Optimizing the Use of Pulsed Fluoroscopy to Reduce Radiation Exposure to Children

Transcription

1 Optimizing the Use of Pulsed Fluoroscopy to Reduce Radiation Exposure to Children Mervyn D. Cohen, MB, ChB, MD Radiologists desire to keep radiation dose as low as possible. Pulsed fluoroscopy provides an opportunity to lower radiation exposure to children undergoing fluoroscopic studies. To optimize the ability of pulsed fluoroscopy to decrease radiation dose to patients during fluoroscopic studies, radiologists need to understand how pulsed fluoroscopy operates. This report reviews the basic physics knowledge needed by radiologists to best use pulsed fluoroscopy to minimize radiation dose. It explains the paradox that the best video frame-grabbed images are obtained when using the lowest fluoroscopy pulse rate and therefore the lowest fluoroscopy radiation dose. Key Words: Radiation dose, pulsed fluoroscopy, patient safety J Am Coll Radiol 2008;5: Copyright 2008 American College of Radiology Department of Radiology, Riley Children s Hospital, Indiana University School of Medicine, Indianapolis, Indiana. Corresponding author and reprints: Mervyn D. Cohen, MB, ChB, MD, Riley Hospital for Children, Department of Radiology, 702 Barnhill Drive, Room 1053, Indianapolis, IN 46202; mecohen@iupui.edu American College of Radiology /08/$34.00 DOI /j.jacr INTRODUCTION There have been many appropriate recent discussions regarding the need to reduce the radiation dose to pediatric patients to as low as reasonably achievable [1-3]. One method of dose reduction that has received recent attention is the use of pulsed fluoroscopy [4-9]. New grid-controlled x-ray tubes have a grid placed between the cathode and the anode [10]. This allows pulses of fluoroscopy to leave the x-ray tube at rates between 1 and 30 frames/s. Radiation no longer enters a patient continuously but in a series of short x-ray flashes [11]. Fluoroscopy at 30 frames/s has been termed continuous fluoroscopy [10,12]. Aufrichtig et al [13] defined pulsed fluoroscopy as 15 frames/s or less and continuous fluoroscopy as 30 frames/s. At my institution we have recently had experience with new pulsed fluoroscopy units from Philips and Siemens and wish to share our important learning during 12 months of experience with these units. Siemens Sireskop SD AXIOM, Siemens Medical Systems, Berlin, Germany and Philips Eleva, Philips Medical Systems, Andover, Massachusetts. Other major equipment manufacturers (GE Healthcare, Milwaukee, Wisconsin; Toshiba Medical Systems Corporation, Tokyo, Japan) also offer pulsed fluoroscopic units. We anticipate that the principles outlined in our experience should be applicable to units manufactured by other vendors. This study was approved by our institutional review board. DEFINITIONS To avoid any misunderstanding, I first define 3 methods of acquiring and keeping permanent images obtained during a fluoroscopic study: 1. Recording the entire study (eg, on CD-ROM or videotape): Disadvantages of this method are that the study is cumbersome to review and cannot be sent to a picture archiving and communication system (PACS). This method is not routinely used, apart from feeding studies. 2. Exposed images: These are traditional fluoroscopically acquired images obtained by pushing an exposure button to specifically expose and acquire an image at a particular moment in time. The image can then be sent to a PACS. These images may also be termed digital spot images. 3. Frame grabbing: At the end of a fluoroscopic period, the fluoroscopic button is released. An image is retained on the video monitor and in the system memory. This retained image can be grabbed and sent to a PACS. There is no additional radiation. These images may also be termed video frame-grabbed images. When using a new pulsed fluoroscopic unit, a radiologist may find that the frame-grabbed images obtained when fluoroscopic pulsing is at its slowest look surpris- 205

