Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

9.1 Ectopic Gastric Mucosa in Meckel’s Diverticulum

Gastrointestinal bleeding in pediatric patients can be suspected for the presence of Meckel’s diverticulum (MD) that is caused by an incomplete closure of the omphalomesenteric duct and more frequently located in the ileum about 50–80 cm from the ileocecal valve. It occurs in 1–3 % of the population, more common in male patients. The ectopic gastric mucosa present in the diverticulum may result in mucosal damage and bleeding due to the production of acid and pepsin [1].

Scintigraphy with 99mTc-pertechnetate is an important and noninvasive test with a low radiation burden and it is easy to perform. In particular, it has a high accuracy for the detection of MD with ectopic gastric mucosa [2]. It is used to localize ectopic gastric mucosa in an MD [3, 4], due to the capacity of this tracer to avidly accumulate in gastric mucosa and then to reveal ectopic gastric mucosa in the diverticulum. The test represents still today the most accurate noninvasive technique for identifying ectopic gastric mucosa in Meckel’s diverticulum, with high specificity and positive predictive value (close to 100 %) both in children and in adults [5, 6]. False-negative results may be due to barium enema, upper GI examination, recent in vivo RBC labeling, or again anatomic causes of error, such as small amount of gastric mucosa in Meckel’s diverticulum, ischemia, or necrosis, or obscured by urinary tract activity (e.g., bladder) [3]. Moreover, false-positive result may be related to laxatives or endoscopy causing bowel irritation, urinary tract activity, lesions with increased blood pool, ulceration, inflammation, irritation, tumor, or intussusception.

Any type of patient preparation to performing the examination is not necessary, but it is important [3] to determine whether the patient has had recent in vivo RBC labeling in which all circulating RBCs were treated with stannous ion by intravenous administration of a “cold” pyrophosphate kit. If so, Meckel’s scan may be compromised, because intravenous 99mTc-pertechnetate will label RBCs rather than concentrate in ectopic gastric mucosa. This may occur for days after the administration of stannous pyrophosphate. This is not a problem with in vitro labeling procedure.

The procedure requires an intravenous injection of 3.7 MBq/kg of body weight of 99mTc-pertechnetate and subsequent serial images of the abdomen, acquired for at least 30 min in the anterior view. A camera with large field of view is preferred and a collimator with a low-energy, all-purpose, parallel hole (photopeak, typically 20 % window at 140 keV; computer, 128 × 128 matrix, single- or two-byte mode). The left lateral decubitus position during the acquisition time can decrease small bowel activity arising from the stomach, increasing the diagnostic accuracy.

The images may be summed to make multiple sets of 10–15 min each, to facilitate interpretation of the data. Additional static images at the end of the dynamic series, as SPECT acquisition, can highly aid the identification of Meckel’s diverticulum [7].

A positive scan demonstrates the appearance of focal activity in the ectopic gastric mucosa at the same time as in the normal gastric mucosa, although a small Meckel’s diverticulum may appear later than the stomach. The most frequently anatomic position where it may be displayed is the right lower quadrant. The anatomic sites that most often can be misinterpreted to be the diverticulum are the kidneys and the ureter or the bladder, in which there is an activity that, however, usually appears after that is seen in the normal gastric mucosa.

Some pharmacologic maneuvers have been reported to improve the detection of Meckel’s diverticulum, for example, the use of pentagastrin, histamine H2 blockers, or glucagon [3]. Pentagastrin is a potent stimulator of gastric secretions and increases gastric mucosa uptake of pertechnetate and also stimulates the secretion of pertechnetate and GI motility, potentially reducing ectopic site activity. Histamine H2 blockers (cimetidine, ranitidine) block secretion from the cells and increase gastric mucosa uptake. Glucagon relaxes the smooth muscles of the GI tract, decreasing peristalsis. However studies with pentagastrin, histamine-H2 blockers, or glucagon are rarely performed [8], because pharmacologic pretreatment is not considered necessary for obtaining a high-quality Meckel’s scan [3].

9.2 Inflammatory Bowel Diseases

Inflammatory bowel diseases (IBD) are diseases that often debut in late childhood or adolescence and include two different clinical entities causing chronic inflammation of the intestines, ulcerative colitis and Crohn’s disease, that are chronic but characterized by alternating periods of relapse and remission [9].

Due to the continuous exacerbations of the disease, it is crucial to have a reliable and easily reproducible diagnostic test. However the endoscopic methods are invasive and require sedation, involving some risks, and often can only evaluate a limited portion of the colon. In addition, often there is a poor correlation between clinical symptoms and endoscopic [10–12] or histological findings, and, finally [13–15], the endoscopic findings cannot predict the response to the treatment [10–12, 14].

