| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 Departments of Radiology, Division of Pediatric Radiology (S.C., D.P.F., G.S.B.)
2 Medicine, Division of Dermatology (N.S.P., C.R.S.)
3 Pathology, Division of Diagnostic Pathology (C.R.S.), Duke University Medical Center, Box 3808, Erwin Rd, Durham, NC 27710
4 Department of Radiology, Children's Hospital, Boston, Mass (T.L.).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
MATERIALS AND METHODS: MR imaging studies and clinical records in six children aged 2
–4 years in whom SGA was diagnosed between 1993 and 1997 were retrospectively reviewed. All MR imaging examinations included T1-weighted and fast spin-echo T2-weighted sequences. Three children received intravenous contrast material. The diagnosis was established by using excisional biopsy results, with confirmation by means of characteristic features in mucin-stained specimens.
RESULTS: All children presented with a nontender, nonmobile mass. The lesion in all six children was confined to the pretibial (n = 5) or prepatellar (n = 1) soft tissues. MR imaging characteristics were uniform. Masses were subcutaneous in location and had ill-defined margins. On T1-weighted images, the lesion was isointense or slightly hyperintense to muscle. On T2-weighted images, the signal intensity was more heterogeneous, but the lesion was predominantly hyperintense. In three children who received contrast material, the lesion enhanced nearly homogeneously.
CONCLUSION: Although SGA can often be diagnosed on the basis of clinical characteristics, MR imaging may be requested by practitioners unfamiliar with the lesion. In these cases, with recognition of typical MR imaging features, a limited biopsy and specific histologic preparation that includes mucin staining can be recommended.
Index terms: Children, extremities, 44.249 • Extremities, MR, 44.121411, 44.121415, 44.12143, 45.121411, 45.121415, 45.12143 • Extremities, abnormalities, 44.249, 45.249 • Extremities, neoplasms, 44.369, 45.369 • Granuloma, 44.249, 45.249
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
SGA occurs most frequently in healthy young children, in whom it manifests as a painless, nonmobile mass usually found in a lower extremity. Because of the subcutaneous location, the diagnosis of SGA can be more difficult than that with other (cutaneous) types of granuloma annulare, and the differential considerations include both benign and malignant processes. The manifestation of SGA also may be atypical and includes pain, rapid enlargement, or unusual location. Consequently, the diagnosis of SGA is more uncertain, particularly for those unfamiliar with the group of annular granulomas.
Diagnostic imaging may be requested to help evaluate a nonspecific soft-tissue mass. Magnetic resonance (MR) imaging is particularly useful in the characterization of soft-tissue masses; its use often leads to a specific diagnosis. To this point, to our knowledge the description of the MR imaging features of SGA has been limited to anecdotal accounts in the surgical, dermatology, or pathology literature (1,3). Therefore, the purpose of this investigation was to determine the MR imaging features of SGA in children. In addition, the literature describing imaging features of SGA was reviewed to determine the role of radiology in the evaluation of these lesions.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
All patients underwent MR imaging with a 1.5-T system (Signa LX; GE Medical Systems, Milwaukee, Wis). All examinations included axial T1-weighted (repetition time msec/echo time msec, 250–500/11–22) and T2-weighted fast spin-echo (repetition time msec/effective echo time msec, 3,000–4,566/85–112) fat-saturated sequences. Other sequences included a combination of sagittal and coronal T1-weighted (250–500/11–22), fast spin-echo T2-weighted (3,000–4,566/85–112 [effective]), or inversion recovery (inversion time, 150 msec) sequences. After intravenous administration of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) in three children, fat-suppressed T1-weighted axial and coronal sequences were performed.
All MR imaging studies were reviewed, and a consensus was rendered by two pediatric radiologists (D.P.F., G.S.B.) with experience in pediatric musculoskeletal MR imaging. Lesion characteristics previously reported (6) for MR evaluation of soft-tissue masses were recorded; these included anatomic location, margination, signal intensity characteristics, presence of edema, and involvement of adjacent structures.
