator No.
Novel X-Ray Technology (n = 204 Procedures)
AK (mGy)a
No. of Procedures
AK (mGy)a
No. of Procedures
Operator 1
15 [7.32–42.07]
54
5 [2.25–23.25]
43
Operator 2
10 [4–22]
72
2 [1–5]
96
Operator 3
13 [6–62]
78
7 [3–33.5]
65
aValues are median [first quartile–third quartile].
AJR:204, June 2015 W715
TABLE 4: Procedure Dose and Acquisition Characteristics Strauss et al. Standard X-Ray Novel X-Ray Technology Technology (n = 204 Procedures)a (n = 204 Procedures)a
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Character Fluoroscopy time (min)
p Median
Variance Distribution
1.6 [0.8–3.4]
1.5 [0.7–3.7]
0.53
0.77
0.54
2 [1–8]
2 [1–9]
0.62
0.49
0.91
1.09 [0.46–4.89]
0.38 [0.17–1.45]
< 0.0001
0.10
< 0.0001
No. of DSA images DAP (Gy ⋅ cm2) Procedure Fluoroscopy
1.02 [0.45–4.77]
0.38 [0.17–1.41]
< 0.0001
0.14
< 0.0001
Digital acquisition
0.43 [0.20–4.06]
0.13 [0.02–1.25]
< 0.05
0.25
< 0.05
0.77 [0.35–1.72]
0.30 [0.15–0.77]
< 0.0001
0.09
< 0.0001
0.26 [0.14–0.46]
0.03 [0.01–0.14]
< 0.0001
0.06
< 0.0001
Fluoroscopy DAP per fluoroscopy time (Gy ⋅
cm2 /min)
Digital acquisition DAP per frame (Gy ⋅ cm2 /frame) Fluoroscopy DAP per fluoroscopy time and patient mass (Gy ⋅ cm2 /min ⋅ kg)
0.027 [0.016–0.047]
0.015 [0.007–0.024]
< 0.0001
0.17
< 0.0001
Digital acquisition DAP per frame and patient mass (Gy ⋅ cm2 /frame ⋅ kg)
0.005 [0.004–0.012]
0.002 [0.001–0.003]
< 0.0001
< 0.0001
< 0.0001
Procedure
12.00 [5.50–33.26]
4.00 [2.00–12.00]
< 0.0001
0.18
< 0.0001
Fluoroscopy
12.00 [5.15–32.00]
3.98 [2.00–10.60]
< 0.0001
0.25
< 0.0001
AK (mGy)
Digital acquisition
3.92 [1.47–20.66]
1.11 [0.19–6.18]
< 0.005
0.22
< 0.01
6.54 [4.00–13.44]
2.81 [1.70–5.03]
< 0.0001
< 0.01
< 0.0001
Digital acquisition AK per frame (mGy/frame)
1.84 [1.07–3.37]
0.35 [0.09–0.62]
< 0.0001
< 0.05
< 0.0001
Fluoroscopy AK per fluoroscopy time and patient mass (mGy/min ⋅ kg)
0.27 [0.15–0.42]
0.10 [0.06–0.17]
< 0.0001
< 0.005
< 0.0001
Digital acquisition AK per frame and patient mass (mGy/frame ⋅ kg)
0.05 [0.03–0.07]
0.01 [0.01–0.02]
< 0.0001
< 0.05
< 0.0001
Fluoroscopy AK per fluoroscopy time (mGy/min)
X-ray beam area at isocenter (cm2) Procedure
105 [68–161]
112 [74–190]
0.04
0.44
0.22
Fluoroscopy
104 [68–161]
112 [75–197]
0.03
0.43
0.14
Fluorography
144 [82–217]
167 [115–263]
0.16
0.56
0.15
Note—DSA = digital subtraction angiography, DAP = dose-area product, AK = air Kerma. aValues are median [first quartile–third quartile].
TABLE 5: Estimates of Diagnostic Reference Levels (DRLs) and Comparison of Dose Between Standard and Novel Systems Value (Third Quartile) Standard X-Ray Technology (n = 204 Procedures)
DRLs
Novel X-Ray Technology (n = 204 Procedures)
Dose Fraction Achieved With Novel Technology Compared With Standard Technology
Procedure DAP (Gy ⋅ cm2)
4.89
1.45
0.30
Fluoroscopy DAP per fluoroscopy time (Gy ⋅ cm2 /min)
1.72
0.77
0. 45
Digital acquisition DAP per frame (Gy ⋅ cm2 /frame)
0.46
0.14
0. 30
Fluoroscopy DAP per fluoroscopy time and patient mass (Gy ⋅ cm2 /min ⋅ kg)
0.047
0.024
0. 51
Digital acquisition DAP per frame and patient mass (Gy ⋅ cm2 /frame ⋅ kg)
0.012
0.003
0. 25
Procedure AK (mGy)
33.26
12.00
0. 36
Fluoroscopy AK per fluoroscopy time (mGy/min)
13.44
5.03
0. 37
Digital acquisition AK per frame (mGy/frame)
3.37
0.62
0. 18
Fluoroscopy AK per fluoroscopy time and patient mass (mGy/min ⋅ kg)
0.42
0.17
0. 40
Digital acquisition AK per frame and patient mass (mGy/frame ⋅ kg)
0.07
0.02
0. 29
Note—DAP = dose-area product, AK = air Kerma.