Autologous leukocytes labeled with 99mTc-hexamethylpropyleneamine oxime (HMPAO) is used to characterize active IBD [16–19]. The pathophysiologic basis of its use is correlated to the recruitment of the circulating leukocytes at the site of inflammation through a multistep process: adhesion to microvascular endothelium, transmigration through the vessel wall, and further migration in extravascular tissue and into the bowel lumen [20].

The diagnostic accuracy of 99mTc-HMPAO leukocyte scintigraphy in IBD is high, with sensitivity and specificity around 90 % [21]. It may be stated that a negative leukocyte scan virtually excludes the presence of active disease and the sensitivity of the method in untreated patients is high even in the early stages of disease, when radiologic or endoscopic findings are often normal or equivocal.

The labeling of leukocytes with 99mTc-HMPAO with 20–45 mL of venous blood has been described [16].

Two to 3 h after injection, and no later, an 8 min static anterior supine view of the abdomen and pelvis is obtained, followed by SPECT imaging to provide better localization of disease distribution [22].

A large field of view gamma camera with a low-energy, high-resolution collimator is usually preferred. Early imaging of the pelvis and abdomen is essential (bowel activity is seen in 20–30 % of children by 1 h and 2–6 % of adults by 3–4 h postinjection). Regional images are obtained for at least 800,000 counts/large field of view of 5–10 min/view. Whole-body images should include the anterior and posterior head, chest, abdomen, pelvis, and extremities when clinically indicated. A limited study to evaluate a particular region of the body is acceptable in select cases. SPECT images of the chest, abdomen/pelvis, or spine may be helpful [23]. Accurate interpretation of labeled leukocyte scintigraphy requires knowledge of the normal and abnormal variants of leukocyte localization in the abdomen. Abnormal bowel localization may be seen by 15–30 min and usually increases in intensity over the next 2–3 h. The degree and extent of bowel disease are usually demonstrated by 1–2 h. Shifting patterns of bowel activity on later images usually indicate distal transit of labeled granulocytes or, at times, bleeding within the bowel lumen. False-positive results can occur from rapid small bowel transit of hepatobiliary secretion and focal accumulation of activity in the cecum, particularly if imaging is done after 1 h in children or 4 h in adults. Active gastrointestinal bleeding or swallowed cells can be mistaken for an inflammatory bowel process. Focal collections of inflamed peritoneal fluid or sites of focal bowel inflammation can be mistaken for abscess [23]. Volume-rendered images using the maximum-activity-projection or maximum-intensity-projection technique also are derived from the SPECT data in order to increase continuity of structures and facilitate comprehension of spatial relationships [24].

Although leukocyte imaging is useful, PET with 18F-FDG is becoming the new standard for nuclear medicine imaging in patients with IBD, based on the known potentiality of this tracer to localize the inflammatory lesions [9, 25, 26]. 18F-FDG has been proposed for imaging [31] infection/inflammation in part because it has been seen at sites of infection/inflammation during routine 18F-FDG imaging of cancer patients. Further studies showed that cells involved in infection and inflammation, especially neutrophils and the monocyte/macrophage family, are able to express high levels of glucose transporters, especially GLUT1 and GLUT3, and hexokinase activity [25, 32–35].

From limited experimental studies, it seems that the ability of the procedure to identify sites of inflammation and infection is related to the glycolytic activity of the cells involved in the inflammatory response. Many types of cells are involved in this process although no single cell was found specifically and consistently involved in all models. In addition, enhanced glucose consumption and subsequent 18F-FDG uptake can also be the result of a stress reaction of the affected cells in response to cell damage (metabolic flare) [36].

For this reasons, 18F-FDG PET and PET/CT have been proposed as noninvasive imaging methods to assess extent, location, and disease activity in adult and pediatric patients with IBD [9, 27, 28].

The diagnostic performance of 18F-FDG PET has proven to be excellent, with studies on pediatric patients where the average sensitivity was 98 % (higher than both endoscopy and abdominal ultrasound) and overall accuracy was comparable to the invasive procedure (83 % vs. 82 %) [25, 26]. Considering also that this technique avoids the long preparation times and the risks of radiolabeled autologous leukocytes, 18F-FDG PET may offer a definite clinical advantage for the patients [29].

Even if 18F-FDG PET/CT may not be able to replace conventional studies, this functional method may be useful when conventional studies cannot be performed or fail to be completed [30].

Patients must fast for at least 4 h before 18F-FDG imaging, during which time they should be encouraged to drink sufficient water to ensure hydration and promote diuresis. Necessary medications are allowed and must be recorded.

Because the effect of antibiotics on 18F-FDG uptake is unknown, it is useful to avoid such drugs, but no general recommendation on withdrawal can be stated. The patient should be advised to avoid strenuous physical exercise within 24 h before injection.

18F-FDG dose to obtain good imaging with a PET scanner operated in three-dimensional mode is optimized according to the EANM pediatric dosage card issued in 2008 [37].