2 206 Journal of the American College of Radiology/Vol. 5 No. 3 March 2008 ingly good, better than frame-grabbed images captured with higher fluoroscopic pulse rates and higher radiation exposure. This is a paradox. One is using the lowest pulse rate and the lowest fluoroscopic dose, yet one obtains the best-quality frame-grabbed images (Figure 1). The following discussion explains this paradox and makes suggestions for the optimal use of pulsed fluoroscopy. POSSIBLE MISUNDERSTANDINGS REGARDING PULSED FLUOROSCOPY Radiologists may not fully understand how pulsed fluoroscopy works and how to minimize radiation exposure when using pulsed fluoroscopy. Radiologists may expect that a reduction in the fluoroscopic radiation pulse rate of 50% will decrease a patient s fluoroscopic radiation exposure by 50%, if no other fluoroscopic parameters under the immediate control of the radiologists are changed. This is not true. There is a reduction in radiation exposure as fluoroscopic pulse rate is decreased, but it is never a 50% dose reduction with a 50% decrease in frame rate. The reason for this is that manufacturers increase the duration or the amplitude or power of each fluoroscopic radiation pulse as fluoroscopic pulse rate is decreased. They do this to retain at least reasonable image quality. Manufacturers usually offer radiologists a choice of 4 pulse rates. These usually range from 2 to 30 frames/s. Radiologists can rapidly switch between these different pulse rates using controls mounted on the image intensifier. The effect of changing the pulse rate on the fluoroscopic kilovolts (kvs) and milliamperes (mas) is illustrated in Tables 1 and 2. The absolute figures will vary from manufacturer to manufacturer and on other settings such as filtering, grid use etc; the relative effect however remains unchanged, i.e, exposure parameters (kv and ma) increase as the pulse rate is increased. Radiologists may intuitively feel that as the fluoroscopic pulse rate is lowered, frame-grabbed images will diminish in quality. This is because the radiation dose over any time period is lower with a slower pulse rate. An understanding of the physics of frame grabbing provides a simple explanation that the reverse is true (Figure 1). If pulse rates of 30, 15, 8, and 3 frames/s are available, the best frame-grabbed images are obtained at the lowest pulse rate of 3 frames/s. The reason for this is that a frame-grabbed image is not an integration of data and images acquired over a fixed period of time (eg, the last 0.5 seconds of fluoroscopic time). What is grabbed is either the very last pulse of radiation (on the Eleva) or an integration of the last 3 pulses (on the Sireskop SD AXIOM). Because both the duration and the energy of each pulse are increased by each manufacturer as the pulse rate drops, the lowest pulse rate provides the highest radiation exposure per pulse of radiation and therefore the Fig 1. Effects of alteration in fluoroscopic pulse rate. (A) Frame-grabbed image obtained from fluoroscopy at 3 frames/s. This image, obtained using a much lower fluoroscopy radiation exposure rate than image B, has much less noise than image B. (B) Frame-grabbed image obtained from fluoroscopy at 30 frames/s. best frame-grabbed image. This is an excellent paradox. The lowest fluoroscopic dose provides the highest quality framegrabbed image to send to the PACS. The difference in

3 Cohen/Optimizing Pulsed Fluoroscopy 207 Table 1. Manufacturer 1: Phantom study showing relative changes in the fluoroscopic kv and ma as changes are made in the fluoroscopic frame pulse rate and the automatic dose setting at the image intensifier Automatic Dose Setting at the Pulse Rate and Fluoroscopic kv-ma Image Intensifier Continuous Low Medium High images will vary between patients. The above is definitely true for our Siemens and Philips fluoroscopic units, but we cannot comment on units from other manufactures. I suggest that radiologists check with the manufacturers of their equipment to confirm that each fluoroscopic pulse increases in strength as the pulse rate is decreased. Reducing the fluoroscopic pulse rate will still have 2 disadvantages. On a real-time fluoroscopic image, as the pulse rate is lowered, the cumulative radiation dose falls, and the real-time fluoroscopic image becomes more noisy or grainy. Second, with very slow pulse rates, motion such as swallowing or peristalsis becomes jerky. The use of the term continuous fluoroscopy is potentially confusing. Continuous radiation to a patient during fluoroscopy is not used on our 2 fluoroscopic units. In the past, the term did apply to continuous radiation production and the display of images every ms (ie, display at a television frame rate of 30 frames/s). The term continuous fluoroscopy is now used by both of our manufacturers to designate the fastest used fluoroscopic rate, usually 30 frames/s. Pulsed fluoroscopy refers to pulses of radiation that are slower than this. Manufacturers usually offer 3 options of lower fluoroscopy pulse rates. It is very easy to switch the pulse rate during a study. EXAMPLES OF SUGGESTED APPLICATIONS OF THE ABOVE PRINCIPLES TO CLINICAL PRACTICE I now discuss several situations that may be encountered in daily clinical practice and suggest how pulsed fluoroscopy can