of the standard and novel fluoroscopes. The x-ray beam area at the isocenter of the fluoroscope is listed at the end of Table 4. The small differences in these estimated areas (whole procedure, fluoroscopy only, or fluorography only) expressed in units of cm2 rep-
W716
resent square x-ray field areas with a dimension that ranges from 10 to 13 cm. Estimates of DRLs and the third quartile of the various dose distributions in Table 4 are listed in the second and third columns of Table 5 for the standard and novel units, re-
spectively. Figures 1–6 illustrate the distributions of the patient dose indexes with both bar charts and boxplots. Estimates of DRLs are provided for both DAP and AK. The total procedure DRL is broken down into dose due to fluoroscopy and fluorography. These
AJR:204, June 2015
Discussion This study provides surveyed estimates of patient dose indexes of clinical pediatric abdominal and peripheral fluoroscopically guided studies from which DRLs normalized for the complexity of the procedure and for the size of the pediatric patient are estimated. These estimates of DRLs are provided for both a standard technology fluoroscope and a novel technology fluoroscope. The area of the x-ray beam at the isocenter of each fluoroscope was also calculated for both fluoroscopic and fluorographic operational modes. The estimated procedure DRLs are for an entire study and are expressed as procedure DAP (line 1 in Table 5) and procedure AK (line 6 in Table 5). These estimates are based on data with a broad distribution due to varying complexities of the procedures and the varying masses of the patients. These estimates are representative for only the median-complexity procedure and the median-sized patient. Adjusting the normalized DAP and AK values in lines 2–5 and 7–10, respectively, in Table 5 by each individual patient’s examination complexity (fluoroscopy time and DSA images) and mass allows a more accurate estimate of an individual patient’s dose indexes. The equations in the Materials and Methods, the normalized values of DAP and AK in Table 5, and the patient’s mass, study fluoroscopy time, and number of DSA images obtained during the study allow a more accurate estimate of an individual patient’s DAP and AK. Estimates of DRLs for both DAP and AK are necessary. Although AK assists in predicting the risk of deterministic radiation injuries (e.g., skin damage), DAP is the better choice for predicting the risk of stochastic radiation injuries (e.g., cancer) [6]. Older fluoroscopes may display DAP but not AK. An estimate of the area of the x-ray beam at the level of the patient, listed at the end of Ta-
No. of Procedures
80
Standard imaging technology Novel imaging technology
60 40 20 0 0
1
2
3
4
0
1
2
3
4
5
6
5 6 DAP (Gy · cm2)
7
8
9
10
11
7
8
9
10
11
Fig. 1—Bar graph and boxplot show distribution of procedure dose-area product (DAP) values by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 40 No. of Procedures
breakdowns are presented normalized by the complexity of the procedure (i.e., fluoroscopy time or number of DSA images) and by both the complexity of the procedure and the size of the patient (i.e., fluoroscopy time or number of DSA images × patient mass [kg]). The fraction of dose achieved with the novel fluoroscope ranged from 0.18 to 0.51 of the dose used with the standard fluoroscope depending on the dose index used for the calculation, as listed in the last column of Table 5. The fraction of dose achieved between the two systems during digital acquisitions always exceeds the dose reduction achieved during fluoroscopy.
Standard imaging technology Novel imaging technology
30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50 AK (mGy)
60
70
80
90
100
Fig. 2—Bar graph and boxplot show distribution of procedure air Kerma (AK) values by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 80 No. of Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
DRLs for Fluoroscopically Guided Procedures
Standard imaging technology Novel imaging technology
60 40 20 0 0
1
2
3
0
1
2
3
4
5
6
4 5 6 DAP Rate (Gy · cm2/min)
7
8
9
10
7
8
9
10
Fig. 3—Bar graph and boxplot show distribution of fluoroscopy dose-area product (DAP) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
AJR:204, June 2015 W717
Strauss et al.
No. of Procedures
Standard imaging technology Novel imaging technology
10 8 6 4 2 0 0
0.5
0
0.5
DAP Rate (Gy · cm2/frame)
1.0
1.5
1.0
1.5
Fig. 4—Bar graph and boxplot show distribution of digital acquisition dose-area product (DAP) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
No. of Procedures
50
Standard imaging technology Novel imaging technology
40 30 20 10 0 0
5
10
0
5
10
15
20
15 20 AK Rate (mGy/min)
25
30
35
25
30
35
Fig. 5—Bar graph and boxplot show distribution of fluoroscopy air Kerma (AK) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 15 No. of Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
12
Standard imaging technology Novel imaging technology
10
5
0 0
2
4
6
0
2
4
6
8
10
12
14
16
18
20
8 10 12 AK Rate (mGy/frame)
14
16
18
20
Fig. 6—Bar graph and boxplot show distribution of digital acquisition air Kerma (AK) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
W718
ble 4, allows estimation of AK at the interventional reference point if only DAP is displayed by the fluoroscope. Estimates of DRLs for both a standard fluoroscope and a novel fluoroscope provide two distinct reference levels, which may be helpful for a department with older, newer, or both types of fluoroscopes. The standard DRLs may provide guidance concerning expected doses from fluoroscopes in the field in the second half of their useful life, whereas the DRLs of the novel unit may provide guidance for expected doses from state-ofthe-art fluoroscopes recently installed. Estimates of DRLs for both fluoroscopy and fluorography may assist with a root analysis of high patient doses at the end of a study. For example, the patient’s dose from fluorography may be appropriate, but the fluoroscopy component of the study may be high. This difference in dose could be due to inefficient technique by the fluoroscopist or poor configuration of the fluoroscopic mode. If the situation is reversed—appropriate dose for fluoroscopy but excessive dose for fluorography, then the number of recorded radiographic images may be excessive or the dose at the image receptor for each recorded image may be excessive. DRLs have not been previously published for pediatric peripheral and abdominal angiography to our knowledge. However, two different studies of adult neuroangiographic procedures [8, 10] and one study of adult electrophysiology procedures [11] compare the doses required by standard and novel fluoroscopes similar to the fluoroscopes used in this study. These studies listed reductions in patient dose with the novel unit that were 35% [8], 25% [10], and 58% [11] of the dose required by the standard system. We report a comparable dose fraction of 34% in our study in comparison with these previous studies. Because our study had the availability of RDSR data collection, our study provides a larger variety of estimates of DRLs than the three previous studies. Although our study provides data suitable for assisting in the development of DRLs for pediatric peripheral and abdominal fluoroscopically guided procedures, several limitations remain. First, one company manufactured both fluoroscopes used in this study. The degree of technology advancement achieved by this company may not be representative of other imaging companies. Second, the estimates of DRLs for fluorography are based on a limited number of recorded DSA images
AJR:204, June 2015
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
