Author + information
- Received March 4, 2013
- Revision received April 26, 2013
- Accepted May 9, 2013
- Published online October 1, 2013.
- Maja Ingwersen, DVM∗,
- Anna Drabik, PhD†,
- Ulrike Kulka, PhD‡,
- Ursula Oestreicher, Dipl Biol‡,
- Simon Fricke, BS§,
- Hans Krankenberg, MD∗,
- Carsten Schwencke, MD∗ and
- Detlef Mathey, MD∗∗ ()
- ∗University Cardiovascular Center, Hamburg, Germany
- †Department of Medical Biometry and Epidemiology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
- ‡Department of Radiation Protection and Health, German Federal Office for Radiation Protection, Oberschleissheim, Germany
- §Philips GmbH, Hamburg, Germany
- ↵∗Reprint requests and correspondence:
Prof. Dr. Detlef Mathey, University Cardiovascular Center, Prof. Schofer, Woerdemanns Weg 25-27, D-22527 Hamburg, Germany.
Objectives This study sought to evaluate differences in radiation exposure of the operator depending on the type of catheterization lab procedure.
Background Invasive cardiologists and angiologists are exposed to long-term, low-dose occupational radiation. Increased workload and specialization require more detailed knowledge of the extent and cause of the radiation exposure.
Methods In this prospective single-center experience, radiation doses of 3 operators were measured by real-time dosimetry for body, neck, and hand during 284 procedures in 281 patients over a period of 14 weeks. To determine the association between the type of procedure and the doses and to draw a pairwise comparison between the procedures, 3 mixed models were used.
Results The type of procedure, the patient's body mass index, and the fluoroscopy time were independently associated with the operator's radiation exposure. Per procedure, the operators were exposed to a mean effective dose (E) of 2.2 ± 5.9 μSv. Compared with coronary angiography, E was 2.3-fold higher in pelvic procedures (95% confidence interval [CI]: 1.7 to 3.0, p < 0.001), 1.7-fold higher in upper limb procedures (95% CI: 1.3 to 2.1, p < 0.001), and 1.4-fold higher in below-the-knee procedures (95% CI: 1.1 to 2.0, p = 0.023). The mean eye dose was 19.1 ± 37.6 μSv. Eye doses were significantly higher in peripheral procedures than in coronary angiography procedures. The mean hand dose was 99.6 ± 196.0 μSv. Hand doses were significantly higher in pelvic than in coronary angiography, upper limb, and below-the-knee procedures.
Conclusions Endovascular procedures for pelvic, upper limb, and below-the-knee disease are accompanied with a higher radiation exposure of the operator than with coronary procedures.
- biological dosimetry
- dose aware system
- occupational radiation exposure
- radiation exposure in cardiology and angiology
- real-time dosimetry
Occupational radiation exposure is a concern of catheterization lab operators. X-ray technology has been notably improved in the last decades, not only in terms of higher resolution, but also in terms of safety. However, nowadays, more patients are treated and more complex cases are performed per day by a single interventionalist. In addition, there is a trend for cardiologists and angiologists to specialize in specific procedures. Because radiation accumulates over the years, even low doses may contribute to the risk of malignant disease, heritable effects, and tissue reactions.
Does the type of procedure affect the radiation exposure of body, hands, and eyes of the physician? Is there a relevant difference between procedures? In case of a better knowledge of the procedural effect on occupational radiation exposure, more targeted protection and monitoring could be possible.
In a prospective single-center experiment, radiation doses were measured for body, neck, and a hand of 3 experienced operators. Between June 7, 2010, and September 20, 2010, the operators' radiation exposure was determined during procedures in consecutive patients who underwent coronary or peripheral pelvic, upper limb (UL), or below-the-knee (BTK) angiography or intervention. Operators had 10 to 20 years of catheterization lab experience and provided periodically renewed certificates of specialized knowledge in the field of radiation protection.
Data were collected for 5 different types of procedures: coronary angiography (CAG); percutaneous coronary intervention (PCI); percutaneous endovascular pelvic; UL; and BTK interventions. For the majority of procedures, the femoral access route was used. All body parts of the operators were kept out of the imaging field at all times. For surveillance of patients' safety, dose area product (DAP) measurements were performed during each procedure.
