(For a printable version without graphics, please click here)
Learning Objective - To develop a more complete understanding of Nuclear Medicine imaging methods and to develop some visual experience with nuclear images.
The practice of Nuclear Medicine includes both diagnostic and therapeutic techniques. Most of the procedures are related to organ imaging using internally distributed radioactive material. There are also diagnostic techniques which quantitatively measure physiologic function (such as thyroid radioiodine uptake or gastric emptying time). Several therapeutic procedures are done using radioactive material and it is sometimes hard to understand the difference between Nuclear Medicine Therapy and Radiotherapy or Radiation Oncology. If a therapeutic procedure using radiation uses the radioactive material in an encapsulated form which can be retrieved from the body, that is usually done in the specialty of Radiotherapy (Radiation Oncology). If the radioactive material is introduced into the body so that it becomes incorporated into some body physiology and cannot be retrieved, that procedure would be done in Nuclear Medicine.
Since the most common procedures in Nuclear Medicine are organ imaging procedures, this discussion will be concentrated on those studies. To understand nuclear imaging, it is necessary to take a very different approach to interpreting the images than was used for conventional radiographic images. Radiographs are produced by transmitting an x-ray beam through the patient and recording the shadow. In nuclear imaging, the photon source is the patient so the beam is radiating outward from the site in which the radiopharmaceutical is localized. For nuclear studies, the images are usually labeled in relationship to the body surface against which the detector is placed. For instance, if the patient is supine and the detector is placed against the anterior abdomen to record the liver image, that image would be called an anterior view of the liver. The terms AP and PA which were so common in describing radiographic images are not used for nuclear images.
The standard detector system for Nuclear Medicine is a scintillation camera. The scintillation camera uses a sodium iodine crystal to convert the energy of the gamma photon into a flash of light that is detected by a photo-multiplier tube array which views the side of the scintillating crystal away from the patient. There is a collimator between the patient and the scintillating crystal so that the image will record only primary photons moving directly from the organ to the crystal (the collimator is analogous to the grid used for conventional radiography).
Scintillation Camera Picture
Scintillation Camera Diagram
The photo-multiplier tubes produce an electrical pulse when they are exposed to light from the scintillating crystal. The output from all of the photo-multiplier tubes can be summed to determine the total energy of the gamma photon and this allows the spectrometer to limit the recorded photons to only those coming from the radiopharmaceutical administered for the current test (in other words, we do not want to include counts from background radiation and other radioactive material in the image). The output from each photo-multiplier tube can be evaluated in relation to the other tubes in the array to determine the exact site where the gamma photon interacted with the scintillating crystal (the amplitude of the electrical pulse from each photo-multiplier tube is inversely proportional to the distance from the center of that tube to the position of the scintillation). Once the position (on an x-y axis) of the scintillation has been determined, a point of light is placed in a corresponding position on a video display. The equipment is usually pre-programmed to record a specific number of events (photon interactions in the scintillating crystal). The number of events to be recorded is determined by the level of statistical reliability needed for accurate image interpretation. Most nuclear medicine images record somewhere between 500,000 and 1,000,000 counts per image.
Two (2) specialized techniques should be mentioned because the acronyms referring to those techniques are so commonly used. The term PET imaging is used for positron emission tomography. This is a very elegant method of doing tomographic (body plane) imaging using positron emitting isotopes. Positrons are positively charged electrons which are emitted in the decay process of some radionuclides. When positrons are emitted, they have a relatively short half-life ending by interaction with negatively charged electrons. This event results in the annihilation of the electron and the positron producing two (2) 511 KeV photons moving in exactly opposite (180 degrees opposed) directions. To record this annihilation event, scintillation detectors can be placed on opposite sides of the patient and electronically connected so that only simultaneous events (occurring simultaneously in both detectors) are recorded.
When an event is recorded in both detectors simultaneously, the interaction must have occurred exactly half way between the two (2) detectors. From that, you can see that the positron image would be inherently tomographic showing the distribution of the radiopharmaceutical in a plane of the body located exactly at the mid-point in the space between the two (2) detectors. Since the detector position can be moved relative to the body, the plane being imaged can be set at any position. The problem with positron emission tomography is not related to the technicalities of imaging but complexity is created by the relatively short physical half-life of most positron emitting radioisotopes. Because of the short half-lives, it is usually necessary to produce the radioactive material on site and this requires the operation of some type of particle accelerator.
