| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Division of Abdominal Imaging and Interventional Radiology, Massachusetts General Hospital, White 270, 55 Fruit St., Boston, MA 02114.
Received January 24, 2000;
accepted after revision May 3, 2000.
Supported by a grant from Philips Medical Systems.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Seventy patients (29 men and 41 women; age range, 24-78 years) were enrolled in one of seven abdominal and pelvic fluoroscopic examinations, including upper gastrointestinal series (n = 20), barium enema (n = 10), voiding cystourethrogram (n = 10), percutaneous abdominal catheter tube injection (n = 10), hysterosalpingogram (n = 10), and percutaneous needle insertion and catheter placement (nephrostomy, percutaneous biliary drainage) (n = 10). Each patient underwent at least 10 sec of continuous fluoroscopy that was randomly and blindly compared with 10-sec periods of pulsed fluoroscopy at 15, 7.5, and 3.75 frames per second. A radiologist outside the examination room, unaware of the frame rate per second, evaluated the procedure in real time on a television monitor. The radiologist assessed image quality and diagnostic acceptability using a scoring system. Statistical analysis was performed using the paired Student's t test.
RESULTS. For all procedures at all frame rates, we found no statistically significant superiority of one frame rate over another. For most procedures, the slower frame rates were considered equivalent to continuous fluoroscopy when the images were assessed for image quality and diagnostic confidence.
CONCLUSION. Our findings suggest that most abdominal and pelvic fluoroscopic procedures can be performed at substantially lower frame rates than those used for continuous fluoroscopy; adopting this procedure may lead to substantial dose savings for the patient and the fluoroscopy operator.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Although the radiologist can use simple dose-saving maneuvers to minimize fluoroscopic doses (X-ray field collimation, image intensifier placed close to the patient, spot films minimized), manufacturers have investigated methods to reduce the exposure dose emitted from the X-ray tube. Most radiographic machines use low-energy filters to remove unwanted low-energy X rays emitted from the anode because these X rays only increase patient dose and do not improve image quality. Several manufacturers have offered intermittent or pulsed fluoroscopy to reduce the dose. The premise is that a conventional continuous beam is not required for most fluoroscopic procedures, which can be adequately performed at slower frame rates.
Although initial studies using pulsed fluoroscopy revealed the potential for reduced radiation dose when compared with continuous fluoroscopy, earlier models of pulsed fluoroscopy machines had significant limitations, and overall patient doses were similar to those of continuous fluoroscopy [5,6,7,8]. Because the pulses to the anode were controlled at the X-ray generator level, high resistance was induced in the long X-ray cables. This lead to an uneven current, with the production of a significant percentage of low-energy X-ray beams, which did not improve the radiographic image [9]. Therefore, many manufacturers abandoned this modification.
More recent modification designs resulted in the pulses originating in the X-ray tube rather than in the X-ray generator [9]. With a grid-controlled technique, the resulting pulses are in the form of discrete and uniform current without wasteful low-energy radiation. This innovation permits multiple frame rates to be used from one to 30 frames per second (continuous fluoroscopy). Although lower frame rates have been shown in phantom and some cardiac models to reduce exposure doses by more than 50%, to our knowledge no study has compared the image quality, diagnostic fluoroscopic information, and accuracy of these lower frame rates with those of conventional continuous fluoroscopy [5,6,7,8].
We evaluated the diagnostic accuracy of a grid-controlled fluoroscopy unit compared with a conventional continuous fluoroscopy unit for a variety of abdominal and pelvic fluoroscopic examinations.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Procedure Protocol
All fluoroscopic procedures were performed using an EZ Diagnost scanner
(Philips Medical Systems, Erlangen, The Netherlands). For all procedures, each
patient underwent standard continuous fluoroscopy using three differing frame
rates: 15, 7.5, and 3.75 frames per second. For each procedure, one of two
radiologists was asked to assess the fluoroscopic image for diagnostic
acceptability and image quality. To achieve this comparison, each patient
underwent fluoroscopic exposures divided into three pairs or segments, with
each segment consisting of a period of continuous exposure for 10 sec that was
then compared with one of three different frame rates for a period of 10 sec.
The order in which the frame rates were selected was random. For each
fluoroscopic segment, the selection of the order of continuous fluoroscopy
versus fluoroscopy at one of the three randomly chosen frame rates was also
random. For example, a three-segment examination might consist of the
following: first segment: continuous fluoroscopy (10 sec), then fluoroscopy at
7.5 frames per second (10 sec); second segment: fluoroscopy at 15 frames per
second (10 sec), then continuous fluoroscopy (10 sec); and third segment:
continuous fluoroscopy (10 sec), then fluoroscopy at 3.75 frames per second
(10 sec).