All children underwent excisional biopsy, and the diagnosis of SGA was determined on the basis of subsequent histopathologic analysis results. SGA was defined by the presence of dermal and subcutaneous collections of epithelioid histiocytes and/or multinucleated giant cells surrounding zones of altered, basophilic, hypocellular connective tissue containing mucin deposits (as demonstrated with the colloidal iron stain) (Fig 1).
|
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
–4 years (mean age, 3.0 years). All patients were referred for imaging by surgical services (orthopedic or pediatric surgery). All children were healthy except for the presence of a nontender mass. This mass was present for 1–6 months in four patients. The duration of the lesion was uncertain in two children. In one of these latter two patients, the mass was discovered after unrelated trauma to the leg. In the five children in whom information about the growth of the mass was available, the mass was either stable (n = 2) or slowly growing (n = 3). All masses were located in the lower extremity; five were pretibial and one was prepatellar in location. Four were located in the left lower extremity, and two were located in the right lower extremity. All children had a single mass, with the exception of the girl with a prepatellar lesion who also had a smaller lesion, which was otherwise identical, located in the soft tissues at the right olecranon.
Laboratory evaluation was performed only in the girl with the prepatellar mass. The erythrocyte sedimentation rate was mildly elevated (26 mm/h), but antinuclear antibody and rheumatoid factor analysis results were negative. These were obtained because a rheumatoid nodule was considered in the patient's preoperative evaluation. An excisional biopsy followed MR imaging (Fig 2). The initial histopathologic diagnosis was "rheumatoid nodule." After a review of the case, mucin staining was performed, and the diagnosis was changed to SGA.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
The lesion most often manifests as a painless, nonmobile, subcutaneous mass with no associated overlying cutaneous abnormality. The mass may rapidly enlarge over the course of weeks but may be present for many months. A history of antecedent trauma is extremely rare. SGA occurs in the lower extremities (often in a pretibial location) in at least 65% of cases (4,5) and is most often a solitary lesion. Other sites of involvement include the upper extremity, buttocks, and face or scalp (3,5,7,8). Despite this distribution, trauma is not an established cause and the etiology of SGA is unclear (3). The lesion will spontaneously regress, even if it is recurrent. Local or distal recurrence occurs in 19%–75% of cases (3,4,10).
Granuloma annulare, including the subcutaneous type, is seen most often by dermatologists. When the manifestation is cutaneous, a biopsy is not necessary to facilitate the diagnosis of coexistent subcutaneous lesions. When lesions are subcutaneous only or the location or history (eg, rapid enlargement or pain) is atypical, however, a biopsy without imaging is appropriate (3,5). Imaging for evaluation of a subcutaneous lesion is more frequent when children are seen by nondermatologic specialists. In fact, in one series of 12 children (1), the number of physicians who evaluated children before SGA was diagnosed ranged up to five, with a mean of three (1). It is in these situations that diagnostic imaging may be requested to determine the nature of a subcutaneous mass. In our series, all six of the children were referred by surgical specialists for imaging of a nonspecific subcutaneous mass.
The differential diagnostic considerations for subcutaneous masses in children are extensive; they include posttraumatic, infectious, neoplastic, and syndromic or systemic disease-associated categories. Of these, the subcutaneous nodules of rheumatoid arthritis (when the diagnosis of rheumatoid arthritis is not clinically apparent) can be difficult to distinguish clinically from SGA. Clinical and laboratory features can help in this differentiation, because children with SGA have no symptoms other than a mass, and their serum laboratory evaluation results (eg, rheumatoid factor) are normal. In a setting where the diagnosis of rheumatoid arthritis is unclear, the imaging features can be identical (11), and histopathologic findings can be very similar. In this situation, positive mucin staining is a characteristic of SGA (8). Although a relationship between SGA and rheumatoid arthritis has been suggested in the past, reports in the contemporary literature support the contention that SGA virtually never progresses to rheumatoid arthritis (1,3,5,9).
