| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reflections |
1 From the Department of Radiology, Division of Diagnostic Ultrasound, Thomas Jefferson University Hospital, 7th Floor Main Bldg, 132 S 10th St, Philadelphia, PA 19107. Received October 28, 1999; revision requested November 24; revision received January 24, 2000; accepted January 27. Address correspondence to the author (e-mail: barry.b.goldberg@mail.tju.edu).
Abstract
Although diagnostic ultrasonography (US) was developing in the late 1940s and early 1950s, it was not until the 1960s, with the availability of commercial equipment, that its usefulness in obstetrics began to be realized fully by radiologists and obstetricians around the world. Advances from A-mode to bistable and then to gray-scale static imaging were followed by the introduction of automated compound imaging and real-time US. Also, the development and initial use of Doppler US for the detection of fetal heart motion and the eventual use of pulsed and color Doppler US for the evaluation of such fetal structures as the major vessels and heart chambers contributed to increasing the usefulness of US in obstetrics. The development of specialized transducers—in particular, endovaginal probes—resulted in images of the early fetus. At the present time, the development of multiplanar, three-dimensional imaging shows great promise for more complete imaging of the fetus. The importance of US in the examination of the pregnant patient and, in particular, of the fetus has led to its worldwide dominance as the imaging modality of choice. The contributions of obstetric US to improving maternal well-being and fetal health have been recognized as a key component in all countries around the world.
Index terms: Radiology and Radiologists, history • Ultrasound (US) • Ultrasound (US), in infants and children
The beginnings of diagnostic ultrasonography (US) as we know it today were in the late 1940s and early 1950s, with pioneering researchers using sonar and ultrasonic flaw detector–based equipment that was developed as a result of the war effort. A key group in this country that participated in the development of equipment that was among the first to be used in obstetrics was directed by a radiologist, Douglas Howry, MD, at the University of Colorado in Denver, who in 1950 recorded the first cross-sectional images with US (1–4).
The other major research group was led by John Wild, MD, a surgeon who immigrated to the United States from England after World War II. He and his associates developed many of the basic concepts of US instruments as we know them today, which included a transducer that was placed within the rectum—a forerunner of endovaginal scanning. However, his group never emphasized the use of US in obstetrics (5,6). Dr Wild had an indirect influence on an obstetrician, Ian Donald, MD, from Scotland, in terms of introducing him to the potential usefulness of US in diagnosis in humans. Dr Donald, in the late 1950s and early 1960s, with his associates, started to explore the usefulness of US in obstetrics and gynecology (7–9). In the United States, Dr Howry and his engineering associates worked with obstetrician Horace Thompson, MD, and his younger associate, Kenneth Gottesfeld, MD, in using the instruments they developed to examine obstetric patients (Fig 1) (10,11).
|
|
I can remember the first time I picked up the bell-shaped transducer that housed the piezoelectric crystal that produced the sound waves and placed it on my abdomen. I saw strange vertical reflections arising from a linear baseline (A-mode display) that had no relationship to any x-ray images I had seen.
Within a month, I had developed a program of research that entailed going to the autopsy room in the morning before work to see if there had been any deaths. At that time, it was much more common to perform an autopsy. I was able to examine the cadaver with the A-mode US transducer, which I placed over the liver, kidneys, pelvis, and heart, and to record reflections that arose at different depths. I placed long needles (up to a dozen) parallel to the ultrasound beam, exactly over where the transducer was placed, in one cadaver. The pathologist noted which structures were at which depths along the course of the needle. In this way, I was able to identify the relationships between the patterns I was seeing and the structures being traversed (14). It was obvious that this information was only one-dimensional, and one had to use one's mind to put together the entire picture.
During the working day, whenever patients were sent for x-ray studies prior to surgery for ovarian, uterine, or renal masses, I would place the US transducer over the area of interest as determined on the x-ray image or by means of palpation and obtain an A-mode US image that appeared to be characteristic of the mass. I then would go to the pathologist's laboratory, where the specimens usually were kept in the refrigerator, remove them, and place them in a water bath (a small fish tank that I had purchased and kept filled with isotonic saline). The specimen was placed within this liquid, and the transducer was placed on the surface. In this way, I was able to correlate the A-mode US patterns that I had seen in the intact body; then, with the help of the pathologist, the specimen was sliced, and photographs were obtained for comparison with the US recordings.