With the current PET/CT scanners, the acquisition is performed in 3D whole-body mode, using steps of 1.5–3 min per bed position. Whole-body acquisition is usually defined as a field of view covering the head to mid thigh, starting in the pelvic area, when the bladder is empty.

In conclusion, 18F-FDG PET may be useful when endoscopic evaluation may not be feasible. In patients with an established diagnosis of IBD, 18F-FDG PET and PET/CT may provide information about disease activity, location, and extent within the intestinal tract, allowing early recognition of disease relapse and possible complications of the disease in association with clinical symptoms, physical exam, and laboratory data. 18F-FDG PET and PET/CT may guide decisions regarding the choice of therapy and may also allow the evaluation of efficacy of the medical therapy in IBD, because metabolic changes after the treatment (assessed by 18F-FDG PET) usually precede morphological changes assessed by conventional imaging methods (Figs. 9.1 and 9.2) [38, 39].

Fig. 9.1
figure 1

A 16-year-old boy with Crohn’s disease. The maximum-intensity-projection 18F-FDG-PET image shows a diffuse and intense radiopharmaceutical uptake in the large bowel

Full size image
Fig. 9.2
figure 2

The same patient in Fig. 9.1; 1 year later, posttreatment maximum-intensity-projection 18F-FDG-PET image shows a complete disappearance of any radiopharmaceutical uptake in the large bowel

Full size image

9.3 Appendicitis

Acute appendicitis is a surgical disease, particularly difficult to diagnosis in children. Delay in diagnosis is associated with morbidity from perforation, abscess, and peritonitis. It is therefore incumbent for the management of the pediatric patients with abdominal pain to correctly diagnose in order to treat patients appropriately.

In particular, in patients with abdominal pain and atypical or equivocal signs, symptoms, or laboratory tests, adjunctive imaging studies often are used to increase early diagnostic accuracy. Other than abdominal x-rays, ultrasound probably is the most commonly used adjunctive test for the patient with an atypical presentation. However, it has an accuracy of only 30 % in patients with early appendicitis because the appendix may not display the changes required for visualization [40].

99mTc-HMPAO-labeled leukocyte imaging is highly sensitive for detecting even small inflammatory processes in the abdomen because of high target-to-background ratio and early rapid uptake at sites of inflammation [41]. In fact, 99mTc-HMPAO is an agent that complexes avidly with polymorphonuclear leukocytes, and it has a rapid uptake into areas of acute inflammation [42, 43].

An anterior image of the pelvis has to be acquired at 30–60 min postinjection, and it needs to be repeated at approximately 60 min intervals until either the scan showed abnormal uptake, indicating a positive scan, or remained negative until 3 h, at which time scanning was terminated.

There are a variety of abdominopelvic inflammatory processes that may be detected with 99mTc-WBC imaging, for example, inflammatory bowel disease, abscess, and even bowel ischemia, and often differentiating between these conditions and appendicitis is not easy to obtain, and this results in a specificity of 85 % [16]. However, this issue can be obviated if the positive scans are interpreted in light of the overall clinical picture in order to consider 99mTc-HMPAO WBC imaging as an accurate, noninvasive test to exclude appendicitis in children with clinical suspicion of the disease but with anomalous presentation.

It is noted that inflammatory lesions show an avid uptake of 18F-FDG. The high resolution of PET, especially associated to CT or MR, together with the high concentration of 18F-FDG in inflammatory tissues, makes PET a potential useful tool for an earlier diagnosis of appendicitis and other abdominal inflammatory diseases. However, the use of 18F-FDG-PET in the detection of appendicitis is rare. More often, it is an incidental finding of high 18F-FDG uptake in the right iliac fossa due to appendicitis during exams performed for malignancies. Familiarity with the normal pattern and physiologic variations of 18F-FDG distribution and with clinical data relevant to the patient can direct to a correct diagnosis, reducing a misleading differential diagnosis with tumors [44, 45].

9.4 Gastroesophageal Reflux and Esophageal Transit

Radionuclide studies for evaluating gastrointestinal transit in adults have been adapted for use in infants and children for assessing esophageal transit, gastroesophageal reflux, and gastric emptying.

9.4.1 Gastroesophageal Reflux

Gastroesophageal reflux, which is a condition characterized by the reflux of gastric and duodenal contents across the gastroesophageal junction into the esophagus, can occur in infants typically from 2 months of age. Typically, children can have symptom resolution by 18 months of age, and 30 % of them may have symptoms until the age of 4 years. The presence of gastroesophageal reflux may be related to severe complications, as strictures and pneumonia [46].

To provide information about esophageal and gastroesophageal function, nuclear medicine technique may be relevant with esophageal and gastroesophageal scintigraphy.

99mTc-sulfur colloid is often used for the assessment of gastroesophageal reflux in children, and the tracer should be added to two thirds of the feeding volume typical for the patient, with a dose of 0.1–1 mCi (3.7–37.0 MBq), in order to leave one third unlabeled to clear any remaining activity from the oropharynx and esophagus [47–49].

