CT Scans, Nuclear Medicine and PET imaging
Last time, I explained a bit about how standard radiography using x-rays was used to form images of the insides of our bodies. Now I want to talk about all the other imaging methods that are available to the present generation of doctors. All I can do in the limited space available is to give you a flavour of the new 'scanning' technology, which has revolutionised medical practice. This time I'll deal with the two techniques which, like x-rays, use radiation (CT and nuclear medicine), next time, the two which don't (ultrasound and magnetic resonance imaging)..
We have seen that conventional x-ray techniques shine a wide beam of x-rays at the patient, and capture the transmitted radiation with its 'shadow' of the internal organs on a piece of fairly standard photographic film. CT scanning uses x-rays as well, but in a completely different way. Instead of the wide beam of rays, CT scanners produce a very narrow 'pencil' beam. This passes through the patient and hits not a film, but an electronic radiation detector which simply records the amount of radiation passing through that bit of the patient. Which is OK, but it just gives you a number, not a picture. The clever part of CT scanning was to mount the x-ray tube and detector on opposite sides of a circular, doughnut-shaped, gantry. If you then rotate the tube and detector though a few degrees and repeat the measurement, and keep on doing that until you have moved through a full circle, a dedicated computer can reconstruct an image of the internal organs in that thin slice of the patient (figure 1- click to enlarge). Advance the bed and patient a few centimetres through the gantry and repeat the process, and you get another slice, and so on until you have a set of images encompassing the part of the patient you are interested in. The early scanners were only big enough to get the head into, but it wasn't long before technology allowed the construction of gantries which would take the whole body.
We often talk of 'revolutionary' developments in medicine and other sciences, but it would be difficult to overestimate the impact of CT scanning on imaging, and on medicine in general. Let's put it into context by thinking about brain imaging. Before CT, there were two methods for demonstrating disease in the brain. One involved the injection of contrast material (see previous article) into the blood vessels supplying it (angiography), and the other depended on the injection of air into the fluid-filled space around and within the brain (pneumoencephalography). The first of these carries some risk, and will only demonstrate the blood vessels, not the 'meat' of the brain, the second only shows the internal and external surface of the brain, and was exquisitely painful. So, dangerous and unpleasant investigations, which still only delivered limited information concerning the structure of the brain and any abnormality within it. Overnight (almost!), CT produced images of the internal structure of the brain which, even with the unsophisticated early machines, far surpassed anything which had been possible previously. Suddenly, we could 'see' tumours and decide whether they were curable by surgery before you operated. We could diagnose infarcts (damage to brain tissue caused by bleeding or blood clots in the blood vessels supplying the brain) in patients with strokes, and decide on the appropriate treatment. With the advent of body scanners the same was true of disease in organs such as the pancreas, which had previously been completely inaccessible to imaging. I'll say a bit at the end of next month's column about the impact all this had on those of us lucky (and old!) enough to have been around at the time, but Sir Godfrey Hounsfield, the inventor ofCT, thoroughly deserved his Nobel prize.
The early scanners could take several minutes to scan one slice, so imaging the whole chest or abdomen took a long time. Thanks to developments in x-ray tube design, computing power and engineering, scan times have come right down to a few seconds. Thanks to slip-ring technology, the x-ray tube and detector can now revolve continuously, and if the bed the patient is lying on moves slowly and continuously into the gantry, the tube actually describes a spiral around the patient, allowing information from a large volume of their body to be acquired in a short time (quick enough, for example, for the whole chest to be scanned during one breath hold, eliminating any blurring due to respiratory movements). The dedicated imaging computer can then reconstruct slices from that volume of material. The resulting images are exquisitely detailed (figure 2), enabling individual organs and their relationship to each other to be demonstrated. For example tumours can be seen, and the exact degree to which they have spread to surrounding structures can be documented. This may, sadly, show that the tumour is inoperable, but at least that will spare the patient the distress of an unnecessary operation; on the other hand, the scan pictures may help the surgeon to plan the operation in some detail before the patient ever gets near to the operating table, increasing the chances of success. If radiotherapy is to be used, the images can be used directly in planning the radiation treatment, guiding the beam to ensure that it hits the tumour without damaging the surrounding normal tissue.
