| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Departments of Radiology (K.C., C.B.C., Y.M.L., D.R.) and Pathology (P.H.), Veterans Affairs Medical Center, University of California–San Diego, 3350 La Jolla Village Dr, San Diego, CA 92161. From the 2000 RSNA scientific assembly. Received July 26, 2000; revision requested September 7; final revision received December 11; accepted January 5, 2001. Address correspondence to D.R. (e-mail: dresnick@ucsd.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Nine female patients with leg pain and imaging features indicative of intramedullary sclerosis were seen during a 25-year period. None of the patients had a history of trauma or infection, familial bone disease, or related abnormal laboratory findings. Imaging studies included radiography (n = 9), CT (n = 4), MR imaging (n = 5), and skeletal scintigraphy (n = 5). Histologic correlation was available in five patients.
RESULTS: Sixteen bone lesions (midtibia, n = 14; distal fibula, n = 1; and proximal femur, n = 1) were evident. Both lower extremities were involved in seven patients, and a single extremity was involved in two. Intramedullary sclerosis was present, as was cortical thickening, mainly in the diaphysis of the long bones, without extensive periosteal reaction or soft-tissue involvement. Findings at bone scintigraphy were positive in all lesions. Histologic analysis showed nonspecific changes of markedly sclerotic bone with a variable degree of mineralization and maturity.
CONCLUSION: Intramedullary osteosclerosis is a distinct disorder that typically affects the diaphysis of one or both tibiae in women. Characteristic imaging findings, when coupled with clinical information, allow precise diagnosis.
Index terms: Bones, CT, 45.1211 • Bones, MR, 45.1214 • Bones, radionuclide studies, 45.12172 • Bones, sclerosis, 45.5322 • Fibula, 45.5322 • Intramedullary osteosclerosis, 45.5322 • Tibia, 45.5322
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Intramedullary osteosclerosis is a rare condition that has no familial history and is not associated with infection, trauma, or systemic illness. Results of biochemical analysis are within normal limits. Although the imaging features of this condition resemble those of many sclerosing bone dysplasias—stress fractures, osteomyelitis, metabolic and endocrine disorders, and bone-forming tumors (1–7)—these features generally allow the correct diagnosis when coupled with clinical information and findings of laboratory analysis.
The purpose of our study was to determine the conventional radiographic, computed tomographic (CT), magnetic resonance (MR) imaging, scintigraphic, and histologic features of intramedullary osteosclerosis and to review the clinical features.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
All nine patients experienced mild to moderate chronic leg pain that was aggregated by physical activity and corresponded to the sites of abnormalities shown on conventional radiographs. Family history in all patients revealed no similar musculoskeletal illness. Only one patient had a history of generalized disease (type 1 diabetes mellitus and Hashimoto thyroiditis). None of these patients gave a history of trauma or infection; however, one patient reported leg pain that had started after a long hiking trip. Physical examination revealed no cutaneous stigmata, leg deformities, or leg length discrepancies. Laboratory data (complete blood count, erythrocyte sedimentation rate, and serum alkaline phosphatase levels) were normal in all patients.
Imaging Studies and Image Review
All imaging studies in each patient were grouped together at the time of analysis. Routine radiographs were available for review in all patients.
CT scans had been obtained in four patients with one of two scanners (DR3, Siemens Medical Systems, Iselin, NJ; 9800 CT scanner, GE Medical Systems, Milwaukee, Wis). The CT technique varied due to the retrospective nature of the study. Scanning was performed with 4–20-mm contiguous transverse sections throughout the lesions. Contrast material was not administered in all patients.
MR images had been obtained in five patients. MR imaging was performed with one of two 1.5-T MR systems (Signa, GE Medical Systems, Milwaukee, Wis; Magnetom, Siemens Medical Systems, Erlangen, Germany). The studies varied and were performed with T1-weighted (repetition time msec/echo time msec of 300–600/8–27), intermediate-weighted (2,075–3,000/15–24), and T2-weighted (2,000–6,500/60–105) spin-echo sequences in the transverse, coronal, and sagittal planes. In one case, a gradient-echo sequence (140/4.4, 80° flip angle) was also used. Gadolinium-enhanced (gadopentetate dimeglumine, Magnevist; Berlex Laboratories, Wayne, NJ) T1-weighted fat-saturated images were available in three patients. All MR studies were performed with a field of view of 14–40 cm, section thickness of 3–6 mm, intersection gap of 1–3 mm, two signals acquired, and matrix of 256 x 192 to 512 x 256.