Three experienced gastrointestinal and genitourinary radiologists participated in the study. For each procedure, two radiologists participated. One radiologist was required to perform the fluoroscopic examination and the other to review the images for analysis. The second radiologist reviewed the images on a separate monitor outside the examination room as the procedure was performed. Only the X-ray technologist knew the order of the fluoroscopic segment pairs. At the beginning of the examination, the technologist selected a sheet that listed a three-segment examination and included all three fluoroscopic pairs. One radiologist performed the examination with the X-ray technologist, who switched the frame rates according to the selected sheet. The X-ray technologist indicated to the radiologist analyzing the study (outside the room) that the frame rate was changing by notifying them when a new fluoroscopic frame rate was started.
Each patient received a minimum of 60 sec of exposure (three segments with a pair of frame rates at 10 sec each). We believe that this procedure did not significantly increase the length of the examination because most fluoroscopic examinations last at least 60 sec. Even if a procedure could have been performed in a shorter time, the overall dose to the patient would likely remain the same because of the geometric design of the X-ray tube and image intensifier position.
Image Analysis
A scoring sheet for each of the three segments involved in the examination
was devised and completed during the examination. The observing radiologists
were requested to evaluate each pair of segments according to overall image
quality and diagnostic acceptability. The scoring system consisted of three
levels: level 1 was superior, level 2 was similar, and level 3 was
inferior.
Statistical analysis of the results was performed for continuous versus pulsed fluoroscopy at 3.75, 7.5, and 15 frames per second using the paired Student's t test.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
More recently, manufacturers have investigated methods of modifying the exposure from the X-ray tube. During conventional fluoroscopy, the X-ray beam is continuous and rapidly contributes to the total patient and operator dose. To minimize the dose, the operator must use fluoroscopy for the minimum time necessary to gain adequate diagnostic information. Unfortunately, because of a combination of operator experience and patient case mix, radiation doses for common procedures are extremely variable. Patient doses can vary markedly from one operator to another for the same procedure, despite adequate radiologic and equipment training. Therefore, manufacturers developed a method of reducing the dose by permitting the fluoroscopist to lower the exposure by reducing the number of exposures per second, so-called pulsed fluoroscopy. Instead of a continuous exposure, the X-ray beam is pulsed at a variable number of times per second, which can be controlled by the operator. Image perception would obviously alter because of the flicker or flicker effect on the images, depending on the number of frames used per second. The number of frames per second that would be acceptable for diagnostic fluoroscopy has never been scientifically established.
Initially, pulsed fluoroscopy did not live up to its expectations, primarily because of limitations in producing a pulsed exposure of uniform current [9]. Conventional pulsed fluoroscopy involved rapid on—off switching at the generator level. Because the generator has to be placed several yards away from the X-ray tube (both for safety reasons and the size of equipment), the cables between the X-ray tube and the generator can often be up to 20 ft (6 m) long. This length of cables induces a strong resistance that causes a strong capacitance in the cables. The effect produces an uneven current in the cables, with a slow rise in current after the exposure is initiated until the expected current is attained, then a slow fall back to zero current until fluoroscopy is turned off. The shape of the exposure curve is similar to a bell-shaped curve (Fig. 1), with the steepness of the curve dependent on the length of cabling. This rise and fall in current has been termed the "ramp and trail effect" of conventional fluoroscopy. Because much of the desired exposure level is below the chosen current, a large proportion of X rays are in the lower energy range. Because these lower energy X rays are absorbed by the patient, the result is an increased radiation dose without improvement in image quality. Therefore, conventional pulsed fluoroscopy offered little if any dose savings over continuous fluoroscopy, and many manufacturers discontinued offering it as an option on fluoroscopy machines.
|
Recently, an innovation called grid-controlled fluoroscopy has been developed. This technique also uses pulsed fluoroscopy, but the pulses are controlled in the X-ray tube and not at the generator. Therefore, the ramp and trailing effect caused by cable capacitance is not observed (Fig. 1). Discrete uniform tube current is maintained by placing a negatively charged grid between the electron-emitting cathode and the X ray—producing anode (Fig. 2). When the grid is switched on, the negative charge repels the electrons emitted from the cathode and prevents a current across the X-ray tube. Multiple pulses can be produced by rapidly switching the grid on and off. Because the on—off switching mechanism is instantaneous, a uniform current can be produced across the X-ray tube (Fig. 2). This ensures a uniform X-ray exposure to the patient and minimizes unwanted radiation. Image quality is also improved because the X rays are produced in the desired diagnostic range. A further advantage of the grid-controlled fluoroscopic system is that the length of the exposure pulse can be varied depending on the clinical application. For standard abdominal applications, a 5- to 20-msec exposure pulse is typically used, whereas in pediatric applications, the pulse length can be reduced to 2-5 msec [10].
|
A secondary advantage of grid-controlled fluoroscopy is that the blooming effect of the image intensifier is reduced [10]. If during continuous fluoroscopy the image intensifier is moved (as it often is), then a small time elapses before the automatic exposure device corrects the new field of view to the correct exposure. As fluoroscopic panning is performed, a temporary overexposure occurs when moving from a dense area (mediastinum) to a less dense area (the lungs) (or from the pelvis to the abdomen), which causes image degradation with streaking and flaring of the image (image blooming). The opposite effect occurs when moving from the lungs to the mediastinum. This blooming effect is significantly reduced using grid-controlled fluoroscopy because each pulse automatically corrects to the desired exposure level.