Gastroesophageal scintigraphy is best performed at the time of a usual feeding. Then, short-lasting 5 s anterior dynamic images are acquired for 60 min, and at the end, anterior and posterior static images of the thorax can be acquired to look for evidence of aspiration.

The dynamic images should be reviewed to recognize any episodes of reflux and regions of interest may be placed on the entire esophagus and on the upper esophagus to aid in the analysis of data and in order to demonstrate the number of episodes of reflux and level reached within the esophagus, and the clearance rate of reflux episodes may then be determined.

9.4.2 Esophageal Transit

Radionuclide-esophageal transit studies have been performed in infants, usually as part of the evaluation for reflux or in the evaluation of patients with esophageal motility disorders such as achalasia, diffuse esophageal spasm, nutcracker esophagus, tracheoesophageal fistula, Down syndrome, esophagitis, systemic sclerosis, and diabetes mellitus.

Other diagnostic methods, as esophageal manometry, contrast radiography, and endoscopy, may be used to assess esophageal peristalsis, anatomical lesions, and mucosal lesions, but the advantages of esophageal scintigraphy include its noninvasive nature, quantifiability, and low radiation burden. Its clinical application has been suggested to be useful when the other procedures are unavailable or not tolerated by the patient or when the results are equivocal [49–55].

A 10 mL bolus of water or milk labeled with 150 μCi (5.55 MBq) of 99mTc-sulfur colloid is administered to the patient in the supine position. Posterior images at 0.4 s intervals for 150 frames are acquired, including the mouth and stomach, in the camera field of view. A radioactive bolus is placed in the mouth and swallowed on command followed by a dry swallow 30 s later. Abnormal studies may be repeated in the upright position to determine the effect of gravity.

Image analysis is performed by the evaluation of time activity curves derived from regions of interest placed on the upper, middle, and lower thirds of the esophagus and on the stomach. This procedure may demonstrate abnormal esophageal transit in pathologic states and may also visualize esophageal transit condensing dynamic images.

Usually, the normal transit time through the esophagus is typically less than 10 s. Esophageal transit ranges from 3.4 ± 1 s for infants to 4.6 ± 1.9 s for patients 8–16 years of age. Gastroesophageal reflux and esophagitis are associated with prolonged transit times [56].

9.4.3 Gastric Emptying

Patients with abnormal gastric emptying may present nausea, vomiting, abdominal discomfort, early satiety, diarrhea, and “dumping.”

The correct assessment of gastric emptying helps to guide treatment decisions, particularly in the neurologically impaired children that frequently demonstrate symptomatic delayed gastrointestinal motility.

Furthermore, delayed gastric emptying may be secondary to several pathological conditions, as pyloric or duodenal stenosis, acidosis, hypothyroidism, autonomic neuropathy associated with diabetes mellitus, central nervous system disease, systemic lupus erythematosus, dermatomyositis/polymyositis, infection, and many others.

The rate of gastric emptying assessed scintigraphically has been shown to depend on the type of meal used: liquids typically empty faster than solids.

In infants, a milk or formula feeding is usually administered and the type of milk and volume fed should be standardized according to patient size. However, many of the patients studied have feeding difficulties and in these cases is difficult to standardize the volume fed.

The dose of 99mTc-sulfur colloid added to the milk depends on whether gastric emptying is performed in conjunction with the reflux study.

If only the rate of emptying is of interest, this dose can be decreased to 100 μCi (3.7 MBq).

In older children, solid gastric emptying may be performed by having the patient eat an egg sandwich containing 99mTc-sulfur colloid (250–300 μCi). The meal should be scaled according to patient size (where adults are given 4 eggs and 50 mL of water). Thirty second anterior images are acquired every 10 min. Between images, the patient should sit upright. Images are acquired until 120 min.

A region of interest is drawn around the stomach. Activity from the bowel should not be included in the region of interest.

The range for normal gastric emptying has been difficult to establish in children for ethical reasons. Furthermore, the test meals have not been standardized. In general, laboratories have to decide on the values to be used based on their own experience. A study performed several years ago in children thought to be normal retrospectively shown, for milk, a residual of 36–68 % at 60 min and 42–56 % in a small number of older children [57].

In another report using dextrose as the test meal, the 60 min residual was 27–81 % in children under 2 years of age and 11–47 % in older children. This age-related difference in emptying rate has been observed by others, although the composition of the meal may also play a role [58].

9.5 Hepatobiliary Scintigraphy

Hepatobiliary scintigraphy is useful in the assessment of patients with right upper quadrant pain, in particular if there is the suspicion of acute cholecystitis. Cholecystitis is less common in children than in adults, but its incidence in pediatric patients has probably been underestimated [59, 60].

It is possible to recognize two types of cholecystitis: acute calculous cholecystitis, which occurs when the cystic duct becomes obstructed by gallstones leading to gallbladder distension and edema, and acalculous cholecystitis, which can occur in prolonged illness, sepsis, or trauma.

