| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 The Mallinckrodt Institute of Radiology, Washington University School of
Medicine, 510 S Kingshighway, St. Louis, MO 63110.
2 Department of Orthopedic Surgery, Washington University School of Medicine,
St. Louis, MO.
Received October 11, 2003;
accepted after revision February 2, 2004.
Address correspondence to W. D. Middleton.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Giant cell tumors of the hand typically appear as solid, homogeneous hypoechoic masses with detectable internal vascularity that are associated with the flexor tendons of the fingers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sonography of all cases was performed using commercially available equipment with high-frequency linear transducers (7.5-13 MHz). The different sonograpic machines used were the HDI 3000 and HDI 5000 (ATL), the Elegra (Siemens), and the Acuson Sequoia (Siemens). The sonograms were all obtained by one of two board-certified radiologists with 5 or more years of experience in musculoskeletal sonography. All patients underwent color or power Doppler sonography to assess vascularity. Two radiologists who were aware of the diagnosis jointly reviewed the sonographic images retrospectively, and a consensus regarding the sonographic features was achieved. Data collection included tumor site (tendon involved and location of joint nearest the center of the lesion), size (average of three orthogonal dimensions), echogenicity (with respect to the predominant adjacent tissues), bone erosions (defined as a concave cortical defect in the surface of the adjacent bone), calcification (defined as hyperechoic regions with shadowing), increased or decreased through-transmission, and cystic areas (defined as discrete anechoic regions within the lesion). The length of contact with the tendon was measured on images obtained along the long axis of the tendon, and the maximum length was recorded in millimeters. The circumferential involvement was measured on images oriented along the short axis of the tendon, and the maximum contact was measured in degrees. The detectability of lesional blood flow was assessed on the basis of color or power Doppler images. Tumors were considered hypervascular if they were substantially more vascular than adjacent tissues. The distribution of flow was assessed and characterized as peripheral, central, or both. Dynamic scans were obtained as part of our routine, and the presence of lesion motion with respect to tendon motion was assessed in real time. Our institutional review board approved this study.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ten of the 12 tumors involved the flexor tendon sheath of the fingers, and the remaining two involved the extensor sheath. The digits involved were the first (four tumors), second (three), third (one), fourth (three), and fifth (one). The tumors were located near the metacarpal–phalangeal joints in three cases, at the proximal interphalangeal joints in four cases, and at the distal interphalangeal joints in five cases. Tumor size ranged from 3 to 15 mm with a mean of 9.7 mm.
All tumors were hypoechoic (Fig. 1). Ten were homogeneous and two were heterogeneous. Posterior acoustic enhancement was seen in four cases (Fig. 2). Bone erosions were noted in two patients, occurring at the metacarpal–phalangeal joint in both cases (Fig. 3). None of the lesions showed hyperechoic regions, cystic regions, calcifications, or shadowing.
|
|
|
All the lesions were in contact with a tendon (Figs. 1, 2, 3, 4A, and 4B). The length of contact with the tendon on long-axis views ranged from 1 to 11 mm, with a mean of 5.7 mm. The amount of circumferential contact with the tendon on short axis views ranged from 30° to 360°, with a mean of 136°. In three cases, a component of the tumor grew between the tendon and the bone. In all cases, it was noted that the tumors did not move with the tendon when the affected digit was flexed or extended.
|
|
Internal vascularity was seen in 11 tumors, and five were hypervascular (Fig. 4B). Five tumors showed peripheral flow, one showed central flow, and the remaining five had both central and peripheral vascularity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because the lesions arise from the tendon sheath, close contact with a tendon is expected. This was a constant finding in our study. All giant cell tumors were in contact with a tendon to a lesser or greater extent, with the average length of contact being 6 mm and the mean circumferential tendon involvement being 140°. In one patient, the tendon was completely encased by the tumor. However, despite the close contact, the tumor did not move with the tendon when the tendon was flexed and extended in any of our cases. This is expected because the lesion arises from the sheath and not from the tendon itself.
In most reports, giant cell tumors of the tendon sheath involve the volar surface of the fingers more often than the dorsal surface [3, 7]. This was true in our patients, with flexor tendon involvement being five times more common than extensor tendon involvement. The predominance of involvement near the distal joints of the fingers in our group of patients is typical [7].
