Technetium-99m

Technetium exists only in the form of radioactive isotopes. The first Tc-99m generator was developed in 1958. This allowed convenient transport of the radionuclide and an explosion in the development of medical applications of Tc-99m.

Tc-99m is the most widely used tracer in nuclear medicine, accounting for over 85% of routine diagnostic procedures in a nuclear medicine department. The following radiation characteristics of technetium-99m make it an ideal agent (Fig. 12-1).

Figure 12-1 Decay scheme showing the isometric transition of Tc-99m.

Gamma rays are attenuated both in tissue and in detector material (thallium-doped NaI). Compton scatter is the predominant interaction in tissue while the photoelectric effect is the interaction occurring in the detector. Radionuclides emitting low-energy gamma rays are likely to be absorbed by photoelectric effect within the tissue and will not reach the detector. This leads to poor resolution images, often requiring larger doses than those with higher-energy gamma rays, which will make it through to the detector.

Tc-99m is obtained from “milking” a molybdenum-99 (Mo-99) generator (Fig. 12-2). Mo-99 is obtained as a by-product from the fission of uranium. Mo-99 is chemically separated from the other radionuclides in the reactor product. Purified Mo-99 as anionic molybdate solution is loaded on to the generator column, which contains alumina. The alumina, which is positively charged, is able to adsorb Mo-99 ions. The assembly is then autoclaved. Several normal saline washings of the column yield eluate containing Tc-99m. These washings are subjected to several quality control tests to determine eluate volume, radionuclide purity (“moly” breakthrough), radiochemical purity (proper chemical form of technetium as pertechnetate), pyrogenicity, sterility, and/or alumina breakthrough. The half-life of Mo-99 is 66 hours allowing for weekly delivery of the generator to elute Tc-99m.

Figure 12-2 Molybdenum-technetium generator.

The eluted technetium can now be tagged to the appropriate pharmaceutical for use. When radionuclides such as technetium are tagged to specific pharmaceuticals, they are called radiopharmaceuticals. The pharmaceutical portion of the diagnostic radiopharmaceutical is present in very small amounts and will not elicit any pharmacologic response in the patient. The radioactive component is present in even smaller amounts. The technetium-99m radiopharmaceuticals for use in the emergency situation are listed in Box 12-1.

The radiopharmaceuticals are made up from “cold kits” supplied by manufacturers. With the exception of HMPAO (hexamethylpropyleneamine oxime), the kits are generally stable for up to 12 months. These kits provide a very convenient method of preparing radiopharmaceuticals in facilities that may not be close to laboratories. Technetium-99m can assume oxidation states ranging from I to VII based on the number of electrons available for reaction with ligands. The lower the oxidation state the less stable it is and the most likely to react with ligands. The most stable electronic configuration for technetium is the state with the value of VII where it is present as the pertechnetate ion and unlikely to tag to a ligand. Consequently, in order to produce a Tc-99m–labeled radiopharmaceutical, the technetium has to be first reduced to a lower oxidation state. This is most commonly accomplished by the addition of tin. As well, some cold kits contain antioxidants such as ascorbic acid or gentisic acid, which not only retard oxidation of the radiolabeled product, but also improve the stability of the kit. Tin II (stannous) supplies electrons and in the process becomes oxidized to tin V (stannic) ion. The various oxidation states of technetium allow the formation of a variety of different radiopharmaceuticals, which adds to the value of technetium as a nuclear medicine staple.

Reagents in the “kit” vial are generally freeze-dried and packed under vacuum. Alternatively, they may be combined with an inert gas such as nitrogen. This ensures that the tin in stannous form will not get oxidized to stannic form by exposure to the atmosphere. The amount of tin in the reagent vial is critical to the preparation of the radiopharmaceutical. Inadequate amounts of tin allow the oxidation of free pertechnetate, which ultimately degrades the image obtained and could potentially render a study nondiagnostic.

Indium-111

In-111 is produced in a cyclotron from the radioactive parent Cd-112. It has a half-life of 67 hours (2.8 days). The principal photons are 178 keV and 247 keV. The relatively long half-life of In-111 allows sequential imaging (Fig. 12-3).

Figure 12-3 Decay scheme for In-111—electron capture.

Xenon-133

Xe is an inert and relatively insoluble gas produced by the fission of uranium-235. It is used in its radioactive form most commonly for ventilation studies. Xe-133 has a physical half-life of 5.3 days and, in the face of normal pulmonary function, a biological half-life of approximately 30 seconds.

The main disadvantages of Xe-133 are the low photon abundance (36%) and low tissue penetration of the 81 keV gamma ray. The low photon energy generally mandates the initial use of the xenon, that is, before the administration of the technetium lung perfusion agent. Overlying soft tissue, such as breast, can produce artifacts on the xenon images.

Gallium-67 Citrate

While Ga-67 citrate cannot be used for an emergent situation requiring a diagnostic test within a few hours of requisition, it is an important radionuclide for the workup of infection, which constitutes an emergency of sorts.

Gallium-67 is produced in a cyclotron from the parent zinc-68. Gallium-67 decays by electron capture to stable zinc-67 in 3.26 days (Fig. 12-4). The transition energy, which is 0.997 MeV, is dissipated by several electron capture transitions. Several gamma photons are emitted that are used for imaging. The principal ones are 93 keV with 37% abundance, 185 keV with 20% abundance, 300 keV with 17% abundance, and 394 keV with 5% abundance.

Figure 12-4 Decay scheme for gallium-67.

The exact mechanism of uptake of Ga-67 citrate is not precisely known. After intravenous administration of Ga-67, the complex dissociates to become bound to transferrin.

The principal organs that localize gallium are the liver, spleen, and bone marrow. The excretion of gallium is bimodal, initially through the kidneys and later through the GI tract. Persistence of gallium activity in the kidneys beyond 24 hours should be cause for further workup. There is some uptake in the lachrymal and salivary glands and the lactating breast. This is attributed to the high concentration of lactoferrin in these tissues. Both transferrin and lactoferrin are metabolized in the liver, which accounts for the uptake in the liver. Gallium is believed to behave like iron and utilizes the transferrin mechanism and responds to procedures like total body irradiation in a manner similar to iron saturation of transferrin.

Pulmonary Embolism

The incidence of pulmonary embolism (PE) and its detection have increased with the increasing frequency of long-distance travel and consequent loss of mobility. This has been further complicated by the aging of the population and the increasing incidence of underlying illness such as heart disease and peripheral vascular (including venous) disease.