The grid-controlled fluoroscopic technique has several advantages. The most obvious is the ability to reduce the overall radiation dose by changing the selected frame rate. Higher frame rates lead to higher exposure. Previous studies have documented a substantially reduced dose using pulsed fluoroscopy in in vitro and in vivo studies [5, 9, 10]. Dose savings to the patient have been calculated to be 22% for frame rates of 15 frames per second, 38% for 10 frames per second, 49% for 7.5 frames per second, 80% for 3.75 frames per second, and as much as 86% for 2 frames per second [5,6,7,8,9, 10]. What is uncertain is whether these reduced frame rates can be used in clinical practice to produce diagnostically acceptable images. Our findings reveal that grid-controlled fluoroscopy can be used as an alternative to continuous fluoroscopy and can produce images of diagnostic quality.
For this study to have relevant implications for abdominal fluoroscopic procedures, the procedures analyzed were chosen to reflect a range of examinations. Furthermore, to compare the effects of slower frame rates on different examinations, we included examinations that required static imaging (i.e., hysterosalpingography) and those in which the patient was required to move (i.e., upper gastrointestinal series).
The grid-controlled fluoroscopic effect is most pronounced for studies involving little patient motion or image intensifier movement (i.e., hysterosalpingography and abdominal catheter injections) in which 3.75 frames per second is likely to be acceptable for most imaging. However, even for studies involving significant patient or image intensifier movement (upper gastrointestinal series and delicate catheter manipulations), the slower frame rates (3.75 and 7.5 frames per second) proved acceptable in most patients. In patients in whom a higher frame rate is desirable, the frame rate can easily be switched to 15 frames per second, which still offers a significant dose reduction of at least 22% when compared with continuous fluoroscopy. Our findings suggest that continuous fluoroscopy should rarely be used for abdominal and pelvic fluoroscopic procedures.
Potential disadvantages of grid-controlled fluoroscopy might be expected because of the perceived flicker effect, particularly at lower frame rates. The flicker effect is observed when reducing the frame rate from continuous fluoroscopy. The image has an increasing tendency to flicker as the frame rate decreases. However, this effect did not appear to affect the ability of the radiologists to observe and diagnose normal and abnormal conditions. We noticed that some initial training and habituation to the lower frame rates may be required before they are considered acceptable. This initiation period is likely to vary from one observer to the next.
A further potential disadvantage of this technique is that fine catheter and needle manipulations might not be seen to their best advantage using the slower frame rates. However, we found that in only three of 10 patients was a higher frame rate desirable; this was not the case for the other seven patients we examined. It is possible that a larger study might reveal that higher frame rates should be used for some procedures.
This is the first clinical report to directly compare grid-controlled fluoroscopy with continuous fluoroscopy. Our results strongly suggest that grid-controlled fluoroscopy is acceptable for all abdominal fluoroscopic procedures, although a frame rate of 15 frames per second may be required for studies that require considerable motion or fine catheter or needle manipulations. Although a recognizable flicker effect occurs at lower frame rates (3.75 frames per seconds), our findings reveal that the lower frame rate does not cause a loss of diagnostic detail or prolonged examinations.
This study did not evaluate potential quantitative dose reductions in each patient, although this has been confirmed by other phantom studies [5]. Further studies should be performed to quantitatively evaluate the actual dose savings to patients using grid-controlled fluoroscopy. Typical fluoroscopic doses for an upper gastrointestinal series range from 2 to 3 rad (0.02-0.03 Gy) [1,2,3,4,5]. This study suggests that these doses can be reduced to less than 1 rad (0.01 Gy) for most examinations. This reduction is particularly important for pediatric patients and patients of reproductive age undergoing pelvic fluoroscopy. Many pelvic fluoroscopic examinations involve static imaging (hysterosalpingography and voiding cystourethrography), and our findings suggest that 3.75 frames per second may be used for most of these examinations.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Hernanz-Schulman Science to Practice: Can Fluoroscopic Radiation Dose Be Substantially Reduced? Radiology, January 1, 2006; 238(1): 1 - 2. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. L. Ward, C. E. Barnewolt, K. J. Strauss, R. L. Lebowitz, V. Venkatakrishnan, M. Stehr, D. L. McLellan, C. A. Peters, D. Zurakowski, P. S. Dunning, et al. Radiation Exposure Reduction during Voiding Cystourethrography in a Pediatric Porcine Model of Vesicoureteral Reflux Radiology, January 1, 2006; 238(1): 96 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T Vehmas Do radiologists' radiation-related opinions predict their fluoroscopy doses? Br. J. Radiol., September 1, 2003; 76(909): 653 - 655. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