After work, my wife and I would often bring the US machine to the obstetrics department's delivery ward. At that time, it was very common to have labor inductions, and the women who had been admitted that afternoon would be receiving oxytocin drips to induce labor. I was able to obtain fetal biparietal diameter measurements whether the fetuses were in a vertex or breech presentation (Fig 3). Many of these women would have undergone radiography to confirm the position of the fetus and to determine the pelvic dimensions and evaluate for dystocia. My wife acted as my research coordinator, recording the patient information and the US measurements that I obtained in each patient. Next, we would go to the nursery to measure with calipers the biparietal diameter in the babies who had been delivered that morning. These data also were recorded and later were compared with the predelivery US measurements.
|
I remember presenting my first paper at the 1965 Radiological Society of North America (RSNA) meeting. Many of you will recall that during that time, the entire RSNA meeting was held at the Palmer House hotel, Chicago, Ill, and the exhibits were all within the rooms of the hotel. The main meeting was held in the ballroom. It was not until the 75th anniversary meeting of the RSNA, at which there were individual plaques listing the accomplishments for each year of the meeting, that I learned that mine had been the first article on US ever presented at the RSNA. It was also at the 1965 meeting that I began to make contact with equipment manufacturers and learned about other ultrasound machines that were being developed. In 1965, there were only two ultrasound companies displaying their products: Picker X-ray and SmithKline Instruments.
In the department of radiology, as one might expect, there was no room available for the US equipment. The machine initially was kept in a hallway and then was moved into any available examining room to perform a study on a patient. It was during the end of my 2nd year of residency that I was able to obtain a room in which I could store the machine. A word of caution to me, however, was that this formerly was the closet in which the radium had been stored. During that period, the use of radium essentially had disappeared, and it was decided to get rid of the safe and the radium it contained. As you can imagine, I insisted on several Geiger counter readings of the area before I would go into it and store the machine.
Vivid in my memory is the smell of the coating that was used for the Polaroid film. The US machine was equipped with a Polaroid camera; each image was duly recorded, the film was pulled from the packet, and a special coating material was applied that was contained within a sponge in a plastic holder, although one inevitably got the coating on one's hands. Images not coated properly would rapidly turn brown. The images were stored in small envelopes, one for each of the cases, with the information recorded for each case. For presentations, the departmental photographer prepared the images on 3 x 4-inch glass slides.
I remember clearly during my residency a number of the staff physicians and other residents telling me that I was wasting my time, that US would go nowhere, that it was witchcraft compared with radiography, and that it was almost impossible to interpret the images and get reliable information. For whatever reasons, I ignored these naysayers and continued to gather data in many areas and to publish the results. It was just as important that I began to find compatriots within the United States and scattered around the world who were hearing similar negative statements but who were still fascinated by the potential usefulness of this new imaging modality.
Approximately 1 year after I started my residency, one of the salesmen from SmithKline Instruments approached me to evaluate a new advance, Doppler US, that was used to detect fetal heart motion and placental flow, as well as flow within peripheral vessels. This small instrument, known as the Doptone, contained two crystals within the transducer, one sending out a continuous sound wave and another receiving any reflections. Thus, I was able to expand my research in obstetrics by using Doppler US to detect fetal heart motion and the site of the placental attachment (Fig 4). This was a revolutionary approach, since the standard had been to use nuclear imaging to identify the placenta and, of course, to use a stethoscope for detecting the fetal heartbeat. With pulsed Doppler US, it was possible to detect fetal heart motion at a much earlier stage of fetal development.
|
By the end of my residency, there had been an upswing in the amount of literature available on all aspects of US from around the world. Pioneering works from many groups of both radiologists and obstetricians on the use of obstetric US began to appear with greater frequency. The first measurement charts for predicting gestational age by using the biparietal diameter were developed, and new equipment, including two-dimensional US, was becoming available commercially.
