Radionuclide Bone Scintigraphy in Pediatric Orthopedics

Radionuclide Bone Scintigraphy in Pediatric Orthopedics

Common Orthopedic Problems 0031-3955/86 $0.00 + .20 Radionuclide Bone Scintigraphy in Pediatric Orthopedics James J. Conway, M.D. * Scintigraphic...

28MB Sizes 0 Downloads 38 Views

Common Orthopedic Problems

0031-3955/86 $0.00

+ .20

Radionuclide Bone Scintigraphy in Pediatric Orthopedics

James J. Conway, M.D. *

Scintigraphic imaging has improved considerably during the last decade. This has been brought about by the introduction of new radiopharmaceuticals as well as better resolution gamma cameras. Improved resolution allows recognition of greater detail in the normal as well as abnormal states in the child. Improved results have also been obtained because of refinements in technique designed for pediatric patients. 9 In particular, the use of magnification for small bony parts depicts pathologic distributions of radionuclides to better advantage. 48 Scintigraphic imaging is now able to recognize subtle alterations in perfusion or metabolism of bone and often the study is designed to answer specific questions as they relate to symptomatology or other imaging findings. The purpose of this article is to emphasize those aspects of pediatric bone scintigraphy that have evolved since 197716 and that offer an advantage to the pediatrician in the diagnosis and management of orthopedic disorders of childhood.

TERMINOLOGY

Scanning Versus Scintigraphy. Radionuclide bone imaging was originally termed "bone scanning" because the images were generated using rectilinear scanning devices. The term "scan" has subsequently been adopted for the technique of computed tomography and ultrasound and thus confusion occasionally arises in ordering studies. Rectilinear scanning is a rarity today, with all nuclear medicine images being generated by gamma cameras. An appropriate term for this method is scintigraphy, with the image derived being termed a scintigram. Three-phase Bone Scintigraphy. The gamma camera allows dynamic acquisition of angiographic and early as well as delayed static images *Chief, Division of Nuclear Medicine, The Children·s Memorial Hospital; Professor of Radiology, Northwestern University Medical School, Chicago, Illinois

Pediatric Clinics of North America-Vol. 33, No.6, December 1986

1313

1314

JAMES J. CONWAY

following the injection of a radiopharmaceutical. The combination of dynamic as well as static images during a study has been termed three-phase bone scintigraphy and is a technique in common usage today. The first, or angiographic, phase records the vascular transit of radionuclide through the primary arterial vessels and the soft tissues within the first few seconds following injection. Each image is only of a few seconds' duration. They provide a qualitative temporal impression of regional vascularity but are of insufficient resolution to perceive individual bone vascularity (Fig. 1A). The second phase generates high-count static images immediately following the angiographic portion of the examination. This has been called the "blood pool phase" but more correctly reflects an "extracellular phase" because the radioisotope moves from the blood vessels into the extracellular spaces of the soH: tissue and bone. Such an image is usually derived within

R

B

Figure 1. A. Radionuclide angiographic images of the lower extremities at 5-second intervals. Arrow points to focal region of hyperemia in the distal metaphysis of the right femur. B, "Extracellular phase" image immediately after completion of the angiographic phase. Physes become defined and arrow points to increased localization of the radionuc!ide in the distal metaphysis of the right femur. C, Static delayed images at 1.5-3 hours after injection of the radionuclide. All activity appears in the bone. The physes are sharp and well defined. The arrow points to a focus of osteomyelitis in the distal femur.

RADIONUCLIDE BONE SCINTIGHAPHY IN PEDIATRIC OHTHOPEDICS

1315

the first minute following injection. The higher count images better resolve individual bone structures such as the highly vascular and metabolically active physes at the ends of growing bones in children (Fig. IB). The third phase is a static high-count image usually obtained between 1 V2 and 3 hours after injection of the radiopharmaceutical. These images are of high count (0.5 to 1 million photons) and good resolution. The distribution of the radionuclide within the bone correctly reflects the regional bone perfusion and metabolism with a minimum of soft-tissue background (Fig. Ie). Occasionally, a very delayed image up to 24 hours is obtained and has been referred to as the fourth phase of bone scintigraphy.2 Such delayed imaging serves to reduce soft-tissue background activity to a minimum. Multiple-phase bone scintigraphy is hailed by proponents as useful in differentiating soft tissue from bone abnormalities. I, 31,40, 41, 55 Others have pointed out the nonspecificity of three-phase scintigraphy and have questioned its value. 56, 58

LOCALIZATION MECHANISMS Radionuclide localization within bone is primarily dependent upon two factors. The first is vascular perfusion and the second is bone metabolism. The extraction of radionuclide from the blood into the bony matrix depends on the concentration as well as the total amount of radiopharmaceutical agent available to the bone. Thus, perfusion is a principal factor for increased localization in hyperemic states, such as those due to osteomyelitis, and for decreased localization in ischemic states, such as in Legg-Calve-Perthes disease. Metabolism as a factor is principally reflected by altered bone turnover such as within neoplasms or in the increased metabolic activity from fracture healing. An important concept is that there are no false-positive or falsenegative bone scintigrams. There are only false-positive or false-negative interpretations. Bone scintigraphy accurately portrays even subtle alterations in perfusion and metabolism. Therefore, a basic understanding of the pathophysiology in pediatric bone disorders is essential. The vascular and metabolic information provided by bone scintigraphy complements the anatomic information flOm roentgenographic studies and allows a more accurate diagnosis of pathology. MUSCULOSKELETAL INFLAMMATION The simplest, most sensitive and specific imaging technique currently available to the pediatrician for differentiating septic arthritis, cellulitis, and osteomyelitis is bone scintigraphy. In pediatric patients, accuracies of 92 per cent have been reported. 34 This is in spite of the fact that roentgenographic examination is frequently entirely normal. On occasion, the bone scintigram may be normal in osteomyelitis. 7, 26 This can occur in early lesions when there is little alteration of perfusion or metabolic change of