A second commonly used acronym for nuclear imaging tomography is SPECT. This is another imaging method in which single photons are used (in contrast to the two (2) photons used for PET imaging). The acronym actually stands for single photon emission computed tomography. In the SPECT technique, commonly used radiopharmaceuticals (most of these are labeled with 99m Technetium) are localized in the organ of interest. While the organ is radioactive, the detector is rotated around the patient so that the gamma photons are collected from many different viewpoints. After sufficient information is collected, the image can be reconstructed using an algorithm which is very similar to the one used for x-ray computed tomography.
The most important single factor in Nuclear Medicine imaging is the radiopharmaceutical. The radiopharmaceutical has two components. There is a radioactive label which allows external detection but the more important part of the radiopharmaceutical is the chemical moiety which determines exactly where and how the radiopharmaceutical will be localized. You might think that you could make the entire body radioactive and then just image the organ that you would like to study. Of course, that approach would not work because you would not be able to distinguish the organ of interest from the surrounding radioactive background. It is absolutely necessary to have a radiopharmaceutical which precisely localizes the radioactivity to the organ and the physiologic function that you want to study. Frequently, there is more than one method by which an organ can be imaged and that is particularly true if the organ has several physiologic processes which might be used to localize the radiopharmaceutical. For example, the liver can be imaged using either the reticuloendothelial system phagocytic function or the hepatocyte function. The lungs are usually imaged to show both ventilation and perfusion.
The technical characteristics of the radiopharmaceutical are extremely important. The gamma photon emission from the radioactive isotope must be an appropriate energy for the instrumentation (scintillation camera) which is currently available. Usually an emission energy in the range of 80-200 KeV is appropriate. It is desirable to have a radioactive label which only produces gamma rays. Beta particles are not energetic enough to reach the detector so they only contribute to radiation exposure of the patient but produce no useable imaging photons. The radioactive isotope should have a sufficiently long physical half-life to allow completion of the procedure but the physical half-life should be relatively short so that the patient radiation exposure can be limited. The radiopharmaceutical must be readily available and reasonably cheap. The specific radioactive nuclide must have good chemistry in the sense of being easy to attach to the chemical moiety of the radiopharmaceutical.
Frequently, the term half-life is used without any descriptive modifier. Most of the time when that is done, physical half-life is the intended meaning of the term. However, there are other half lives and it usually improves communication if you will specify exactly which half-life you mean if you use the term. The most important half-life in determining radiation exposure to the patient is the effective half-life. The effective half-life is calculated using the physical and biologic half-lives of the radioactive material. The formula is given below but the formula is not very important. The information you need to remember is that the effective half-life is always shorter than the shorter of either the biologic or physical half-life.
The remainder of the Nuclear Medicine discussion is to demonstrate some commonly used images. For convenience, these will just start at the top of the body and move down.
Although several different types of brain scans have been done over the history of nuclear imaging, the currently used brain scan is a study based on the distribution of cerebral perfusion. These images are usually displayed in para-sagittal and trans-axial tomographic sections (SPECT). Since the brain is symmetrical, the study can usually be interpreted by comparing the sides. However, some abnormalities affect the brain in a rather symmetrical way so it is important to have some standard reference for normal brain perfusion. The images below show the normal perfusion pattern.
Normal perfusion pattern
The usefulness of brain perfusion imaging is not clearly established. It can be used to demonstrate, in a relatively quantitative way, what is the current perfusion pattern and that is sometimes useful to distinguish multi-infarct dementia from other causes of dementia. There is some evidence that it might be useful as a prognostic indicator of deterioration in Alzheimer's disease and there is a hope that it might be able to predict the development of schizophrenia.
You should also understand that the Nuclear Medicine imaging approach has the ability to demonstrate the blood flow through specific vessels which are important to cerebral circulation. The image below is an anterior image obtained relatively soon after the injection of the radiopharmaceutical. The positioning of the scintillation camera allows flow through the neck vessels to be imaged and it also shows the regional distribution of perfusion in the brain, clearly defining the perfusion areas for the middle cerebral arteries as well as the anterior cerebral arteries. The flow study shows a patient immediately after the onset of a "stroke" showing significantly reduced flow to the left cerebral hemisphere. Some of this flow reduction is obviously related to obstruction of the carotid artery in the neck.
