Image Wisely, a joint initiative of ACR, RSNA, ASRT and AAPM,
provides information to the medical community to promote safety in medical imaging.
Steven Y. Huang, MD; A. Kyle Jones, PhD, University of Texas, MD Anderson Cancer Center, Houston, TX
Image-guided procedures can be both diagnostic and therapeutic. Despite their utility, the effective radiation dose to patients can be quite high. Based on estimates available from the literature and reports to the FDA, the frequency of major radiation injury is likely between 1:10,000 and 1:100,000 fluoroscopy procedures .
Radiation dose to the patient is influenced by many factors, such as the nature and complexity of the procedure, patient, prior radiation exposure, operator, and fluoroscopic equipment. In this section, we focus on how the procedure and patient may affect radiation dose.
Compiling reference levels for radiation exposure during fluoroscopy in interventional radiology (IR) is difficult. There can be substantial variation in similar procedures because of patient, operator and equipment-related factors. The largest registry of dose data in the United States is the Radiation Doses in Interventional Radiology Procedures (RAD-IR) study [2,3]. This study contains data on 2,142 procedures, and reference levels were proposed for certain IR procedures (table 1). These reference levels should serve as a general guide. Dogmatic adherence to reference levels fails to take important factors into account.
The goal of all IR procedures is patient care. A low radiation dose may degrade image quality and potentially sacrifice diagnostic and therapeutic capability. At the other extreme, high-dose cases expose patients to the risk of radiation injury. It is important to note, however, that in some circumstances, a planned intervention to achieve a life-preserving outcome may exceed the threshold for deterministic radiation injuries. Prior to cases in which a substantial radiation dose is expected, a discussion of radiation injury should be part of the informed consent process.
Nature and Complexity of Procedure
One of the difficulties in applying reference levels to IR is that the procedures are unique, potentially complex and complications may arise during the course of the procedure. Furthermore, IR procedures are increasingly performed for therapy and, as a result, continue until treatment is completed.
At our institution, it is not uncommon for radiation dose from hepatic tumor embolization to exceed the proposed 1.9 Gy reference level published by the RAD-IR study. Procedure complexity is increased by 1) the size and number of tumors, 2) quantity, caliber and tortuosity of supplying arteries, 3) and associated problems with vascular access.
To mitigate the use of radiation exposure in complex cases, operators should thoroughly review all pre-procedure imaging. Review of cross-sectional imaging can assist in treatment planning, device selection, and, in the case of peripheral arterial disease, may minimize or exclude the necessity of angiography altogether.
In addition, the adjunctive use of ultrasound during fluoroscopic procedures should be used whenever possible. When we compared our institutional database of patient radiation dose to those reported by the RAD-IR study, we noted that cumulative air kerma for nephrostomy and biliary access was lower at our institution, a finding we attributed to our routine use of ultrasound guidance .
Optimizing Fluoroscopic Study
There may be moments during a fluoroscopic procedure when high spatial resolution is necessary. Spatial resolution can be improved by using the magnification mode of the image intensifier on most modern flat panel detectors. Magnification almost always results in higher patient radiation dose.
Electronic, geometric and digital techniques are available to magnify the image in fluoroscopy. Electronic magnification occurs within the image intensifier, with most units having three to five magnification modes. Generally, increasing magnification levels results in a decrease to the brightness gain of the image intensifier.
The automatic brightness control algorithms used in analog image intensifiers will compensate by increasing the radiation dose rate by the square of the magnification factor. Thus, if magnification increases by a factor of four, the radiation dose rate increases by a factor of 16. In flat panel systems, the increased dose rate with magnification is typically less .
In geometric image magnification, the image receptor is moved further from the patient (increasing scatter radiation) or the patient is moved closer to the source (increasing patient dose). Another factor to consider is that unless a very small focal spot is used (e.g., 0.3 mm), geometric magnification results in a larger penumbra, which can degrade spatial resolution.