Both of these conditions may be associated with inflammation and edema and consequently with blood flow impairment and bacterial infection advancement.

Typically, the most common symptoms in children are abdominal pain localized in the right upper quadrant or epigastric region. Other symptoms include nausea, vomiting, anorexia, and fever for several days.

The most common clinical signs are represented by a right upper quadrant tenderness, jaundice, and enlarged and palpable gallbladder. Regarding the laboratory tests, leukocytosis is a common finding [61–63]. In the diagnosis of cholecystitis, ultrasound is basic in the evaluation of the biliary tract and detecting gallstones; however, its diagnostic accuracy presents a substantial margin of error. In a recent systematic review of different imaging modalities in patients with suspected acute cholecystitis [64], hepatobiliary scintigraphy is the most accurate diagnostic imaging modality for acute cholecystitis, with a sensitivity and specificity for hepatobiliary scintigraphy of 96 % and 90 %, respectively. The sensitivity of hepatobiliary scintigraphy results significantly higher than ultrasound (81 %).

99mTc-disofenin or 99mTc-mebrofenin is administered intravenously, with a dose for infants and children of 1.85 MBq/kg (0.05 mCi/kg). Mebrofenin is always preferred in jaundiced infants with hyperbilirubinemia, with a minimum administered activity of 37 MBq (1.0 mCi), as up to 24 h delayed images are often required. The patient should be fasting for 4 h prior to the test. Immediately after the injection, dynamic imaging is acquired for 60 min (0.5–1 min/frame) in anterior projects using preferably a high-resolution collimator and 128 × 128 matrix. Additional views such as right lateral and left or right anterior oblique may be performed, if required. When acute cholecystitis is suspected and the gallbladder is not seen within 60 min, delayed images for up to 4 h should be obtained [65].

In adult patients, acute cholecystitis is associated with non-visualization of the gallbladder on hepatobiliary scintigraphy, and the visualization of gallbladder activity excludes the diagnosis of acute cholecystitis with high accuracy. However, in children, the presence of cholecystitis is not entirely excluded if there is gallbladder visualization, because this is possible in acalculous cholecystitis [46]. After 60 min images, if the gallbladder is visualized, it is possible to perform additional dynamic imaging for 60 min following infusion of 0.02 μg/kg sincalide, a synthetic C-terminal octapeptide of cholecystokinin. Poor contraction and emptying of the gallbladder following sincalide may occur in partial cystic duct obstruction, acalculous cholecystitis, or chronic cholecystitis.

Hepatobiliary scintigraphy in children can also be used for the evaluation of choledochal cyst and biliary leak. Biliary atresia is characterized by obliteration or discontinuity of the extrahepatic biliary system, resulting in obstruction of bile flow [66]. The disorder represents the most common surgically treatable cause of cholestasis encountered during the newborn period. If not surgically corrected, secondary biliary cirrhosis invariably results. In infants, hepatobiliary scintigraphy is used to differentiate biliary atresia from hepatocellular disease [67]. The disorder, if not surgically corrected, results in secondary biliary cirrhosis. Thus early diagnosis of biliary atresia remains imperative, and the gold standard within the diagnostic modalities is liver biopsy and/or intraoperative cholangiogram. Studies have demonstrated 100 % sensitivity of 99mTc-mebrofenin hepatobiliary scintigraphy (HIDA scan) for picking up biliary atresia, and its specificity has been reported to range from 88.6 to 92 % [68, 69].

The premedication with phenobarbital or ursodeoxycholic acid before hepatobiliary scintigraphy may increase bile secretion and improves the diagnostic differentiation between biliary atresia and neonatal hepatitis [65]. The administered activity for infants and children is 1.8 MBq/kg (0.05 mCi/kg), with a minimum administered activity of 18.5 MBq (0.5 mCi). Mebrofenin is always preferred in neonates with hyperbilirubinemia, with a minimum administered activity of 37 MBq (1.0 mCi), as up to 24 h delayed images are often necessary [67].

For the image acquisition [67], a large-field-of-view γ-camera equipped with a low-energy all-purpose or high-resolution collimator is recommended. Whenever possible, continuous (dynamic) computer acquisition (usually in the anterior or left anterior oblique view) should be performed (1 frame/min). The image matrix of 128 by 128 is optimal on a standard large-field-of-view camera. In pediatric patients, an appropriate electronic acquisition zoom should be used. Initial images are usually acquired dynamically, starting at injection and continuing for 60 min. When visualization of the gallbladder is the end point of the study, it can be stopped earlier when activity is seen in the gallbladder. Additional views (e.g., right lateral, left or right anterior oblique) may be obtained as needed to clarify anatomy. To resolve concern about common bile duct obstruction (highly unlikely in the presence of gallbladder visualization), demonstration of tracer activity in the small bowel may need to be pursued.