Thyroid imaging was the historical origin for Nuclear Medicine imaging. After the Second World War, radioactive iodine (131 iodine) was readily available from reactors used to produce fissionable material (Oakridge and Hanford). Medical researchers began using it to evaluate thyroid function. Although 131 iodine is no longer used for thyroid scanning (because the beta particle produces too much radiation exposure when enough is given to provide sufficient gamma photons for imaging) there are many other options for imaging the thyroid gland. The most commonly used radiopharmaceutical is technetium in the pertechnetate form (TcO4). The element technetium is chemically very similar to iodine and it is actively localized in the thyroid gland. In the human thyroid, no significant organification of technetium occurs so it does not persist in the gland in the same way that iodine would. It is useful for thyroid imaging because it is very inexpensive, and has a favorable gamma emission. The relatively short physical half-life results in a low level of radiation exposure. The image below is a normal 99m technetium image of the thyroid gland. You can see that technetium is not only localized in the thyroid gland but in the salivary glands and other mucous secreting glands as well. Thyroid imaging is commonly done with several radioiodine isotopes. 123 iodine is probably the most satisfactory although 125 iodine can still be used.
Normal technetium 99m image
The thyroid scan, in addition to showing the size and position of the gland, gives information about regional function. The image below is a normal thyroid.
The image below shows a relatively hypo-functional area in the right lower lobe which corresponds in position to a palpable nodule. If a palpable mass in the thyroid is relatively hypo-functional, it has a somewhat increased probability of being neoplastic. If all hypo-functional nodules were considered without regard to their other characteristics, it could be said that approximately twenty percent (20%) would be neoplastic.
The next image (below) shows a diffusely hyper-functional gland and this would be the appearance of the thyroid with Graves Disease or diffuse toxic goiter. This patient would have increased uptake of radioiodine, probably more than 50% of the administered dose at 24 hours.
Ventilation and perfusion imaging of the lungs has most commonly been used to detect the presence of pulmonary embolization. In routine clinical practice, it is also used to give a quantitative indication of regional lung function when pulmonary resection is planned. For instance, if it is necessary to remove the left lower lobe in an individual who has bronchiectasis, it might be helpful to learn exactly what percentage of total lung function would be lost with the removal of the that part of the lung. Using the lung scan, it is relatively easy to determine exactly what percentage of total perfusion and ventilation is attributable to each lung. This kind of quantitative information is sometimes used to decide on the extent of the surgical resection.
Localization of the radioactivity for perfusion imaging is done by using the pulmonary capillary bed as a filter to trap radioactive particles which are placed into the venous circulation. The specific radiopharmaceutical is a 99m technetium labeled macro-aggregated albumen. The albumen is heat aggregated so that ninety percent (90%) of the particles are between ten (10) and ninety (90) microns in maximum dimension. As you know, these particles are too large to pass through the pulmonary capillary bed so they are "strained out" with their site localization being proportional to blood flow. It is interesting that most perfusion lung scans are done to detect the presence of pulmonary embolization and yet we actually use embolization to localize the radiopharmaceutical. This is a very effective localization technique resulting in almost all of the radioactivity being trapped within the lungs. Although most nuclear medicine images show some background radioactivity in body areas other than the organ which is being imaged, the perfusion lung scan shows very little body background radioactivity.
After the radiopharmaceutical is localized in the lungs, six (6) images are made including the anterior, posterior, both laterals and both posterior oblique views. The radiopharmaceutical has a persistence half time (biologic half time) in the lungs of about three and one half (3.5) hours. Areas of reduced perfusion will be shown as focal areas of reduced radioactivity. The images below show the anterior and posterior perfusion images of a patient with normal perfusion.
