| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, National Naval Medical Center, 8901 Wisconsin Ave.,
Bethesda, MD 20889.
2 Uniformed Services University of the Health Sciences, Bethesda, MD
20889.
3 Department of Radiology, University of Maryland, Baltimore, MD 21201.
4 Department of Orthopedic Pathology, Armed Forces Institute of Pathology,
Washington, DC 20306.
5 Department of Orthopaedics and Rehabilitation, University of Miami, Coral
Gables, FL 33124.
6 Department of Radiology, Washington Hospital Center, Washington, DC
20010-2976.
7 Department of Radiology, Mayo Clinic, Jacksonville, FL 32224.
Received September 23, 2002;
accepted after revision April 21, 2003.
Address correspondence to D. J. Flemming.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed 50 cases of calcific tendinitis involving underlying bone. Clinical data reviewed included patient age and sex and lesion location. Images reviewed included radiographs (n = 44), CT scans (n = 13), MRIs (n = 16), and bone scintigrams (n = 13). Radiologic examinations were evaluated for the presence of cortical erosion, periosteal reaction, and marrow extension. Pathology confirmation was available in 37 cases.
RESULTS. The average age of patients was 50 years (range, 16–82 years), with 29 female patients (58%). Calcific tendinitis with associated bone involvement was seen most commonly in the femur (40%) and the humerus (40%). Concretions were most commonly solid-appearing (50%). Cortical erosion was the most common manifestation of osseous involvement (78% of cases). Marrow involvement was shown in 18 (36%) of 50 cases. Marrow extension was most commonly seen in the lesser and greater tuberosities of the humerus, which accounted for 61% (11/18) of cases. Focal increased radionuclide uptake was seen in 13 (100%) of 13 cases.
CONCLUSION. Calcific tendinitis presenting with osseous destruction, marrow changes, and soft-tissue calcifications may be confused with neoplasm both radiologically and pathologically. Recognition of the atypical presentation of this common disease may prevent unnecessary biopsy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CT scans were unenhanced studies performed with slice thickness varying from 1.5 to 5 mm. MRIs were unenhanced examinations obtained from a variety of MR scanners that operated from 0.35 to 1.5 T. MRIs consisted of T1-weighted (TR range/TE range, 300–1000/9–30), T2-weighted (1600–3000/40–120), and fast spin-echo T2-weighted (2000–4800/60–120) sequences in all 16 patients. Slice thickness varied between 3 and 8 mm, with a 10–20% interslice gap. Fat saturation was used on T2-weighted images in 10 patients.
Image Evaluation
The presence or absence of soft-tissue calcification was assessed on
radiographs, CT scans, and MRIs, and the size was measured on the imaging
modality that best depicted the abnormality. Soft-tissue calcifications were
categorized by character (solid, stippled, or amorphous) and shape (round or
elongated) on radiographs. The character of calcification (solid, stippled, or
amorphous) was also evaluated on CT scans. Soft-tissue calcification was
assessed (on the basis of radiographic or CT localization) on MRIs for
predominant signal-intensity characteristics on T1-weighted images (low [less
than that of muscle], intermediate [equal to that of muscle], high [greater
than that of muscle]) and T2-weighted images (low [similar to or less than
that of muscle], intermediate [similar to that of fat], high [greater than
that of fat]) and for the presence or absence of surrounding edema on long TR
sequences. The presence or absence of a soft-tissue mass associated with
calcification was evaluated on CT scans and MRIs.
The presence or absence of cortical erosion was assessed on radiographs, CT scans, and MRIs. In cases with multiple modalities available for review, the ability to detect cortical erosion was compared.
Radiographs were used to assess the presence or absence of periosteal reaction. The character of periosteal reaction was classified as aggressive (perpendicular or lamellated) or nonaggressive (solid). These features were evaluated only in cases with radiographs of the abnormality in profile.