1316

JAMES J. CONWAY

the bone due to the infection. Repeat scintigraphy after one or two days may define the infection, particularly if the patient has not received antibiotics (Fig. 2A and B). Needle aspiration does not affect bone scintigraphy. In more emergent circumstances, an immediate injection of an infection-specific radiopharmaceutical such as gallium-67 citrate or indium111 tagged white blood cells will depict the true nature of the disorder within 6 to 8 hours. 39, 57 The normal bone scintigram in osteomyelitis should not be equated with the photopenic (cold) lesion of ischemic osteomyelitis. 46 Such a finding is ominous because antibiotic as well as the radiopharmaceutical cannot enter the site of infection due to the ischemic nature of the lesion. Surgical drainage should be performed on an emergent basis to minimize bone damage and deformity. Ischemia occurs because of a compromised vascular supply, for example, in the hips, or due to necrosis and compression of the vasculature by soft-tissue edema. Neonatal Osteomyelitis Earlier studies did not advocate the use of bone scintigraphy for osteomyelitis in the neonate. 4, 44 Current studies routinely utilize magnification techniques and current generation gamma cameras have better resolution. Our recent study depicted all suspected sites of infection in a large series of neonates with osteomyelitis. 9 A common clinical presentation is the lack of movement of an extremity. Fever or elevated white blood cell count are often absent. The importance of bone scintigraphy in the neonate suspected of osteomyelitis is that the recognition of multiple sites can be made in spite of little symptomatology or radiographic findings. It

R ·10-22 A Figure 2. A, A lO-year-old female patient with cellulitis of the forearm and Staphylococcus au reus positive culture. Normal sedimentation rate. Temperature 40°C. Slight tenderness to palpation of right ankle. Ankles appeared normal clinically and on roentgenogram. Bone scintigram is normal. B, Abnormal bone scintigram in the distal metaphysis of the right tibia (arrow) consistent with osteomyelitis 2 days following the original normal bone scintigram. Ankle much more tender to palpation.

RADIONUCLIDE BONE SCINTICRAPHY IN PEDIATRIC ORTHOPEDICS

1317

is important to monitor all sites of involvement in order that appropriate therapy can be instituted to prevent deformity. Septic Arthritis of the Hip Specific attention is directed towards the recognition of ischemic osteomyelitis in children with septic arthritis of the hip. Avascular necrosis of the proximal femoral epiphysis is a well-recognized complication of septic hip joint. Early recognition and appropriate treatment of bone involvement can prevent crippling deformity. Our experience indicates that untreated septic hip of greater than 5 days' duration has a high incidence of developing avascular necrosis. Any lateral displacement of the femoral head in the acetabulum on roentgenogram should alert the clinician to the possibility of a septic hip joint. Needle aspiration is essential for diagnosis and should be performed on an emergent basis. Bone scintigraphy may demonstrate a "cool or cold" femoral epiphysis as evidence of the sepsis which has compromised the blood supply to the epiphysis (Fig. 3A and B). Flat Bone Osteomyelitis Osteomyelitis of flat bones is very difficult to recognize radiographically. Characteristically, lesions of flat bones go unrecognized for weeks to months and often present clinically with intermittent low-grade fever or a minimally elevated white blood cell count. In osteomyelitis, erythrocyte sedimentation rates are invariably elevated after 3 days. Bone scintigraphy not only recognizes the locus of involvement but indicates specific placement for needle aspiration of the most likely site to obtain a positive culture. 22 Computed tomography assists in localizing complex anatomic sites for biopsy or drainage. 32 Sickle Cell Crisis Versus Osteomyelitis The child with sickle cell crisis often presents with focal pain, swelling, tenderness, warmth, and erythema, which is identical to the clinical presentation of infection. Bone scintigraphy often depicts abnormal increased localization even in the earliest stages of sickle cell crisis and, therefore, cannot be used to differentiate sickle cell crisis from osteomyelitis; 67Ga citrate likewise can localize abnormally in sickle cell crisis. However, the sequential combination of bone sCintigraphy followed by gallium-67 citrate imaging and their specific patterns is reported to be efficacious in differentiating osteomyelitis from sickle cell crisis. 3 It may be a moot point, however, as it is estimated that sickle cell crisis is 50 times more common than osteomyelitis in those patients with sickle cell disease who present in such a fashion. 36 An important differentiation is the clinical response to conservative therapy. Imaging studies are relegated to those patients with persistent symptoms of several days' duration. Bone scintigraphy is helpful if a photon-deficient (cold) region is found, which is more probably due to sickle cell disease than infection and is also useful in localizing multiple or unsuspected sites for needle aspiration and culture. An abnormal bone scintigram requires follow-up (j7Ga scintigraphy to differentiate crisis from osteomyelitis (Fig. 4A and B).