The procedures were conducted in 2 catheterization labs, equipped with Philips (Hamburg, Germany) x-ray systems Allura Xper FD 10 (MRC200 0508 ROT-GS 1003 x-ray tube) for coronary interventions and Allura Xper FD 20 (MRC200 0407 ROT-GS 1004 x-ray tube) for peripheral interventions, both equipped with automatic dose regulation according to procedural needs. The generators were set to 50 to 120 kV (75 kV on average) depending on the patient's weight and the current was adapted accordingly to 50 to 1,000 mA (800 mA on average). Fields of view were 15 to 25 cm of diagonal square for coronary and 15 to 48 cm for peripheral interventions with automatic collimation to the minimum required fluoroscopy field.
In the fluoroscopy mode, the detector entrance dose rate varied between 42 and 84 mGy·min−1 for coronary and between 21 and 84 mGy·min−1 for peripheral interventions. The inherent filtration of the x-ray systems was 2.6 mm Al equivalent. For coronary interventions, additional spectral filters of 0.4 mm Cu/1.0 mm Al equivalent or 0.1 mm Cu/1.0 mm Al equivalent were used, depending on the pulsed fluoroscopy frame speed and the entrance fluoroscopy dose rate. For peripheral interventions, additional spectral filters of 0.9, 0.4, and 0.1 mm Cu/1.0 mm Al equivalent were applied. Pulse fluoroscopy frame speed was 7.5·s−1 or 15·s−1 for both x-ray systems.
The detector entrance cine/exposure dose rate was 20 to 20,000 mGy.frame−1 for both x-ray systems. Additional spectral filters of 0.9, 0.4, and 0.1 mm Cu/1.0 mm Al equivalent were used in the cine/exposure mode. The frame rate was 15·s−1 for coronary procedures in the cine mode and 6·s−1 to 1·s−1 for pelvic, UL, and BTK procedures in the exposure mode, depending on the velocity of the contrast medium.
The operators used lead aprons, thyroid collars, and protective glasses equivalent to 0.5 mm lead. Lead glass screens and above- and below-table shield systems of 0.5 mm lead equivalence were used during almost all procedures. Additional bismuth drapes were not commonly used.
Each of the operators was equipped with 3 personal dosimeters that are part of the DoseAware system (Philips) to measure their radiation exposure in real time. The dosimeters were placed at shoulder level above the protective lead collar, at chest height under the lead apron, and at the wrist. The radiation exposure was detected every second and wirelessly sent to a base station. Accumulated radiation doses of each operator at body, neck, and hand were recorded at the beginning and the end of every single procedure to calculate the procedural radiation dose.
Effective dose, eye dose, hand dose
The effective dose (E) was estimated according to the method of Niklason et al. (1) using the data collected from the body dosimeter under the apron (Hu) and the neck dosimeter above the apron (Hos): E = 0.02 (Hos – Hu) + Hos. This equation reflects the exposure of sensitive organs in the trunk and considers the exposure of unprotected parts of the body. Therefore, E expresses the total risk of a person, taking into account the radiosensitivity of each organ.
The eye dose was estimated by the equation: eye dose = 0.75 × neck dose, recommended by Martin (2) following available evidence. The hand doses were measured by a dosimeter worn on the wrist.
In order to verify the radiation doses received, dicentric chromosome assays using fluorescence plus giemsa staining and fluorescence in situ hybridization (FISH technique) analyzing symmetrical translocations in stable cells were carried out in 2 of the participating interventionalists according to standard protocols (3). The dicentric chromosome assay has a limit of determination of 100 mSv whole body dose for low linear energy transfer radiation. The assay detects a recent exposure best because of an intrinsic exponential removal rate of dicentrics (half-time between 6 months and 3 years) (4). The FISH technique detects translocations of chromosomes. It is suitable for the assessment of the radiation dose received chronically over many years or for a past exposure a long time ago because translocations persist for years in peripheral lymphocytes. The detection limit is 300 to 500 mSv lifetime dose (4) depending on the age of the person.
For descriptive analyses, we used means and standard deviations. To determine the association between effective dose, eye dose, or hand dose and the type of procedure, we used 3 mixed models with each dose type representing a dependent variable. All models included the investigator as a random factor, and procedure type, patients' body mass index (BMI), and fluoroscopy time (in minutes) as independent variables and were analyzed for significant 2-way interactions. Following logarithmic transformation, pairwise comparisons of parameters are presented as ratios of geometric means and their corresponding 95% confidence intervals (CIs). Effect sizes were reported in form of marginal (partial) R2 values with variance explained by fixed factors and a conditional R2 with variance explained by fixed and random factors (5). Due to asymmetric distributed dose values, logarithms of doses were used as dependent variables.