Some soft-tissue masses can be sufficiently diagnosed on the basis of clinical features alone. For example, rheumatoid nodules, foreign-body reactions, fat necrosis, hematomas, abscesses, and inflammatory (ie, infectious) granulomas are often suspected on the basis of a history of trauma, systemic symptoms (ie, fever or other constitutional symptoms), or laboratory abnormalities (elevated rheumatoid factor level or erythrocyte sedimentation rate, leukocytosis). When the etiology of a soft-tissue mass is clinically uncertain and further noninvasive evaluation is indicated, imaging is warranted.
MR imaging is well recognized as providing the greatest amount of diagnostic information about the characteristics of soft-tissue abnormalities in children (6). Despite the primary role of MR imaging for soft-tissue masses, descriptions of the MR features of SGA have been anecdotal. MR imaging evaluation of SGA in children has been reported (1,3,8) at least three times in the past (Table 1). The MR imaging features that were described were similar to those we encountered in our series; however, no images were provided (1). In a series of three patients with lesions involving the scalp (8), performance of MR imaging was reported in only one patient, and the description was limited to "soft-tissue swelling." In a recent series of 47 children with SGA (3), the authors commented that two patients underwent CT and MR imaging; the lesions were described only as being subcutaneous in location.
In our six cases of SGA, all lesions had indistinct margins, with signal intensity abnormalities that extended into adjacent subcutaneous fat. The pathologic correlate to the ill-defined margin is probably the poorly circumscribed inflammatory response. In all of our cases, the signal intensity on T1-weighted images was equal to or slightly greater than that of muscle, with heterogeneous but predominantly high signal intensity on T2-weighted images. In three lesions, there was contrast enhancement, which was homogeneous in one lesion and minimally heterogeneous in the other two. The enhancement likely reflected the inflammatory nature and increased vascularity of SGA. The explanation for the extension of abnormal signal intensity on T2-weighted images beyond enhancement in one child is uncertain; this signal intensity abnormality may have represented associated edema. Enhancement on CT scans has been reported to be more variable. Argent et al (4) noted enhancement in only one of four lesions evaluated with CT. The reason for this variability is not known. However, since there were no data available about precontrast attenuation in the SGA lesions evaluated with CT, subtle enhancement could have been overlooked.
Features of SGA seen on images obtained with other modalities have also been described. Conventional radiography demonstrates a nonspecific soft-tissue mass without calcification or bone involvement (1,3,4). Sonography was also mentioned in reports of two investigations (1,4), and lesions were noted to be hypo- or anechoic. Bone scintigraphic images obtained in one study (8) were reported to be normal. In the series by Argent et al (4), the lesions were all found in the lower extremities. As with the MR imaging findings in our six patients, lesions were localized to the subcutaneous tissues, with no extension deep to the fascia or involving underlying bone. Argent et al (4) concluded that although this appearance was nonspecific, SGA should be included in the differential diagnosis of subcutaneous masses.
Other subcutaneous masses that are relatively asymptomatic include hemangioma, lymphangioma, lipoma, fibroma or fibromatosis, granuloma, foreign body reaction, infection, fat necrosis, and vascular malformations (5,12). The following MR features can help distinguish some of these entities: fat signal intensity (lipomas, hemangiomas), large cystic spaces (lymphangiomas), and prominent vascularity (vascular malformations). The appearance on MR images of fibrous tissue, granulomas, true rheumatoid nodules, and fat necrosis can have an appearance similar to that of SGA (6,11,13,14).
SGA was offered as a diagnostic consideration in two of six children in our series. In four individuals, the MR imaging interpretation at the time of diagnosis did not include SGA as a consideration. Causes listed were reaction to a foreign body, fibromatosis, chronic infection, hemangioma, and fat necrosis. In one patient, the MR images from another institution were reviewed after biopsy. Consideration of SGA as the cause of a subcutaneous mass is important from the standpoints of both biopsy technique and specimen evaluation.