PE is the third most common cause of death in the United States with 650,000 occurring each year. PE seems to present with greater preponderance in hospitalized patients. Given the high prevalence of the condition, its lethality, and the fact that a large number of patients with PE have atypical presentations, it is recommended that every patient with chest pain be worked up for PE.

Venous thrombosis, in contrast to arterial thrombosis, is caused by problems with the plasma clotting system. There is minimal platelet participation in the venous versus the arterial process. Furthermore, thrombus formation in the cardiac chambers is seen most commonly with slow flow conditions similar to venous thrombosis.

There is increasing evidence that an underlying coagulopathy may be responsible for spontaneous deep venous thrombosis (DVT) and PE. Hypercoagulability may be congenital or acquired. Primary or acquired deficiencies in protein C, protein S, or antithrombin III are known to be common causes of DVT and PE.

There are many risk factors for DVT and PE. It is well known that, among these, prolonged immobility, recent surgery, pregnancy, and underlying malignancy initiate DVT.

Radiopharmaceuticals and Techniques

The major function of the lungs is to effect the exchange of CO₂ for O₂ from the blood. This is accomplished by perfusion of the capillaries in the walls of the alveoli where inspired air brings in O₂ and is exchanged for CO₂ in the deoxygenated blood. The CO₂ in the alveoli is then discharged to the outside via expiration. These two aspects of the respiratory anatomy and physiology are assessed in nuclear medicine. Pulmonary perfusion is assessed by the administration intravenously of technetium-99m macroaggregated albumin, whereas the various phases of ventilation are studied using gases or aerosols.

In pregnant women, a perfusion-only lung scan should be considered if the chest x-ray is normal and the patient has no history of smoking or lung disease. Since Tc-99m is excreted in the breast milk, the patient should be encouraged to express and store milk for 2 days until the radionuclide has decayed sufficiently.

Perfusion

Technetium-99m macroaggregated albumin (MAA) is the agent of choice for assessing pulmonary perfusion. The mechanism of accumulation with this agent is the blockade of capillaries in the pulmonary arterial circulation.

A minimum of 100,000 particles is necessary to obtain readable images. To obtain optimal images, approximately 200,000 to 600,000 particles are required. The size of the particles ranges from 5µ to 100µ with the majority measuring around 30µ. The desired particle concentration can be obtained by diluting the kit with normal saline. It is important to check the particle size using a microscope or a hemocytometer. Smaller fragments enter into the general circulation and are phagocytized by the liver and spleen.

Approximately 0.1% of capillaries are blocked with one injection. Ordinarily this degree of blockade is of little or no consequence. However, in patients whose pulmonary arterial circulation is tenuous, such as patients with pulmonary hypertension, blockade of the reduced capillary bed can precipitate cardiac complications or exacerbate the underlying condition. In patients with right-to-left shunts, transfer of particles into the systemic arterial circulation could potentially cause adverse coronary or cerebral events. For these patients as well as pediatric patients, we recommend the administration of a lower dose of particles (100,000 to 200,000).

Technetium-99m MAA is produced by the addition of Tc-99m pertechnetate to a sterile “cold kit” containing stannous albumin aggregates as lyophilized powder. Tc-99m pertechnetate is added to the cold kit. The tagged kit is allowed to stand at room temperature for 15 minutes to ensure maximum tagging. The usual administered activity in adults is 3 to 5 mCi (111–185 MBq). The pediatric dose is 25 to 50 mCi/kg of body weight. The lung is the critical organ in this procedure, receiving an absorbed radiation dose of 0.22 rad/mCi.

Tc-99m MAA is administered intravenously in a peripheral vein. Withdrawal of blood back into the syringe could produce clots, which could appear as “hot spots” on the lung scans. Injection into central venous access such as Swan-Ganz catheters is to be avoided. The syringe should be gently agitated to avoid the settling out of the MAA particles. Injecting the dose slowly over three to five deep breaths, as well as having the patient supine, will ensure the even distribution of particles.

The patient is imaged in a sitting position, although images could be obtained with the patient supine. Ideally, the patient is imaged on a large field-of-view (FOV) gamma camera using a parallel-hole collimator. A diverging collimator may be necessary in larger patients to encompass both lungs on the anterior and posterior views. The standard views are the anterior, posterior, right and left laterals, both right and left posterior oblique, and both right and left anterior oblique (Fig. 12-6). Generally, a minimum of 500,000 counts is accumulated per image.

Figure 12-6 Normal Tc-99m perfusion lung scan. Images show a standard six-view study: anterior, posterior, and left anterior oblique views (top row, left to right); left posterior oblique, right posterior oblique, and right anterior oblique views (bottom row, left to right).

In our institution, we have found single photon emission CT (SPECT) to be of use in determining whether the perfusion defect is segmental or nonsegmental.

Ventilation

Inert gases can be used to perform ventilation studies of the lung in tandem with technetium perfusion studies. Their short biologic half-life allows relatively complete clearance of the lung providing a clean slate, so to speak, for lung imaging with a longer-lived perfusion agent.

Xenon-133

Xenon-133 is relatively inexpensive and is the most commonly used agent for ventilation. Overlying soft tissue easily attenuates the principal gamma emission of 81 keV. Consequently, images of the lungs are generally obtained in the posterior projection. Additional images may be obtained in the posterior oblique projections. Images are rarely obtained in the anterior, anterior oblique, and lateral projections because of the degradation caused by overlying soft tissues such as breast. Consequently, any abnormalities that are seen only in the anterior areas of the lungs will be missed. Exhausting the xenon is a cumbersome and complex process requiring a charcoal trap or a venting system. To avoid contamination with xenon, which is heavier than air, a negative pressure room is required. Furthermore, xenon is fat soluble and will adhere to plastic drapes, floor wax, and instrument grease. Consequently, the background level will gradually rise during the workweek.

Xenon can be used to assess all phases of ventilation. The most commonly used technique involves having the patient breathe xenon through a spirometer. The patient exhales as deeply as possible and then inhales 10 to 20 mCi of Xe-133. The respiration is suspended at the end of the inhalation for 15 seconds, while the first image is obtained. The patient breathes xenon out into a spirometer, which constitutes a closed system. Approximately 2 L of oxygen are used to dilute the expired xenon (Fig. 12-7). The patient rebreathes this mixture for 2 to 3 minutes, at which time another static image is obtained. This constitutes the equilibrium image. After equilibrium has been reached, fresh air is breathed in until the xenon is completely washed out. Images are obtained every 15 seconds for 2 to 3 minutes. For patients with chronic obstructive pulmonary disease (COPD), the washout phase may be delayed up to 5 minutes to image areas of regional airway trapping. This entire process presupposes that the patient is able to cooperate with breathing into a spirometer or a closed system. The initial/single breath reflects the regional ventilatory rate. The equilibrium phase depicts the aerated volume of the lungs, while the washout phase delineates trapping. Xenon is fat soluble and partially soluble in blood, which will cause deposition in the liver, particularly in patients with fatty replacement in the liver.