The presence or absence of bone marrow involvement was assessed on radiographs, CT, and pathology as shown by calcification in the marrow space with surrounding trabecular and cortical destruction. Marrow involvement was determined to be present on MRIs if fat signal replacement was seen on T1- or T2-weighted images. The predominant signal intensity of marrow involvement on T2-weighted images was evaluated as low, intermediate, or high (same criteria as described for calcification). Marrow involvement on T2-weighted images was also assessed as homogeneous or heterogeneous (mild, moderate, or marked). The relationship of marrow involvement and soft-tissue calcification was evaluated in each case for contiguity.
Delayed bone scintigraphy uptake was characterized by intensity (mild, moderate, or marked), extent (focal or diffuse), and location relative to bone (eccentric or central).
Pathology Evaluation
Calcific tendinitis was pathologically confirmed in 37 cases. Specimens
were obtained by open biopsy in all cases. H and E–stained slides were
available for reevaluation by an experienced orthopedic pathologist in 27
cases. Pathology material was reviewed for presence and character of
calcifications, fibrosis, chronic inflammation, giant cell reaction, chondroid
metaplasia, and evidence of marrow involvement. All reevaluated pathology
material was also examined under polarized light for the presence or absence
of birefringent crystals.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Calcific tendinitis associated with osseous involvement was most commonly seen in the humerus or femur, with each representing 40% of cases (20/50 patients). The sites of involvement included the femoral diaphysis (n = 19), humeral tuberosities (n = 11), humeral diaphysis (n = 9), hand and wrist (n = 3), foot (n = 3), acetabulum (n = 2), distal femur (n = 1), cervical spine (n = 1), and clavicle (n = 1). The diaphysis was involved in 56% (n = 28) of patients, followed by juxtaarticular sites in 44% (n = 22).
Femoral involvement was posterior and subtrochanteric along the linea aspera at the level of or within 6 cm of the lesser trochanter in 95% of cases (n = 19) (Fig. 1A, 1B). The one case of distal femoral involvement was seen at the medial femoral condyle. Humeral diaphysis involvement in all cases (n = 9) was anterior, at or near the insertion of the pectoralis major muscle. Proximal humerus involvement was also seen in the lesser tuberosity in six cases and in the greater tuberosity in five cases. Hand or wrist involvement was shown dorsal to the capitate in two cases and at the first metacarpophalangeal joint in one case. Foot involvement was at the first metatarsophalangeal joint in all cases (n = 3). Hip involvement (n = 2) was at the inferior medial aspect of the acetabular rim in one case and the superior lateral acetabulum in the other. Cervical spine involvement (n = 1) was anterior to the dens and body of C2. Clavicular involvement (n = 1) was seen at the level of the origin of the supracoracoid bursa and the deltoid muscle.
|
|
The calcifications appeared solid on radiographs in 50% (n = 22) of patients, stippled in 25% (n = 11), and amorphous in 20% (n = 9). In two cases (5%) involving the proximal humerus, there was minimal soft-tissue calcification not easily detected or characterized by radiographs adjacent to cortical erosion, and pathology evaluation confirmed calcific tendinitis. The soft-tissue calcifications in intratendinous locations were suggested by an elongated or comet-tail shape in 50% (n = 22) of cases but were round or nonspecific in others. The calcifications were small, averaging 2.0 x 1.2 cm, with sizes ranging from 4.0 cm x 2.0 cm to 0.5 cm x 0.5 cm. On CT, the character of calcifications appeared solid in 46% (n = 6) of cases, stippled in 38% (n = 5), and amorphous in 15% (n = 2). The calcific concretions on MRIs were low in signal intensity on both T1- and T2-weighted images in all cases (n = 16) and were separated from the involved tendon in 11 (73%) of 15 cases by surrounding edema on T2-weighted MRIs. No significant soft-tissue mass was seen at the site of cortical erosion in patients examined with cross-sectional imaging, but increased signal in adjacent enlarged tendons was seen in all cases.