1318

JAMES J. CONWAY

Figure 3. A, Lateral displacement of the hip out of the acetabulum, as seen on the right side of this pelvis roentgenogram, associated with fever and limited use of the leg, strongly suggests the diagnosis of septic hip. Immediate needle aspiration is recommended to make the diagnosis and obtain a specimen for culture. B, Follow-up bone scintigraphy depicts absent localization of radionuclide in the proximal femoral epiphysis on the right and normal localization on the left (arrows). Avascular necrosis is a complication of septic arthritis of the hip.

RADIONUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1319

Figure 4. A, An 18-year-old patient presented with sickle cell disease, low grade fever and shoulder pain of four days' duration. There is increased localization in the proximal epiphysis and physis of the right humerus (arrow) on bone scintigraphy. B, Normal gallium67 citrate scintigraphy 24 hours later excludes the diagnosis of osteomyelitis.

Multifocal Osteomyelitis

The condition commonly called "chronic nonsuppurative multifocal osteomyelitis," and which has been given several other names,8. 30. 51, 60 is appearing with increasing frequency in the pediatric population, These children typically present with focal swelling, erythema, and mild bone pain. Occasionally, a pustular eruption of the palms and soles occurs. 6 There is often roentgenographic evidence of bone destruction associated with periosteal bone reaction suggesting chronicity. Almost any bone but frequently the clavical with other multiple sites is involved. Typically, bone scintigraphy demonstrates variable increased localization of radionuclide at multiple sites. Bone biopsy which includes numerous plasma cells suggests a reactive inflammatory process of a chronic nature. Cultures are negative for virus as well as bacteria. Therapy with antibiotics is of questionable help and the lesions gradually subside only to recur at later intervals or at different sites. 67Ga citrate imaging depicts normal as well as focal increases

1320

JAMES J. CONWAY

at the sites recognized on bone scintigraphy. This finding suggests an infectious process of a self-limited nature but the organism or agent remains a mystery. The findings require differentiation from neoplastic processes such as lymphoma of bone. Diskitis of Childhood Low back pain, manifested by refusal to walk, with low-grade fever in an irritable toddler or young child should alert one to the possibility of diskitis of childhood. 52, 59 Bone scintigraphy demonstrates focal increased localization of radio nuclide in the disk space as well as in the vertebral plates above and below the disk (Fig. 5A and B).27 Roentgenographic

Figure 5. A, A 2-year-old male patient presented with a history of "stiff gait" for 2 weeks. Initial roentgenograms of the s'pine were interpreted as being normal. Bone scintigraphy was recommended if symptoms persisted. There is prominen t localization of radionuclide in the vertebral bodies and the disk space of T7- T8 consistent with diskitis. B,' Follow-up roentgenogram of the spine demonstrated progressive narrowing of the disk space between T7-T8 (arrow).

RADIO NUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1321

findings initially are normal but eventually show narrowing of the joint space and irregular margins of the vertebral plates. 67Ga citrate has been reported abnormal prior to bone scintigraphy.1O Needle aspiration of the disk and the vertebral bodies fails to culture bacteria or other infectious agents in the majority of the patients studied. Treatment is symptomatic with bed rest. Antibiotics are used only if there is a poor response to conservative treatment.

AVASCULAR NECROSIS Legg-Calve-Perthes Disease Legg-Calve-Perthes disease (LCPD) is a well-known disorder; however, the factors that determine its onset and the mechanisms for revascularization and healing are poorly understood. Symptomatology lags significantly behind the onset of the disorder. Initial radiographs usually depict LCPD in its late stages. Prognostic indicators based on radiographic changes therefore are of limited value since the factors that produced the radiographic alterations have already occurred. Treatment methods have varied and conservative as well as surgical management has produced similar outcomes. Radionuclide bone scintigraphy is ideal for the study of avascular necrosis of bone because the mechanisms for localization of the radiopharmaceutical are primarily bone perfusion and metabolism. Even mild ischemia of bone significantly decreases localization of the radionuclide and there are several reports testifying to the value of bone scintigraphy in the early diagnosis of LCPD. 5, 11, 12, 13, 20, 24 Earlier attempts to use radionuclide scintigraphy for staging LCPD have created confusion relating to the extent of ischemia within the epiphysis. 23, 25, 38, 45 The normal overlying acetabular radioactivity has often been confused with that of the epiphysis. Thus, simply dividing the area of the femoral epiphysis into thirds or quadrants to measure radioactivity will overestimate the degree of revascularization. In addition, the stages of revascularization based upon the known anatomic blood supply and mechanisms of revascularization were not considered in determining the extent of avascularity, Early studies did not utilize magnification techniques to correctly identify the subtle changes occurring within the epiphysis during revascularization, We have defined the various scintigraphic patterns associated with revascularization of LCPD. 18, 19 Correlation with roentgenographic staging determines if there are any delays or complications of revascularization, which are associated with a poorer prognosis. It is postulated that there are two primary revascularization processes. The first is recanalization of the original vascular bed, This process can occur rapidly over an interval of only a few weeks to several months because the blood vessels remain architecturally intact (Fig, 6A and B), The structural integrity of the femoral epiphysis is maintained, Complications such as fracture, collapse, or extrusion of the epiphysis are less likely to occur and little deformity is produced, The second method of revascularization is via collateral circulation, most likely through the physis via the

1322

JAMES J. CONWAY

Figure 6. A, This 4-year-old male patient had pain in his left hip for 2 weeks. Bone scintigraphy demonstrates total ischemia of the left proximal femoral epiphysis (closed arrow). B, A follow-up bone scintigram at 4 months demonstrates complete revascularization of the left proximal femoral epiphysis. Roentgenograms of the hips were normal.