Figure 12-7 Normal xenon-133 ventilation lung scan. The ventilation is usually performed before the perfusion study. Due to rapid clearance of lung activity of xenon-133, a posterior view study is performed to maximize area of the lung seen. Comparison to perfusion is less than optimal.

Krypton-81m

Kr-81m has also been used for ventilation imaging. The photon emissions of 176 and 192 keV and the short half-life of 13 seconds allow equilibrium imaging of the lungs in multiple projections. However, the short half-life precludes obtaining single-breath and washout images. Kr-81m is obtained from rubidium-81, which has a half-life of 4.6 hours. While krypton generators are available, the cost of the generator can be quite high, due to the short half-life of rubidium-81. The activity administered is 1 to 10 mCi.

Radiolabeled Aerosols

Radiolabeled aerosols are used for studying lung ventilation. Aerosol studies do not allow dynamic imaging. Dynamic imaging, while excellent for assessment of respiratory function, is not required for PE studies.

The most commonly used aerosol is Tc-99m diethylene triamine penta-acetic acid (DTPA) in a volume of 2 mL put into the nebulizer of an aerosol delivery system. The aerosol is prepared by injecting 30 mCi of Tc-99m DTPA into the nebulizer of the delivery system. Side tubing in the system allows oxygen to flow through a flow meter at the rate of 8 to 10 L/min. Generally, only a small amount of activity (1 to 2 mCi) is delivered. The trachea is the critical organ in the aerosol ventilation study. One of the major advantages of this system is that it can be attached to an endotracheal or tracheostomy tube. In patients without airway tubing, the aerosol is inhaled through the mouth while a nose clip is applied. In either circumstance, little or no patient effort or cooperation is required to deliver the aerosol into the airway. The aerosol is usually delivered with the patient in a supine position so that it is distributed evenly. No special venting is required with Tc-99m DTPA as it is with xenon. The second major advantage of using Tc-99m DTPA is that the imaging characteristics are ideal. Therefore, the Tc-99m DTPA study can be performed either before or after the Tc-99m MAA study. If the ventilation study is performed before the perfusion, the amount of activity required is between 1 and 2 mCi. However, if the perfusion study is obtained first, the number of counts accumulated for the ventilation study has to be 3 to 4 times greater than for the perfusion.

In the workup for PE, Tc-99m DTPA aerosol is ideal, since images can be obtained in projections to match the ventilation images. After inhalation, aerosol particles are deposited in the distal airways and not the alveoli. Following inhalation, the Tc DTPA particles dissolve in the fluids within the alveoli and ultimately diffuse across the epithelial barrier into the circulation. As long as the epithelial barrier is intact, the aerosol diffuses relatively slowly into the circulation. The half-time disappearance of the aerosol from the alveoli is about 80 minutes. This is much faster in patients whose epithelial barrier may be deficient, as in COPD or in smokers. The Tc DTPA that has entered into the circulation is cleared via the kidneys. Larger particles are deposited in the central airways, the mouth, and the alimentary tract (from swallowed particles) (Fig. 12-8).

Figure 12-8 Tc-99m DTPA aerosol images. Normal study shows six views, and corresponding six views are obtained for perfusion. Comparable images make interpretation of lung scan much easier than with single posterior view xenon-133 study.

Technegas

Technegas is composed of ultrafine particles of Tc-99m-labeled carbon particles produced by combustion of pertechnetate at 2500° C in a graphite crucible. A technegas generator is portable and produces fine images. The smaller particle size makes this a better agent than Tc DTPA aerosol, particularly in patients with COPD.

Chest X-Ray

It is important to have a chest x-ray for evaluation before performing the ventilation-perfusion (VQ) scan (Fig. 12-9). It is good practice to have one that has been obtained within 24 hours of performing the VQ scan. It would be ideal to have a full-inspiration posteroanterior (PA) and lateral chest x-ray available for interpretation. However, a large percentage of the patients who are at risk for PE in the hospital or the intensive care unit setting may have several other cardiopulmonary pathologic processes that could interfere with the reading of the VQ scan. In these patients, one has to be satisfied with the portable radiograph, which is rarely of the quality of the standard PA and lateral radiographic study.

Figure 12-9 Normal chest x-ray. Interpretation of the VQ study for pulmonary embolism.

Symptomatology and chest x-ray findings in patients with PE are frequently nonspecific. Hampton’s hump and the Westermark sign are findings on chest x-rays that are seen in PE. However, these signs are not often visualized on the chest radiograph. With the introduction of multidetector CT scanners, the majority of patients for PE workup are directly imaged using CT pulmonary angiography. In patients with impending renal compromise, CT angiography with contrast may be fraught with danger. Increasingly, the CTPE has become the gold standard for diagnosis of pulmonary embolism. As well, VQ lung scanning is important for establishing the diagnosis of PE. These studies can indicate whether or not the patient is likely to have PE.

The Prospective Investigation Of Pulmonary Embolism Diagnosis (PIOPED) study, completed in the 1990s, provided a comprehensive look at the value of ventilation and perfusion scans in acute PE. The PIOPED criteria were based on the use of Xe ventilation scans. With the development of CT pulmonary angiography, the algorithms proposed by the PIOPED study have had to be modified.

A pragmatic approach to the interpretation of VQ scans has been proposed, and the following facts have to be taken into consideration:

• The most important diagnostic feature of PE on VQ scans is the mismatched defect—where there is a perfusion defect with no abnormality on the ventilation scan.

• The basis of the mismatch is that with an uncomplicated PE, the ventilatory architecture remains intact.

• Complete matching of the VQ abnormalities is seen in obstructive lung disease.

• In pulmonary infarction, there is incomplete matching of the VQ defects and a corresponding abnormality on the radiograph.

The first step is to obtain a chest x-ray for comparison with the VQ scans.

The second step is to scrutinize the perfusion scan for perfusion defects. These should be characterized based on whether they are segmental/subsegmental or nonsegmental. Nonsegmental defects are those that do not correspond anatomically to a segment and are therefore unlikely to be secondary to a pulmonary embolus.