DRLs for Fluoroscopically Guided Procedures because fluoroscopy stored images are substituted for DSA images whenever possible during pediatric studies to reduce patient dose. Third, the limited number of procedures in this study prevents more accurate estimates of DRLs for individual types of studies or for different groups of patients of similar sizes. More studies are needed to improve the accuracy of pediatric DRLs in interventional fluoroscopy. A larger sample size would allow patients to be grouped by the type of study and by a limited range of patient sizes. A study with more than four operators and a large sample size would allow investigation of the influence of individual operators on patient doses. Data should also be collected for all the major manufacturers of interventional fluoroscopes. In conclusion, this study provides data suitable for assisting in the development of DRLs for pediatric peripheral and abdominal fluoroscopically guided procedures. The normalized estimates of DRL allow calculation of more accurate estimates of dose for an individual patient than a single median DRL for all procedures and patients. These estimates of DRL should help departments develop pediatric dose guidelines. The double set of estimates of DRLs—one for a standard fluoroscope in the marketplace for more than 10 years and one for a fluoroscope with advanced technology relatively new to the marketplace—should help departments manage their pediatric radiation doses as the currently installed equipment is replaced with new units with the latest technology. Acknowledgments We thank Joanne Lovelace for her skillful help with references and the management of the preparation of this manuscript and Melinda Dighe for her help in collecting the subject’s mass. References 1. Sidhu M, Coley BD, Goske MJ, et al. Image Gently, Step Lightly: increasing radiation dose awareness in pediatric interventional radiology. Pediatr Radiol 2009; 39:1135–1138 2. Frush D, Denham CR, Goske MJ, et al. Radiation protection and dose monitoring in medical imaging: a journey from awareness, through accountability, ability and action...but where will we arrive? J Patient Saf 2013; 9:232–238 3. Miller DL, Balter S, Schueler BA, Wagner LK, Strauss KJ, Vano E. Clinical radiation management for fluoroscopically guided interventional procedures. Radiology 2010; 257:321–332 4. Miller DL. Efforts to optimize radiation protection in interventional fluoroscopy. Health Phys
2013; 105:435–444 5. Schueler BA, Balter S, Miller DL. Radiation protection tools in interventional radiology. J Am Coll Radiol 2012; 9:844–845 6. National Council on Radiation Protection and Measurements. Radiation dose management for fluoroscopy-guided interventional medical procedures. Bethesda, MD: NCRP, 2010: Report 168 7. [No authors listed]. Radiological protection and safety in medicine: a report of the International Commission on Radiological Protection. Ann ICRP 1996; 26:1–47 [Erratum in Ann ICRP 1997; 27:61] 8. Söderman M, Mauti M, Boon S, et al. Radiation dose in neuroangiography using image noise reduction technology: a population study based on 614 patients. Neuroradiology 2013; 55:1365–1372 9. Chida K, Ohno T, Kakizaki S, et al. Radiation dose to the pediatric cardiac catheterization and intervention patient. AJR 2010; 195:1175–1179 10. Söderman M, Holmin S, Andersson T, Palmgren C, Babic D, Hoornaert B. Image noise reduction algorithm for digital subtraction angiography: clinical results. Radiology 2013; 269:553–560 11. Dekker LR, van der Voort PH, Simmers TA, et al. New image processing and noise reduction technology allows reduction of radiation exposure in complex electrophysiologic interventions while maintaining optimal image quality: a randomized clinical trial. Heart Rhythm 2013; 10:1678–1682 12. Sanchez RM, Vano E, Fernández JM, Moreu M, Lopez-Ibor L. Brain radiation doses to patients in an interventional neuroradiology laboratory. AJNR 2014; 35:1276–1280 13. D’Ercole L, Thyrion FZ, Bocchiola M, Mantovani L, Klersy C. Proposed local diagnostic reference levels in angiography and interventional neuroradiology and a preliminary analysis according to the complexity of the procedures. Phys Med 2012; 28:61–70 14. Alexander MD, Oliff MC, Olorunsola OG, BrusRamer M, Nickoloff EL, Meyers PM. Patient radiation exposure during diagnostic and therapeutic interventional neuroradiology procedures. J Neurointerv Surg 2010; 2:6–10 15. Suzuki S, Furui S, Matsumaru Y, et al. Patient skin dose during neuroembolization by multiplepoint measurement using a radiosensitive indicator. AJNR 2008; 29:1076–1081 16. D’Ercole L, Mantovani L, Thyrion FZ, et al. A study on maximum skin dose in cerebral embolization procedures. AJNR 2007; 28:503–507 17. Thierry-Chef I, Simon SL, Miller DL. Radiation dose and cancer risk among pediatric patients undergoing interventional neuroradiology procedures. Pediatr Radiol 2006; 36(suppl 2):159–162 18. Bor D, Cekirge S, Türkay T, et al. Patient and staff doses in interventional neuroradiology. Radiat Prot Dosimetry 2005; 117:62–68 19. Struelens L, Vanhavere F, Bosmans H, Van Loon R, Mol H. Skin dose measurements on patients for diag-
nostic and interventional neuroradiology: a multicentre study. Radiat Prot Dosimetry 2005; 114:143–146 20. Pitton MB, Kloeckner R, Schneider J, Ruckes C, Bersch A, Duber C. Radiation exposure in vascular angiographic procedures. J Vasc Interv Radiol 2012; 23:1487–1495 21. Glatz AC, Patel A, Zhu X, et al. Patient radiation exposure in a modern, large-volume, pediatric cardiac catheterization laboratory. Pediatr Cardiol 2014; 35:870–878 22. McFadden SH, D’Helft CI, McGee A, et al. The establishment of local diagnostic reference levels for paediatric interventional cardiology. Radiography 2013; 19:295–301 23. Verghese GR, McElhinney DB, Strauss KJ, Bergersen L. Characterization of radiation exposure and effect of a radiation monitoring policy in a large volume pediatric cardiac catheterization lab. Catheter Cardiovasc Interv 2012; 79:294–301 24. Dragusin O, Gewillig M, Desmet W, Smans K, Struelens L, Bosmans H. Radiation dose survey in a paediatric cardiac catheterisation laboratory equipped with flat-panel detectors. Radiat Prot Dosimetry 2008; 129:91–95 25. Al-Haj AN, Lobriguito AM, Rafeh W. Variation in radiation doses in paediatric cardiac catheterisation procedures. Radiat Prot Dosimetry 2008; 129:173–178 26. Martinez LC, Vano E, Gutierrez F, Rodriguez C, Gilarranz R, Manzanas MJ. Patient doses from fluoroscopically guided cardiac procedures in pediatrics. Phys Med Biol 2007; 52:4749–4759 27. Onnasch DG, Schroder FK, Fischer G, Kramer HH. Diagnostic reference levels and effective dose in paediatric cardiac catheterization. Br J Radiol 2007; 80:177–185 28. Orbach DB, Stamoulis C, Strauss KJ, et al. Neurointerventions in children: radiation exposure and its import. AJNR 2014; 35:650–656 29. Thierry-Chef I, Simon SL, Land CE, Miller DL. Radiation dose to the brain and subsequent risk of developing brain tumors in pediatric patients undergoing interventional neuroradiology procedures. Radiat Res 2008; 170:553–565 30. Swoboda NA, Armstrong DG, Smith J, Charkot E, Connolly BL. Pediatric patient surface doses in neuroangiography. Pediatr Radiol 2005; 35:859–866 31. Lin P, Schueler BA, Balter S, et al. Accuracy and calibration of integrated radiation output indicators in diagnostic radiology. College Park, MD: American Association of Physicists in Medicine, 2015 32. Racadio J, Strauss K, Abruzzo T, et al. Significant dose reduction for pediatric digital subtraction angiography without impairing image quality: a preclinical study in a piglet model. AJR 2014; 203:904–908 33. McFadden S, Hughes C, D’Helft CI, et al. The establishment of local diagnostic reference levels for paediatric interventional cardiology. Radiography 2013; 19:295–301
AJR:204, June 2015 W719
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Pediatric Imaging Original Research
Estimates of Diagnostic Reference Levels for Pediatric Peripheral and Abdominal Fluoroscopically Guided Procedures Keith J. Strauss1 John M. Racadio1 Neil Johnson1 Manish Patel1 Rami A. Nachabe2 Strauss KJ, Racadio JM, Johnson N, Patel M, Nachabe RA Keywords: diagnostic reference level, pediatric, peripheral and abdominal angiography, radiation dose DOI:10.2214/AJR.14.13630 Received August 12, 2014; accepted after revision September 28, 2014. The Department of Radiology, Division of Interventional Radiology, Cincinnati Children’s Hospital Medical Center has a master research agreement with Philips Healthcare. K. J. Strauss has had his travel expenses paid to perform research and development activities for Philips Healthcare and has performed paid medical physics consulting services for Philips Healthcare. J. M. Racadio has had his travel expenses paid to participate in Phillips Healthcare– sponsored symposia. R. A. Nachabe is an employee of Philips Healthcare; however, the authors who are not Philips employees had full control of inclusion of any data and information that might present a conflict of interest for the author who is a Philips employee. Presented at the 2014 annual meeting of the Radiological Society of North America, Chicago, IL. 1 Department of Radiology, University of Cincinnati School of Medicine, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave, MLC 5031, Cincinnati, OH 45229-3026. Address correspondence to K. J. Strauss ([email protected]). 2 Interventional X-Ray Department, Philips Healthcare, Best, The Netherlands.