The digital data can be reformatted to 4 to 6 min images for filming or digital display. Cinematic display of the data may reveal additional information not readily apparent on reformatted display. Image intensity scaling should be study relative rather than individual frame relative. The former allows for appreciation of activity changes over the duration of the study.

If there is visualization of the biliary drainage of the radiotracer into the bowel, scintigraphy can exclude the presence of biliary atresia. If no bowel activity is detected up to 24 h and liver uptake is normal, the diagnosis of biliary atresia is suspected [46, 67]. If the patient is being studied for a biliary leak, 2 to 4 h delayed imaging (or longer delays in some cases) and patient-positioning maneuvers (e.g., decubitus views) may be helpful. Any drainage bags should be included in the field of view if the biliary origin of a leak or fistula is in question. In patients with a suspected leak, it may be helpful to acquire simultaneous right lateral or other views on a multihead camera.

However, normal liver uptake of tracer with no excretion up to 24 h can occur in severe cases of neonatal hepatitis, Alagille syndrome, dehydration, sepsis, TPN cholestasis, and bile plug syndrome in cystic fibrosis. Neonatal hepatitis typically demonstrates reduced hepatocyte uptake and delayed hepatobiliary transit of tracer into the bowel [46]. Failure of tracer to enter the gut is consistent with biliary atresia but can also be caused by hepatocellular disease or immature intrahepatic transport mechanisms. Renal or urinary excretion of the tracer (especially in a diaper) may be confused with bowel activity and is a potential source of erroneous interpretation [67].

An interesting application of 18F-FDG PET in this setting is its use in children with biliary cirrhosis and fever of unknown origin (FUO) on the waiting list for liver transplantation. Infection imaging with FDG-PET relies on the fact that granulocytes and mononuclear cells use glucose as an energy source specifically during their metabolic burst upon activation by triggers. The advantages of FDG-PET are early imaging after injection, higher resolution and higher target-to-background ratio, sensitivity to chronic low-grade infections, and high interobserver agreement [70].

Systemic infections are considered to be contraindications for liver transplantation; however, hepatic infections are frequently only cured by the removal of the infected organ during transplantation. Therefore the information obtained by FDG-PET imaging to identify intrahepatic infections may be crucial for the management of patients with FUO awaiting liver transplantation [71].

9.6 Hyperinsulinism

Congenital hyperinsulinism (CHI) is a rare disease of hypoglycemia due to dysregulated and excessive insulin secretion, with an incidence of severe cases estimated at 1:50,000 and the incidence rising in consanguineous populations [72, 73].

In infants, the treatment of congenital hyperinsulinism is required to prevent possible neurologic complications. Forty percent of cases of hyperinsulinism in children results from pathologic adenomatous pancreatic β-cells. Diffuse hyperinsulinism involves the whole pancreas with enlarged abnormal β-cell nuclei. Ten percent of cases of infantile hyperinsulinism are atypical and fit neither classification. Focal hyperinsulinism is cured by resection of the adenoma, whereas diffuse pancreatic involvement may be treated with subtotal pancreatectomy. [74, 75].

Positron emission tomography (PET) using 6-L-(18)F-fluorodihydroxyphenylalanine ((18)F-DOPA) may be useful for classifying pancreatic involvement in infantile hyperinsulinism as focal or diffuse and can be used preoperatively to identify the two forms of infantile hyperinsulinism, differentiating between patients who should receive curative focal pancreatic resection and those who should receive medical management [76, 77]. (18)F-DOPA is a radioactive isotope of L-dihydroxyphenylalanine (L-DOPA), an intermediate in the catecholamine synthesis pathway. L-DOPA is avidly taken up by neuroendocrine cells, including CHIb-cells, while uptake is minimal in the normal pancreas. It is not understood why CHI pancreatic tissues exhibit more a differential uptake and retention pattern than normal tissue, particularly as CHI is not apparently linked with altered catecholamine metabolism.

Initial observations showed a high sensitivity (88–94 %) and specificity (100 %) of (18)F-DOPA PET imaging in differentiating focal from diffuse CHI [78, 79] which was further enhanced by the concurrent use of CT angiography using iodine-based dyes. The latter is useful to localize the site of the lesion, in relation to anatomical structures, which is particularly useful at the time of focal lesionectomy. Although meta-analyses of (18)F-DOPA PET imaging studies indicate good diagnostic performance [80, 81], recent studies have suggested that the predictive value of scanning may not be as accurate as initially estimated [82, 83].

9.7 Protein-Losing Enteropathy

Protein-losing enteropathy can be the result of primary intestinal lymphangiectasia or secondary intestinal lymphangiectasia in association with cardiac diseases or obstructed lymphatics. Abnormal or inflamed mucosal surface secondary to intestinal inflammation/infection and immunologic, inflammatory, and vasculitic disorders can also cause enteric protein loss [84].