Cortical erosion was the most common manifestation of osseous involvement and was seen in 78% (n = 39) of the 50 patients; the remaining patients showed marrow involvement with no radiographically definable cortical defect. Cortical involvement was shown in 75% (n = 33) of patients examined with radiography (n = 44) and in all patients examined with CT (Fig. 2A, 2B). CT detected cortical involvement not identified on radiographs in three patients. MRI showed cortical erosion in 50% (n = 8) of patients but did not reveal surface involvement in patients not examined with other modalities.
|
|
Periosteal reaction was identified in only 32% (n = 14) of all radiographs (n = 44) and was seen only with diaphyseal involvement of long bones. Periosteal reaction on radiographs was imaged in profile in 19 cases and was seen in 74% (n = 14) of cases with optimal radiographic examination of the diaphysis of the proximal femur or humerus (n = 19). Periosteal reaction was characterized as nonaggressive (solid) in 64% (n = 9) of cases and as aggressive (lamellated) in 36% (n = 5).
Bone marrow involvement was shown by radiology or pathology in 18 (36%) of 50 patients. Pathologic confirmation of radiologic evidence of bone marrow extension was obtained in six cases. Bone marrow involvement was revealed at pathology in only two cases involving the hand that were not examined with advanced imaging. Marrow extension in these two cases was seen dorsally in both, with involvement of the capitate in one case and of a carpal boss in the other. Affected bone marrow was encountered in 11 cases near the tuberosities of the proximal humerus and was nearly equally divided between the lesser (n = 6) and greater (n = 5) tuberosities. Extension into marrow was documented in three cases affecting the linea aspera of the femur (Fig. 3A, 3B, 3C) and in one case affecting the medial femoral condyle. Marrow involvement was seen in single cases affecting the cervical spine at C2 and the acetabulum.
|
|
|
Of the 18 cases with marrow involvement, 17 had radiographs for review. Marrow involvement was shown in eight of these cases (44%) by radiography and in six (86%) of seven cases imaged with CT. In one case, spontaneous partial resolution of intraosseous calcification was seen in the greater tuberosity on subsequent radiographs (Fig. 4A, 4B, 4C, 4D, 4E, 4F). Of the 18 cases with intraosseous extension, 12 had MRI, all of which showed marrow involvement. On T1-weighted images, marrow involvement was predominantly low in signal intensity in 58% (n = 7) of cases and of intermediate intensity in 42% (n = 5). On T2-weighted images, bone marrow involvement was predominantly low in signal intensity in 58% (n = 7) of cases and high in 42% (n = 5). Marrow extension was heterogeneous on T2-weighted images in 75% (n = 9) of cases and homogeneously high in signal intensity in 25% (n = 3). The marrow process was directly adjacent to calcification in tendon or bursa in all cases (Fig. 5A, 5B, 5C).
|
|
|
|
|
|
|
|
|
Focal increased radionuclide uptake in the affected bone was shown in all 13 patients examined with bone scintigraphy. The degree of uptake was classified as marked in 54% (n = 7) of cases, moderate in 31% (n = 4), and mild in 15% (n = 2). The increased radionuclide activity was eccentric in all cases, which suggests a cortical process.
Microscopically identifiable calcifications were present in all 27 cases with pathology material rereviewed. The most common pattern of calcification on light microscopy was small granular and psammoma-like bodies in the dense connective tissue of a tendon in 67% of cases (n = 18). The remaining 33% (n = 9) of cases displayed histologic features of tumoral calcinosis with large amorphous calcifications and cystic changes. The calcifications were presumably hydroxyapatite crystals, although X-ray diffraction studies or special staining was not performed to confirm this presumption. In addition to the conspicuous calcific deposits, all cases showed varying degrees of stromal fibrosis, chronic inflammation, and histiocytic proliferation with multinucleated giant cells. Other histologic elements observed included hemosiderin deposition, granulation tissue, ganglion cyst–like changes, tenosynovial hyperplasia, and reparative new bone. Of particular interest was the presence of chondroid metaplasia in three cases (11%), leading to the concern for chondrosarcoma by the referring pathologist. Birefringent crystals were not identified in any cases.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pathogenesis of calcium hydroxyapatite crystal deposition in or near a tendon is unclear. Several theories have been proposed, including degeneration of a tendon as a result of recurrent trauma [15], local hypoxia leading to alkaline pH [16], and neurologic or metabolic factors. The leading theory at present is that calcific tendinitis is a primary disorder in susceptible tendons [17]. Degeneration of the tendon has been considered to be the result of the reparative response to crystal deposition rather than the cause of presentation [17]. Local hypoxia is thought to lead to fibrocartilaginous metaplasia, and it is the cartilaginous tissue that produces the radiographically and pathologically identifiable calcification. However, interestingly, only three (11%) of our 27 cases with pathology material for review showed chondroid metaplasia.