WEB This is a web exclusive article. AJR 2015; 204:W713–W719 0361–803X/15/2046–W713 © American Roentgen Ray Society
OBJECTIVE. The objective of our study was to survey radiation dose indexes of pediatric peripheral and abdominal fluoroscopically guided procedures from which estimates of diagnostic reference levels (DRLs) can be proposed for both a standard fluoroscope and a novel fluoroscope with advanced image processing and lower radiation dose rates. MATERIALS AND METHODS. Radiation dose structured reports were retrospectively collected for 408 clinical pediatric cases: Half of the procedures were performed with a standard imaging technology and half with a novel x-ray technology. Dose-area product (DAP), air Kerma (AK), fluoroscopy time, number of digital subtraction angiography images, and patient mass were collected to calculate and normalize radiation dose indexes for procedures completed with the standard and novel fluoroscopes. RESULTS. The study population was composed of 180 and 175 patients who underwent procedures with the standard and novel technology, respectively. The 21 different types of pediatric peripheral and abdominal interventional procedures produced 408 total studies. Median ages, mass and body mass index, fluoroscopy time per procedure, and total number of recorded images for the standard and novel technologies were not statistically different. The area of the x-ray beams was square at the level of the patient with a dimension of 10–13 cm. The dose reduction achieved with the novel fluoroscope ranged from 18% to 51% of the dose required with the standard fluoroscope. The median DAP and AK patient dose indexes were 0.38 Gy ⋅ cm2 and 4.00 mGy, respectively, for the novel fluoroscope. CONCLUSION. Estimates of dose indexes of pediatric peripheral and abdominal fluoroscopically guided, clinical procedures should assist in the development of DRLs to foster management of radiation doses of pediatric patients.
T
he Image Gently campaign urges careful management of the pediatric patient’s radiation dose from any imaging procedure that uses ionizing radiation, including complex angiographic studies [1]. Radiation management in the pediatric interventional fluoroscopic suite is a multifaceted process that needs to occur for each individual patient [2– 5]. The patient dose at the end of a clinical procedure can be estimated and compared with established standard-dose guidelines within the department as a function of the type of study and patient size. Estimated patient doses that exceed departmental standards should be investigated to ensure that appropriate steps to manage patient dose were followed throughout the procedure [6]. Departmental standard-dose guidelines are typically adapted from diagnostic reference levels (DRLs), which are based on na-
tional survey results of estimated patient doses during clinical studies [7]. Over the past decade, 13 published studies on the estimated doses during adult vascular angiographic procedures, the majority from neurovascular laboratories, were identified in the literature [8–20]. During the same time period, the estimated doses surveyed at seven pediatric cardiac catheterization departments were reported [21–27], and three different groups reported estimated patient doses during pediatric neurointerventions [28–30]. However, information on the estimated doses during pediatric peripheral and abdominal fluoroscopically guided studies from which appropriate DRLs could be established could not be identified in the published literature. We hypothesized that a novel image-processing technology capable of higher image quality at lower patient doses might reduce the required radiation dose during fluoro-
AJR:204, June 2015 W713
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Strauss et al. scopically guided studies of the trunk and limbs of pediatric patients ranging in age from neonates to 21 years. The purpose of this investigation was to survey the radiation dose levels during pediatric peripheral and abdominal fluoroscopically guided procedures at a tertiary care pediatric hospital. Data were collected for both a standard fluoroscopic unit (technology introduced in 2003 with typical patient dose rates) and a novel fluoroscopic unit (technology available in the United States in 2013) with advanced image processing and lower patient radiation dose rates. The surveyed radiation dose levels should assist in the development of DRLs, which foster management of radiation dose rates of pediatric patients during interventional fluoroscopic procedures. Materials and Methods Study Patients and Data Collection The institutional review board of Cincinnati Children’s Hospital Medical Center approved the collection of clinical data for this retrospective study. Patient informed consent was waived; however, patient confidentiality was secured according to HIPAA guidelines. Dose indexes from 204 interventional radiology (IR) procedures performed between September 2013 and January 2014 using a standard technology system were collected. All procedures performed using the department’s two standard fluoroscopes during this 5-month period except neurointerventional procedures were included. Dose indexes from an equal number of IR procedures (excluding neurointerventions) performed between December 2013 and March 2014 using a novel technology were collected. The novel technology fluoroscope (commissioned in December 2013) replaced one of the department’s standard technology units. The department’s three pediatric IR operators performed the 408 procedures. Radiation dose structured reports (RDSRs) generated by the fluoroscopes were collected via a networked third-party radiation dose database server (PEMNET, Clinical Microsystems). The RDSRs contain radiation dose values such as air Kerma (AK) at the interventional reference point, dose-area product (DAP), fluoroscopy time, and total number of images created for each radiation acquisition (e.g., fluoroscopy, digital subtraction angiography [DSA], rotational acquisitions, and cone beam CT acquisition). In addition, the patient medical record number, patient demographics (e.g., age, sex, mass, height, date of birth), date of the procedure, and study type were available. The accuracy of the DAP and AK values, which are indexes of estimates of the patient’s radiation
W714
dose, that are displayed on the fluoroscopes was verified with a calibrated ionization chamber and electrometer by the department’s staff medical physicist. A standard calibration protocol developed by the American Association of Physicists in Medicine [31] to measure the discrepancy between the displayed values on the fluoroscope and the calibrated ionization chamber positioned at the interventional reference point was used by the medical physicist. The measured correction factors, 1.03 and 1.05 for the standard and novel units, respectively, were applied to all dose indexes displayed on the units and were recorded in each unit’s RDSRs.