The third step is to determine the size of the segmental perfusion defect. A classic segmental perfusion defect corresponds anatomically to a bronchopulmonary segment. These are pleural based and wedge shaped. The defect is categorized as large when it occupies 75% or more of the segment, moderate when occupying 25% to 75% of the segment, and small when it is less than 25%.

The fourth step is to ascertain whether the segmental or subsegmental defects are matched on the ventilation study.

Unmatched perfusion defects are not likely due to acute or chronic pulmonary emboli. Pulmonary emboli are multiple in 90% of cases and bilateral in 85%. In the first few days, the defects may disappear or become smaller. New defects may occur because of fragmentation. Changes in regional perfusion pressure could transform a partially obstructing clot into a fully obstructing one.

The fifth step is to discuss the likelihood of PE based on the scans in light of the pretest likelihood of PE. If no abnormalities are noted on the perfusion study, the study is considered normal. The likelihood of PE is less than 5%.

Abnormal exams are classified by the size and the number of perfusion abnormalities and concomitant abnormalities (or absence of abnormalities) on the ventilation scan, as follows:

A high-probability scan has two or more large, mismatched segmental defects (or equivalents in moderate/large defects) with no abnormality on the ventilation study (Fig. 12-10). In the clinical setting where PE is highly likely, a high-probability VQ scan indicates a probability of PE greater than 90%. If the mismatched perfusion defects should resolve within days or weeks, the probability of recent embolism is higher.

Figure 12-10 High-probability lung scan with Tc-99m DTPA aerosol and MAA. Superior segment of the right lower lobe and a major portion of the left lower lobe show perfusion deficit with good ventilation, a VQ mismatch.

A low-probability scan is one in which the perfusion scans are smaller than 25% of a segment regardless of the ventilation scan or chest x-ray appearance, matched on ventilation scan, and accompanied by larger radiographic abnormalities (Fig. 12-11).

Figure 12-11 Xenon-133 ventilation study (A) and Tc-99m MAA study (B) show nonsegmental matching abnormalities suggesting a low probability for pulmonary embolism.

An intermediate-probability scan is one that does not fit into the high or low categories.

Repeat Scans

Because the perfusion picture can evolve over the ensuing weeks or months, it is advisable to repeat the study after 3 weeks. Defects persisting at 3 months are unlikely to resolve. The larger the defect and the older the patient, the less likely is the perfusion scan to revert to normal. In patients with diffuse lung disease, emboli are less likely to resolve completely. Patients with high-probability or intermediate-probability scans who have been treated for PE should be followed up with a 3-month scan.

Radiopharmaceuticals

The radiopharmaceuticals used for the diagnosis of brain death are Tc-99m DTPA and Tc-99m HMPAO or ECD. Tc-99m DTPA is used on the premise that it provides a radionuclide angiogram. We prefer Tc-99m HMPAO or ECD since they are brain perfusion agents and will not be picked up in a nonperfused brain. Furthermore, a nuclide angiogram can be performed dynamically while injecting the Tc-99m HMPAO or ECD and additionally followed up with planar static or SPECT images.

Tc-99m DTPA is used in this indication as a nondiffusible blood pool agent. Therefore, it is best used for the radionuclide angiographic portion of the study. The Tc-99m DTPA does not cross the blood–brain barrier.

Tc-99m HMPAO (hexamethyl-propylene oxime or exametazime) can be prepared with or without stabilizer for brain imaging. The Tc-99m HMPAO is lipophilic and is preferentially retained by the brain because it is converted in vivo by glutathione to the hydrophilic complex. This could also occur in vitro with high stannous ion concentration, heat, and free radicals. Hence, only small amounts of stannous chloride must be used. The Tc-99m pertechnetate should be fresh and not more than 30 minutes from eluting. The stabilizer used is methylene blue, which acts as a free-radical scavenger.

Technique

In some nuclear medicine laboratories, a scalp tourniquet is applied to minimize the external carotid perfusion. This measure also reduces scalp perfusion, which could be mistaken for cerebral perfusion. The most commonly used tourniquet is an elastic band placed above the level of the orbits. This is not recommended in children since the application of the tourniquet may increase the intracranial pressure.

In most instances the use of Tc-99m DTPA will suffice. It is injected intravenously as a bolus, and dynamic images may be obtained with the camera centered over the face. Occasionally, it may be difficult to administer a good intravenous bolus. In this case, using Tc-99m DTPA is convenient, in that a repeat injection can be performed.

The use of Tc-99m HMPAO or ECD is recommended in difficult cases. Since the uptake of these agents relies on both perfusion and active uptake by a viable brain, a dynamic flow study may not be necessary and delayed static images will suffice, particularly if the intravenous injection of bolus was unsatisfactory and subsequent dynamic images could not be obtained for a variety of reasons including malfunctioning of the computer.

Since both dynamic flow and static images are used in this study, we prefer to use the low-energy all-purpose (LEAP) collimator, which can be used in both the dynamic and static studies. With Tc-99m DTPA 10 mCi are given intravenously, whereas with Tc-99m HMPAO, 15 to 20 mCi of activity are administered. Because both sets of compounds are tagged with Tc-99m, the window setting is at 15% at a 140 keV peak. The flow study involves obtaining 1- to 2-second images for 30 seconds. Immediate or early planar static images are acquired for 500k to 750k counts in the standard anterior, posterior, and both lateral projections. It is vital to image over the injection site to verify that the dose has not been infiltrated. With Tc-99m HMPAO delayed static images are obtained 2 hours post injection in the standard projections.

Interpretation

On the radionuclide angiogram, flow will be seen in the common carotid arteries. The flow will not be seen beyond the base of the skull. If there is no cerebral blood flow, no uptake will be seen in the brain.

Tc-99m DTPA is used strictly as a blood pool agent in the documentation of brain death. It does not cross the blood–brain barrier, remaining in the blood pool even on delayed static images. Consequently, the delayed static images, which are obtained 5 to 10 minutes after completion of the “flow” study, will not reveal any activity in the territory of the cerebral arteries.

On HMPA and ECD studies, in a normal, viable brain there will be uptake of radionuclide in the brain—preferentially in the cortex. With brain death, no activity will be seen in the brain. Occasionally, there is increased uptake of the radionuclide on the immediate static images in the nasopharyngeal region. This was originally described by Mishkin. However, it has been documented in other instances such as internal carotid occlusion without brain death. This is attributed to the shunting of blood flow into the external carotid circulation. In the presence of clinical signs of brain death, the “hot nose” sign could be used as a secondary or corroborative finding. The sagittal sinus may be visualized faintly due to filling from the external carotid circulation.