blood supply to the metaphysis. Since entirely new vascular channels must be developed, the process is prolonged and complications such as collapse and extrusion are more prone to occur. Delayed revascularization via collateral channels, therefore, heralds a poor prognosis. Disruption of the normal zone of provisional calcification and growth columns of the femoral metaphysis produces growth disturbances and deformity of the femoral neck. Premature uneven closure of the physis with subsequent varus or valgus deformities occurs. The scintigraphic pattern of recanalization via existent vascular channels is manifested initially by the appearance of a "lateral column." This has a characteristic scintigraphic appearance and occurs within the first few months after the ischemic insult. The appearance is due to the unique blood supply of the proximal femoral epiphysis in the child via the superior capsular arteries which arise from the medial circumflex artery. This vessel arises usually from the femoral artery, passes posteriorly behind the femoral neck, pierces the femoral capsule, and then the superior capsular branches pass through the hip joint space and enter the epiphysis in a posterolateral locus. 14 The lateral column is best seen in an anterior or posterior projection and becomes obscured in the frog lateral projection due to overlap of the acetabulum (Fig. 7A and B). Continued revascularization by recanalization is manifested by gradual filling in of radioactivity into the anterior aspect of the epiphysis ,(Fig. 8AE). This has been termed "anterior extension." Thus, there is a gradual return to normal of the epiphyseal activity. If the activity in the epiphysis exceeds that of the acetabular rim or femoral physis in the <;hild, then it is considered to be abnormally increased, which heralds a complication such

RADIONUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1323

Figure 7. A, Ther~ is a very prominent "lateral column" (arrow) in this lO-year-old boy with a limp of 7 months' duration. This finding is felt to be evidence of revascularization via recanalization of the original vessels to the epiphysis. 8, Note that the "lateral column" is not defined in the lateral projection of the hip ( arrow) since the anterior portion of the epiphysis is the last to revascularize.

as a superimposed fracture upon the revascularization process. On nonmagnified views of the entire pelvis, there may appear to be a generalized increased radioactivity about the hip and acetabulum. This may obscure the epiphyseal activity and the revascularization pattern. Some authors have ascribed this generalized increased localization of the radioisotope about the hip to healing of the avascular necrosis of bone, but it is more likely due to hyperemia within the soft tissues and acetabulum from the irritative process of the deformed head and abnormal ambulation. The scintigraphic pattern associated with neovascularization via collateral circulation is manifested by a gradual widening of the physeal activity as the blood supply fills in from the base of the epiphysis. This has been termed "base filling" (Fig. 9A and B). In our analysis of a large series of children with LCPD using serial scintigraphic and roentgenographic evaluations, a poor prognosis as manifested by deformity of the femoral head was predicted by delayed revasculization, collapse of the femoral epiphysis, and extrusion of the epiphysis out of the acetabular confines. 63 It is now recognized that the earliest roentgenographic signs are usually late findings in LCPD and that the initial ischemic vascular insult occurs 3 to 6 months before any roentgenographic changes are observed. Symptoms and signs of pain and limp lag behind the onset of the disord~r and are probably related to complications of fracture or collapse. An important concept is that of the "whole-head" avascularity at the onset of LCPD. In all studies of early LCPD with normal radiographs of the hips, scintigrams have displayed a total avascular appearance of the epiphysis. This finding is consistent with the usual vascular supply to the pediatric hip. Methods which attempt to determine the degree of avascu-

1324

JAMES J. CONWAY

Figure 8. A, Advanced changes are noted in the left proximal femoral epiphysis in this 4-year-old boy who was totally asymptomatic. B, The finding of Legg-Calve-Perthes disease involving the left proximal femoral epiphysis is barely evident on this routine pelvis scintigram. C, The right hip has a normal scintigraphic appearance. D, There is almost total revascularization of the epiphysis (arrow). The "lateral column" has filled in anteriorly. E, The "anterior extension" is noted in the lateral projection (arrow).

RADIONUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1325

Figure 9. A, There is total head avascularity of the right proximal femoral epiphysis consistent with Legg-Calve-Perthes disease. Roentgenogram of the pelvis at that time (1982) only demonstrated the right epiphysis to be slightly smaller than the left epiphysis. B, Two years lat~r (1984), there is partial revascularization of the epiphysis through "base filling" (arrow). This pattern of revascularization is thought to be due to neovascularization through the growth plate via the lateral circumflex artery.

larity based upon roentgenographic changes which have occurred six months to a year later are fallacious. Roentgenograms at later stages of the disorder reflect the changes and complications superimposed upon the type of revascularization process. Radionuclide bone scintigraphy is accurate for the early detection of LCPD. Roentgenographic changes appear very late and more often reflect complications of the disorder. Avascular necrosis of bone cannot be diagnosed by anatomic changes depicted on roentgenograms. Our studies imply that there are two mechanisms of revascularization. The first mechanism, through recanalization of the previous blood vessels, is rapid and there is little subsequent deformity. The second revascularization pattern is via neovascularization, which requires a prolonged interval for reconstitution of the vascular bed. Complications such as collapse, extrusion, and growth disturbances of the femoral neck produce deformity and poor results. Correlation of the scintigraphic patterns with roentgenograms even at the initial study can determine a prognosis for healing without deformity. Factors and mechanisms affecting revascularization are now better understood anq this knowledge should be of assistance in the management of LCPD. Osteochondroses There are a number of bone disorders signaled by pain that are associated with roentgenographic changes ascribed to avascular necrosis of bone. These have been given various eponyms, usually by whomever first described the lesion. Among these are Kohler's disease (tarsonavicular), Kienbock's disease (carpal lunate), Panner's disease (capitellum of the

1326

JAMES

J.

