Radiation burden of patients undergoing endovascular abdominal aortic aneurysm repair

Article Outline

• Abstract
• Materials and methods
  • Patients – EVAR
  • Dosimetry-EVAR
• Results
  • EVAR procedure
  • CT angiography
• Discussion
  • Stochastic risks
  • Deterministic risks
• Conclusion
• Author contributions
• References
• Copyright

Introduction

Endovascular repair of abdominal aortic aneurysm (EVAR) requires the patient's extended exposure to x-rays, before, during, and after the intervention. The aim of this study was to determine the radiation exposure of patients undergoing EVAR and to assess the probability for the induction of both late and early radiation-related effects.

Methods

During the period of May 2006 to December 2007 EVAR was carried out in 62 patients using a mobile C-arm unit. The following dosimetric quantities were assessed: fluoroscopy time, cumulative dose in air, dose-area product, field area, and peak skin dose.

Results

The duration of fluoroscopy and the body mass index were found to be the main factors that influence the radiation burden in our hospital. The mean effective dose per procedure, 6.2 mSv, was between that from a planar coronary angiography and a coronary angioplasty. Taking into account the computed tomography (CT) procedure-related angiographies carried out during the first year, patients receive a total effective dose of about 62 mSv within the first year. In vivo dosimetry showed that the peak skin dose was linearly correlated with cumulative dose in air and did not exceed 1.0 Gy, ie, it was less than the threshold for any acute skin reaction.

Conclusion

Repair of abdominal aortic aneurysm results in substantial radiation burden. Radiation-related risks for carcinogenesis and skin injuries are factors that have to be taken into account in the selection of the strategy of each facility.

Fluoroscopy-guided therapeutic procedures, such as endovascular aneurysm repair, are an essential part of the contemporary practice in medicine. A drawback of these procedures is the associated radiation risk. Taking into account that complications may appear even many years after endovascular aneurysm repair of the abdominal aorta (EVAR), the life-long follow-up often includes computed tomography (CT) imaging, a modality that requires a substantial radiologic burden. Therefore, the assessment of the corresponding radiologic burden to the patient and the definition of the steps required to keep it as low as reasonably achievable is one of the many prerequisites for the choice of the optimum treatment strategy in each facility. In addition, almost anything that helps to reduce patient exposure will also reduce staff exposure.

Health effects induced by ionizing radiations, such as x-rays, can be divided into two categories: (1) stochastic effects, ie, induction of cancer in exposed individuals and heritable diseases in their offspring (so far, there is no direct evidence that in humans exposure of parents to radiation leads to excess heritable disease in offspring);¹ and (2) deterministic or non-stochastic effects (tissue reactions), observable only if the dose is above some threshold.

In radiologic protection, stochastic effects are widely assumed related with DNA damage response processes in single cells. Protection measures are employed under the assumption that the probability of induction of such effects in an organ or tissue is directly proportional to the mean dose absorbed in the irradiated organ or tissue (the quotient of absorbed energy to the organ or tissue to its mass, expressed in Gy, J/kg) with no dose threshold (“linear-non-threshold” model). The quantity effective dose (assessed in mSv), defined as the weighted mean value of the doses absorbed in a number of organs or tissues,¹, ² is widely used as an estimator of the radiologic risk related to stochastic effects (7.3% per Sv for the whole population).² However, taking into account the age-dependent radiosensitivity and the time interval between exposure and possible clinical appearance of these effects, the age distribution of the exposed patients and their health status should to be taken into consideration. For example, the life expectancy of patients with an aortic aneurysm drastically reduces the probability of clinical manifestations of some types of cancer,¹, ² as well as the induction of heritable diseases.

Deterministic effects, sometimes also called tissue reactions, are due in a large part to killing/malfunction of many cells in an organ or tissue following doses above a specific dose threshold. Early harmful tissue reactions can be inflammatory-type reactions as a result of cell permeability changes and histamine release, such as early transient erythema, seen in skin a few hours after a dose of at least 2 Gy from x-rays within a short time period when the exposed area of the skin is relatively large.¹, ², ³, ⁴, ⁵ The threshold of temporary epilation and the main erythematous reaction, whose onset is after approximately 10 days, are larger, 3 and 6 Gy, respectively.⁴ The induction of more severe effects, such as dermal atrophy, necrosis, and desquamation, that are related to cell loss, can be induced by even higher skin doses. Taking into account the threshold values of the organ and tissues irradiated directly during EVAR, peri-procedure imaging, and the rapid attenuation of the diagnostic x-ray beams inside the human body, a peak dose less than 2 Gy in any skin area within a short time period can exclude the induction of serious harmful deterministic EVAR-related effects in men and non-pregnant women.