Although osseous involvement associated with calcific tendinitis has been reported in multiple small series [3–12], it is unusual despite the high incidence of calcific tendinitis in the general population. The propensity for osseous involvement reported previously at the proximal linea aspera of the femur and at the level of the proximal humerus was also reflected in our series. The reason for this site predilection is uncertain. Calcium hydroxyapatite deposition and resultant inflammatory response in large muscle tendons may produce focal hypervascularity, leading to local bone resorption at the osseous junction. Cortical resorption coupled with large mechanical forces may account for the observed tendency for osseous changes at these characteristic sites. Interestingly, bilateral tendon involvement was seen in three cases, but the osseous involvement and symptoms were unilateral in these cases. Periosteal reaction is presumably caused by the local inflammatory response incited by the deposition of calcium in the affected tendon insertions. The periosteal reaction may have an aggressive appearance that can be confused with malignancy, particularly juxtacortical lesions, if it is not recognized as a manifestation of calcific tendinitis.
High-quality, properly positioned radiographs are necessary to diagnose calcific tendinitis using imaging. Tangential radiographs of the affected cortex are required to detect and characterize the calcifications and their association with cortical or marrow involvement. The previously reported comet-tail appearance of the calcifications [3] helps to confirm their intratendinous location, although this appearance was seen in only 51% of our cases. The findings of soft-tissue calcifications associated with cortical erosion or periosteal reaction are distinctive when the observations are made at the proximal linea aspera and the anterior proximal humeral diaphysis. Not surprisingly, CT was superior to radiography for the detection of both cortical erosion and soft-tissue calcification. In fact, osseous involvement is probably underestimated on radiography; a subsequent study would be required to assess its true incidence. CT, in our experience, is the optimum imaging modality to depict the continuity of the tendinous, cortical, and medullary processes. MRI offers superior evaluation of the marrow involvement, but calcification in the adjacent tendon may not be appreciated, leading to the false presumption of neoplasm and emphasizing the need for correlation with radiographs.
Radiographic evidence of calcific tendinitis extending beyond the cortex into adjacent marrow has not been described in a large series of patients, to our knowledge. Two reports with a total of five patients have detailed radiologic identification of calcific deposits in the tuberosities of the humeral head associated with calcific tendinitis similar to those in our series [10, 11]. One report [10] lacked pathologic descriptions despite confirmation of calcification in cysts in the humeral head at surgery, similar to observations made by Mosley [18]. The other report [11] also described humeral head involvement with presumptive evidence of calcium deposition in the humeral lesser tuberosity of a 58-year-old woman. The presumptive evidence was based on rapid resolution of the radiopacity similar to a case in our series (Fig. 4A, 4B, 4C, 4D, 4E, 4F). The relationship of calcific deposits in the humeral tuberosities and "cyst" formation is not clear. Humeral head "cysts" are not uncommon in patients imaged with MRI for evaluation of rotator cuff disorders or instability and are presumed to be degenerative in nature. Therefore, "cysts" may exist before the development of calcium deposition in the rotator cuff, and marrow involvement may reflect communication with the joint or the rotator cuff. This theory is supported by the relatively large medullary collections that were seen despite small tendon concretions and small cortical erosions in these cases. Alternatively, "cyst" formation may be a result of the intramedullary deposition of calcium and subsequent reparative response, although more reactive marrow changes might be expected than were seen in our patients. Reactive marrow edema without direct extension of concretions has also been described [12].