I became aware of other radiologists who were pioneers in US, including Donald King, MD, Frederick Winsberg, MD, Roger Sanders, MD, Michael Johnson, MD, Edward A. (Ted) Lyons, MD, and Arthur Fleischer, MD; all had a strong influence on the development of the obstetric and gynecologic applications of US. Other pioneers such as George Leopold, MD, and Atis Freimanis, MD, developed uses for US in the abdomen, while Raymond Gramiak, MD, concentrated on the use of US within the heart. It should be noted that one of the first fellows of Dr Freimanis was Roy Filly, MD, who has become a leader in the development of the use of obstetric US in the United States. Many others were to follow, including Jason Birnholtz, MD, whose fellow, Beryl Benaceraf, MD, also is recognized as a leading researcher in obstetric US. The same can be said of one of my first fellows, Alfred Kurtz, MD. These individuals and others they trained or influenced helped to make obstetric US an established part of imaging in radiology. At the same time, obstetrician-gynecologists who also were pioneers in obstetric US influenced many in their specialty to perform US. There is a division of obstetric cases between radiologists and obstetrician-gynecologists, with the predominance of obstetric US being performed in offices by obstetricians and within hospitals by radiologists (17).
After completion of my residency and 1 year on the staff at Einstein, I had the opportunity to move to the Department of Radiology at Hahnemann Medical College, Philadelphia, Pa, in 1968. At that time, a radiologist who was also chairman of the department, Jay Stauffer Lehman, MD, worked with radiologists George Evans, MD, and Marvin Ziskin, MD, to conduct research in US (18). They were working with SmithKline Instruments to develop a two-dimensional–US machine. Luther Brady, MD, was instrumental in bringing SmithKline together with Dr Lehman on this project. It was during this period that I was introduced to two-dimensional US equipment.
As with many of the early machines, the initial equipment used was a water bath in which an automated transducer was moved. This allowed for larger transducers that could be better focused to improve resolution. However, the water bath proved cumbersome for routine patient use (Fig 5) (19,20).
|
|
|
|
Two-dimensional gray-scale US came about as a result of the development during the Vietnam War of scan converters, which made it possible to display images in shades of gray. With this advance, it was also possible to introduce the concept of compound imaging. This required one to move the transducer in a sectorlike motion as it was advanced across the skin surface. The reason for this was that it increased the chances for the ultrasound beam to strike a reflector perpendicularly and thus increased the chances of recording more complete information from the region of interest. As a result, more detailed US images of the fetus and of other structures throughout the body became possible. This resulted in improved resolution, which made it easier to interpret both normal and abnormal structures. This same concept was incorporated in an automated water bath US machine (Octoson), which I had an opportunity to evaluate when I first arrived at Thomas Jefferson University Hospital, Philadelphia, Pa, in the late 1970s.
In the early 1970s, we still were using Polaroid cameras to record the information, but by the end of that decade, the first film cameras became available, which was important to obtain the best display of shades of gray. By this time, details of the fetus and measurements of all areas of the fetus were being published worldwide by both radiologists and obstetrician-gynecologists. It also was during this time that programs for the training of sonographers and physicians were being developed. In the early 1970s, I started teaching formal training courses and had the opportunity to invite some of the pioneers from around this country and the world to participate.
In the mid-1970s, I had the opportunity to evaluate one of the earliest real-time machines marketed by Siemens Ultrasound, known as the Vidoson (23). It was a large machine with the transducer housed within a liquid-filled bag that made contact with the skin and that produced images at 1–2 frames per second. The company had effectively marketed this machine throughout Europe and South America especially for use in the evaluation of the fetus and had given it to Dr Winsberg in Canada and to me in the United States for evaluation (Fig 9). However, the equipment was not selling in North America, and I was asked to go to Germany to speak to the engineers about why they were not having success here.
|
Toward the end of the 1970s, the first successful real-time US machines were developed and were made available commercially. (Advanced Diagnostic Research later was purchased by Advanced Technology Laboratories). This revolutionized the use of US for obstetrics, since the fetus, which moves, could be visualized easily, and information about the fetal heart could also be obtained. In the 1980s, real-time, two-dimensional US essentially replaced the static gray-scale imaging machines. With each advance in transducers, equipment, and technology, the use of US to depict more details of the fetus led to its rapid acceptance throughout the radiology and obstetrics-gynecology community.
By the mid-1970s, with the rapid expansion of the uses of US and with its acceptance by the referring physicians, the volume of referred cases had increased to the point where we now had three B-scan gray-scale US machines, and I limited my radiologic practice to US.