CONWAY

elbow), Osgood-Schlatter's disease (proximal tibial tubercle), and several others. Bone scintigraphy readily differentiates avascular necrosis from other metabolic bone disorders which are most likely secondary to trauma. 49 Avascular necrosis of the capitellum (Panner's disease) presented clinically as an example of little league elbow (Fig. lOA-C). Kienbock's disease has typically presented with a marked increase in radionuclide localization which excludes the diagnosis of avascular necrosis. These observations point out the inability of roentgenography to define avascular necrosis from other disorders, as the anatomic changes may be similar in appearance. Radionuclide scintigraphy better defines the underlying pathophysiology, allowing a more rational approach to therapy.

Figure 10. A, This lO-year-old little league pitcher was unable to fully extend the right elbow. Roentgenograms suggested osteochondritis dissecans of the capitellum. Tomography demonstrated fragmentation (arrow). B, The radionuclide does not localize in the capitellum (closed arrow) of the distal right humerus, indicating avascular necrosis of this epiphyseal ossification center. C, A magnification anterior projection of the left elbow demonstrates normal radionuclide localization in the capitellum of the humerus (open arrow).

RADIO NUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1327

TRAUMA The greater sensitivity and specificity for defining the changes in bone secondary to trauma mandate the use of radionuclide scintigraphy in any child with complaints of pain who has normal roentgenographic findings. The ordinary wear and tear of childhood play does not produce significant changes even on bone scintigraphy. Essentially, significant or repeated trauma is required in order to produce metabolic bone changes. The following are specific conditions which warrant the use of bone scintigraphy on a routine basis. The Toddler's Fracture This is a well-known condition in the young toddler that is clinically manifested by a limp or refusal to walk. 21 Routine roentgenograms are often normal while bone scintigraphy defines increased localization of radionuclide at the injury site (Fig. 11). A fracture line or periosteal changes may be visualized on fine detailed follow-up radiographs. The lesion may represent an occult undisplaced fracture or plastic bowing injury. Plastic bowing injuries produce profound microtrabecular changes within the bone, and as a result scintigraphy is very abnormal. 43 Treatment is by casting or immobilization for an appropriate length of time. There are other forms of "toddler's fracture" which are only depicted with bone scintigraphy. These include calcaneal injuries 61 and compression injuries, particularly of the cuboid bone in the foot (Fig. 12). Compression injuries in particular cause profound scintigraphic changes, presumably due to microtrabecular disruption. These characteristically occur at impact sites such as the patella, the heel, the sesamoids, and the midtarsal bones. The localization of the

Figure 11. A 9-month-old "toddler" had a limp for 2 days. Multiple roentgenograms of the legs were normal in appearance. There is diffuse increased radionuclide localization throughout the diaphysis and distal metaphysis of the right tibia (arrows) consistent with recent trauma.

1328

JAMES J. CONWAY

Figure 12. This 2V2-year-old boy was favoring his left leg for 8 weeks. Roentgenographic studies were normal. Bone scintigraphy depicts intense radionuclide localization in the cuboid bone of the left foot. The finding is thought to be related to compression injury of the bone.

radionuclide in the bone often parallels the impact surface. A limping child with a normal roentgenographic examination requires bone scintigraphy to further evaluate the cause of the limp.

Child Abuse There is controversy as to the role of bone scintigraphy in detectirig child abuse. l7 , 42 Correlative studies between bone scintigraphy and roentgenographic survey examinations define a greater sensitivity of scintigraphy over roentgenography for recognizing significant bone trauma in child abuse.35, 62 Roentgenography better defines evidence of healed fractures, physeal injuries, and skull fractures, whereas scintigraphy better defines occult bone and periosteal injuries, particularly of the diaphyses and flat bones such as the ribs, pelvis, and spine. In addition, bone scintigraphy frequently depicts soft tissue injury. Roentgenography and scintigraphy, therefore, can act in a complementary fashion. Their combined use is warranted since there is evidence that the number of lesions present is instrumental in determining the care of the child. 35 Because the majority of child abuse occurs in infants and young children under 2 years of age, the quality of bone scintigraphy and roentgenography is critical. A recent unpublished study at The Children's Memorial HospitaP5 defined a higher error rate in recognizing bone trauma in infants with either roentgenography or scintigraphy. The interpretation of bone scintigrams was more variable than that of roentgenograms, and the accuracy

RADIONUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1329

of the interpretation correlated directly with the experience of the interpreter. Our current practice is to obtain immediate regional roentgenograms for obvious injuries to enable prompt and appropriate treatment. Bone scintigraphy is routinely obtained after the emergent conditions are under control or if there is significant evidence of abuse without obvious bone trauma. If there is any evidence of head trauma, skull roentgenograms are obtained in addition to the bone scintigram. Sports Injuries Bone scintigraphy has become an important diagnostic tool in sports medicine. 29. 50, 53, 54, 64 Sports injuries even in the child are associated with chronic and repeated stress. Bone scintigraphy depicts the focal metabolic changes prior to anatomic changes from stress. It is often specific for the type of injury. For example, stress injuries of the proximal portion of the tibia (Fig. 13A and B) and shin splints are found in runners.