With brain death, there is no uptake of HMPAO or ECD radiopharmaceuticals in nonviable brain tissue. This can be further verified by obtaining, if necessary, SPECT images, which will show an “empty skull.”

Tomographic SPECT images allow better visualization of activity distribution within the skull, and allow one to see brainstem and cerebellar activity that is hard to see with planar imaging shown above. Figure 12-13 shows the same patient as in Figure 12-12 with SPECT imaging clearly demonstrating activity in the brainstem and cerebellum. The cerebellar and brainstem activity usually clears in 24 to 36 hours and confirms diagnosis of brainstem death as well. The patient shown in Figures 12-12 and 12-13 showed no activity in the cerebellum or brainstem when study was repeated after 20 hours.

Figure 12-13 Tc-99m ECD brain perfusion SPECT imaging shows complete absence of perfusion in the cerebral hemispheres and minimal uptake in the brainstem and cerebellum. The brainstem and cerebellar activity usually disappears within 24 hours and a second study is required to confirm complete absence of perfusion.

Radiopharmaceuticals

The radiopharmaceuticals used include Tc-99m methylene diphosphonate for part of the workup of musculoskeletal infection, gallium-67 citrate, and indium-111 oxime and Tc-99m HMPAO for autologous leukocytes.

Leukocyte Labeling

Leukocytes can be labeled with indium-111 oxime or with technetium-99m HMPAO. Infections that mount a neutrophilic response are best visualized with the leukocyte label. Opportunistic infections, however, are not associated with neutrophilic response; hence these are not ideal for imaging with radionuclide-labeled leukocytes.

The labeling process is performed in vitro, is cumbersome, and requires meticulous attention to detail. Furthermore, the in vitro process carries with it the risk to personnel of handling blood products. The process itself takes 2 to 3 hours. Forty milliliters of blood is withdrawn into a syringe containing anticoagulant. The syringe is kept upright for 1 to 2 hours to allow the red blood cells (RBCs) to sediment. This is further accelerated by the addition of hydroxyethyl starch and the hypotonic lysis of the RBCs. The next step is to separate the leukocytes from platelets by centrifugation. The leukocyte pellet that forms after the separation of the RBCs and platelets is incubated with the radiopharmaceutical, washed, and injected into the patient. The doses of Tc-99m HMPAO-labeled and In-111–labeled leukocytes are 5 to 10 microCi and 300 to 500 µCi, respectively.

A total white blood cell (WBC) count of at least 2000/mL is essential for obtaining satisfactory images. Neutrophils are the majority of WBCs labeled; hence the images obtained reflect neutrophil-mediated response. Mature granulocytes are highly specialized, short-lived nondividing cells. They generally measure around 12 to 15 microns, and possess a multilobulated nucleus and multiple cytoplasmic granules. These mature in the bone marrow for about 15 days and then reside in the circulation for 10 hours. If they migrate into tissues, they survive about 4 days.

Images obtained with leukocytes labeled with indium or technetium immediately following injection show intense pulmonary activity that clears rapidly. This is believed to be due to leukocyte activation during labeling, which slows their movement through the pulmonary vasculature. Tc-99m HMPAO–labeled WBC studies, on the other hand, can be performed within a few hours of imaging, making this a more desirable study for “emergency” purposes. The advantages and disadvantages of each of these techniques are discussed in a later section.

There are some conditions presenting as emergencies in the workup of infection, particularly osteomyelitis, where the labeling of autologous leukocytes could present a problem. The first of these is sickle cell crises. Here the sickling RBCs cannot be effectively separated from WBCs. The second is leukemia, where the WBCs encountered are abnormal and functionally not responsive to infection.

Indium-Labeled Leukocytes

A large FOV gamma camera is used with a medium-energy parallel-hole collimator. A 15% window is centered on the 174-keV photopeak and a 20% window is centered on the 245-keV photopeak.

With indium-111–labeled WBC studies, images are obtained 24 hours after injection, by which time the pulmonary activity would have already cleared. It is important to note that the 24-hour time lag required for indium-labeled infection imaging would remove it from consideration as a “true emergency.” At 24 hours following injection, there is normal distribution through the spleen, liver, and bone marrow (Fig. 12-14).

Figure 12-14 Normal distribution of In-111 oxine-labeled leukocytes. A large portion of the activity is sequestered in the liver and spleen.

One disadvantage of indium-111 imaging is that the amount of radioactivity administered—0.3 to 0.5 mCi of indium—results in low photon flux and makes it more difficult to accumulate the necessary number of counts to achieve a satisfactory image. Furthermore, the photopeaks are not optimal in comparison with technetium-99m.

The advantages of using the indium label are its long half-life and the fact that the label is stable. Another advantage is that indium can be used in conjunction with other radionuclides such as Tc-99m sulfur colloid bone marrow imaging, particularly with musculoskeletal infection. Here simultaneous dual isotope scans can be acquired, or, alternatively, they can be acquired in tandem. Since some infections can be indolent, such as those involving joint prostheses, the longer half-life of indium makes it a much more useful compound.

Technetium-Labeled Leukocytes

A low-energy, all-purpose, high-resolution collimator is used with a 15% window centered on the 140-keV photopeak of Tc-99m. Granulocyte labeling is selective with the Tc-99m HMPAO complex. Urinary activity and renal parenchymal activity appear soon after injection. Biliary activity is seen in a variable number of patients as early as 3 hours following injection. This accounts for bowel activity appearing at 4 hours. Bone marrow uptake is also seen with Tc-99m–labeled leukocytes, although none is seen in normal bone. Images are routinely obtained at 2 and 4 hours post injection, although 45-minute and 24-hour images may also be obtained if needed (Fig. 12-15).

Figure 12-15 Tc-99m HMPAO WBC imaging shows two foci of collection in the left upper quadrant of the abdomen corresponding to suspected abscess on CT scan of the abdomen.

Technetium Methylene Diphosphonate

Tc-99m methylene diphosphonate (MDP) is one of the agents of choice for evaluation of osteomyelitis. This is the agent used for the triple-phase bone scan. It is injected intravenously, and flow/perfusion studies of the affected part are obtained at one image per second for a total of 30 seconds. For a perfusion study, ideally a high-sensitivity collimator is used. This is followed by the acquisition of “blood pool” images or immediate static images of the area in question. Two to four hours after intravenous administration of the Tc-99m MDP, delayed static images are acquired for 500 K counts per projection utilizing a high-resolution collimator (Fig. 12-16). In smaller nuclear medicine departments it is not uncommon to use a LEAP collimator for both the perfusion and the static images.