This pathophysiologic condition generally results from an abnormal mucosal permeability, desquamation, inflammation, or back pressure in the intestinal lymphatic network [85].

The protein loss is nonselective and includes plasma proteins such as albumin, globulins, and transferrin. Radiolabeled proteins that have been used for determining enteric protein loss include 131I-albumin, 51Cr-albumin, and 67Cu-ceruloplasmin [84].

99mTc-human serum albumin (HSA) has been successful in localizing the site of enteric protein loss in adults [86–92]; however, the literature is lacking concerning pediatric patients [93–95].

A study performed on children [96] reported that the scan has a higher sensitivity in patients with lower albumin and total protein values, presumably reflecting a higher rate of protein loss; they submitted all patients to an anterior abdominal scintigraphy after the intravenous injection of freshly prepared 99mTc-HAS. The administered age-adjusted doses were 185–503 MBq based on an adult dose of 740 MBq. They acquired dynamic images every minute for 1 h, using an all-purpose collimator and a large-field-of-view γ-camera. Additional delayed images were obtained at 2–6 h in most patients and 24 h in a few patients.

They found enteric 99mTc-HSA uptake in 67 % of the children, suggesting the site of protein loss. The location of the uptake was most likely in the small bowel in 91 % of the early images and in 27 % of the delayed images. Colonic activity was noted in 72 % of the delayed images, most likely representing transit of activity rather than a second site of protein loss.

9.8 Colonic Transit

Two different types of chronic functional constipation have been identified in children based on colonic transit time measurement: a more generalized and severe form known as slow transit constipation and a segmental type known as functional fecal retention [97].

Both entities present with similar symptomatology but involve different pathophysiological mechanisms and require different treatment strategies.

It has been shown that children thus classified respond to different treatment strategies. Only a small proportion of cases require surgical intervention such as appendicostomy, colostomy, or colonic resection [99–101]. The different types of abnormal colonic transit can be identified using radiopaque markers: slow transit (pancolonic or globalized delay), normal transit, and functional fecal retention (outlet obstruction or distal obstruction) [98]. Colonic transit scintigraphy can aid in the identification and therapeutic decision-making in patients with functional fecal retention, the most common cause of chronic constipation in children [102].

This method has been used to determine colonic transit in adults with chronic constipation, and the reported advantages, compared with radiopaque marker studies, include a low radiation dose and the acquisition of multiple images allowing estimates of gastric, small bowel, and segmental colonic transit to be made [103, 104].

Intake of laxatives has to be stopped 5 days before the transit studies, and fasting is required for 4 h before the start of the test. The radiopharmaceutical 99mTc-calcium phytate colloid, suspended in 20 mL of milk, can be administered by mouth. The dose is determined according to each patient’s weight and is based on an adult dose of 250 MBq. Anterior and posterior view images are obtained immediately after ingestion and during the subsequent 2 h to estimate gastric emptying. Three categories of colonic transit could be readily distinguished by visual assessment of the acquired images. In studies considered to demonstrate normal transit, the tracer reached the cecum by 6 h, passed through the colon, and was largely excreted by 48 h. Slow transit was identified when the tracer reached the cecum at 6 h, but most radioactivity was retained in the proximal colon and transverse colon at 24, 30, and 48 h. Patients in whom the tracer reached the rectosigmoid by 24–30 h but was not passed at 48 h were appreciated. This pattern was defined as consistent with functional fecal retention or outlet obstruction.

9.9 Gastrointestinal Bleeding

Gastrointestinal (GI) bleeding is often encountered in daily clinical settings. Dramatic advances in endoscopic technology in recent years have facilitated diagnosis and treatment of bleeding from the esophagus, stomach, and duodenum, as well as most cases of bleeding from the large intestine. Although it is now possible to observe the small intestine using video capsule endoscopy and double-balloon enteroscopy [105], diagnosing the source of GI bleeding and providing treatment remain the challenges in some patients [106].

GI bleeding scintigraphy is a noninvasive examination that enables detection with a bleeding rate as low as 0.1 mL/min [107]. In children, while most gastrointestinal bleeding may not be life threatening, it is necessary to determine the source, degree, and possible cause of the bleeding and to distinguish minor from major bleeding [108]. In the stable child with occult bleeding, management is geared toward making the diagnosis and excluding more serious conditions. Investigations are directed by the history and physical examination, but most diagnostic tests can be performed on an outpatient basis [109].

99mTc-red blood cell (RBC) scintigraphy generally is useful for assessing GI bleeding in patients. Meckel’s diverticula also may be identified through this technique, although other causes of bleeding may include intussusception, IBD, Henoch-Schonlein purpura, gastritis, duodenitis, Mallory-Weiss tear, infectious enterocolitis, allergic enterocolitis, midgut volvulus, polyps, tumors, vascular malformations, enteric duplication cysts, nodular lymphoid hyperplasia, hemolytic uremic syndrome, and foreign body and trauma [110].