Reported p values are 2-sided using a 5% significance level. Statistics were performed with SPSS (version 21, Chicago, Illinois).
The 3 operators performed 284 procedures in 281 patients over 14 weeks.
Radiation dose per procedure
Table 1 shows the radiation doses of the operators during the different types of procedures. Coronary artery procedures show lower effective doses and less radiation exposure to eyes and hands than peripheral artery procedures do. In all types of procedures, the mean effective dose to the body was 2.2 ± 5.9 μSv (n = 266), and the mean radiation dose to eyes and hands were 19.1 ± 37.6 μSv (n = 278) and 99.6 ± 196.0 μSv (n = 273), respectively.
In all mixed models, type of procedure, BMI, and fluoroscopy time were significantly associated with the effective dose and the doses to eyes and hands (all p values < 0.05).
Effect of the type of procedure
Endovascular procedures involving pelvic, UL or BTK arteries were associated with significantly higher effective doses than CAG procedures were. The mean effective dose during pelvic procedures was 2.3-fold higher (95% CI: 1.7 to 3.0, p < 0.001), during UL procedures, 1.7-fold higher (95% CI: 1.3 to 2.1, p < 0.001), and during BTK procedures, 1.4-fold higher (95% CI: 1.1 to 2.0, p = 0.023) than the dose the operator received from CAG (Fig. 1). The effective doses during pelvic and UL procedures were also significantly higher than those during PCI procedures (2.2-fold, 95% CI: 1.5 to 3.3, p < 0.001, and 1.6-fold, 95% CI: 1.1 to 2.3, p = 0.008, respectively) (Table 2). There was no significant difference in radiation exposure between coronary procedures with or without stenting and between lower or upper limb procedures. In contrast, pelvic procedures were accompanied with significantly higher effective doses for the operator than were upper and lower limb procedures.
The eye dose during peripheral procedures was higher than that during CAG. The mean eye dose in pelvic procedures was 2.4-fold higher (95% CI: 1.4 to 4.1, p = 0.002), in UL procedures, 2.0-fold higher (95% CI: 1.2 to 3.1, p = 0.004), and in BTK procedures, 2.7-fold higher (95% CI: 1.4 to 5.1, p = 0.003) than the dose from CAG. The eye dose did not differ between peripheral procedures (Fig. 1, Table 2).
Effect of BMI
Patients' mean BMI was 27.0 ± 4.9 kg/m2. Each additional unit of patients' BMI increased the operators' effective doses by 3.6% (95% CI: 1.8% to 5.4%, p < 0.001), their eye doses by 4.8% (95% CI: 1.4% to 8.4%, p = 0.002), and their hand doses by 5.9% (95% CI: 2.9% to 8.9%, p < 0.001), irrespective of the type of procedure and the fluoroscopy time.
Effect of fluoroscopy time
Fluoroscopy took 7.2 ± 5.7 min on average. Every additional minute of fluoroscopy increased the effective dose of the operators by 3.2% (95% CI: 1.7% to 4.7%, p < 0.001), their eye doses by 4.6% (95% CI: 1.7% to 7.6%, p = 0.002), and their hand doses by 8.7% (95% CI: 6.0% to 11.5%, p < 0.001), independent of the type of procedure and the patient's BMI.
In the blood samples of 2 operators, 1 showed no dicentric chromosome and 1 had 3 dicentric chromosomes in 1,000 cells scored, not indicating a significant difference to the normal background level and not exceeding the detection level of 100 mSv in the recent past. The FISH technique assessed 3.0 ± 1.2 and 3.8 ± 1.7 translocations per 1,000 stable cells, respectively. Although the number of translocations was in accordance with working life of the operators of about 10 and 20 years, respectively, none of them differed significantly from healthy control subjects of the same age group.