Figure 13. A, A 4%-year-old female patient with congenital heart disease presented with right tibial pain and limp of 1 week's duration. Erythrocyte sedimentation rate was mildly elevated. Roentgenographic examination of the leg was normal. The bone scintigram demonstrates a marked proximal diaphyseal localization of the radionuclide in the right tibia (arrow). B, The radionuclide localization is primarily in the posterior cortex of the tibia as seen on this lateral projection of the leg. Stress injuries typically involve the cortex in an asymmetric fashion. Osteomyelitis usually involves the metaphysis. Tumors usually involve the medullary space.

1330

JAMES J. CONWAY

Figure 14. Arrows point to focal areas of increased radionuclide localization on both sides of the fifth lumbar vertebra. The findings are consistent with bilateral stress injuries of the pars interarticularis in this 15-year-old male patient who experienced back pain after lifting weights. The scintigrams returned to normal after 8 months of wearing a back brace.

Spondylolysis Low back pain in the adolescent without fever or elevated sedimentation rate is often a manifestation of spondylolysis. Abnormal bone scintigraphy (Fig. 14) is associated with back pain28 and thus this condition is thought to be due to stress upon the pars interarticularis of the vertebra. With therapy such as the use of a back brace, bone scintigraphy gradually returns to normal with alleviation of the pain. The focal defect on roentgenography mayor may not resolve. The bone scintigram, therefore, more correctly depicts the stress relationships to the anatomic defect of spondylolysis.

MISCELLANEOUS Reflex Sympathetic Dystrophy A puzzling and difficult condition to diagnose and treat is that of reflex sympathetic dystrophy. A focal sensitivity to touch, erythema, edema, trophic skin changes, and pain are noted particularly in the distal portions of extremities following trauma. These findings are believed to represent exaggerated but poorly understood responses of the sympathetic nervous system to trauma. Bone scintigraphy frequently depicts either decreased localization (ischemia) or increased localization (hyperemia) of the distal portion of the extremity (Fig. 15).31.33,37 Treatment usually is with systemic corticosteroids. 37 Myositis Ossificans Recent studies have suggested a significant role for radionuclide scintigraphy in the early diagnosis of myositis ossificans following trauma, 47

RADIO NUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1331

R Figure 15. A 14-year-old female patient experienced pain in her right lower extremity for 7 weeks, necessitating the use of crutches and cane. There is diffuse increased localization of the radionuclide throughout the ankle and foot consistent with reflex sympathetic dystrophy.

Localization of radionuclide is seen in the soft tissues (Fig. 16) prior to any calcification appearing on roentgenography. The prompt recognition of this disorder enables earlier treatment and also differentiates the occasional misleading appearance of myositis ossificans, which simulates neoplastic periosteal changes extending from bone.

Figure 16. This 14-year-old female patient developed progressive loss of elbow motion following treatment for an undisplaced supracondylar fracture of the left humerus. The bone scintigram depicts soft tissue localization (arrows) of the radionuclide consistent with early myositis ossificans.

1332

JAMES J. CONWAY

CONCLUSIONS Radionuclide' bone scintigraphy has become an important part of pediatric nuclear medicine. Improved radiopharmaceuticals and gamma cameras along with improved techniques such as magnification enable examination of even the smallest premature neonate. A wide spectrum of bone abnormalities in childhood is readily depicted prior to any anatomic roentgenographic changes. The underlying pathophysiology of many disorders is now better understood. The quality of bone scintigraphy and the experience of the interpreter are direct determinants of the accuracy of radio nuclide bone scintigraphy. The pediatrician should be aware of these factors and act accordingly in obtaining adequate bone scintigraphy for the orthopedic disorders of childhood. ACKNOWLEDGMENTS The author thanks his orthopedic colleagues, Luciano S. Dias, M.D., Clare Giegerich, M. D., Armen S. Kelikian, M. D., and Mihran O. Tachdjian, M. D., for their continuing support and enlightening insight into the applications of radionuclides in pediatric orthopedics. I also thank my chief nuclear medicine technologist, Sue Weiss, C.N.M.T., for her expertise in pediatric nuclear medicine technology and technological assistance in obtaining the images illustrating this chapter. Finally, I fully appreciate the superb secretarial assistance provided by KathyLynn Pollack in the preparation of this article.

REFERENCES 1. Alazraki, N., Dries, D., Lawrence, P., et al.: Assessment of skin ulcer healing capability by technetium-99m phosphate angiogram and blood-pool imaging. J. Nucl. Med., 26:586-591, 1985. 2. Alazraki, N., Dries, D., Datz, F., et al.: Value of a 24-hour image (four-phase bone scan) in assessing osteomyelitis in patients with peripheral vascular disease. J. Nucl. Med., 26:711-717,1985. 3. Amundsen, T. R., Siegel, M. J., Siegel, B. A.: Osteomyelitis and infarction in sickle cell hemoglobinopathies: Differentiation by combined technetium and gallium scintigraphy. Radiology, 153:807-812, 1984. 4. Ash, J. M., and Gilday, D. L.: The futility of bone scanning in neonatal osteomyelitis: Concise communication. J. N ucl. Med., 21 :417-420, 1980. 5. Bensahel, H., Bok, B., Cavailloles, F., et al.: Bone scintigraphy in Perthes disease. J. Pediatr. Orthop., 3:302-305, 1983. 6. Bergdahl, K., Bjorksten, B., Gustavson, K. H., et al.: Pustulosis palmoplantaris and its relation to chronic recurrent multifocal osteomyelitis. Dermatologica, 159:37-45, 1979. 7. Berkowitz, 1. D., and Wenzel, W.: Normal technetium bone scan in patients with acute osteomyelitis. Am. J. Dis. Child., 134:828-830, 1980. 8. Bjorksten, B., and Boquist, L.: Histopathological aspects of chronic recurrent multifocal osteomyelitis. J. Bone Joint Surg., 62:376-380, 1980. 9. Bressler, E. L., Conway, J. J., and Weiss, S. C.: Neonatal osteomyelitis examined by bone scintigraphy. Radiology, 152:685-688, 1984. 10. Bruschwein, D. A., Brown, M. L., and McLeod, R. A.: Gallium scintigraphy in the evaluation of disk-space infections: Concise communication. J. Nucl. Med., 21:925927, 1980. 11. Calver, R., Venugopal, V., Dorgan, J., et al.: Radionuclide scanning in the early diagnosis of Perthes' disease. J. Bone Joint Surg., 63:379-382, 1981.