The main advantage of the use of 99mTc-RBCs is the possibility of visualization of GI bleeding over the course of several hours. Bleeding rates as low as 0.1–0.4 mL/min may be detected. Large bowel endoscopy of actively bleeding patients has a low diagnostic yield and is potentially harmful to the patient, and the small bowel is not successfully visualized endoscopically. Angiography typically localizes bleeding when the rate is greater than 1 mL/min. However, for bleeding to be identified, it should occur during the 20 to 30 s time interval during which contrast is administered. Scintigraphy permits the visualization of the entire GI tract.

Labeling of RBCs is most efficient by the in vitro method (98 %) as compared with the in vivo (70 %) and modified in vivo (90–95 %) methods. The disadvantage of lower labeling efficiencies is the possibility of secretion of free pertechnetate within the gastric mucosa into the duodenum and also the excretion of free pertechnetate into the urinary collecting system increasing the likelihood of false-positive studies.

Positive studies demonstrate tracer activity outside of normal vascular structures with antegrade or retrograde motion of tracer through bowel. The motion is best detected on cinematic display and may occur very rapidly. Small bowel bleeding may be distinguished from a colonic source by the demonstration of rapid distal progression through a series of multiple small, centrally located, curvilinear segments on cinematic display of the abdomen. Large bowel bleeding has a more elongated pattern with peripheral location within the abdomen compared with bleeding within small bowel. Stationary activity is more likely to represent a vascular abnormality and urinary or penile activity. In rare cases, a stationary site may represent adherent blood clot to the bowel wall.

99mTc-sulfur colloid also has been used in the assessment of GI bleeding. It has the advantage of detecting rates as low as 0.05–0.1 mL/min through the achievement of a high bleeding to background ratio. Its disadvantages include a short duration of imaging time to localize bleeding of approximately 20 min and limited interpretation of potential bleeding sites in proximity to the liver and spleen as these structures accumulate 99mTc-sulfur colloid during the test.

Both 99mTc-RBC scintigraphy and 99mTc-sulfur colloid scintigraphy detect sources of venous and arterial GI bleeding, whereas contrast angiography only detects arterial sources [111–114].

9.10 Hepatoblastoma

Hepatoblastoma (HB) is the most common primary malignant hepatic tumor in childhood [115, 116]. According to the histology, HB can be in pure epithelial and mixed type, the latter being formed by epithelial and mesenchymal components. The clinical onset is characterized by an abdominal mass, while in the advanced disease, anorexia and weight loss can also be present. Ninety percent of cases present increased level of serum alpha-fetoprotein (α-FP).

At present, complete resection is possible in more than 50 % of cases, and preoperative chemotherapy has been successfully used in converting unresectable to resectable tumors [117–119]. Liver transplantation has been proposed as an option in patients with unresectable tumors.

The follow-up is based on the association between laboratory analysis and imaging.

In particular, although a rising level of serum α-FP is generally associated with tumor recurrence, most protocols require additional imaging follow-up in order to localize neoplastic lesions, as US, CT, or RM [120–122]. However, early detection of recurrent hepatoblastoma is not always possible with conventional imaging methods such as CT and MRI.

A further dimension of information based on the regional biochemical and physiological abnormalities can be provided by positron emission tomography (PET) using F18-fluorodeoxyglucose (FDG), which has been successfully used for the last two decades in localizing primary and metastatic tumors in adults [123–125]. It is known that 18F-FDG uptake in tumors is proportional to the metabolic rate of viable tumor cells, which have an increased demand for glucose than normal tissue. It has already been proven that hepatoblastoma cells have demonstrated prominent glycogen granules in the cytoplasm [126, 127], and this fact may suggest active accumulation of glucose and its transformation and accumulation in glycogen granules [128] and consequently can explain the uptake of 18F-FDG.

The whole-body PET/CT scan has to be performed as a standard examination, following the guidelines for the administered dose and for the acquisition parameters of the PET imaging in pediatrics.

18F-FDG PET/CT could provide incremental diagnostic value in the initial evaluation of patients affected by hepatoblastoma, helping to detect additional metastatic sites at diagnosis.

However, 18F-FDG PET/CT has no established role in the initial diagnosis of hepatoblastoma [129], but it is helpful in detecting early recurrence [130], and few studies have evaluated its role in follow-up and restaging of patients after chemotherapy and surgery [131–135].

Since experience is so far limited in the literature [136], multicenter and prospective studies are warranted to suggest the introduction of 18F-FDG-PET/CT in the routine imaging workup for hepatoblastoma staging and in case of suspicion of relapse (Fig. 9.3).

Fig. 9.3
figure 3

A 4-year-old boy affected by hepatoblastoma. PET evaluation was performed to establish liver transplantation eligibility. Axial PET/CT fusion images show 18F-FDG-avid lesions in the liver

Full size image