Imaging Technology This study was completed using two C-arm imaging systems: a novel advanced image-processing and dose reduction technology (AlluraClarity, Philips Healthcare), which was approved by the U.S. Food and Drug Administration for sale in the United States in 2013, and a standard technology (Allura Xper, Philips Healthcare), which has been available in the United States since 2003. The novel system uses real-time image noise reduction algorithms driven by parallel processing hardware to improve image quality in real time. This improved image quality allows the manufacturer to reduce patient entrance dose with changes to configuration settings [10, 32]. The added beam filter, reduced pulse width, smaller focal spot size, managed x-ray tube voltage and tube current, and
reduced detector AK differ appropriately as a function of the body region to be imaged and the physical size of the pediatric patient. Therefore, the configuration of the acquisition parameters on the novel system reduces patient entrance dose while maintaining image quality at levels similar to the standard technology [31].
Data Analysis Procedural dose indexes, expressed as both AK in units of mGy at the interventional reference point and DAP in units of Gy · cm2, were obtained from the RDSR. Indicators of the complexity of an interventional fluoroscopic procedure are total fluoroscopy time and total number of recorded DSA images [33]. To compress the distribution of AK and DAP due to the varying complexities of the 408 procedures, fluoroscopy and fluorography radiation doses of individual cases were normalized by the fluoroscopy time and the number of recorded DSA images for each procedure, respectively, to provide four rates: DAP/min and AK/min for fluoroscopy; and DAP/number of DSA images and AK/number of DSA images for fluorography. To further reduce the distribution of the dose rate indexes due to varying thicknesses of the body part imaged, the rates calculated in the previous paragraph were normalized by the patient’s mass [27] to provide DAP/min ⋅ kg, AK/min ⋅ kg, DAP/number of DSA images · kg, and AK/number of DSA images ⋅ kg.
TABLE 1: Patients’ Characteristics and Baseline Standard X-Ray Technology (n = 204 Procedures)
Novel X-Ray Technology (n = 204 Procedures)
p
180
175
NA
12.6 [5.5–17.2]a
11.6 [3.4–15.9]a
0.07, 0.17, 0.39b
90:114
113:91
0.03
Mass (kg)
39.3 [19.2–60.6]a
35.9 [15.1–57.9]a
0.27, 0.58, 0.39b
Body mass indexc
18.2 [15.6–22.0]a
18.3 [16.4–22.3]a
0.21, 0.14, 0.10b
Characteristics No. of patients Age (y) Male-female ratio
No. of examinations per patient (no. of patients)
0.69
1
163
155
2
11
15
3
5
3
4
1
1
5
0
0
6
0
1
37
34
No. of procedures performed in 31 patients who were treated with both technologies
0.79
Note—NA = not applicable. aValues are median [first quartile–third quartile]. bThe p values are listed for median, variance, and distribution, respectively. cWeight in kilograms divided by the square of height in meters.
AJR:204, June 2015
DRLs for Fluoroscopically Guided Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
TABLE 2: Procedure Types Standard X-Ray Technology (n = 204 Procedures)
Novel X-Ray Technology (n = 204 Procedures)
Angiography or venography
6
5
Pyelography
1
4
Central line injection
6
2
Biliary or percutaneous transhepatic cholangiography
2
3
Biopsy
2
2
Procedure
Chest tube placement
11
8
Embolization
1
1
Direct sclerosing
27
18
Drain placement
8
9
ERCP
2
10
Esophageal dilatation
25
15
Fistulography
5
14
Gastrojejunostomy
20
18
Nasogastric or nasojejunal tube placement
17
12
Peripherally inserted central catheter
7
5
Steroid injection
17
23
Lymphangiography
1
1
Vascular access
15
13
Arthrography
14
24
Lumbar Puncture
11
11
Nephrostomy
6
6
release R2011a), and a p value of < 0.05 was considered statistically significant.
Note—Data are number of procedures.
The following equations allow the estimate of total procedure (TP), fluoroscopy (FS), or fluorography (FG) DAP or AK for an individual patient from the normalized estimates of DAP and AK. In each of the following three equations, substitute either DAP or AK for X: X / (min ⋅ kg) × FT × PM = FS X
X / (DSA ⋅ kg) × DSA × PM = FG X TP X = FS X + FG X
(1), (2), (3),
where min ⋅ kg indicates units of the estimated fluorography DRL, FT is fluoroscopy time in minutes, PM is patient mass in kilograms, DSA ⋅ kg are units of the estimated fluoroscopy DRL, and DSA is the number of DSA images. Finally, division of DAP by AK estimates the area of the x-ray beam at the isocenter of the fluoroscope.
Statistical Analysis The Pearson chi-square test was applied to categoric variables such as sex, clinical study type, number of procedures per operator, and patient inter- and
intramodality replicates. Two-sample Mann-Whitney U, Levene’s, and Kolmogorov-Smirnov tests were applied to compare medians, variances, and distributions. The variables such as age, mass, body mass index (BMI [defined as weight in kilograms divided by the square of height in meters]), procedure dose, fluoroscopy time, number of acquired frames, fluoroscopy dose rate, and acquisition dose rate were investigated with the aforementioned three tests. Given the nongaussian distribution of the collected data, variables are reported as medians along with the first and third quartiles. Statistical calculations were performed using statistics software (Matlab, MathWorks, version 7.12,
Results The age, mass, and BMI of patients did not show any statistical difference in median, variance, and distribution for both technologies (Table 1). The fraction of males in the standard group is smaller (44%) than that in the novel group (55%). The total number of patients treated with the standard technology and the novel technology showed no statistical difference (p = 0.69) in intramodality replicates (Table 1). Thirty-one patients were treated with both technologies (37 and 34 procedures with the standard and novel technology, respectively) showing no statistical difference in intermodality number of repeated patients (p = 0.79) (Table 1). Twenty-one different types of interventional procedures were performed (Table 2). Despite the variety of clinical procedures, the proportion of clinical study types performed with both technologies was not statistically significant (p = 0.22). Table 3 summarizes the distribution of the number of procedures among the three operators. This table also lists the distribution of the procedure AK per operator. There is no statistical difference between the distribution of the procedures among the operators (p = 0.06). A significant procedure dose reduction (p < 0.05) between the two technologies was observed for each of the three operators. No statistical differences in the median, variance, and distribution were observed in fluoroscopy time and number of acquired DSA images for both technologies (Table 4). DSA acquisitions were performed in 35 and 31 procedures with the standard and novel technology, respectively. Thus, no statistical difference exists in the number of procedures with DSA in both groups (p = 0.69). This statistical analysis validates the comparison of the two types of fluoroscopes. Table 4 lists the median, first quartile, and third quartile of the surveyed dose indexes
TABLE 3: Distribution of Procedure Air Kerma (AK) Values Among Operators Standard X-Ray Technology (n = 204 Procedures) Operator No.
