This time, I want to say a bit about the two imaging techniques we haven't covered yet, ultrasound (USS) and magnetic resonance imaging (MRI). Unlike the scanning methods we looked at in the previous two columns (5 and 6) these do not rely on the use of ionising radiation, so they are free of the small risks (discussed in column 3) associated with exposure to x-rays and gamma rays.
As its name suggests, this technique uses high frequency sound waves instead of ionizing radiation to form images. The sound waves are emitted by a hand-held transducer, or probe, and the returning wave, reflected from the interfaces between different tissues in the body, is detected and used to form the image.
You just can't imagine how much excitement was generated by the advent of ultrasound imaging - you had to be there, as they say. At the time the technology was really taking off (in the late seventies) I was a junior surgeon, and it was the developments in ultrasound more than anything else that led me to change direction and start training all over again, this time as a radiologist. The technology had actually been around for a long time. In the 1950s, Ian Donald, an obstetrician in Glasgow, began to experiment with the use of adapted industrial sonar equipment to examine the pregnant uterus.
Initial commercially available scanners were clumsy, ceiling-mounted affairs, which produced low resolution black and white images. Even so, the potential of the technique was obvious, and during the 1970s and eighties, there were enormous advances. First, the development of grey scale imaging began to allow the differentiation of different types of tissue, and then real-time imaging made the whole process of acquiring images much quicker and easier, as well as allowing the operator to watch moving structures within the body. Use of the technique has extended beyond obstetrics, and is now applied to the examination of just about every organ system in the body, resolving structures down to a couple of millimetres in size (see the image of a foetus in the uterus - figure 1 ).
The importance of ultrasound lay in the fact that it made possible the examination of organs which had previously been completely inaccessible, or which had only been imperfectly visualized using complex and potentially risky interventional procedures.
What is more, no ionizing radiation was involved, and so ultrasound has had an enormous impact on paediatric imaging, because we are particularly keen to minimize the radiation exposure of children. And the technology continues to improve; for example, use of the Doppler effect (change in wavelength of the sound when it reflects off a moving column of blood) means that blood flow to organs or tumours can be monitored (see figure 2 (left) - image of flow in carotid artery ).More recently, we have seen the introduction of ultrasound contrast agents (microscopic bubbles of air), which will extend the range of applications for the technique even further, and new software enables the machines to produce three-dimensional images (see figure 3 (right) - a foetus still in the womb - it's a boy! ). In addition to its clinical utility, we shouldn't underestimate the importance of the fact that ultrasound equipment is relatively cheap compared to modalities such as CT, MR and PET (and I mean relatively - a good ultrasound machine still costs in the region of £120,000!).
Magnetic Resonance Imaging (MRI)
While ultrasound uses sound waves to form images, MR uses the combination of a strong magnetic field, and radiofrequency waves. The patient lies at the centre of a really strong magnet (see figure 4 - left), which causes all the hydrogen molecules in the body to line up with the magnetic field. A radio frequency signal is then used to knock them out of alignment; when the RF field is removed, the molecules all flick back and emit a signal which is amplified to form the image. It's actually more complicated than that, but that's about as much as I understand.
MR developed out of the work done with nuclear magnetic resonance (NMR) used to analyze small amounts of material in physics labs in the middle of the 20th century. By the 1960s, this type of NMR spectroscopy was well established, and in the 1970s a number of groups began to be interested in the production of images of human tissue using the same technology. In the UK, there were separate teams in London, Nottingham and Aberdeen working on this, and small prototype scanners were produced. The first commercial machine was installed in Manchester in 1983. Since then, the technology has developed in leaps and bounds, as was the case with CT (see last month's column), and now images of exquisite detail can be obtained in any plane with relatively short scan times. I said earlier that it would be difficult to overestimate the impact of the first ultrasound images. In the case of MR, it would be an understatement to say that we were gobsmacked (see figure 5, for images of our webmaster's brain - said to be normal! - coming shortly). Scanners now produce images of real people that look just like the illustrations in anatomy textbooks, and it is even possible to talk about 'MR post-mortems' as a possible alternative to actually cutting up a body after death and looking for yourself. A bit grim, I know, but it illustrates just how close these MR images are to the real thing. One of the main areas of application of MR at the moment is in the diagnosis of musculoskeletal problems. Images of the interior of joints can be produced which reveal in exquisite detail the muscles and ligaments, enabling orthopaedic surgeons to diagnose disease or injury and plan their surgical treatment before they ever put knife to skin.
Not surprisingly, MR has made an impact in just about every area of medicine and surgery. It can, for example, be used to demonstrate blood flow without the use of intravascular contrast agents, and will undoubtedly eventually replace conventional angiography in a number of areas (see figure 6 (right) , showing the aorta and its main branches).
MR is also the only imaging modality other than nuclear medicine (see previous column) which can provide useful functional information; this aspect of MR is particularly exciting, and is likely to develop rapidly in the near future.
Over this and the previous two columns we've looked at the enormous advances in imaging technology over the past thirty years or so. All very well, but apart from making radiology much more expensive, has it had any real effect on how patients are managed?
The answer is yes, and that's the reason for the excitement doctors experienced as these new methods emerged. Let's take just one example. When I was still a surgeon, back in the seventies, nearly every operating list would include an 'exploratory laparotomy'. A laparotomy is a major procedure to open up the patient's abdomen and look inside, and it was 'exploratory' because the surgeon usually had no idea what he or she would find. They would be doing this either because the patient had abdominal symptoms that might be due to serious disease, or because they had a known problem like lymphoma (cancer of the lymphatic system) and the only way to find out if the spleen and abdominal lymph nodes were involved was to get in there and have a look. That just doesn't happen any more. When today's surgeon opens an abdomen, chest or head, they nearly always know exactly what they are going to find, because pre-operative scanning has confirmed that disease is present, demonstrated its extent, and often its exact nature. This means that the operation can be planned in detail in advance, and there are seldom any nasty surprises.
Not only that, but we can avoid the situation, all too frequent thirty years ago, where the surgeon would operate only to find no sign of any disease requiring surgery or disease which had progressed so far that surgical treatment wasn't possible. In either case, the patient had undergone an unnecessary major operation.
Thirty or forty years ago, the diagnosis of disease was chiefly made by clinical means (that is: the taking of a history and direct examination of the patient by the doctor) with radiology sometimes providing some help in confirming the doctor's suspicions. At the start of the 21st century, imaging occupies a much more central role in diagnosis, so much so that concern is sometimes expressed that doctors are losing the ability to examine patients properly for themselves, and are relying too much on the radiologist's scanners!
In addition, the rise of interventional radiology means that radiologists don't always stop at making the diagnosis; they are often able to go on and treat the patient, using techniques which can avoid the major surgery that used to be required.
In Star Trek, Bones made his diagnosis in a matter of seconds by running a hand-held scanner over the patient. Well, we're not quite there yet, but the possibility is much less fanciful than it was when the series first appeared on our screens!