Figure B shows a series of images acquired during a digital subtraction angiography examination performed using an anthropomorphic pelvis phantom. The first column shows the initial images that are used as a mask for subtraction from later images that also contained iodinated contrast material. The second column shows an example of images obtained after an operator added thin tubes with diluted iodinated contrast material that are intended to simulate blood vessels. Note that in each of the images shown in the second column there are three (horizontal) tubes, which contain iodinated contrast material that has been diluted by differing degrees. The last column in Figure B shows the resultant subtracted image, namely the image in column two subtracted from the image in column one. Subtracted images show only the added tubes, and the patient anatomy is no longer visible. Note that the tube visibility is markedly improved in the subtracted images once the anatomy has been removed. The four rows shown in Figure B relate to the choice of "dose level" that is made at the operator at the beginning of each DSA acquisition. The first row (dose level of 0.24) corresponds to the lowest radiation doses, which increase to the highest level (dose 12.0) in the last row.
Figure B. Images from a DSA run, showing the mask image (left), image with iodine (middle) and subtracted images (right). The first row was acquired at the lowest radiation level (0.24 dose setting), and each subsequent row used more radiation with the fourth row using the highest exposures (12.0 dose setting).
The acquisition of any DSA run requires the operator to select a dose level that ranges from 0.24 to 12. When the operator selects 0.24 for this phantom, the system uses 70 kV, a tube current of 187 mA and an exposure time per frame of 10.1 ms (i.e., 0.010 s) that corresponds to a total of 1.9 mAs. The measured entrance skin air kerma at this setting was 0.042 mGy/frame, which is very low given that a typical entrance skin air kerma in radiography would be of the order of 3 mGy per image. Increasing the selected dose level for the series of images shown in Figure B kept the x-ray tube voltage constant, but progressively increased the x-ray tube current as well as the exposure time per frame. For a dose level of 1.2, the tube current was 286 mA and the exposure time 30.4 ms (i.e., 8.7 mAs), which resulted in an entrance skin air kerma of 0.202 mGy/frame. At the dose setting of 2.4, the tube current increased to 351 mA and the exposure time increased to 49.6 ms (i.e., 17.4 mAs) and the entrance skin air kerma was 0.406 mGy/frame. At the highest dose level of 12, the mA was 444, the exposure time for each frame was 105 ms (i.e., 46.6 mAs), and the entrance skin exposure was 3.3. mGy/frame.
Imaging equipment in interventional radiology requires the system to select parameters such as x-ray tube voltage, x-ray tube current, and exposure time per digital frame. Information about patient size is normally obtained during the initial fluoroscopy, which is used to help identify an optimal x-ray tube voltage. However, it is the operator's responsibility to be aware of the available choices in dose setting (0.24 to 12 in this case), and identify the one that is most appropriate for a given patient. In the example depicted here, the selected dose setting was approximately proportional to the patient dose. The choice of the amount of radiation used is the responsibility of the radiologist in charge of the procedure and should be determined by the specific diagnostic task at hand.
Figure C shows enlarged regions of interest of the subtracted images acquired at differing dose levels for the simulated vessel that contains the lowest concentration of iodinated contrast material. The four images in Figure C clearly illustrate how increasing the exposure level reduces the amount of random noise (mottle) in the DSA images. Radiologist should strictly adhere to the principle of As Low As Reasonably Achievable (ALARA), and use no more radiation than is required to perform any specific imaging task. During any given procedure, it is appropriate (AND desirable) to progressively reduce the amount of radiation used if this will not have any adverse effect on diagnostic and therapeutic performance.
Figure C. Enlarged regions of interest of DSA images at four dose levels ( 0.4, upper left, 1.2, upper right, 2.4, lower left, and 12.0, lower right). Note how increasing the radiation exposure reduces the noise (mottle) and improves the visibility of the simulated blood vessels, particularly the on containing the lowest concentration of iodinated contrast material (uppermost vessel in each image).
DSA images contain little anatomical structure, and therefore the random noise around the visible vessels is the factor that limits lesion visibility. The amount of radiation dose used to acquire DSA images is therefore higher than in conventional digital photospot imaging. One DSA frame would likely have a radiation dose that is a factor of four or so higher than any corresponding digital photospot image (see fluoroscopy above). A good approximate rule that is used is that DSA frames have doses that are comparable to those of conventional radiography, whereas digital photospot images have doses that are four to five times lower than in conventional radiography.
Because Interventional Radiography has the potential for high patient doses, state of the art equipment is normally provided with additional dose monitoring systems that will provide additional patient dose information to the operator. Dose information may be provided regarding the rate at which radiation is being delivered to the patient (mGy per second or mGy per frame), as well as the total amount of radiation delivered during the complete procedure. On the system that was used to generate the DSA images depicted in Figures B and C, there are two patient dose quantities that are available to operators for review:
A. Dose at the Interventional Reference Point (IRP). One important radiation risk to patients undergoing interventional procedures is the possibility of a deterministic effect such as a skin burn. The possibility of any deterministic effect (e.g., skin burn, epilation) is determined by the maximum skin dose. An interventional reference point has recently been defined by the IEC, which is located 15 cm from the imaging isocenter (i.e., closer to the x-ray tube) and is intended as a surrogate for the patient skin dose. For the images shown in Figure B (dose level 12) the acquisition of 11 images resulted in a dose of 56 mGy to the Interventional Reference Point.
Skin burns and epilation have a threshold dose below which the effect does not occur; this threshold dose value is normally taken to be at least 2 Gy, and an important goal in any radiological exam is to keep doses below this threshold value. It is important to note that the IRP dose is a surrogate for the patient skin dose; the true patient skin dose will depend on the size and location of the patient relative to the imaging system, as well as any motion of the imaging system that would minimize the maximum skin dose. Nonetheless, the IRP dose is a useful indicator of the possible maximum skin dose, and can guide operators as to the possibility of deterministic effects for any given procedure.
B. Kerma Area Product (KAP). The overwhelming majority of patients (i.e., > 99.999%) undergoing any type of diagnostic radiological examination will have NO possibility of suffering from a deterministic effect such as skin burns or epilation. For these examinations, the patient risk is primarily the induction, which is related to the total amount of ionizing radiation energy that is absorbed by the patient. This carcinogenic risk can be quantified by measuring the Kerma Area Product, which is the product of the average air kerma (mGy) of the x-ray beam multiplied by the corresponding x-ray beam area (cm2). In the highest dose example shown in Figure B, the KAP value for 11 images in the DSA acquisition at the highest dose setting (i.e., 12) was 15 Gy-cm2.
The single DSA acquisition (i.e., Figure B; Dose level 12; 11 frames) may be compared with typical KAP values in other areas of radiological imaging: (a) Radiography where a skull radiographic examinations is ~ 1 Gy-cm2, a chest radiographic examination is ~ 0.5 Gy-cm2, and an abdominal radiographic examination is ~ 5 Gy-cm2; (b) fluoroscopy, where a barium swallow examination is ~ 10 Gy-cm2, a barium meal is ~ 20 Gy-cm2, and a barium enema is ~ 40 Gy-cm2; (c) CT where a typical single head CT scan is ~18 Gy-cm2 and a typical single phase abdominal CT scan is ~ 25 Gy-cm2; (c) Interventional Radiology, where cerebral angiography is ~ 60 Gy-cm2, abdominal angiography is ~ 90 Gy-cm2, and high dose procedures such as TIPS can result in patient doses of the order of 260 Gy-cm2.