Ventilation imaging of the lungs is designed to show the distribution of inspired air. Several different approaches have been used. Historically, the most common technique was to have the patient inhale a radioactive gas, 133 xenon. This was an effective method of demonstrating ventilation but the procedure was very dynamic only allowing images from one direction for each inhaled dose of radioactivity. Another disadvantage of this approach was the difficulty in controlling the xenon gas. Although an attempt was made to contain it in a closed system, unless the patient was completely cooperative, it was likely that some radioactive xenon would be liberated into the environment of the imaging area. Since this radioactive gas has an affinity for a number of materials (plastic drapes, floor wax, instrument grease, etcetera) it was common to find the xenon background level gradually rising during the working week in Nuclear Medicine departments which performed a large number of ventilation lung scans.
The more modern method of ventilation imaging uses the inhalation of a 99m technetium labeled aerosol. Using the aerosol, the radioactivity stays in the area into which it was inhaled long enough to allow multiple images to be produced. Using this technique, a six-view ventilation lung scan can be produced with images which match the perfusion images. The images below are the anterior and posterior images of the patient shown in the previous perfusion scan.
Although the ventilation/perfusion lung scan can be quantitatively precise in showing lung physiology, the interpretation to determine the presence or absence of embolization has always been problematic. If the patient has perfusion abnormalities, it is possible (actually, very likely) that the perfusion abnormalities are caused by a ventilation problem. The body will shift pulmonary artery blood flow away from areas in the lung which are hypo-ventilated. Because of that, if perfusion abnormalities are present, a ventilation scan must be done to determine the pattern of ventilation. If the patient has focal perfusion defects but has a normal ventilation pattern, it is assumed that the perfusion abnormalities are due to pulmonary embolization. If there are ventilation abnormalities which match the perfusion abnormalities, it is assumed that the perfusion abnormalities are caused by the ventilation abnormalities and this would be an indication of some type of intrinsic lung disease. Of course, ventilation abnormalities can be caused by a wide range of pulmonary pathology, anything from an acute asthmatic attack to lobar pneumonia. What is your diagnosis for the ventilation/perfusion lung scan below?
When the term "liver scan" is used we are usually talking about the anatomic liver scan which has historically been done using the reticuloendothelial system of the liver to localize the radiopharmaceutical. The liver can also be imaged using the hepatocyte function (bile formation) and both of these techniques will be briefly discussed below.
The most frequently requested liver scan is for the purpose of finding space occupying lesions within the liver. The best way to anatomically examine the liver using nuclear imaging is to localize the radioactivity using the reticuloendothelial system phagocytes (Kupffer's cells). The most frequently used radiopharmaceutical is 99m technetium labeled "sulfur colloid" (technetium heptasulfide). During the preparation of this radiopharmaceutical, a colloidal sulfide particle is formed which is labeled with 99m technetium. Since the particle is small enough to pass through capillaries, it circulates generally until removed from circulation by the reticuloendothelial system. Although some of this material ends up in the spleen, lungs, and bone marrow, if the liver is functioning normally, the great majority of it will be diffusely localized throughout the liver.
After the localization of the radiopharmaceutical, anterior, posterior, and both lateral views of the liver are obtained with the scintillation camera. The distribution of radioactivity should reflect the organ anatomy although liver size and shape are so variable that it is sometimes difficult to clearly distinguish normal from abnormal.
Space occupying lesions are usually easily detected. The anterior and right lateral images below clearly show evidence of multiple space occupying lesions throughout the liver and this appearance is most commonly produced by metastatic involvement of the organ. Remember, occasionally liver cysts will produce a similar scan appearance.
Another frequently seen abnormal liver scan pattern shows poor localization of the radiopharmaceutical in the liver with excessive localization in the spleen. The spleen is usually enlarged and there may some shrinkage of the right lobe and enlargement of the left lobe of the liver. This scan pattern indicates the presence of advanced hepatocellular disease and can be seen as an end result of severe viral hepatitis or alcohol abuse.
Anterior image of diffuse liver disease
The liver can also be imaged using the hepatocyte function. Originally, this method was developed using 131 Iodine labeled rose bengal to detect biliary atresia in infants. More recently, iminodiacetic acid compounds have been developed. These are labeled with 99m technetium and produce much better images of the liver and biliary tract. These radiopharmaceuticals have gone through several evolutions beginning with iminodiacetic acid or "IDA", then HIDA, followed by PIPIDA and finally DECIDA. Although you will see this study requested as a "pipida" scan, it is better to call it a biliary tract image. The Nuclear Medicine department will use the most up to date radiopharmaceutical.