RADIONUCLIDE BONE SCINTIGRAPHY IN PEDIATRIC ORTHOPEDICS

1333

12. Carty, H., Maxted, M., Fielding, J. A., et al.: Isotope scanning in the "irritable hip syndrome." Skeletal Radiol., 11:32-37, 1984. 13. Cavailloles, F., Bok, B., and Bensahel, H.: Bone scintigraphy in the diagnosis and follow up of Perthes' disease. Eur. J. Nucl. Med., 7:327-330, 1982. 14. Chung, S. M. K.: The arterial supply of the developing proximal end of the human femur. J. Bone Joint Surg., 58:961-970, 1976. 15. Collins, M., Tanz, R. R., and Conway, J. J.: Diagnostic imaging problems in abused children. Unpublished data, 1985. 16. Conway, J. J.: Radionuclide bone imaging in pediatrics. Pediatr. Clin. North Am., 24:701711, 1977. 17. Conway, J. J.: Letter to the editor. Further comments on the role of the radionuclide skeletal survey in the diagnosis of the suspected abused child. Radiology, 148:574-576, 1983. 18. Conway, J. J., Weiss, S. C., and Maldonado, V.: Scintigraphic patterns in Legg-CalvePerthes disease. Radiology, 149:102, 1983. 19. Conway, J. J.: Pediatric Imaging I. Syllabus, a categorical course on nuclear medicine. Oak Brook, Illinois, Radiological Society of North America, Inc., 1985, pp. 165-171. 20. Danigelis, J. A., Fisher, R. L., Ozonoff, M. B., et al.: -Tc-polyphosphate bone imaging in Legg-Perthes disease. Radiology, 115:407-413, 1975. 21. Dunbar, J. S., Owen, H. F., Nogrady, M. B., et al.: Obscure tibial fracture of infants-the toddler's fracture. J. Can. Assoc. Radiol., 15:136-144, 1964. 22. Farley, T., Conway, J., and Shulman, S. T.: Hematogenous pelvic osteomyelitis in children. Clinical correlates of newer scanning methods. Am. J. Dis. Child., 139:946950, 1985. 23. Fasting, O. J., Bjerkreim, I., Langeland, U., et al.: Scintigraphic evaluation of the severity of Perthes' disease in the initial stage. Acta Orthop. Scand., 51 :655-660, 1980. 24. Fasting, O. J., Langeland, N., Bjerkreim, I., et al.: Bone scintigraphy in early diagnosis ofPerthes' disease. Acta Orthop. Scand., 49:169-174,1978. 25. Fisher, R L., Roderique, J. W., Brown, D. C., et al.: The relationship of isotopic bone imaging findings to prognosis in Legg-Perthes disease. Clin. Orthop., 150:23-29, 1980. 26. Fleisher, G. R, Paradise, J. E., Plotkin, S. A., et al.: Falsely normal radionuclide scans for osteomyelitis. Am. J. Dis. Child., 134:499-502, 1980. 27. Gates, G. F.: Scintigraphy of discitis. Clin. Nticl. Med., 2:20-25, 1977. 28. Gelfand, M. J., Strife, J. L., and Kereiakes, J. G.: Radionuclide bone imaging in spondylolysis of the lumbar spine in children. Radiology, 140:191-195, 1981. 29. Geslien, G. E., Thrall, J. H., Espinosa, J. L., et al.: Early detection of stress fractures using 99mTc-polyphosphate. Radiology, 121:683-687, 1976. 30. Giedion, A., Holthusen, W., Masel, L. F., et al.: Subacute and chronic "symmetrical" osteomyelitis. Ann. Radiol., 15:329-342, 1972. 31. Greyson, N. D., and Tepperman, P. S.: Three-phase bone studies in hemiplegia with reflex sympathetic dystrophy and the effect of disuse. J. Nucl. Med., 25:423-429, 1984. 32. Hernandez, R. J., Conway, J. J., Poznanski, A. K., et al.: The role of computed tomography and radionuclide scintigraphy in the localization of osteomyelitis in flat bones. J. Pediatr. Orthop., 5:151-154', 1985. 33. Holder, L. E., and Mackinnon, S. E.: Reflex sympathetic dystrophy in the hands: Clinical and scintigraphic criteria. Radiology, 152:517-522, 1984. 34. Howie, D. W., Savage, J. P., Wilson, T. G., et al.: The technetium phosphate bone scan in the diagnosis of osteomyelitis in childhood. J. Bone Joint Surg., 65:431-437, 1983. 35. Jaudes, P. K.: Comparison of radiography and radionuclide bone scanning in the detection of child abuse. Pediatrics, 73:166-168, 184. 36. Keeley, K., and Buchanan, G. R: Acute infarction of long bones in children with sickle cell anemia. J. Pediatr., 101:170-175, 1982. 37. Kozin, F., Soin, J. S., Ryan, L. M., et al.: Bone scintigraphy in the reflex sympathetic dystrophy syndrome. Radiology, 138:437-443, 1981. 38. LaMont, R L., Muz, J., Heilbronner, D., et al.: Quantitative assessment offemoral head involvement in Legg-Calve-Perthes disease. J. Bone Joint Surg., 63:746-752, 1981. 39. Lisbona, R, and Rosenthall, L.: Observations on the sequential use of 99mTc-phosphate complex and 67Ga imaging in osteomyelitis, cellulitis, and septic arthritis. Radiology, 123:123-129, 1977.