be optimally used. Situation 1 Fluoroscopy is being done at 15 frames/s. The real-time fluoroscopic image looks great, and the radiologist feels that the radiation exposure can be lowered and still retain adequate image quality. Should the radiologist lower the automatic exposure dose needed at the image intensifier to produce the image? The correct answer is no. Lowering the fluoroscopic dose needed at the image intensifier or lowering the fluoroscopic pulse rate will both lower the total fluoroscopic dose. However, lowering the dose at the intensifier will also result in a decrease in the quality of any framegrabbed image. Lowering the pulse rate, instead of the intensifier dose, will also lower the fluoroscopic dose but will actually result in a better quality frame-grabbed image. Thus, if one wants to decrease the fluoroscopic dose, lowering the pulse rate is always preferable to lowering the intensifier dose. This is true only if one is using predominantly frame grabbing and not image exposure to capture images to send to the PACS. Situation 2 The quality of the real-time fluoroscopic image is adequate. The frame-grabbed image at the lowest fluoroscopic pulse rate is still not of adequate quality. What should be done? There are 3 options. The best solution is to use an exposed image. One will always obtain a better quality image by using a specific exposure. This will use extra radiation but will result in a better image (Figure 2). Table 2. Manufacturer 2: Phantom study showing relative changes in the fluoroscopic kv and ma as changes are made in the fluoroscopic frame pulse rate and the automatic dose setting at the image intensifier Automatic Dose Setting at the Pulse Rate and Fluoroscopic kv-ma Image Intensifier Continuous Low Medium High

4 208 Journal of the American College of Radiology/Vol. 5 No. 3 March 2008 However, it will not increase the radiation dose from further real-time fluoroscopy. The other 2 options will both result in an increase in dose during fluoroscopy. One could increase the automatic exposure dose needed at the image intensifier. This will result in a small increase in the frame grab image quality, but not as much as capturing an exposed image. The final option, in larger patients, is to bring in the grid. This will also result in a small increase in the frame grab image quality, but not as much as capturing an exposed image. Situation 3 If a low fluoroscopic pulse rate is being used, and it is desirable to improve the quality of the real-time fluoroscopic image, what is the best way to do this? One could simply increase the fluoroscopic pulse rate. Increasing the pulse rate will improve the fluoroscopic image but worsen the quality of the frame-grabbed image. Instead, try bringing in the grid. This will increase the fluoroscopy dose approximately the same as doubling the fluoroscopic pulse rate. Both will improve the quality of the real-time fluoroscopic image. However, only the grid will improve the quality of the frame-grabbed image. Even in patients as young as 5 or 6 years, we have found that bringing in the grid will, surprisingly, improve image quality. If this does not have the desired effect, take the grid out and increase the automatic exposure dose needed at the image intensifier. Fig 2. Effect of alteration in method of obtaining the image. (A) Frame-grabbed image obtained from fluoroscopy at 3 frames/s. (B) Exposed image. This image required more radiation than image A and has much less noise. EDUCATION AND TRAINING Another important lesson during a year of experience with the new Philips and Siemens pulsed fluoroscopic units is that the applications personnel, in general, are not well aware of all of the parameters discussed in this article. They are very knowledgeable regarding training technologists in the things that are important for technologists to know. They are much less knowledgeable with regard to important information in training radiologists in the optimal use of the functionality to decrease radiation dose provided by these powerful new pulsed fluoroscopy units. It is important for manufacturers leadership to recognize these deficiencies and to remedy them. A recent ACR White Paper on radiation in medicine also draws attention to this problem [14], stating that vendors need to continually train and update their application specialists to make sure that improved validated parameter choices are introduced to its entire customer base. It is also very important for each radiology department to make sure that every faculty and trainee radiologist is very well trained in using a pulsed fluoroscopy unit.