Figure 12-16 Tc-99m MDP scan shows increased flow, blood pool, and delayed activity in the right first MTP joint and great toe region suggesting an osteomyelitis. Increased delayed activity in the left MTP region is due to osteoarthritis and does not show increased flow or blood pool counts.

Indications

Infections generally tend to be indolent and rarely present as emergencies. However, when the postsurgical patient presents with symptoms suggestive of infection, there is an excellent case for the workup utilizing nuclear medicine. The accuracy of WBC-labeled studies in abdominal abscesses is greater than 85%, which is comparable with CT.

With the advent of fusion imaging or SPECT CT, the work of the diagnostician is easier and the results far more valuable. The complementary use of the anatomic strengths of CT with the metabolic sensitivity of nuclear medicine will serve to vastly and incrementally increase the usefulness of each of these modalities by overcoming the limitations of each modality.

Abdominal Abscesses

Abdominal abscesses are a common and difficult problem in the postsurgical patient. Although ultrasound is probably the easiest and the most expedient method of diagnosis, surgical dressings, wounds, edema, and the patient’s body habitus often provide deterrents to diagnosis using this modality. While CT provides exquisite resolution, it carries with it the inconvenience of administration of oral contrast and risk of precipitating or worsening renal compromise with the administration of intravenous contrast. Furthermore, the radiation exposure with multidetector CT is not trivial, particularly in patients who might need repeated studies. Here nuclear medicine with the use of radiolabeled WBCs can prove invaluable in pinpointing the pathology. The additional advantage of using nuclear medicine procedures is that the entire body can be imaged without additional radiation exposure.

The most common causes of intra-abdominal abscesses include the following:

The most commonly encountered organism is Escherichia coli. Skin flora may cause abscesses following penetrating wounds to the abdomen. Neisseria gonorrhoeae and chlamydia are commonly involved in pelvic abscesses in females who might have pelvic inflammatory disease. Microbial flora of the GI tract varies from the proximal to the distal with small numbers of aerobic streptococci in the stomach and proximal small bowel, and large numbers of these organisms as well as large numbers of anaerobic gram-negative bacilli and anaerobic gram-positive flora in the terminal ileum and colon. The differences in the types and concentrations of these microorganisms partially account for differences in septic complications based on the location of the injury or disease. Septic complications from upper GI perforations are far less ominous than those from colonic insults.

With indium-labeled WBC scans, whole body imaging is recommended 4 and 24 hours after intravenous administration of the agent; and with Tc-99m–labeled WBC studies, imaging is done at 1 and 4 hours. Abscesses demonstrate progressively increasing activity, often greater than the liver and outside areas of normal accumulation.

In the case of hepatic abscesses, In-labeled WBCs are preferred because the biliary excretion seen with Tc-99m–labeled WBCs could complicate the interpretation of the image. It is important to note that activity in an abscess will increase on delayed images, while normal uptake regresses. Hence, when searching for hepatic abscesses, we recommend that In-111–labeled WBCs be used. Analysis of the images should definitely not be made on the basis of the early images alone.

Pancreatic abscesses are best studied with Indium-111–labeled studies because of the increased likelihood of spontaneous communication of the abscess with the gut. As a note of caution, unfortunately, WBC scans are positive with pancreatic abscesses and fat necrosis as well as pancreatic pseudocysts.

In patients on peritoneal dialysis with catheter tunnel infection, nuclear medicine provides the ability to distinguish between early peritonitis and exit site infection. This is of special importance where the clinical management is different based on the site of the infection.

There are six functional compartments within the peritoneal cavity:

Inflammatory Bowel Disease

Tc-99m–labeled WBC studies are believed to be useful in the diagnosis of inflammatory bowel disease (IBD). Early views of the abdomen supply the diagnosis with uptake in the diseased bowel greater than noted in the spleen. Delayed images are less reliable, because with Tc-99m–labeled WBC studies, there may be excretion of Tc-99m–labeled secondary complexes into the bowel. The configuration of the uptake in the bowel is usually sufficient to distinguish IBD from abdominal abscesses.

Musculoskeletal Infections

The major problems of imaging for osteomyelitis with WBC-labeled studies are that (1) osteomyelitis is frequently more chronic and (2) WBCs normally accumulate in the bone marrow where they are destroyed by the reticuloendothelial system. In the imaging of a prosthesis, nuclear medicine imaging is not affected by the prosthesis, whereas the prosthesis is a deterrent to imaging with any of the other imaging modalities. Bone scintigraphy is useful, in that a negative scan essentially excludes a prosthetic complication. Adding Ga-67 citrate increases the accuracy of bone scintigraphy. The combination of leukocyte labeling combined with marrow imaging is believed to have the highest accuracy of all the imaging studies available (Fig. 12-17).

Figure 12-17 Tc-99m MDP bone scan shows soft tissue infection but no evidence of osteomyelitis. Patient is status post removal of right hip prosthesis.

Joint Prostheses

Infection and loosening are the most frequent complications of joint replacement. It is vital to be able to differentiate between these, especially since the management strategy is totally different in each case. With aseptic loosening, the patient requires a single hospital admission for the single-stage revision joint replacement. With an infected prosthesis multiple hospital admissions are required: initially to perform an excisional arthroplasty, then for a prolonged course of antibiotic therapy, and eventually for a revision arthroplasty.

When there is no Tc-99m MDP uptake in and around the prosthesis, the possibility of any actual abnormality is remote. Diffuse periprosthetic uptake is seen in infection as well as aseptic loosening. With the new porous-coated prostheses, periprosthetic bony ingrowth is stimulated; therefore, diffuse periprosthetic uptake is seen because of programmed bony ingrowth and is not necessarily indicative of pathology. This type of uptake when seen in nonporous prosthetics is associated with increased blood flow and osteolysis seen with infection. Tc-99m MDP imaging is of tremendous value because of high negative predictive value.

The most important differentiator between the two clinical entities—loosening and infection—is the level of neutrophils, which are always present in large numbers in infection and absent in loosening. Therefore, Tc-99m–labeled leukocytes would seem to be intuitively the agent of choice. However, since leukocytes are also present in the bone marrow, one has to search for concordance and discordance with accompanying Tc-99m sulfur colloid images. For instance, when the WBC-labeled images show more activity than the Tc-99m sulfur colloid, the discordance is due to infection. The combined technique has an accuracy of greater than 90%.

Ga-67 citrate imaging is also useful because of the high negative predictive value. Gallium uptake is related to inflammation and not necessarily infection alone. This makes this a less specific test. However, Ga-67 may prove to be more useful if the infection is believed to be more chronic and therefore less likely to be mediated by granulocytes.