All of our patients underwent an excisional biopsy. An excisional (as opposed to limited incisional) technique was also elected for the two patients in whom SGA was included in the differential diagnosis. However, a wide excision is not necessary to help diagnose SGA. Complications of excisional biopsy to evaluate for SGA include infection and scar formation (1). In one investigation of 12 children (1), parents were unhappy with the cosmetic results after excisional biopsy in four (25%) cases. In another article (9), skin grafting was reported to be necessary after wide excision. Fine-needle aspiration is occasionally used for soft-tissue masses. However, fine-needle aspiration results are not reliable for SGA, because the typical palisaded architecture may not be identifiable. In a recent discussion of fine-needle aspiration to evaluate for SGA (1), it was noted that this method may lead to an incorrect diagnosis of malignancy. Finally, mucin staining of the specimen may not be routine and is helpful in confirming the diagnosis of SGA (7). To direct appropriate specimen preparation, therefore, it is important for the pathologist to be aware that SGA is a diagnostic consideration. In one of our patients (Fig 2), the prepatellar nodule was misdiagnosed as a rheumatoid nodule until the specimen was reexamined with mucin staining.
What is the role of diagnostic imaging in the evaluation of a lesion that may be subcutaneous in location? As mentioned before, a mass (particularly one that manifests in early childhood) that is clinically suspected to be SGA may be diagnosed on the basis of clinical features (3) or the results of limited biopsy. In these circum-stances, imaging is not necessary. In those children who have a mass with clinical features atypical of SGA (or in circumstances where clinical features of SGA are not recognized), MR imaging can be helpful. If the MR imaging features are not specific (eg, lipoma, lymphangioma, hemangioma, vascular malformation), then a biopsy should be considered (12). If the lesion has indistinct margins but a central mass, is limited to the subcutaneous tissues, and is isointense to muscle on T1-weighted images and has mixed but predominantly high signal intensity on T2-weighted images, then a limited (nonaspiration) incisional biopsy with mucin staining can be recommended. An excisional biopsy is not necessary.
In summary, with recognition of the clinical and imaging features associated with the entity of SGA in children, the radiologist may have a pivotal role in determining the most appropriate subsequent evaluation.
|
ADDENDUM |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
|
Footnotes |
|---|
From the 1997 RSNA scientific assembly.
Abbreviation: SGA = subcutaneous granuloma annulare
Author contributions: Guarantors of integrity of entire study, D.P.F., S.C.; study concepts, D.P.F., G.S.B., S.C.; study design, D.P.F., G.S.B.; definition of intellectual content, D.P.F., G.S.B.; literature research, S.C.; clinical studies, N.S.P., C.R.S.; data acquisition and analysis, T.L., D.P.F., S.C., G.S.B., C.R.S.; manuscript preparation, D.P.F., S.C.; manuscript editing and review, D.P.F., S.C., G.S.B., T.L., C.R.S., N.S.P.
Received February 17, 1998;
revision requested April 26, 1998; revision received August 3, 1998;
accepted October 19, 1998.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONADDENDUMReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Hacihamdioglu, A. Ozcan, and S. Kalman Subcutaneous Granuloma Annulare in a Child: A Case Report Clinical Pediatrics, April 1, 2008; 47(3): 306 - 308. [PDF]
|
|
|
|
 |
|
 M. F. Blacksin, D.-H. Ha, M. Hameed, and S. Aisner Superficial Soft-Tissue Masses of the Extremities RadioGraphics, September 1, 2006; 26(5): 1289 - 1304. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Pham, D. P. Fessell, J. E. Femino, S. Sharp, J. A. Jacobson, and C. W. Hayes Sonography and MR Imaging of Selected Benign Masses in the Ankle and Foot Am. J. Roentgenol., January 1, 2003; 180(1): 99 - 107. [Full Text] [PDF]
|
|
|
|
 |
|
 I. GRANT, P. MOHAMMED, and P. J. MAHAFFEY Granuloma Annulare in the Hand J Hand Surg Eur Vol., December 1, 2002; 27(6): 556 - 558. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