The aim of this study was to determine the radiation exposure of patients undergoing EVAR and to assess the corresponding radiologic risks. Reduction in the patient's radiation burden during fluoroscopically-guided interventions usually results in a reduction of the staff radiation burden.

Back to Article Outline

Materials and methods 

Patients – EVAR 

During the period of May 2006 to December 2007, 62 EVAR consecutive procedures were carried out in at the Ioannina University Hospital (IUH) in 62 patients (60 male and 2 female, Table I). The mean age and body mass index (BMI) of the patients were 74-years-old (range, 50-87-years-old) and 28.9 kg/m² (range, 21-40 kg/m²), respectively. All repairs were carried out by one of the authors (M.M.) assisted by another certified vascular surgeon and a trainee. A young medical physicist was also physically present in the operation theater during some of the procedures. The procedures were performed using a mobile C-arm unit BV Libra (Philips, Best, The Netherlands). The unit used is equipped with a 3.15 kW generator, a 6.35 mm Al_eq filter (1.5 mm Al + 0.1 mm Cu), an iris-type collimator, and a 9” image intensifier. Both the tube current and high voltage were selected automatically in each mode and their mean values during the continuous low dose rate mode used for fluoroscopy in the present study, were 3.1 mA and 81 kV, respectively, corresponding to a 5.3 mm Al half-value layer of the x-ray beam. During the procedure, the x-ray source was placed under a surgical table (Getinge, AlphaMaxx, Rochester, NY).

Table I. Patients' characteristics (n = 62)

Dosimetry-EVAR 

The dosimetric data were obtained and analyzed by scientists of the IUH Medical Physics Laboratory (J.K-E., D.D., and K.D.). More specifically, the following quantities from the C-arm unit were registered: (1) duration of fluoroscopy, t; (2) dose-area product (DAP), sometimes also called air-kerma area product, a quantity related to the total number of photons emitted from the x-ray unit; (3) cumulative absorbed dose in air, D, at a 69.5 cm distance from the x-ray source (30 cm from the detector) with no patient in place (free-in-air geometry).

The ratio of the last two quantities, DAP/D, ie, the area of the radiation field at a distance 70 cm from the x-ray source, was calculated.

A low sensitivity Kodak X-Omat V radiographic film (Eastman Kodak Co, Rochester, NY), 14” × 17”, was positioned between the patient's body and the mattress to document the sizes and the locations of the radiation fields at the patient's skin, assess the source to skin distance, and provide an immediate qualitative assessment of spatial skin-dose distribution. The film did not add discomfort to the patient lying on the film for extended procedure times, since it does not have to be in a rigid cassette. The reduction of the skin dose due to the presence of the radiographic film along the beam path was neglected in the present study.

The possibility of exceeding a threshold for induction of a deterministic effect was based on peak skin dose assessment. For this reason, the cumulative doses absorbed in air were correlated with the peak skin doses assessed by in vivo measurements in a subgroup of patients. The empirical relationship between the directly measured peak skin doses and the indications of the fluoroscopic unit can be used for the assessment of the radiation burden of the entire group of patients. More specific, in vivo dose measurements on the patient's back were carried out during the EVAR procedure in 5 randomly chosen patients using individually calibrated LiF:Mg,Ti thermoluminescent dosimeters (TLDs) and a Harshaw 3500 reader (Harshaw-Global, Solon, Ohio). Thirty-two dosimeters divided into eight groups were positioned between the radiographic film and the patient's back forming an array 30 cm–long along the patient's spine. Optical density measurements of the radiographic film were used to confirm the adequate coverage of the patient's back to identify the location of the peak skin dose. A known radiation dose was given in 12 TLDs. The remaining six were not irradiated and were used to determine the background signal.