Finally, in lieu of geometric and electronic magnification, digital magnification can be used on fluoroscopic images without any additional radiation dose to the patient. This magnification technique, however, can only be applied after image acquisition, and, therefore, is best used as a problem-solving tool.
Similar to other digital media, digital magnification crops a portion of the image and enlarges it to size. When this is performed, image quality (i.e., resolution) decreases, but may still be acceptable depending on the indication for the exam.
Most studies performed in IR utilize some form of contrast (e.g., iodine, carbon dioxide or room air). High-contrast images are optimized by enhancing the sharpness of the edges. This enhancement, however, also increases quantum mottle. The degree of edge enhancement is often set by the manufacturer of the fluoroscopy equipment or medical physicist. In practice, low-contrast images are rarely needed for diagnostic purposes in IR (e.g., vascular access or placement of pleural and intraperitoneal drainage catheters). In these cases, fluoroscopy is used to ensure adequate device position, and fine anatomical detail is unnecessary.
Temporal resolution is the ability to detect that an object has moved over time. IR procedures require varying degrees of temporal resolution. Rapidly dynamic systems, such as the thoracic aorta, will need a higher imaging frequency than more static systems, such as the biliary tree.
During pulsed fluoroscopy, temporal resolution increases with pulse rate assuming constant pulse width. Typical pulse rates on most modern fluoroscopy units range from 3.75 to 30 pulses per second. Pulsed fluoroscopy at the lowest pulse rate should be used to decrease patient dose. In our practice, pulse rates of 7.5 to 15 are generally used for dynamic systems while lower pulse rates are used for relatively static systems.
Compared to continuous fluoroscopy, dose savings to the patient have been calculated to be 49 percent for pulsed fluoroscopy operated at 7.5 frames per second and 80 percent for pulsed fluoroscopy operated at 3.75 frames per second .
Table 1: Proposed patient reference levels, not corrected for body habitus, for certain interventional radiologic procedures 
|Procedure||Reference Dose (Gy)||KAP (Gy cm2)||Fluoroscopy Time (min)||No. of Images|
|Transjugular intrahepatic portosystemic shunt creation||3.00||525||60||300|
For stone access
|Renal or visceral angioplasty|
|Bronchial artery embolization||2.00||240||50||450|
|Uterine fibroid embolization||3.60||450||36||450|
|Other tumor embolization||2.60||390||35||325|
|Gastrointestinal hemorrhage localization and treatment||3.80||520||35||425|
|Embolization in the head|
|Pelvic artery embolization for trauma or tumor||2.50||550||35||550|
|Embolization in the spine for AVM or tumor||8.00||950||130||1500|
|Reprinted from Radiology;253(3), Miller DL, Kwon D, Bonavia GH, Reference level for patient doses in interventional radiology: proposed initial values for U.S. practice — Table 3, p760, 2009, with permission from RSNA.|
Fluoroscopy is restricted to patients who do not exceed the table weight limit, which varies by manufacturer but is generally 350 pounds. The maximum clearance between the table and image intensifier is approximately 45 cm. For larger patients, radiation dose is severely influenced by the patient’s body habitus.
Recall that half value layer is the amount of tissue that will reduce the quantity of X-rays to half the original number. At 60 kV, an increase in patient thickness by 3.5 cm doubles the number of X-rays required to penetrate the patient. It is important to ensure that the protocol selected prior to the procedure, which sets variables such as automatic brightness control and tube current, is appropriate for both the patient and procedure.
The website imagegently.org provides an extensive discussion of fluoroscopy and radiation reduction techniques particular to the pediatric population.
Patient positioning during fluoroscopy is important to visualize anatomy, enhance image quality and optimize patient radiation dose. Radiation exposure is influenced by path through the body. Thus, orientations which yield high dose rates (i.e., tube angulation) should be used only when absolutely necessary.
The post-operative patient is a unique challenge for interventional radiologists. Prone position for a patient following laparotomy may not be feasible. In these cases, patient safety and comfort must be weighed against technical success of the procedure and radiation dose. If a decubitus position is necessary, the same methods to minimize radiation dose apply with the added caveat that the source should be furthest away from the operator.