Bone scintigraphy (single photon emission CT, or SPECT) was performed in five patients with one of two gamma cameras (GCA 40 A, Toshiba, Tokyo, Japan; Orbiter model 7500, Siemens Medical Systems) equipped with a general-purpose collimator. Images were obtained 3–6 hours after intravenous injection of 550–740 MBq (15–20 mCi) of technetium 99m methylene diphosphonate, or 99mTc MDP.
Images were interpreted in a retrospective fashion by consensus of two musculoskeletal radiologists (K.C., D.R.) who had 6 and 25 years of experience, with particular attention to lesion location, intramedullary and cortical involvement, periosteal reaction, soft-tissue involvement, length and degree of sclerosis, symmetry, uptake of radionuclide on scintigraphic images, and signal intensity and contrast material enhancement on MR images.
Review of Histologic Findings
Histologic findings were available in five patients. All histologic specimens were stained with hematoxylin and eosin and were reviewed by a bone pathologist (P.H.), who had 30 years of experience.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
Skeletal scintigraphy was performed in five patients with nine lesions. The images showed intense tracer uptake that corresponded to the sites of abnormalities shown on conventional radiographs. In one patient with bilateral extremity lesions, an asymptomatic lesion was incidentally found on the radionuclide study (Fig 2) in an area that appeared normal at conventional radiography.
Histologic findings were available in five patients and were nonspecific. Some of the histologic features of this entity are reported to overlap those of melorheostosis (Fig 3). The peripheral and major part of the lesion was characterized by the replacement of normal spongiosa by markedly sclerotic and thickened trabeculae that encompassed the marrow cavity with a variable degree of mineralization and maturity and were seen radiographically as osteosclerosis. There were focal areas of osteoblastic rimming of the trabeculae. More centrally, a fairly cellular spindle cell proliferation with bland fibroblastic cytomorphology associated with immature collagen deposition was noted.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The clinical, radiographic, scintigraphic, and histologic findings in the nine patients in our study support those in previous studies (1–3). However, our study was somewhat different from that of Horwitz (2), who reported asymptomatic lesions that affected pubic bones, and Sotelo-Ortiz (3), who reported a lesion with associated bone deformity.
Although there were limitations to our study, such as the small number of patients and the lack of histologic results for all patients, we believe that our study and our review of the literature can allow some conclusions. To our knowledge, only seven cases of intramedullary osteosclerosis have been reported (1–3). Five of these seven patients were women. All nine patients in our study were women. These data suggest that this condition is more common in women than in men. Findings in our study also indicate that bone scintigraphy can depict lesions, even asymptomatic lesions or those in patients with conventional radiographs interpreted as normal.
The cause of intramedullary osteosclerosis is poorly understood. The disease has been classified previously by some investigators (1–3) as one of the sclerosing bone dysplasias, which are a group of developmental anomalies that reflect disturbances in the formation and modeling of bone. In many ways, correct diagnosis of intramedullary osteosclerosis is based on the exclusion of disorders known to be associated with bone sclerosis, including osteosarcoma, lymphoma, osteoblastic metastasis, osteoid osteoma, chronic osteomyelitis, healing stress fractures, metabolic disorders, and many sclerosing bone dysplasias (1–7).
Radiographically, the benign appearing intramedullary sclerosis that occurs without extensive periosteal new bone formation and soft-tissue abnormality, which are characteristic of idiopathic intramedullary osteosclerosis, is different from the typical findings of malignant tumors such as osteosarcoma, lymphoma, and osteoblastic metastasis. Although osteosarcoma, lymphoma, and osteoblastic metastasis in the diaphysis of a tubular bone have some characteristics that resemble those of intramedullary osteosclerosis, periosteal elevation and soft-tissue extension are often evident (7–9). The lack of a classic clinical history and the lack of a small radiolucent nidus on imaging studies generally allow the elimination of osteoid osteoma as an alternative diagnostic choice (10,11). The possibility of chronic sclerosing osteomyelitis usually can be discarded owing to the absence of clinical and laboratory findings of infection (12).
Intramedullary osteosclerosis is mainly located in the midtibia, and patients typically experience activity-related lower leg pain; therefore, the initial diagnosis of a stress injury or stress fracture often is considered. The absence of a fracture line and the presence of medullary sclerosis as the dominant imaging finding generally are sufficient to suggest an alternative diagnosis (13–15).
Metabolic and endocrine disorders that need to be distinguished from idiopathic intramedullary osteosclerosis are renal osteodystrophy, hypervitaminosis A, pseudohypoparathyroidism, and pseudopseudohypoparathyroidism (4,7). Laboratory analysis is essential in eliminating the diagnoses of these disorders. Furthermore, generalized osteosclerosis is more characteristic of renal osteodystrophy, pseudohypoparathyroidism, and pseudopseudohypoparathyroidism. Hypervitaminosis A may cause periosteal new bone formation, especially in the diaphysis of tubular bones of the lower extremities, but the medullary cavity is typically spared.