Time, distance, projection
A high radiation exposure is supposed to be a consequence of a high DAP (6), observed in complex, long lasting procedures and in pelvic procedures. Sanchez et al. (7) found a significant correlation (Spearman rho value 0.729, p = 0.002) between the operator's personal dose and the patient's DAP. Known determinants of the DAP are a patient's BMI and radiation exposure time (8). We showed that the operator's dose is associated not only with the patient's BMI and fluoroscopy time, representing lesion complexity, but also with the type of procedure. Compared with CAG, effective dose was increased 1.4- to 2.3-fold in BTK, UL, or pelvic procedures. Similar results were found in comparison to PCI. This increase may be attributable to the distance of the operator to the patient, the frequency of oblique projections, the source to detector distance, the field of view, and the frame rate. Theocharopoulos et al. (9) showed that the radiation exposure varies by a factor of 40, depending on the position of the operator. Compared with UL and BTK procedures, in pelvic procedures, E was increased 1.4- and 1.6-fold. In these procedures, the operator has to stand closer to the lesion and, therefore, is more intensely exposed to the scatter. Kim and Miller (10) showed that decreasing the distance from 1 m to 0.75 m doubles the occupational radiation dose.
The operator's eye dose is increased 2.0- to 2.7-fold in UL, pelvic, and BTK procedures compared with doses for CAG. Hand doses are highest in pelvic procedures even if compared to those for UL and BTK procedures. A shorter distance and a lack of effective shielding may explain these findings.
Estimated annual radiation exposure
With a supposed workload of 500 to 1,000 procedures per year for an operator conducting mainly CAG and PCI, we can assume a total effective dose of 0.25 to 0.5 mSv, an eye dose of 3.2 to 6.4 mSv, and a hand dose of 33.1 to 66.2 mSv per year. In contrast, an operator mainly involved in peripheral procedures would be exposed to an effective dose of 1.6 to 3.1 mSv, an eye dose of 13.1 to 26.2 mSv, and a hand dose of 59.5 to 119 mSv per year. These doses, except for the eye dose in peripheral procedures, are well below the dose limit recommendations of the International Commission on Radiological Protection (ICRP, European Union) (11,12) and the National Council on Radiation Protection and Measurements (NCRP, United States) (13) as shown in Table 3 (14,15). The estimated annual effective dose of catheterization lab operators is comparable to the natural background exposure of about 2.4 mSv/year (16) and the mean cumulative effective dose from medical imaging procedures in the United States in nonelderly adults (2.4 ± 6.0 mSv/year) (17).
Whereas the occupational doses for coronary operators are similar to previously described doses for U.K. diagnostic radiologists in 1993 and 2001 of 0.5 and 0.15 mSv/year, respectively, the doses for peripheral operators correspond to doses described 40 to 50 years ago (1960 to 1976) for radiologic technicians in hospitals with 3.6 mSv/year (18). If we assume a distribution of about 65% UL, 25% pelvic, and 10% BTK procedures for a peripheral operator, as in our study, the radiation exposure would be increased by a factor of about 6 for the body, 4 for the eyes, and 2 for the hands compared with the exposure of a coronary operator (Table 1).
Although dicentric chromosomes indicate acute exposure of moderate to high radiation doses (0.1 to 5 Sv), they are not suitable to assess doses from long-term, low-dose exposure in individual probands because of their removal over time. The FISH technique is also not suited for routine monitoring of low-dose occupational radiation exposure in individuals because of its limited sensitivity. Also, the size of the irradiated volume and the amount of the dose have a significant influence on the yield of aberrations found.
In this context, an extended study including a higher number of operators is urgently needed to reveal a correlation between the number of radiation-induced chromosome aberrations and different types of catheterization lab procedures.
Estimated effective doses of the operators did not exceed dose limit recommendations. However, threshold doses are defined as estimated doses for 1% incidence of a detectable tissue reaction and refer to deterministic effects, such as inflammation, ulceration, or necrosis, which could occur in patients. In contrast, it is not possible to define threshold doses for stochastic effects from mutagenesis, such as cancer or hereditary effects, which might develop in operators resulting from long-term low-dose radiation exposure (19).