Patient cooperation is necessary when performing fluoroscopy. Patients are not only expected to remain motionless during the procedure, but also they may be asked to suspend respiration for periods up to 10-20 seconds.
Any patient movement creates motion artifact. To combat this, operators may repeat the imaging or increase the fluoroscopic frame rate, which both increase radiation dose. The patient’s ability to cooperate during the exam should be evaluated during the pre-procedure consent with anesthesia consultation, if necessary.
On occasion, fluoroscopically guided procedures may be required during pregnancy. A consensus statement was produced by the Society of Interventional Radiology and the Cardiovascular and Interventional Society of Europe regarding radiation management for interventions using fluoroscopic or CT-guidance during pregnancy .
As with all medical procedures, the benefits of the fluoroscopically guided procedure must exceed the potential radiation risks to the patient and fetus. The magnitude of the risk to the fetus is dependent on the fetus’ gestational age and the absorbed dose.
Radiation risks are most pronounced during pre-implantation, organogenesis and during the first trimester. Table 2 lists specific radiation effects to the fetus and their corresponding threshold doses. The techniques to reduce radiation dose to the fetus are generally the same radiation reduction techniques used elsewhere. Additional considerations are to consider arm/neck access, decrease tube current (mA)/increase tube voltage (kVp) and use intravascular ultrasound.
Some patients are at risk for exaggerated radiation response because of an underlying condition. Prior exposure to high doses of radiation is perhaps the single greatest predictor of radiation injury. Repair of DNA injuries is usually complete by 24 hours, but repair and repopulation of damaged cells may take several months.
During the pre-procedure evaluation, patients should be screened for prior high-dose radiation, location and skin changes. Patients with active skin injury from prior radiation do occasionally require additional fluoroscopy. In this scenario, Balter et al. described the use of solder wire to delineate the area so that fluoroscopy could be avoided .
In practice, radiopaque wires, catheters or dilators could also be used. Also, patients with diabetes mellitus, connective tissue disease and homozygosity for ataxia telangiectasia should be informed of an elevated risk of radiation injury .
Radiation injury during fluoroscopic procedures is influenced by many factors, including the procedure and patient. Adequate pre-planning and attention to techniques to minimize radiation dose during the case are integral components of mitigating the risk of radiation injury.
Table 2: Deterministic Radiation Effects at Different Stages of Gestation 
|Stage of Gestation (wk)||Possible Radiation Effect||Dose Characteristic||Estimated Threshold Dose (mGy)|
|3-4||Most sensitive period for the induction of embryonic death||Minimum lethal dose (from animal studies)||100-200|
|4-8||Embryo is also predisposed to the induction of major malformations and growth retardation||Minimum lethal dose (from animal studies)||250 (at 18 d), >500 (at >50 d)|
|Minimum dose for growth retardation||200-500|
|8-15||Most sensitive period for irreversible whole-body growth retardation, microcephaly, and severe mental disability||Minimum dose for growth retardation||250-500|
|Threshold for severe mental disability||60-500|
|Decrease in IQ can occur at lower doses||~ 100|
|16-term||Higher exposures can produce growth retardation and decreased brain size and intellect, although the effects are not as severe as occurs from similar exposures during midgestation||Minimum lethal dose (from animal studies)||>1,500|
|Minimum dose for severe mental disability||>1,500|
|Decrease in IQ can occur at lower doses||>100|
Note: IQ=intelligence quotient
Reprinted from J Vasc Interv Radiol;23, Dauer LT, Thornton RH, Miller DL, Damilakis J, Dixon RG, Marx MV, et al., Radiation management for interventions using fluoroscopic or computed tomographic guidance during pregnancy: a joint guideline of the Society of Interventional Radiology and the Cardiovascular and Interventional Radiological Society of Europe with Endorsement by the Canadian Interventional Radiology Association, p22, 2012, with permission from the Society of Interventional Radiology.