In the current study, the 1990 definition of the effective dose by the International Commission on Radiological Protection was used, that takes into account the ratio of the energy absorbed in 22 organs or tissues to the organ or tissue mass.² The effective dose to DAP ratio proposed by Geijer et al,⁶ 0.145 mSv/Gy cm², was used for the calculation of the effective dose in all patients, despite the fact that the exact value depends among other things on the anatomic location of the aneurysm to be treated, beam penetration (mainly depended on beam filtration and the automatically selected kV), and body-build.

A Philips MX8000 IDT CT unit (Philips Medical Systems, Cleveland, Ohio) equipped with 24 rows of detectors was used for preoperative planning and post-EVAR surveillance (120 kV, 200 mA, 1 second, reconstruction collimation 16 × 1.5 mm, pitch 1.2, and current modulation according to the body habitus). The effective dose was calculated using the ImPACT CT Patient Dosimetry Calculator (version 0.99v) (ImPACT, London, UK) based on Computed Tomography Dose Index measurements carried in air and a pencil-type ionization chamber. The skin dose to the patient's back was assessed using thermoluminescent dosimetry.

Back to Article Outline

Results 

EVAR procedure 

The quantity DAP was linearly correlated with fluoroscopy time, and BMI (n = 62, r = 0.90, standard error of estimate (SEE) = 13.6).

About 22% of the entire DAP value was related to image registration, while the remaining to fluoroscopy. Taking into account that a 42.5 Gy cm² mean DAP value was found (range, 9 and 139 Gy cm², Table II), the mean effective dose per EVAR procedure was 6.2 mSv (range of values 1.3 to 20 mSv).

Table II. Dosimetric parameters (n = 62)

DAP, Dose area product.

The cumulative dose in air, D, ranged between 49 and 898 mGy (Table II) and increased linearly with t and BMI (n = 62, r = 0.95, SEE = 7.1).

Similar linear relationships were found between either DAP or D and time and waist circumference.

Film dosimetry showed that the “source-skin” distance, ∼65 cm on average, was smaller than the 69.5 cm distance, where D was assessed, free-in-air. Assuming a 35% dose increase in patient's skin due to photons backscattered in the human body⁴, ⁷, ⁸, ⁹ and a 27% attenuation of the x-ray beam at the surgical table (according to the direct measurements), the mean and the maximum anticipated cumulative doses (in water) were 0.30 and 1.01 Gy, respectively. However, such an approach overestimates the peak skin doses. Non-overlapping and partially overlapping beam portals (locations where the beam enters into the human body) were used. From the other hand, this approach assumes that the dosimetric parameters provided by the fluoroscopy units are accurate.

In vivo dosimetry, carried out in a subgroup of patients with DAP and D values representative of the entire group, showed that the peak skin dose, D_peak, increased linearly with D (r = 0.976, SEE = 33).

Therefore, the mean and the maximum peak skin doses were 227 and 773 mGy, respectively, ie, about 30% lower than the calculated peak skin doses based on the registered cumulative dose values.

CT angiography 

CT angiographs (two-phase angiography before repair, and a single-phase within the first month after repair) contribute to the same skin area an additional dose of ∼50 mGy. Therefore, the mean and the maximum peak skin doses within a short time period were 0.28 and 0.82 Gy, respectively. This finding is in accordance with the clinical observation of absence of any radiation-related deterministic skin reactions in the studied group of patients.

Back to Article Outline

Discussion 

Fluoroscopic imaging is extensively used in interventional procedures, such as EVAR, to localize the lesion, monitor the procedure, and control and document the end result. Procedures requiring large fluoroscopy times are associated with a significant radiation hazard. Therefore, quantities that can be assessed easily, such as fluoroscopic time, the DAP and D, should be evaluated and recorded in the patient's file. Patient size and procedural aspects, such as the location(s) of the beam, beam angle, distance of the patient's skin from the x-ray tube, type of x-ray unit used, size of the radiation field, irradiation mode (normal or high dose continuous dose rate, pulse rate during both fluoroscopy and image registration), beam attenuators between the tube and the patient's skin are factors that influence the radiological burden.