Novel X-Ray Technology (n = 204 Procedures)
AK (mGy)a
No. of Procedures
AK (mGy)a
No. of Procedures
Operator 1
15 [7.32–42.07]
54
5 [2.25–23.25]
43
Operator 2
10 [4–22]
72
2 [1–5]
96
Operator 3
13 [6–62]
78
7 [3–33.5]
65
aValues are median [first quartile–third quartile].
AJR:204, June 2015 W715
TABLE 4: Procedure Dose and Acquisition Characteristics Strauss et al. Standard X-Ray Novel X-Ray Technology Technology (n = 204 Procedures)a (n = 204 Procedures)a
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
Character Fluoroscopy time (min)
p Median
Variance Distribution
1.6 [0.8–3.4]
1.5 [0.7–3.7]
0.53
0.77
0.54
2 [1–8]
2 [1–9]
0.62
0.49
0.91
1.09 [0.46–4.89]
0.38 [0.17–1.45]
< 0.0001
0.10
< 0.0001
No. of DSA images DAP (Gy ⋅ cm2) Procedure Fluoroscopy
1.02 [0.45–4.77]
0.38 [0.17–1.41]
< 0.0001
0.14
< 0.0001
Digital acquisition
0.43 [0.20–4.06]
0.13 [0.02–1.25]
< 0.05
0.25
< 0.05
0.77 [0.35–1.72]
0.30 [0.15–0.77]
< 0.0001
0.09
< 0.0001
0.26 [0.14–0.46]
0.03 [0.01–0.14]
< 0.0001
0.06
< 0.0001
Fluoroscopy DAP per fluoroscopy time (Gy ⋅
cm2 /min)
Digital acquisition DAP per frame (Gy ⋅ cm2 /frame) Fluoroscopy DAP per fluoroscopy time and patient mass (Gy ⋅ cm2 /min ⋅ kg)
0.027 [0.016–0.047]
0.015 [0.007–0.024]
< 0.0001
0.17
< 0.0001
Digital acquisition DAP per frame and patient mass (Gy ⋅ cm2 /frame ⋅ kg)
0.005 [0.004–0.012]
0.002 [0.001–0.003]
< 0.0001
< 0.0001
< 0.0001
Procedure
12.00 [5.50–33.26]
4.00 [2.00–12.00]
< 0.0001
0.18
< 0.0001
Fluoroscopy
12.00 [5.15–32.00]
3.98 [2.00–10.60]
< 0.0001
0.25
< 0.0001
AK (mGy)
Digital acquisition
3.92 [1.47–20.66]
1.11 [0.19–6.18]
< 0.005
0.22
< 0.01
6.54 [4.00–13.44]
2.81 [1.70–5.03]
< 0.0001
< 0.01
< 0.0001
Digital acquisition AK per frame (mGy/frame)
1.84 [1.07–3.37]
0.35 [0.09–0.62]
< 0.0001
< 0.05
< 0.0001
Fluoroscopy AK per fluoroscopy time and patient mass (mGy/min ⋅ kg)
0.27 [0.15–0.42]
0.10 [0.06–0.17]
< 0.0001
< 0.005
< 0.0001
Digital acquisition AK per frame and patient mass (mGy/frame ⋅ kg)
0.05 [0.03–0.07]
0.01 [0.01–0.02]
< 0.0001
< 0.05
< 0.0001
Fluoroscopy AK per fluoroscopy time (mGy/min)
X-ray beam area at isocenter (cm2) Procedure
105 [68–161]
112 [74–190]
0.04
0.44
0.22
Fluoroscopy
104 [68–161]
112 [75–197]
0.03
0.43
0.14
Fluorography
144 [82–217]
167 [115–263]
0.16
0.56
0.15
Note—DSA = digital subtraction angiography, DAP = dose-area product, AK = air Kerma. aValues are median [first quartile–third quartile].
TABLE 5: Estimates of Diagnostic Reference Levels (DRLs) and Comparison of Dose Between Standard and Novel Systems Value (Third Quartile) Standard X-Ray Technology (n = 204 Procedures)
DRLs
Novel X-Ray Technology (n = 204 Procedures)
Dose Fraction Achieved With Novel Technology Compared With Standard Technology
Procedure DAP (Gy ⋅ cm2)
4.89
1.45
0.30
Fluoroscopy DAP per fluoroscopy time (Gy ⋅ cm2 /min)
1.72
0.77
0. 45
Digital acquisition DAP per frame (Gy ⋅ cm2 /frame)
0.46
0.14
0. 30
Fluoroscopy DAP per fluoroscopy time and patient mass (Gy ⋅ cm2 /min ⋅ kg)
0.047
0.024
0. 51
Digital acquisition DAP per frame and patient mass (Gy ⋅ cm2 /frame ⋅ kg)
0.012
0.003
0. 25
Procedure AK (mGy)
33.26
12.00
0. 36
Fluoroscopy AK per fluoroscopy time (mGy/min)
13.44
5.03
0. 37
Digital acquisition AK per frame (mGy/frame)
3.37
0.62
0. 18
Fluoroscopy AK per fluoroscopy time and patient mass (mGy/min ⋅ kg)
0.42
0.17
0. 40
Digital acquisition AK per frame and patient mass (mGy/frame ⋅ kg)
0.07
0.02
0. 29
Note—DAP = dose-area product, AK = air Kerma.