Osteomyelitis

Osteomyelitis can present following trauma, as a consequence of bacteremia, or as a result of vascular insufficiency such as in a diabetic foot. The common organisms vary with the age of the patient: Staphylococcus aureus and Streptococcus in neonates and infants, S. aureus in adults, and gram-negative bacteria in the elderly. With the changing face of disease, immune status of patients, and exposure to “exotic” microorganisms from previously inaccessible geographic locations, the list of organisms producing osteomyelitis has become much more extensive.

In children osteomyelitis is often hematogenous. Since the epiphyses and metaphyses have separate blood supplies in children, it is common to have only metaphyseal involvement with sparing of the epiphysis. In most instances there is increased accumulation of the radionuclide (Tc-99m MDP) on all three phases, increasing and becoming more focal on the delayed images. Ewing’s sarcoma can mimic osteomyelitis on scans. Consequently, clinical correlation is important to differentiate one entity from the other.

In diabetic patients, infection is often complicated by neuropathy. Tc-99m MDP scans are used for anatomic localization. The diagnosis of infection is effected by using WBC-labeled agents. Since there is uptake of WBC label in the bone marrow, one looks for discordant versus concordant uptake to make the diagnosis. Tc-99m sulfur colloid is used to locate the position of the bone marrow. With the use of SPECT-CT, we can anticipate that the use of a single radiopharmaceutical and anatomic correlation with CT will make the diagnosis and follow-up of osteomyelitis easier.

Septic Arthritis

Septic arthritis tends to be monoarticular and shows increased uptake with Tc-99m MDP, WBC-labeled scanning, and Ga-67. This uptake, however, is nonspecific and can be seen in instances of gout, which can manifest as a monoarticular, sterile inflammatory arthropathy. Uptake is seen on both sides of the joint. With associated cellulitis, radionuclide uptake may be detected in areas around the joint or adjacent soft tissues.

Septic arthritis in pediatrics is a true emergency and often affects the hip. The most common mode of presentation with imaging is as a photopenic defect with normal radiograph. The diagnosis is made following hip aspiration.

Transient synovitis can affect the hip in children with a fair degree of frequency. Bone scans are often negative or only mildly positive in this entity.

Infection in Immunocompromised Patients

Infection imaging in immunocompromised patients is best effected using gallium-67 citrate rather than labeled WBCs. A normal chest x-ray (CXR) with a normal gallium scan excludes the presence of infective pathology in the chest. A normal gallium scan in the face of an abnormal CXR is indicative of the inability of the patient to mount a response and carries a poor prognosis. Pneumocystis carini and/or bacterial pneumonia show diffuse intense uptake of gallium. Faint uptake in the chest may be seen in mycobacterial or cytomegalovirus (CMV) infections. Lymph node uptake is seen most often with mycobacterial infection or lymphoma. Kaposi’s sarcoma will not take up gallium; hence, in the absence of Ga-67 uptake, an abnormal CXR should raise the suspicion of Kaposi’s sarcoma.

Infections in the Lung and Other Pulmonary Pathology

Tc-99m HMPAO–labeled WBC scans are ideal for lung abscesses. Surprisingly, lobar pneumonia is usually negative on white cell scanning except early in the course of the pneumonia. This is believed to be due to early termination of WBC migration in the disease. In acute respiratory distress syndrome (ARDS), there is persistent increasing WBC uptake in the lungs, such that delayed images will present with increasing uptake as opposed to the normal transient early uptake described previously in this section.

Infection associated with obstruction in bronchogenic carcinoma and with bronchiectasis shows abnormal uptake with labeled WBCs.

Radiopharmaceuticals

An ideal agent for GI bleeding would be one that quickly accumulates extravascularly at the bleeding site. The ideal agent would also be rapidly cleared from the blood pool providing good target-to-background ratios (TBRs). Since GI bleeding in most instances is slow and intermittent, the ideal agent would detect slow bleeding and allow for reimaging without reinjection.

The two agents that are used for detection of the source of GI bleeding are Tc-99m–labeled RBCs (Fig. 12-18) and Tc-99m sulfur colloid. Neither of these agents is ideal, and each agent has advantages and disadvantages that the other may not possess.

Figure 12-18 Tc-99m RBC study shows significant progressive accumulation of activity in the descending colon due to bleeding.

Tc-99m–labeled RBCs remain in the vascular compartment, which allows repeated imaging of the GI tract. However, it is the very persistence in the blood pool of the agent that precludes obtaining an optimal TBR. With the more popular but technically more challenging in vitro technique, the appearance of free pertechnetate can be avoided with meticulous attention to washing. The presence of free pertechnetate can present a problem since it is picked up by the gastric mucosa and can obscure bleeding in the gastric cavity as well as the overlying transverse colon. Since the Tc-99m RBCs stay in the blood pool, images are generally difficult to read because of the presence of the tagged RBCs in the large and capacitance vessels.

Tc-99m sulfur colloid is picked up by the reticuloendothelial system and shows uptake in the liver, spleen, and bone marrow. The advantage of the Tc sulfur colloid is that the high TBR allows the detection of active bleeding at low bleeding rates of 0.5 to 0.1 mL per minute. If the patient is not bleeding at the time of the injection and for the next 15 minutes, and bleeds later, no Tc sulfur colloid will be remaining in the blood pool to allow reimaging. Furthermore, the normal uptake in the liver and spleen can also obscure the flexures should they be the sites of bleeding.

Meckel’s Scan

Meckel’s diverticulum is the commonest GI anomaly occurring in approximately 2% of the population. This is more common in men than in women. Five to seven percent of these diverticula contain ectopic gastric mucosa, which can present with bleeding. Most of these patients present with rectal bleeding. The diverticula vary in position, frequently located approximately 2 feet from the ileocecal junction. They are most frequently visualized in the suprapubic region as a small focus of intense Tc-99m uptake.

Hepatobiliary Agents—Mechanism

Bile is produced by the hepatocytes; approximately two thirds of this is excreted into the duodenum via the common bile duct. The remaining one third flows via the cystic duct into the gallbladder where it is stored. When a cholecystagogue (CCK) is administered or is produced in response to a fatty meal, the gallbladder contracts and bile is emptied into the duodenum. If the patient has recently had a meal, the endogenous CCK will prevent the gallbladder from filling. Therefore, the patient is encouraged to not eat for 4 hours before the test. If the patient has been fasting for more than 24 hours, the gallbladder is generally full of viscous bile, which prevents accumulation of tracer. In contrast to bilirubin, the hepatobiliary agents are not conjugated; therefore, abnormalities of the conjugation mechanism are not a contraindication to the use of the hepatobiliary iminodiacetic acid (HIDA) scan or the diisopropyl iminodiacetic acid (DISIDA) scan.