5 Cohen/Optimizing Pulsed Fluoroscopy 209 IMAGE REPROCESSING In addition to pulsed fluoroscopy, many manufacturers now include many forms of image reprocessing in their systems [15]. These include image edge enhancement, image smoothing, recursive filtering, and so on. These operate independently of radiation pulsing. It is not the intent of this article to discuss these. CONCLUSION The recent ACR White Paper on radiation dose in medicine draws attention to the important role that radiologists must play in radiation protection [14], stating that it is incumbent on radiologists to assume the responsibility for their patient s safety with regard to radiation exposure. By understanding and correctly using pulsed fluoroscopy, radiologists have the opportunity to reduce fluoroscopic radiation dose to children. Lowering the fluoroscopic frame rate lowers total radiation over any fixed time period and also lowers the signal/noise ratio on the real-time fluoroscopic image [16]. Our most important lesson is that if one is using pulsed fluoroscopy, the best frame-grabbed images are paradoxically obtained with the lowest fluoroscopic rate and the lowest radiation dose. This finding is not reported in the literature. It a wonderful paradox, because the lowest fluoroscopic pulse rate is also providing the lowest dose during fluoroscopy. As long as one can accept the somewhat jerky images of motion, such as swallowing, with very low pulsed radiation frame rates, this is the setting that should be routinely used. One should always use the lowest frame rate that gives an acceptable fluoroscopic image. Pulsed fluoroscopy can always decrease radiation dose. There is always a potential penalty of a decrease in image quality. Only by fully understanding the mechanisms of operation of pulsed fluoroscopy can radiologists maximize the dose reduction benefits of pulsed fluoroscopy and minimize its disadvantages. REFERENCES 1. Strauss KJ, Kaste SC. The ALARA (as low as reasonably achievable) concept in pediatric interventional and fluoroscopic imaging: striving to keep radiation dose as low as possible during fluoroscopy of pediatric patients a white paper executive summary. Pediatr Radiol 2006;36: Ward VL. Patient dose reduction during voiding cystourethrography. Pediatr Radiol 2006;36: Geijer H. Radiation dose and image quality in diagnostic radiology. Optimization of the dose-image quality relationship with clinical experience from scoliosis radiography, coronary intervention and a flat-panel digital detector. Acta Radiol 2002;43:S Lederman HM, Khademian ZP, Felice M, et al. Dose reduction fluoroscopy in pediatrics. Pediatr Rad 2002;12: Hernandez RJ, Goodsitt MM. Reduction of radiation dose in pediatric patients using pulsed fluoroscopy. AJR Am J Roentgenol 1996;167: Schueler BA, Julsrud PR, Gray JE, et al. Radiation exposure and efficacy of exposure-reduction techniques during cardiac catheterization in children. AJR Am J Roentgenol 1994;162: Stueve D. Management of pediatric radiation dose using Philips fluoroscopy systems DoseWise: perfect image, perfect sense. Pediatr Radiol 2006;36: Strauss KJ. Pediatric interventional radiography equipment: safety considerations. Pediatr Radiol 2006;36: Brown PH, Thomas RD, Silberberg PJ, et al. Optimization of a fluoroscope to reduce radiation exposure in pediatric imaging. Pediatr Radiol 2000;30: Hahn H, Farber D, Allmendinger H, et al. Grid-controlled fluoroscopy in paediatric radiology. Medicamundi 1997;41: Nagel HD. Technical note: dose management. Medicamundi 1997;41: Boland GW, Murphy B, Arellano R, et al. Dose reduction in gastrointestinal and genitourinary fluoroscopy: use of grid-controlled pulsed fluoroscopy. AJR Am J Roentgenol 2000;175: Aufrichtig R, Zue P, Thomas CW, et al. Perceptual comparison of pulsed and continuous fluoroscopy. Med Phys 1994;21: Amis ES Jr, Butler PF, Applegate KE, et al. American College of Radiology white paper on radiation dose in medicine. J Am Coll Radiol 2007; 4: Detorie N, Mahesh M, Schueler BA. Reducing occupational exposure from fluoroscopy. J Am Coll Radiol 2007;4: Cohen MD. Are we doing enough to minimize fluoroscopic radiation exposure rate in children? Pediatr Radiol. In press.