of the standard and novel fluoroscopes. The x-ray beam area at the isocenter of the fluoroscope is listed at the end of Table 4. The small differences in these estimated areas (whole procedure, fluoroscopy only, or fluorography only) expressed in units of cm2 rep-
W716
resent square x-ray field areas with a dimension that ranges from 10 to 13 cm. Estimates of DRLs and the third quartile of the various dose distributions in Table 4 are listed in the second and third columns of Table 5 for the standard and novel units, re-
spectively. Figures 1–6 illustrate the distributions of the patient dose indexes with both bar charts and boxplots. Estimates of DRLs are provided for both DAP and AK. The total procedure DRL is broken down into dose due to fluoroscopy and fluorography. These
AJR:204, June 2015
Discussion This study provides surveyed estimates of patient dose indexes of clinical pediatric abdominal and peripheral fluoroscopically guided studies from which DRLs normalized for the complexity of the procedure and for the size of the pediatric patient are estimated. These estimates of DRLs are provided for both a standard technology fluoroscope and a novel technology fluoroscope. The area of the x-ray beam at the isocenter of each fluoroscope was also calculated for both fluoroscopic and fluorographic operational modes. The estimated procedure DRLs are for an entire study and are expressed as procedure DAP (line 1 in Table 5) and procedure AK (line 6 in Table 5). These estimates are based on data with a broad distribution due to varying complexities of the procedures and the varying masses of the patients. These estimates are representative for only the median-complexity procedure and the median-sized patient. Adjusting the normalized DAP and AK values in lines 2–5 and 7–10, respectively, in Table 5 by each individual patient’s examination complexity (fluoroscopy time and DSA images) and mass allows a more accurate estimate of an individual patient’s dose indexes. The equations in the Materials and Methods, the normalized values of DAP and AK in Table 5, and the patient’s mass, study fluoroscopy time, and number of DSA images obtained during the study allow a more accurate estimate of an individual patient’s DAP and AK. Estimates of DRLs for both DAP and AK are necessary. Although AK assists in predicting the risk of deterministic radiation injuries (e.g., skin damage), DAP is the better choice for predicting the risk of stochastic radiation injuries (e.g., cancer) [6]. Older fluoroscopes may display DAP but not AK. An estimate of the area of the x-ray beam at the level of the patient, listed at the end of Ta-
No. of Procedures
80
Standard imaging technology Novel imaging technology
60 40 20 0 0
1
2
3
4
0
1
2
3
4
5
6
5 6 DAP (Gy · cm2)
7
8
9
10
11
7
8
9
10
11
Fig. 1—Bar graph and boxplot show distribution of procedure dose-area product (DAP) values by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 40 No. of Procedures
breakdowns are presented normalized by the complexity of the procedure (i.e., fluoroscopy time or number of DSA images) and by both the complexity of the procedure and the size of the patient (i.e., fluoroscopy time or number of DSA images × patient mass [kg]). The fraction of dose achieved with the novel fluoroscope ranged from 0.18 to 0.51 of the dose used with the standard fluoroscope depending on the dose index used for the calculation, as listed in the last column of Table 5. The fraction of dose achieved between the two systems during digital acquisitions always exceeds the dose reduction achieved during fluoroscopy.
Standard imaging technology Novel imaging technology
30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50 AK (mGy)
60
70
80
90
100
Fig. 2—Bar graph and boxplot show distribution of procedure air Kerma (AK) values by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 80 No. of Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
DRLs for Fluoroscopically Guided Procedures
Standard imaging technology Novel imaging technology
60 40 20 0 0
1
2
3
0
1
2
3
4
5
6
4 5 6 DAP Rate (Gy · cm2/min)
7
8
9
10
7
8
9
10
Fig. 3—Bar graph and boxplot show distribution of fluoroscopy dose-area product (DAP) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
AJR:204, June 2015 W717
Strauss et al.
No. of Procedures
Standard imaging technology Novel imaging technology
10 8 6 4 2 0 0
0.5
0
0.5
DAP Rate (Gy · cm2/frame)
1.0
1.5
1.0
1.5
Fig. 4—Bar graph and boxplot show distribution of digital acquisition dose-area product (DAP) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
No. of Procedures
50
Standard imaging technology Novel imaging technology
40 30 20 10 0 0
5
10
0
5
10
15
20
15 20 AK Rate (mGy/min)
25
30
35
25
30
35
Fig. 5—Bar graph and boxplot show distribution of fluoroscopy air Kerma (AK) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range. 15 No. of Procedures
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
12
Standard imaging technology Novel imaging technology
10
5
0 0
2
4
6
0
2
4
6
8
10
12
14
16
18
20
8 10 12 AK Rate (mGy/frame)
14
16
18
20
Fig. 6—Bar graph and boxplot show distribution of digital acquisition air Kerma (AK) rates by technology. In boxplot, lower limit of boxes is first quartile, middle line is median, and upper limit is third quartile. Dashed lines show upper and lower limits of range.
W718
ble 4, allows estimation of AK at the interventional reference point if only DAP is displayed by the fluoroscope. Estimates of DRLs for both a standard fluoroscope and a novel fluoroscope provide two distinct reference levels, which may be helpful for a department with older, newer, or both types of fluoroscopes. The standard DRLs may provide guidance concerning expected doses from fluoroscopes in the field in the second half of their useful life, whereas the DRLs of the novel unit may provide guidance for expected doses from state-ofthe-art fluoroscopes recently installed. Estimates of DRLs for both fluoroscopy and fluorography may assist with a root analysis of high patient doses at the end of a study. For example, the patient’s dose from fluorography may be appropriate, but the fluoroscopy component of the study may be high. This difference in dose could be due to inefficient technique by the fluoroscopist or poor configuration of the fluoroscopic mode. If the situation is reversed—appropriate dose for fluoroscopy but excessive dose for fluorography, then the number of recorded radiographic images may be excessive or the dose at the image receptor for each recorded image may be excessive. DRLs have not been previously published for pediatric peripheral and abdominal angiography to our knowledge. However, two different studies of adult neuroangiographic procedures [8, 10] and one study of adult electrophysiology procedures [11] compare the doses required by standard and novel fluoroscopes similar to the fluoroscopes used in this study. These studies listed reductions in patient dose with the novel unit that were 35% [8], 25% [10], and 58% [11] of the dose required by the standard system. We report a comparable dose fraction of 34% in our study in comparison with these previous studies. Because our study had the availability of RDSR data collection, our study provides a larger variety of estimates of DRLs than the three previous studies. Although our study provides data suitable for assisting in the development of DRLs for pediatric peripheral and abdominal fluoroscopically guided procedures, several limitations remain. First, one company manufactured both fluoroscopes used in this study. The degree of technology advancement achieved by this company may not be representative of other imaging companies. Second, the estimates of DRLs for fluorography are based on a limited number of recorded DSA images
AJR:204, June 2015
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/03/15 from IP address 128.122.253.228. Copyright ARRS. For personal use only; all rights reserved