The hepatobiliary agents used most commonly are Tc-99m mebrofenin (HIDA) or Tc-99m disofenin (DISIDA). These radiopharmaceuticals are taken up by the hepatocytes and excreted into the biliary tract. Elevated levels of bilirubin associated with hepatocyte dysfunction may preclude the use of DISIDA since there may be prolonged retention of the tracer in the liver. Following intravenous administration of 3 to 5 mCi of the tracer (HIDA), initial visualization of the cardiac blood pool is noted. Within 5 minutes, the agent appears in the liver. During the next 5 to 7 minutes images of the liver can be obtained in multiple projections. The agent then enters into the biliary tract with “opacification” of the major bile ducts and the common bile duct. At this time, there is appearance of the tracer in the duodenum. The cystic duct and gallbladder are the next to fill. Similar to the bile mentioned above, some of the tracer enters into the gallbladder as long as the cystic duct is patent.

Acute Cholecystitis

The vast majority of patients with acute cholecystitis have cystic obstruction. It is based on this fact that hepatobiliary scintigraphy is used for the diagnosis of acute cholecystitis. Absence of visualization of the gallbladder for up to 4 hours after injection of the radiotracer is considered diagnostic for acute cholecystitis (Fig. 12-19). The appearance of the tracer in the gallbladder effectively excludes the presence of acute cholecystitis. However, in some cases of acalculous cholecystitis, there may be filling of the gallbladder. Additionally, the presence of an accessory cystic duct can cause a filling of the gallbladder through an “alternate” route. Structures that simulate a gallbladder can produce false negatives; these include duodenal diverticula and biliary duplication cysts.

Figure 12-19 Acute cholecystitis. Nonvisualization of the gallbladder for 60 minutes with Tc-99m Choletech. The gallbladder fossa is empty, without any activity. Normal appearance of common bile duct and duodenum.

Not infrequently in chronic cholecystitis a gallbladder may not accumulate tracer. Some pharmaceutical interventions may actually prompt the filling of the gallbladder.

If the patient has fasted for a prolonged period of time, the gallbladder may be full of viscous bile. In this case, the administration of CCK prior to or early in the study (while tracer is still available in the liver) will make the gallbladder contract and empty its contents. If the cystic duct is not obstructed, then tracer will enter the gallbladder, effectively excluding the diagnosis of acute cholecystitis.

Intravenous administration of morphine (0.04 mg/kg diluted in 10 mL of normal saline administered over 5 minutes) causes constriction of the sphincter of Oddi, raising intraductal pressure and forcing the filling of the gallbladder (Fig. 12-20). Occasionally there may not be enough tracer in the liver or the biliary tree to fill the gallbladder. This might necessitate a second injection of the tracer. Administration of morphine is a measure that should be used with caution where there is obstruction to the proximal portion of the cystic duct, as evidenced by the “cystic duct” sign. Enough pressure may be generated in the system to force the tracer into the gallbladder, producing a false negative study.

Figure 12-20 Post morphine scan shows that gallbladder is not filled, indicating complete cystic duct obstruction. Retention of activity in the common bile duct is due to contraction of the sphincter of Oddi.

An important note of caution is to not use CCK after administration of morphine. The CCK causes the gallbladder to contract while the morphine raises the intraductal pressure. This can result in severe pain.

There are certain ancillary signs that are of use in the diagnosis of cholecystitis:

Tc-99m HIDA/DISIDA scans are of paramount importance in the detection of biliary leaks, postsurgical or post-traumatic (Fig. 12-21). These agents are highly sensitive in detecting biliary leaks, which are not seen with other modalities. Additional views of the paracolic gutter and the pelvis must be obtained. A thorough knowledge of the anatomy of the peritoneal spaces is useful in helping the surgeon with the postoperative care of the patient.

Figure 12-21 Tc-99m DISIDA study shows biliary leak with tracer along the right paracolic gutter.

Biliary Atresia Versus Neonatal Hepatitis

Hepatobiliary scintigraphy is useful in the differential diagnosis of biliary atresia versus neonatal hepatitis. Preparation of the patient for a week of phenobarbital (2.5 mg/kg orally twice a day) will stimulate the uptake of the tracer, allowing visualization of a patent biliary tree. This will help rule out atresia.

Biliary atresia occurs after birth in very young infants or neonates. This is characterized by the development of progressive inflammation in the biliary tract causing irreversible damage and disappearance of the bile ducts. The neonate appears normal at birth and then starts to develop obstructive jaundice 2 to 3 weeks after birth. Distinguishing between biliary atresia and neonatal hepatitis is critical since early surgical intervention is mandatory in effecting a good outcome. The procedure known as the Kasai procedure is a roux-en-Y or hepatoportojejunostomy. The chances of success are seriously diminished after the infant is over 3 months old. Hence, every infant with jaundice in the neonatal period should be studied with Tc-99m hepatobiliary agents (HIDA/DISIDA) to ensure that the surgical treatment is instituted before it is too late.

Radiopharmaceuticals and Technique

The patient is placed in a frog-leg position with the scrotum supported on a sling made by using a towel straddled across the front of the patient’s thighs. A conventional flow study with dynamic images is obtained after the intravenous administration of 5 to 10 mCi of Tc-99m pertechnetate or Tc-99m DTPA. These are essentially “blood pool” agents. After the dynamic study is completed, blood pool static images are obtained in the anterior projection. Delayed static images are obtained, which will provide additional confirmation of the lack of vascularity to the affected testis in torsion. It is extremely important to have the technologist mark the side of symptomatology distinctly on the images submitted to the picture archiving and communication system (PACS). Personally, we prefer to use a converging collimator to obtain the images.

Interpretation

Testicular Torsion

In the early stage of torsion, the perfusion to the affected side may be normal. A “nubbin” sign is believed to be pathognomonic for torsion and represents the site of the actual torsion of the testicular vessels. This is believed to be due to visualization of the spermatic artery, which is being filled by the iliac artery (Fig. 12-22). In the later stages of the torsion, that is, 24 hours later, the relative avascular testis is thrown into high relief by the increased flow through the pudendal vessels to the scrotal sac. The now perfused scrotal sac with the absent vascularity in the testis provides a “Hello” sign.

Figure 12-22 Left testicular torsion shows increased peripheral flow and absence of central activity (“Hello” sign) consistent with testicular torsion. Minimal activity is noted in the right testicle, which was normal.