The relationship between long-term, low-dose radiation exposure and the risk of cancer and deoxyribonucleic acid damage as well as tissue deterioration is not yet known completely. Because of the complex mechanism of deoxyribonucleic acid repair, as well as genetic differences in radiation response, a supposed linear-no-threshold dose-effect relationship, proven for doses above 50 to 100 mSv (20,21), may lead to a false estimation of the impact of long-term, low-dose radiation exposure (22). Russo et al. (23) reported an enhanced antioxidant defense by overproduction of glutathione in erythrocytes and caspase-3 in lymphocytes in interventional cardiologists, chronically exposed to low-dose radiation. The longer the time interval between each radiation exposure, the more effective the biological repair (19). However, results from the 15-country study (24) (radiation workers, n = 407,391) show an excess relative risk for cancer of about 2% in workers with a mean radiation exposure of 19.4 mSv. In addition, Eisenberg et al. (25) found a 3% increase in the risk of cancer over 5 years for every 10 mSv of low-dose radiation exposure in patients after acute myocardial infarction. Recently, Roguin et al. (26) reported several cases of brain cancer on the left side of the head in interventional cardiologists who received annual head doses of 20 to 30 mSv. In contrast, it was not possible to identify an increased incidence of cancer in Japanese atomic bomb survivors at dose levels less than 100 mSv. In summary, the generally accepted estimated excess relative cancer risk from an acute single dose of 100 mSv is 1% (Biological Effects of Ionizing Radiation [BEIR] VII report) (20). In case of chronic low-dose exposures, it is recommended to reduce the estimated risk by using a “dose and dose-rate effectiveness factor” of 1.5 (BEIR VII Committee, NCRP) or 2.0 (ICRP), respectively. The risk of heritable effects due to radiation is much lower than that of cancer. Such effects have not been demonstrated in humans to date (21).
Eyes and hands
During peripheral endovascular procedures, the estimated occupational eye dose of operators may achieve about 26 mSv/year, exceeding the recently tightened ICRP recommended dose limit of 20 mSv/year. However, the NCRP has not followed this recommendation so far (Table 3). Damage and mutation in the germinative region of the lens epithelium is proposed to be the most important cause of radiation-induced cataract (27). For this reason, the use of lead glasses with an expected dose reduction by a factor of 10 at least is essential, particularly for peripheral procedures.
The estimated annual dose to the hands is 33.1 to 119 mSv/year. In this study, hand doses were measured at the wrist. Thus, the dose to an area of skin on the hand may be underestimated by a factor of 3 (2). Maybe protective drapes should be taken into consideration more frequently to reduce scatter notably in pelvic procedures. Patient drapes may reduce radiation exposure to the hands by a factor of 29 (28).
Operators' eyes and hands will benefit from an appropriate above- and below-table shielding most certainly.
In this single-center experience data were obtained from three operators. Further multicenter, multinational studies with a higher number of participants are needed to show more precise correlations between procedural characteristics and radiation exposure of the operators.
Endovascular peripheral procedures are accompanied with a higher radiation exposure of catheterization lab operators than coronary procedures are, irrespective of a patient's BMI or fluoroscopy time. Although recommended dose limits were not exceeded in our study, radiation exposure may have biological effects depending on the type and frequency of procedures performed, the time lag between the procedures, the fluoroscopy time, and a patient's BMI. Therefore, protection devices, spectral filtration, pulsed fluoroscopy, and low frame rates should be used whenever possible.
The authors thank Philips Deutschland GmbH for providing the DoseAware system and Naser Ahmeti for essential data collection.
Mr. Fricke is an employee of Philips Healthcare. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- body mass index
- below the knee
- coronary angiography
- confidence interval
- dose area product
- estimated effective dose
- fluorescence in situ hybridization
- International Commission on Radiological Protection
- National Council on Radiation Protection and Measurements
- percutaneous coronary intervention
- upper limb
- Received March 4, 2013.
- Revision received April 26, 2013.
- Accepted May 9, 2013.
- American College of Cardiology Foundation
- Martin C.J.
- Ainsbury E.A.,
- Bakhanova E.,
- Barquinero J.F.,
- et al.
- Hirshfeld J.W. Jr..
- Theocharopoulos N.,
- Damilakis J.,
- Perisinakis K.,
- Manios E.,
- Vardas P.,
- Gourtsoyiannis N.
- Kim K.P.,
- Miller D.L.
- Cousins C.,
- Miller D.L.,
- Bernardi G.,
- et al.
- ↵(2000) WHO. Efficacy and Radiation Safety in Interventional Radiology (World Health Organization, Geneva, Switzerland).
- RöV. Verordnung über den Schutz vor Schäden durch Röntgenstrahlen [Regulation for the protection against harm caused by X-rays]: Röntgenverordnung [German X-ray regulation], 2003
- Hendry J.H.
- Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation
- Russo G.L.,
- Tedesco I.,
- Russo M.,
- Cioppa A.,
- Andreassi M.G.,
- Picano E.
- Eisenberg M.J.,
- Afilalo J.,
- Lawler P.R.,
- Abrahamowicz M.,
- Richard H.,
- Pilote L.
- Kleiman N.J.