Immediately after arriving at Thomas Jefferson University Hospital, I was asked to evaluate an automated water-bath US machine (Octoson). It had been developed by a group of Australian engineers led by George Kossoff, DSc Eng, and was being commercialized by an Australian US company known as Ausonics (24–26). General Electric was considering marketing this equipment in the United States, and I was the first to have such a machine to evaluate. This instrument contained eight large transducers housed within a water tank, which could produce two-dimensional, gray-scale compound US scans automatically. The individual would lie on a large membrane that separated him or her from the transducers that moved within the water (Fig 10). While this equipment could image many areas, one of the more exciting uses was in evaluating the pregnant uterus and in producing gray-scale images of higher resolution than was possible with equipment available then (Fig 11). However, General Electric decided not to go ahead with the commercialization of this product, since the image quality of real-time US was improving rapidly. Both gray-scale static and real-time US equipment had advantages over the more complex and cumbersome large water bath approach that had been used by some of the early pioneers and then was reincarnated, with more sophisticated technology, by the Australian group. An automated machine with real-time capability would have been the best approach, but this pathway never was pursued.
|
|
At the end of the meeting, the consensus was that US was not cancer producing. However, it was thought that since nobody could guarantee that US was 100% safe, it should be used only when indicated clinically in the developing fetus (27). As a result, the government and the third parties who paid for the studies stated that US should be used only when indicated clinically. Nonetheless, the many advantages of US in the examination of the developing fetus have led to its use in more than 80% of all pregnancies.
In the 1980s, additional new technologies were developed and became widespread, including endovaginal US. Its development can be traced back to the beginnings of US and to the initial research of Dr Wild and his engineering associate, John Reid, PhD, and to their development of miniature transducers in the 1950s, as well as to Alfred Kratochwil, MD, in association with the Kretz Instrument Company in the 1960s (28). The development of real-time US, combined with advances in transducer technology, led to its widespread clinical use in the 1980s. It now was possible to examine the early developing fetus and to evaluate for ectopic pregnancy and other abnormalities of the pelvis. This further increased the usefulness of obstetric US and led to earlier diagnoses. By the late 1980s, it was possible to guide needles through the vaginal canal with direction from endovaginal transducers to perform chorionic villus biopsy and to retrieve eggs from stimulated follicles.
By the late 1980s, color Doppler US had been introduced and used in many areas of the body. It was not until the early 1990s that its usefulness in the fetus was fully developed and included the demonstration of placental and umbilical cord abnormalities. More recently, it has been used as a supplement to real-time gray-scale imaging in the evaluation of congenital cardiac abnormalities (29).
In the 1990s, the greatest advance was the development of three-dimensional gray-scale and color Doppler US imaging. Our center has had the opportunity to evaluate some of the earliest three-dimensional–US equipment, which can be used to obtain a volume of US data and display it in multiple planes and to produce reconstructed three-dimensional images (Fig 12). In the United States, Dolores Pretorius, MD, and Thomas Nelson, PhD, have been leaders in developing its uses in obstetrics, as well as in other areas in the body (30). My associate, Anna Lev-Toaff, MD, has taken the lead and has developed a technique—three-dimensional hysterosonography—for better defining the uterine cavity and for demonstrating causes of pregnancy failure, such as submucosal myomas and polyps (31). Three-dimensional US often provides additional information about the fetus and includes exquisite detail of the fetal face, limbs, and genitalia. Image reconstruction has improved considerably with advances in computer capabilities. Three-dimensional color imaging allows for the visualization of vessels in many areas of the body. While this is still of limited value in the fetus, one can predict that with improvement in reconstruction, as well as with the development of real-time capabilities, the advantages of displaying vessels and cardiac chambers in the fetus will become possible in the not-too-distant future.
|
The contributions of radiologists, as well as of obstetrician-gynecologists, to the improvement of maternal well-being and fetal health have been recognized as a key component in all countries around the world. The World Health Organization has placed the importance of US second only to general x-ray imaging in helping to improve health care and the quality of life throughout the world (32). Radiologists in the past and in the present have played important roles in the development and evaluation of new US technology, which has led to improved diagnoses in the evaluation of the developing fetus.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