This study can be used to demonstrate biliary tract obstruction. It is frequently used as an adjunctive study to an ultrasound examination of the gallbladder when the clinical picture suggests acute cholecystitis. If cholecystitis is present, the gallbladder will not fill with radioactivity by retrograde flow through the cystic duct. There is an assumption that the cystic duct is inflamed and too edematous to allow the retrograde passage of radioactivity from the common bile duct through the cystic duct into the gallbladder. So, if the ultrasound examination shows the presence of gallstones with a thickened, edematous gallbladder wall and the nuclear medicine study shows absence of gallbladder filling, the diagnosis of acute cholecystitis has a high probability of being correct in the appropriate clinical situation. This radiopharmaceutical moves through the liver fairly rapidly so it is not possible to get multidirectional views and most of these studies are done with sequential imaging in the anterior projection. In the normal situation, the radioactivity will be localized and removed from the liver within one (1) hour after injection. The films below show an early and late image from a normal biliary tract study.
Nuclear Medicine imaging of the skeleton has been one of the most durable nuclear imaging techniques. The first skeletal images were made using 85 strontium as the radiopharmaceutical and during the 1970s, these images were thought to be satisfactory for demonstrating metastases. An improvement occurred when 18 fluorine became available but that imaging technique was limited because of the relatively short physical half-life of the isotope (approximately two (2) hours). A major improvement in skeletal imaging occurred when technetium labeled compounds were developed. Originally, the radiopharmaceutical was based on a polyphosphate complex (the same polyphosphate produced by the Calgon Company as a dishwasher detergent). Since the polyphosphate chain length was somewhat unpredictable and that did have significant effect on localization in the skeleton, other phosphate compounds were rapidly developed. Currently, diphosphonates are used and they are labeled with 99m technetium.
Since the organ localization dynamics are relatively slow in the skeleton (Slow is a relative term. Actually, it is amazing that the radioactive phosphate is incorporated into the skeleton so rapidly.), following the intravenous injection of the radiopharmaceutical, a two (2) to three (3) hour delay is allowed before the imaging procedure begins. After the radiopharmaceutical is localized in the skeleton, anterior and posterior images of the entire skeleton are obtained. Any focal increase in osteoblastic activity results in an area of increased localization of radioactivity.
|Normal Bone Scan|
You might think that the bone scan would be diagnostically non-specific because so many varieties of pathology produce an abnormality. Increased osteoblastic activity could result from bone metastases, osteomyelitis, or even fractures. However, in clinical practice, the important information is usually the number and site of the abnormalities. Evaluating that information in the specific clinical context usually results in a very specific bone scan result. For instance, if a patient were in an automobile accident and continued to complain about posterior lower rib pain on the right side, three focal abnormalities in that area would certainly indicate the presence of rib fractures. If the patient's clinical problem is the recent onset of severe deep bone pain involving several areas in the axial skeleton and the bone scan shows multiple axial skeleton lesions, the scan result would be highly specific for metastatic skeletal involvement.
One of the most favorable aspects of bone scanning is that the abnormalities are shown as hot (more radioactive) spots in a relatively cold (less radioactive) background. An important characteristic of nuclear imaging is that detection of hot lesions is not size dependent. Actually, a microscopic area of increased radioactivity can be detected if it is at least twice background activity. The limitation of this concept is that the focal hot spot might not be recorded as the correct size. If the focal lesion is smaller than the resolving diameter of the collimator being used, the small hot spot will be displayed with a size equal to the resolving diameter of the collimator.
The anterior and posterior images of the skeleton shown below clearly demonstrate the effectiveness of this study in demonstrating metastatic skeletal involvement.
During the past decade, imaging procedures which are specific to the heart have become a major part of the total Nuclear Medicine imaging volume. Three (3) cardiac imaging procedures are frequently done using the nuclear medicine method.
The least frequently used nuclear cardiology procedure is the direct imaging of a myocardial infarct. This method uses a radioactive phosphate compound which will be localized in areas of necrotic tissue. This localization is not dependent on any organ function but is just a phenomenon of the chemical changes that occur when tissue dies. Although the method is technically very good with approximately an eighty five percent (85%) sensitivity, it is only infrequently used clinically. This is probably related to the fact that the diagnosis of myocardial infarction is usually reasonably certain without the use of this procedure.