The mean fluoroscopy time found in the present study, 22.6 minutes, was comparable or lower than the mean values of 39.4, 28.4, 23.7, and 21 minutes reported by Lipsitz, Geijer, Ferrar, and Weerakkody, but higher than the 13.0 minutes reported by Ho et al.⁶, ¹⁰, ¹¹, ¹², ¹³ The mean DAP value found in the present study, 42.5 Gy cm², was in between the 34 and 84 cm² mean values for coronary angiography and angioplasty at IUH during the study period. The median DAP value, 37.4 Gy cm², was lower than the 150 and 60.1 Gy cm² median values reported by Weerakkody¹² and Geijer,⁶ respectively. In this comparison, one has to take into account that Weerakkody et al excluded from their analysis ∼7% of the carried procedures, that resulted in values several orders of magnitude greater than the typical values recorded. Moreover, the median effective dose, 27 mSv, was about four times higher than the one found in the present study, 6.2 mSv, neglecting beam attenuation in the surgical table and mattress. The calculated peak skin dose ranged between 0.51 to 3.74 Gy, with a median value of 0.85 Gy, (the 2 Gy threshold for skin damage was exceeded in 27% of the patients) vs 0.19 Gy in the present study. In addition, the effective dose due to the three CT scans carried out at IUH during the first year after the EVAR procedure resulted in an effective dose of 33 mSv, almost 40% of the effective dose reported by Weerakkody et al, 79 mSv, mainly due to the elimination of the unenhanced phase in the present study.

Stochastic risks 

The procedure-related 6.2 mSv mean effective dose, a quantity assumed to be proportional to the probability for induction of stochastic effects, was between the mean doses related to planar coronary angiographies and angioplasties at IUH, 5.5 and 15 mSv, respectively. However, EVAR patients in practice receive higher radiation burden than patients with severe coronary artery stenosis, due to the procedure-associated CT aortographies. The two-phase preoperational CT abdominal/pelvic angiography carried out at IUH, adds an effective dose of ∼22 mSv, while the three single-phase enhanced aortographies prescribed during the first post-EVAR year require a total dose of about 33 mSv. Therefore, patients are getting within a single year an effective dose of about ∼62 mSv, ie, ∼30 times higher than that received annually from natural background radiation in most of the countries. Assuming that a risk factor of 4% per Sv can be used (a value obtained by matching the age distribution data of the present study with the excess lifetime mortality data of the American National Academic of Science),¹⁴ the excess mortality due to radiation-induced carcinogenesis from the procedures carried out during the first year was estimated to be 1 per ∼400. Under the assumption that the follow-up period is extended up to 10 years post-procedure with a single-phase CT angiography annually, the repair and its surveillance (pre-EVAR CT angiography, endovascular procedure, and 12 post-EVAR CT angiographies) induce a total effective dose of ∼160 mSv, corresponding to excess mortality of 1 per ∼155.

Deterministic risks 

The skin, the largest tissue of the body, is irradiated non-uniformly, and some of its sections may receive high doses. Although there are some reports on radiation-induced basal cell and squamous cell skin cancers,⁴ skin sensitivity for stochastic effects is considered to be low.¹, ² On the contrary, the thresholds for skin reactions are relatively low. Both mild and severe iatrogenic skin injuries have been reported in patients undergoing fluoroscopically-guided procedures,⁴, ⁵ mainly as a result of the use of inappropriate equipment and poor operational techniques. In the present study, the maximal peak skin dose assessed in the studied group of patients, 0.82 Gy within a short time period, was almost half of the threshold for transient erythema.

The use of the indications provided by the fluoroscopic unit, D and DAP, for the calculation of the peak skin dose introduces in practice large errors, such as inaccuracies related to the calibration of the meter with beam spectra similar to those used during clinical practice, attenuation of the photon beam in the surgical table and the mattress, use of more than one irradiation ports, uncertainties in the exact focus to skin distance, and the values of the backscattering factors to be used. According to the present study, peak skin doses to individual patients treated at IUH can be assessed under daily clinical conditions indirectly based on the cumulative dose in air provided by the fluoroscopic unit (Fig). However, the ∼15% numerical difference between the peak skin dose and D found in the present study is applicable only under conditions similar to those in the present study. For example, use of a different type of surgical table than the one used in the present study (it causes a 28% dose reduction at 81 kV beams) and different focus to skin distance, may modify this value.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 

  Peak skin dose assessed by in vivo dosimetry vs cumulative dose in air. TLD, thermoluminescent dosimeters.