1334

JAMES J. CONWAY

40. Mackinnon, S. K, and Holder, L. K: The use of three-phase radionuclide bone scanning in the diagnosis of reflex sympathetic dystrophy. J. Hand Surg., 9:556--563, 1984. 41. Maurer, A. H., Chen, D. C. P., Camargo, K K, et a!.: Utility of three-phase skeletal scintigraphy in suspected osteomyelitis: Concise communication. J. Nuc!. Med., 22:941949, 1981. 42. Merten, D. F., Radkowski, M. A., and Leonidas, J. C.: The abused child: A radiological reappraisal. Radiology, 146:377-381, 1983. 43. Miller, J. H., and Osterkamp, J. A.: Scintigraphy in acute plastic bowing of the forearm. Radiology, 142:742, 1982. 44. Mok, P. M., Reilly, B. J., and Ash, J. M.: Osteomyelitis in the neonate. Radiology, 145:677-682, 1982. 45. Morley, T. R, Short, M. D., and Dowsett, D. J.: Femoral head activity in Perthes' disease: Clinical evaluation of a quantitative technique for estimating tracer uptake. J. Nuc!. Med., 19:884-890, 1978. 46. Murry, I. P. C.: Photopenia in skeletal scintigraphy of suspected bone and joint infection. Clin. Nucl. Med., 7:13-20, 1982. 47. Orzel, J. A., and Rudd, T. G.: Heterotopic bone formation: Clinical, laboratory, and imaging correlation. J. Nucl. Med., 26:125-132, 1985. 48. Paul, D. J., Gilday, D. L., Gurd, A., et al.: A better method of imaging abnormal hips. Radiology, 113:466-467, 1974. 49. Pavlov, H., Torg, J. S., and Freiberger, R H.: Tarsal navicular stress fractures: Radiographic evaluation. Radiology, 148:641-645, 1983. 50. Prather, J. L., Nusynowitz, M. L., Snowdy, H. A., et a!.: Scintigraphic findings in stress fractures. J. Bone Joint Surg., 59:869-873, 1977. 51. Probst, F. P.: Chronic multifocal cleido-metaphyseal osteomyelitis of childhood. Acta Radiol [Diagn.], 17:531-537, 1976. 52. Rocco, H. D., and Eyring, E. J.: Intervertebral disk infections in children. Am. J. Dis. Child., 123:448-451, 1972. 53. Rosen, P. R, Micheli, L. J., and Treves, S.: Early scintigraphic diagnosis of bone stress and fractures in athletic adolescents. Pediatrics, 70:11-15, 1982. 54. Roub, L. W., Gumerman, L. W., Hanley, K N., et al.: Bone stress: A radionuclide imaging perspective. Radiology, 132:431-438, 1979. 55. Rupani, H. D., Holder, L. K, Espinola, D. A., et al.: Three-phase radionuclide bone imaging in sports medicine. Radiology, 156:187-196, 1985. 56. Sayle, B. A., Fawcett, H. D., Wilkey, D. J., et al.: Indium-1ll chloride imaging in the detection of infected prostheses. J. Nuc!. Med., 26:718-721, 1985. 57. Schauwecker, D. S., Park, H., Mock, B. H., et al.: Evaluation of complicating osteomyelitis with Tc-99m MDP, In-1ll granulocytes, and Ga-67 citrate. J. Nucl. Med., 25:849853, 1984. 58. Shafer, R B., and Edeburn, G. F.: Can the three-phase bone scan differentiate osteomyelitis from metabolic or metastatic bone disease. Clin. Nucl. Med., 9:373-377, 1984. 59. Smith, R F., and Taylor, T. K.: Inflammatory lesions of intervertebral discs in children. J. Bone Joint Surg., 49:1508-1520, 1967. 60. Solheim, L. F., Paus, B., Liverud, K., et al.: Chronic recurrent multifocal osteomyelitis. Acta Orthop. Scand., 51:37-41, 1980. 61. Starshak, R J., Simons, G. W., and Sty, J. R.: Occult fracture of the calcaneus-another toddler's fracture. Pediatr. Radiol., 14:37-40, 1984. 62. Sty, J. R, and Starshak, R J.: The role of bone scintigraphy in the evaluation of the suspect abused child. Radiology, 146:369-375, 1983. 63. Tremaine, C. L., Dias, L. S., Conway, J. J., et al.: Legg-Calve-Perthes disease: A scintigraphic classification and its relation to end result. Unpublished data, 1984. 64. Wilcox, J. R, Moniot, A. L., and Green, J. P.: Bone scanning in the evaluation of exercise-related stress injuries. Radiology, 123:699-703, 1977. Division of Nuclear Medicine Children's Memorial Hospital Chicago, Illinois 60614