The procedure is done by creating images of the heart in several different projections immediately after the radiopharmaceutical is injected and then repeating the image sequence after a delay of a least an hour. In the earlier images, it is as easy to see the distribution of the left ventricular blood pool. In the delayed images, there is greater skeletal localization and less blood pool localization. In the anterior images displayed above, the normal distribution on the left shows no localized myocardial collection of radioactivity. The cardiac blood pool is more radioactive than the remainder of the chest. In the abnormal image on the right side, there is a well defined collection of radioactivity related to the posterolateral wall of the left ventricle. That abnormality was caused by an area of infarction.
Myocardial Function Study
Another cardiac procedure which was frequently utilized ten years ago but is now being used somewhat less often is the myocardial function study. This method which is sometimes referred to as a MUGA (multiple gated acquisition) scan, measures the ability of the left ventricle to pump blood. The report of the study is usually given as an ejection fraction expressed as a percentage of end diastolic volume. If the procedure is carefully done, it can also show abnormal motion of the left ventricular wall.
The myocardial function study is done by making the blood pool radioactive. The contraction pattern of the left ventricular wall is inferred by the change in shape and volume of the left ventricular blood pool during the cardiac cycle. Since the nuclear medicine image is inherently three dimensional, it is very easy to calculate the difference between the end systolic volume and the end diastolic volume of the left ventricle. This volume difference is the amount of blood ejected during systole .
|Good Function||Poor Function|
The myocardial function study can be reported with a number of useful formats. The results of the study are reported as an ejection fraction but the study can also be used to evaluate the regional contraction of the left ventricular myocardium. In the images above, the area of the blood pool of the left ventricle is labeled in heavy black by the computer. Then in the lower images, the end diastolic edge is superimposed on the end systolic edge of the left ventricle. As you can see, the image set on the left side is essentially normal with a 65% ejection fraction and good contraction of the left ventricular myocardium. The image set on the right side shows an ejection fraction of only 41% with relatively poor contraction of the left ventricle.
Myocardial Perfusion Study
The myocardial perfusion study is currently the most frequently done nuclear cardiology procedure. Historically, this method used 201 thallium to show the distribution of myocardial perfusion. Currently, other chemicals are used and these are labeled with 99m technetium as the radioactive tracer. There is good evidence to show that these chemicals are localized in the myocardium proportional to blood flow and the results of the study are similar to those obtained using 201 thallium.
The myocardial perfusion study is usually done using a tomographic technique (Single Photon Emission Computed Tomography, or SPECT). With tomographic imaging, it is possible to look at the perfusion of the myocardium in relatively thin planes which are usually displayed in the coronal, para-saggital, and transaxial directions. Using this method, the patient's myocardial perfusion is usually imaged at rest and immediately after exercise. With this approach, relatively small areas of ischemia can be detected.
In the images above, the pair on the left side is a normally perfused left ventricle viewed in cross section across the short axis of the left ventricle. This can be thought of as a coronal section of the left ventricular myocardium. The superior image is an image done at rest while the inferior is an image done immediately after stress. The image pair on the right side is the same set of images but in that set you can see that the stress image shows markedly reduced relative perfusion to the anteroseptal wall of the left ventricle. This abnormal perfusion pattern with stress is usually caused by atherosclerotic involvement of one or more coronary arteries.
If you would like to learn more about the specific image patterns produced by these nuclear cardiology methods, there is a package of reference films in the Radiology Department. In this package, there are examples of all of the procedures described above.
1. Why are the quantitative nuclear procedures so infrequently used?
2. What is the relative organ radiation exposure of the nuclear scan compared to
radiography? For example, a renal scan
compared to an intravenous pyelogram?
3. What is the risk(s) from the radiopharmaceutical?
4. Are patients a risk to others after a nuclear imaging procedure?
1. Wagner, HN. Nuclear Medicine: The Road to Smart Medicine and Surgery. JNM August 1999;39;8:13N-34N.
Send us comments: Dr. David Adcock, DAVID@uscmed.sc.edu.
Return to the School of
Return to the University of South Carolina homepage.