DRLs for Fluoroscopically Guided Procedures because fluoroscopy stored images are substituted for DSA images whenever possible during pediatric studies to reduce patient dose. Third, the limited number of procedures in this study prevents more accurate estimates of DRLs for individual types of studies or for different groups of patients of similar sizes. More studies are needed to improve the accuracy of pediatric DRLs in interventional fluoroscopy. A larger sample size would allow patients to be grouped by the type of study and by a limited range of patient sizes. A study with more than four operators and a large sample size would allow investigation of the influence of individual operators on patient doses. Data should also be collected for all the major manufacturers of interventional fluoroscopes. In conclusion, this study provides data suitable for assisting in the development of DRLs for pediatric peripheral and abdominal fluoroscopically guided procedures. The normalized estimates of DRL allow calculation of more accurate estimates of dose for an individual patient than a single median DRL for all procedures and patients. These estimates of DRL should help departments develop pediatric dose guidelines. The double set of estimates of DRLs—one for a standard fluoroscope in the marketplace for more than 10 years and one for a fluoroscope with advanced technology relatively new to the marketplace—should help departments manage their pediatric radiation doses as the currently installed equipment is replaced with new units with the latest technology. Acknowledgments We thank Joanne Lovelace for her skillful help with references and the management of the preparation of this manuscript and Melinda Dighe for her help in collecting the subject’s mass. References 1. Sidhu M, Coley BD, Goske MJ, et al. Image Gently, Step Lightly: increasing radiation dose awareness in pediatric interventional radiology. Pediatr Radiol 2009; 39:1135–1138 2. Frush D, Denham CR, Goske MJ, et al. Radiation protection and dose monitoring in medical imaging: a journey from awareness, through accountability, ability and action...but where will we arrive? J Patient Saf 2013; 9:232–238 3. Miller DL, Balter S, Schueler BA, Wagner LK, Strauss KJ, Vano E. Clinical radiation management for fluoroscopically guided interventional procedures. Radiology 2010; 257:321–332 4. Miller DL. Efforts to optimize radiation protection in interventional fluoroscopy. Health Phys
2013; 105:435–444 5. Schueler BA, Balter S, Miller DL. Radiation protection tools in interventional radiology. J Am Coll Radiol 2012; 9:844–845 6. National Council on Radiation Protection and Measurements. Radiation dose management for fluoroscopy-guided interventional medical procedures. Bethesda, MD: NCRP, 2010: Report 168 7. [No authors listed]. Radiological protection and safety in medicine: a report of the International Commission on Radiological Protection. Ann ICRP 1996; 26:1–47 [Erratum in Ann ICRP 1997; 27:61] 8. Söderman M, Mauti M, Boon S, et al. Radiation dose in neuroangiography using image noise reduction technology: a population study based on 614 patients. Neuroradiology 2013; 55:1365–1372 9. Chida K, Ohno T, Kakizaki S, et al. Radiation dose to the pediatric cardiac catheterization and intervention patient. AJR 2010; 195:1175–1179 10. Söderman M, Holmin S, Andersson T, Palmgren C, Babic D, Hoornaert B. Image noise reduction algorithm for digital subtraction angiography: clinical results. Radiology 2013; 269:553–560 11. Dekker LR, van der Voort PH, Simmers TA, et al. New image processing and noise reduction technology allows reduction of radiation exposure in complex electrophysiologic interventions while maintaining optimal image quality: a randomized clinical trial. Heart Rhythm 2013; 10:1678–1682 12. Sanchez RM, Vano E, Fernández JM, Moreu M, Lopez-Ibor L. Brain radiation doses to patients in an interventional neuroradiology laboratory. AJNR 2014; 35:1276–1280 13. D’Ercole L, Thyrion FZ, Bocchiola M, Mantovani L, Klersy C. Proposed local diagnostic reference levels in angiography and interventional neuroradiology and a preliminary analysis according to the complexity of the procedures. Phys Med 2012; 28:61–70 14. Alexander MD, Oliff MC, Olorunsola OG, BrusRamer M, Nickoloff EL, Meyers PM. Patient radiation exposure during diagnostic and therapeutic interventional neuroradiology procedures. J Neurointerv Surg 2010; 2:6–10 15. Suzuki S, Furui S, Matsumaru Y, et al. Patient skin dose during neuroembolization by multiplepoint measurement using a radiosensitive indicator. AJNR 2008; 29:1076–1081 16. D’Ercole L, Mantovani L, Thyrion FZ, et al. A study on maximum skin dose in cerebral embolization procedures. AJNR 2007; 28:503–507 17. Thierry-Chef I, Simon SL, Miller DL. Radiation dose and cancer risk among pediatric patients undergoing interventional neuroradiology procedures. Pediatr Radiol 2006; 36(suppl 2):159–162 18. Bor D, Cekirge S, Türkay T, et al. Patient and staff doses in interventional neuroradiology. Radiat Prot Dosimetry 2005; 117:62–68 19. Struelens L, Vanhavere F, Bosmans H, Van Loon R, Mol H. Skin dose measurements on patients for diag-
nostic and interventional neuroradiology: a multicentre study. Radiat Prot Dosimetry 2005; 114:143–146 20. Pitton MB, Kloeckner R, Schneider J, Ruckes C, Bersch A, Duber C. Radiation exposure in vascular angiographic procedures. J Vasc Interv Radiol 2012; 23:1487–1495 21. Glatz AC, Patel A, Zhu X, et al. Patient radiation exposure in a modern, large-volume, pediatric cardiac catheterization laboratory. Pediatr Cardiol 2014; 35:870–878 22. McFadden SH, D’Helft CI, McGee A, et al. The establishment of local diagnostic reference levels for paediatric interventional cardiology. Radiography 2013; 19:295–301 23. Verghese GR, McElhinney DB, Strauss KJ, Bergersen L. Characterization of radiation exposure and effect of a radiation monitoring policy in a large volume pediatric cardiac catheterization lab. Catheter Cardiovasc Interv 2012; 79:294–301 24. Dragusin O, Gewillig M, Desmet W, Smans K, Struelens L, Bosmans H. Radiation dose survey in a paediatric cardiac catheterisation laboratory equipped with flat-panel detectors. Radiat Prot Dosimetry 2008; 129:91–95 25. Al-Haj AN, Lobriguito AM, Rafeh W. Variation in radiation doses in paediatric cardiac catheterisation procedures. Radiat Prot Dosimetry 2008; 129:173–178 26. Martinez LC, Vano E, Gutierrez F, Rodriguez C, Gilarranz R, Manzanas MJ. Patient doses from fluoroscopically guided cardiac procedures in pediatrics. Phys Med Biol 2007; 52:4749–4759 27. Onnasch DG, Schroder FK, Fischer G, Kramer HH. Diagnostic reference levels and effective dose in paediatric cardiac catheterization. Br J Radiol 2007; 80:177–185 28. Orbach DB, Stamoulis C, Strauss KJ, et al. Neurointerventions in children: radiation exposure and its import. AJNR 2014; 35:650–656 29. Thierry-Chef I, Simon SL, Land CE, Miller DL. Radiation dose to the brain and subsequent risk of developing brain tumors in pediatric patients undergoing interventional neuroradiology procedures. Radiat Res 2008; 170:553–565 30. Swoboda NA, Armstrong DG, Smith J, Charkot E, Connolly BL. Pediatric patient surface doses in neuroangiography. Pediatr Radiol 2005; 35:859–866 31. Lin P, Schueler BA, Balter S, et al. Accuracy and calibration of integrated radiation output indicators in diagnostic radiology. College Park, MD: American Association of Physicists in Medicine, 2015 32. Racadio J, Strauss K, Abruzzo T, et al. Significant dose reduction for pediatric digital subtraction angiography without impairing image quality: a preclinical study in a piglet model. AJR 2014; 203:904–908 33. McFadden S, Hughes C, D’Helft CI, et al. The establishment of local diagnostic reference levels for paediatric interventional cardiology. Radiography 2013; 19:295–301
AJR:204, June 2015 W719