Erdheim-Chester disease is a very rare condition that is usually depicted radiographically as bilateral symmetric inhomogeneous osteosclerosis in the diametaphysis of tubular bones (16,17). The femur and tibia are the most common sites of involvement. Laboratory examination often reveals abnormal lipid metabolism, moderate anemia, and, sometimes, increased C-reactive protein and erythrocyte sedimentation rates. Histologic findings of Erdheim-Chester disease are distinctive.
The sclerosing bone dysplasias are a poorly understood group of developmental disorders. Their cause and pathogenesis are still unknown. In the majority of cases of sclerosing dysplasia of bone, the diagnosis is made on the basis of characteristic radiographic manifestations rather than the histologic findings of the lesion (4–6). Specific sclerosing bone dysplasias include osteopetrosis (Albers-Schönberg disease), pyknodysostosis (Maroteaux-Lamy disease), enostosis (bone island), osteopoikilosis (spotted bone disease), osteopathia striata (Voorhoeve disease), melorheostosis, metaphyseal dysplasia (Pyle disease), hyperostosis corticalis generalisata (Van Buchem disease), autosomal dominant osteosclerosis (Worth disease), sclerostosis, metaphyseal dysplasia, progressive diaphyseal dysplasia (Camurati-Engelmann disease), and hereditary multiple diaphyseal sclerosis (Ribbing disease) (1–7,10). Only two of these dysplasias realistically enter into the differential diagnosis of intramedullary osteosclerosis, namely Camurati-Engelmann and Ribbing diseases (7,10).
Intramedullary osteosclerosis, Camurati-Engelmann disease, and Ribbing disease not only are almost identical in their imaging appearances but also all show intense radiotracer uptake on bone scans (1,18,19). Nevertheless, differences in their clinical manifestations, laboratory findings, and histologic features facilitate correct diagnosis. Camurati-Engelmann disease is an autosomal dominant disease that occurs in the 1st decade of life, and in boys more frequently than in girls (4,20–23). The clinical symptoms include bone and muscle pain, weakness, and atrophy; a waddling and broad-based gait; and delayed puberty. The important radiologic features are fusiform thickening of the cortical bone in the diaphyses of long tubular bones, with a bilateral and symmetric distribution. The medullary cavities may be narrowed, and resulting anemia has been reported (24,25). Hyperostosis of the calvaria is observed almost as frequently as are changes in the long bones (21).
In contrast to Camurati-Engelmann disease, Ribbing disease is of autosomal recessive inheritance and is seen clinically after puberty (8,26–28). The only clinical finding is pain in the involved extremity. Moreover, unilateral or bilateral abnormalities of the long bones are observed with neither skull involvement nor anemia. More recent investigators, however, have noted that Camurati-Engelmann and Ribbing diseases represent phenotypic variations of the same disorder (22,29).
At histologic examination, Ribbing disease is characterized by osteoblastic activity and new bone formation, whereas Camurati-Engelmann disease has striking osteoblastic and osteoclastic activities that reflect both active bone resorption and bone formation. Intramedullary osteosclerosis and Ribbing disease are indistinguishable on the basis of clinical manifestations, laboratory findings, and histologic features, and both are characterized radiographically by asymmetric intramedullary diaphyseal scleroses of the long tubular bones of the lower extremities in adult patients (7,10). However, intramedullary osteosclerosis is a nonhereditary condition that is found commonly in women, whereas Ribbing disease is an autosomal recessive inheritance and has no sex predominance.
In conclusion, patients seen for the first time with radiographically evident intramedullary sclerosis that affects long tubular bones in one or both lower extremities pose a diagnostic dilemma. Because the distinction among reactive, dysplastic, benign, and malignant bone lesions may occasionally be extremely difficult, accurate clinical history and laboratory analysis combined with imaging findings are essential components that allow accurate diagnosis.
Intramedullary osteosclerosis is a diagnosis of exclusion. The essential findings in this disease are onset in adults, female predominance, lack of a similar condition in family members, and an asymmetric distribution of intramedullary osteosclerosis without extensive periosteal new bone formation and soft-tissue involvement that affects a portion of the diaphysis of one or more long bones of one or both lower extremities.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P O'Donnell and P O'Donnell Evaluation of focal bone lesions: basic principles and clinical scenarios Imaging, December 1, 2003; 15(4): 298 - 323. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