Taking into account that additional procedures might be required to be carried out in EVAR patients within a short time period exposing the same skin area, dosimetric data have to be recorded in the patient's file. In those patients that the EVAR-related cumulative dose, D, exceeds 1.1 Gy, patient follow-up procedures are carried out at IUH to address the potential for serious skin injuries according to the recommendations of the International Commission on Radiological Protection.⁴

Back to Article Outline

Conclusion 

Justification and optimization of the strategies used for the treatment of aortic aneurysms must commensurate with the medical purpose. The moderate radiation exposure due to an EVAR procedure is coupled with substantial exposure due to CT angiographies, according to the recently proposed characterizations of the various procedures by the International Commission on Radiological Protection.¹ In addition to the radiation-related risk for carcinogenesis, the induction of harmful skin injuries has to be considered, mainly in obese EVAR patients undergoing additional procedures, such as a coronary angioplasty closely spaced in time with EVAR. The size of the radiation burden related to the treatment and the follow-up calls for further investigations on strategies to reduce the radiological risk as low as reasonably achievable consistent with the medical need. The optimum strategy in each facility has to be studied and justified based on solid clinical and radiobiologic evidences.

Back to Article Outline

Author contributions 

Back to Article Outline

References 

1. International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection, ICRP Report 103. Annals of the ICRP. 2007;37:1–332
   • View In Article
   • Full-Text PDF (45 KB)
   • CrossRef
2. International Commission on Radiological Protection. 1990 Recommendations of the International Commission on Radiological Protection, ICRP Report 60. Annals of the ICRP. 1991;21:1–201
   • View In Article
   • MEDLINE
   • CrossRef
3. International Commission on Radiological Protection. Nonstochastic effects of ionizing radiation. ICRP Report 41. Oxford, UK: Pergamon Press; 1984;
   • View In Article
4. International Commission on Radiological Protection. Avoidance of radiation injuries from medical interventional procedures (ICRP Report 85). Annals of the ICRP. 2000;30:1–70
   • View In Article
   • Full Text
   • Full-Text PDF (35 KB)
   • CrossRef
5. Geleijns J, Wondergem J. X-ray imaging and the skin: radiation biology, patient dosimetry, and observed effects. Radiat Prot Dosimetry. 2005;114:121–12
   • View In Article
6. Geijer H, Larzon T, Popek R, Beckman KW. Radiation exposure in stent-grafting of abdominal aortic aneurysm. Br J Radiol. 2005;78:906–912
   • View In Article
   • MEDLINE
   • CrossRef
7. National Radiological Protection Board. National protocol for patient dose measurements in diagnostic radiology. Didcot, UK: National Radiological Protection Board; 1992;
   • View In Article
8. Ma CM, Coffey CW, DeWerd LA, Liu C, Nath R, Seltzer SM, et al. AAPM protocol for 40-300 kV x-ray dosimetry in radiotherapy and radiobiology. Med Phys. 2001;28:868–892
   • View In Article
   • MEDLINE
   • CrossRef
9. Petoussi-Henss N, Zankl M, Drexler G, Panzer W, Regulla D. Calculation of backscatter factors for diagnostic radiology using Monte Carlo methods. Phys Med Biol. 1998;42:2237–2250
   • View In Article
10. Lipsitz EC, Veith FJ, Ohki T, Heller S, Wain RA, Suggs WD, et al. Does endovascular repair of aortoiliac aneurysms pose a radiation safety hazard to vascular surgeons?. J Vasc Surg. 2005;32:704–705
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (64 KB)
    • CrossRef
11. Ferrar AV, Mohan IV, Steel TS, Harris JP, White GH, Stephen JS. Radiation dose received by patients undergoing AAA repair. ANZ J Surg. 2005;75:A115;(abstract)
    • View In Article
    • CrossRef
12. Weerakkody RA, Walsh SR, Cousins C, Goldstone KE, Tang TY, Gaunt ME. Radiation exposure during endovascular aneurysm repair. Br J Surg. 2008;95:699–702
    • View In Article
    • CrossRef
13. Ho P, Cheng SWK, Wu PM, Ting ACW, Poon JTC, Cheng CKM, et al. Ionizing radiation absorption of vascular surgeons during endovascular procedures. J Vasc Surg. 2007;46:455–459
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (238 KB)
    • CrossRef
14. National Academy of Sciences. Health effects of exposure to low levels of ionizing radiation. BEIR V report. Bethesda, Maryland: National Academy of Sciences; 1990;
    • View In Article