Bone marrow involvement can be difficult to appreciate on radiographs alone. As one would expect, MRI and CT show marrow involvement by calcific tendinitis more clearly than radiographs do but, paradoxically, may lead to clinical confusion. The depiction of marrow involvement associated with soft-tissue calcification and cortical destruction may raise the clinical concern for neoplasm, particularly chondroid lesions. The recognition of the spectrum of advanced imaging manifestations of calcific tendinitis is critical. This is particularly true with MRI, which is often used as the first imaging modality to evaluate pain or suspected neoplasm. We reemphasize the importance of obtaining radiographic correlation in the setting of suspected neoplasm. Although soft-tissue mass due to inflammation has been described in advanced imaging of acute calcific tendinitis, this finding was not present in our series. The characteristic location of involvement in and near the major tendon attachments of the proximal femur and humerus, as well as the lack of a discrete soft-tissue mass, are important clues leading to the correct diagnosis of calcific tendinitis when the underlying bone is affected. Calcific tendinitis should also be considered in the differential diagnosis of discrete cortical erosions at these muscle insertions even if calcifications are not readily seen on radiography because the calcifications may rapidly and spontaneously reabsorb, leaving the osseous defect behind [19]. Bilateral involvement by calcific tendinitis may also be an important observation in excluding neoplasm.
Treatment of calcific tendinitis is usually limited to the use of nonsteroidal antiinflammatory drugs [7]. Supportive therapy is all that is necessary in most cases because the disease is self-limiting and typically spontaneously resolves. Surgical resection of the concretion or intralesional injection of steroids have also been used but are not necessary in most cases.
Biopsy may be avoided when the radiologic findings are diagnostic of calcific tendinitis despite evidence of osseous involvement. When the clinical situation requires biopsy confirmation, it is imperative that the pathologist recognize chondroid metaplasia as a component of the histology as opposed to suggesting chondroid neoplasm. Indeed, six (43%) of 14 of our patients with an initially rendered diagnosis were referred to exclude chondrosarcoma or other chondroid matrix–producing neoplasm or neoplastic-like process because of the histologic appearance. The presence of psammoma-like calcifications in pathology material confirms the diagnosis of calcific tendinitis, particularly when the histologic findings are correlated with radiologic results. The role of the radiologist cannot be overemphasized in these cases, because the correct diagnosis was not suggested by the referring pathologist in any patient with a rendered diagnosis. In our experience, reparative, traumatic, and inflammatory processes are often suggestive of neoplasm to a pathologist at the initial histologic evaluation.
This study has limitations. The study is retrospective, and clinical data were often incomplete because most of our patients were referred. Imaging and pathologic material for review were not standardized or formatted optimally but were considered adequate in all cases. The true incidence of calcific tendinitis with osseous involvement cannot be determined from our retrospective data. However, despite these limitations, this series studies a large number of cases of calcific tendinitis, and we believe it adds to our understanding of the association with osseous involvement.
In conclusion, we describe osseous manifestations that may be associated with calcific tendinitis in a large series of patients. The aggressive appearance of cortical destruction, periosteal reaction, and soft-tissue calcification often leads to an initially more ominous concern about soft-tissue or juxtacortical sarcoma. Extension of calcific tendinitis to involve the underlying marrow may heighten concern as to a neoplastic cause for the radiologic observations. Identification of soft-tissue calcification in a tendon and a characteristic location with no associated discrete soft-tissue mass should provide the necessary clues to the correct diagnosis. Correlating MRI findings with radiographic findings is important because the identification of calcification in a tendon might not be readily appreciated on MRI alone. Recognition of these unusual manifestations of this common disease may prevent unnecessary biopsy or may guide the pathologist to the correct diagnosis in situations in which biopsy cannot be avoided.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. M. Olsen and F. S. Chew Tumoral calcinosis: pearls, polemics, and alternative possibilities. RadioGraphics, May 1, 2006; 26(3): 871 - 885. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