Nuclear Medicine Also known as isotope imaging, scintigraphy or radionuclide imaging, nuclear medicine (NM) has actually been around for a while. This is a completely different way of seeing inside patients. All the techniques we have mentioned so far shine a beam of radiation through the patient, and use the transmitted radiation to form an image. In NM, the source of radiation is introduced into the patient's body, and the emitted rays are used to paint the picture. The radiation is produced by the decay of relatively short-lived radioactive isotopes with the production of gamma rays. These are essentially the same as x-rays, but with shorter wavelengths and higher energy. The trick with NM is to choose a chemical which will be taken up in the organ we are interested in, then 'label' it with a suitable radioactive isotope. The isotope we use most is technetium, because it is readily available, has suitable physical characteristics for imaging, and also has the ability to link readily with other molecules to provide a whole range of radiopharmaceuticals suitable for investigating all the major body systems. The radiopharmaceutical is injected (usually) into a vein.
The radiation that emerges from the patient is detected by a gamma camera. The business end, or head, of the gamma camera contains a large single crystal of sodium iodide about a centimetre thick and up to 40cm across. When gamma rays are absorbed by the crystal, a tiny flash of visible light is emitted. These flashes are detected and amplified by an array of photomultiplier tubes, and the resulting image is stored in a computer. Cameras can have two heads, and most modern scanners have the ability to rotate around the patient, rather like a CT scanner, and produce images of a slice through the patient, as well as producing static images (figure 3). The beauty of NM is that it provides functional imaging. In other words, it produces images that are related to the function of the organ concerned, rather than simply telling us about its appearance. In fact, the anatomical information in NM scans is fairly low resolution in comparison with other imaging techniques, but often we don't care too much about the size and shape of an organ, we're much more interested in how well it's working. It all depends on the clinical situation.
Take the kidney, for example. In a patient with a malignant tumour of the kidney, we want to know exactly how big the tumour is, how far it has spread and whether it will be possible to remove it surgically. For that, we need a CT scan. For a patient with a kidney stone that is causing obstruction to the passage of urine down the ureter to the bladder, on the other hand, we want to know how bad the obstruction is, and what effect it is having on the function of the kidney. In this case, we don't care what it looks like, we need to know what it is doing; we need a NM scan. After injecting a radiopharmaceutical which is rapidly cleared from the blood by the kidneys and excreted in the urine, we can position the gamma camera over the kidneys and capture rapid-sequence images over the course of half an hour or so, following the changes in levels of radioactivity as the pharmaceutical is excreted. This will give us a measurement of the relative function of the kidneys and also tell us just how severe the obstruction is.
Often, of course we want to know about structure and function, and so NM techniques are complementary to the other imaging methods we have been looking at. Almost any organ or tissue and many of the physiological processes in the body can be imaged in this way. Cardiac efficiency can be measured by tagging the patient's blood cells with an isotope and observing the passage of blood through the pumping chambers of the heart, and by using a different radiopharmaceutical, blood flow to the muscle of the heart itself can be mapped, allowing us to detect and quantify coronary heart disease (figure 4). Perhaps the simplest application of nuclear medicine is the bone scan. By injecting a labelled phosphate compound which is taken up by actively metabolising bone cells, we get an image which doesn't show the detailed bone structure like ordinary x-rays, but instead maps the activity of the skeleton. Why is that helpful? Well, any damage to a bone, whether caused by injury, infection or tumour, will result in increased activity by the bone cells in the area, as part of the repair process. This will be seen as a 'hot spot' on the bone scan, often long before x-rays become abnormal. For example, the x-rays of the athlete's leg seen in figure 5 (right) were normal, but the bone scan clearly shows the hot area due to a stress fracture.
PET imaging PET imaging (positron emission tomography) is a new technique in nuclear medicine which is rapidly gaining clinical acceptance. In PET, isotopes which emit positrons are used. A positron is a positively charged electron: it is anti-matter. As soon as a positron meets an electron, which it does almost immediately, there's a lot of them about, they annihilate each other with the production of two gamma rays moving in directly opposite directions. These hit a ring of detectors surrounding the patient and form an image. Some of these isotopes are very short-lived, and can be used to label molecules which play a central role in cellular metabolism; glucose, for example. There is no space to go into PET in detail here, but it is having a huge impact in oncology (cancer medicine) due to its ability to detect small areas of active tumour, allowing early treatment (figure 6 - below), and in cardiology. It is also producing exciting developments in the study of the brain, as it is possible to image changes in brain activity in real time, as patients perform different tasks,so that we can 'see' patients thinking.