The sonographic appearance of giant cell tumor has been described in a largely anecdotal fashion as both hypoechoic [10–13] and hyperechoic [14, 15]. All the lesions in our study were hypoechoic. Despite previous descriptions [15, 16], only two cases (17%) in our series were heterogeneous. The others were homogeneous. Also contrary to a previous report [15], posterior acoustic enhancement was seen in a limited number of our cases. None of the lesions in our study showed cystic elements or hyperechoic areas. No calcifications were encountered. Detectable bone erosions were present in two patients. It is important to realize that giant cell tumors are superficial lesions usually located within the first centimeter of the field of view. Therefore, the use of high-frequency transducers to optimize imaging is necessary. Additionally, when available, transducers equipped with acoustic lenses or 2D arrays that focus the sound in the near field are now available and further optimize scanning of these lesions.
Giant cell tumors typically showed vascularity on color Doppler sonography. Only one case was completely avascular on color Doppler sonography, and in nearly half they were distinctly hypervascular. No consistent pattern of vascularity was present. As with the gray-scale evaluation, the use of high transmit frequencies is important when performing color and power Doppler sonography to maximize sensitivity to blood flow.
The major clinical differential in patients with palpable masses along the flexor tendons of the fingers is giant cell tumor and ganglionic cyst. Distinguishing these two lesions is useful because giant cell tumor can only be treated with resection, although ganglionic cysts can be treated with both resection and needle puncture with steroid injection. The sonographic appearance of ganglion cysts has been described [17] and differs from the findings we encountered in these patients with giant cell tumor. All the giant cell tumors in our series appeared solid, and ganglionic cysts are reported to appear cystic. One potential pitfall that has been described is a ruptured ganglionic cyst, which can appear solid and simulate a giant cell tumor [18]. In addition to characterizing giant cell tumor as solid-appearing, sonography provides information about the extent of contact and circumferential involvement of the tendon that can be helpful during the surgical dissection of giant cell tumor.
In addition to sonography, MRI is capable of identifying and characterizing giant cell tumor of the tendon sheaths. [19]. Although we made no attempt to compare sonography and MRI, we believe that the diagnosis can be strongly suggested with either technique. Given the relative cost, availability, and time necessary to perform the examinations, our hand surgeons routinely image patients with expected giant cell tumors using sonography and reserve MRI for problem cases.
As with all retrospective studies, this one has limitations. Selection bias is a possible confounding factor. A single hand surgeon referred most of the cases in this series, and not all patients with giant cell tumors were examined with sonography. However, we know of no reason to suspect significant differences in the appearance of giant cell tumors seen by different clinicians, nor in patients who may or may not undergo sonography. Another limitation is that no defined, standardized protocol was used to scan all these patients. This might lead to lack of uniformity in some of the measurements. However, all patients were imaged with both longitudinal and transverse views, and measurements were routinely obtained in orthogonal planes. We believe that differences due to this limitation would almost certainly be minor and would not change the overall impression of our analysis.
In conclusion, giant cell tumors of tendon sheath have a characteristic sonographic appearance and occur in predictable locations. The diagnosis should be strongly considered whenever a solid-appearing, hypoechoic mass with detectable blood flow is identified adjacent to the flexor tendons of the fingers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Bianchi, D. D. Santa, T. Glauser, J.-Y. Beaulieu, and J. van Aaken Sonography of Masses of the Wrist and Hand Am. J. Roentgenol., December 1, 2008; 191(6): 1767 - 1775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Murphey, J. H. Rhee, R. B. Lewis, J. C. Fanburg-Smith, D. J. Flemming, and E. A. Walker From the Archives of the AFIP: Pigmented Villonodular Synovitis: Radiologic-Pathologic Correlation RadioGraphics, September 1, 2008; 28(5): 1493 - 1518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, J. Tang, and Y. Luo The Value of Sonography in Diagnosing Giant Cell Tumors of the Tendon Sheath J. Ultrasound Med., October 1, 2007; 26(10): 1333 - 1340. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. C. Roberts, P. T. Liu, and F. S. Chew Imaging Evaluation of Tendon Sheath Disease: Self-Assessment Module Am. J. Roentgenol., March 1, 2007; 188(3_Supplement): S10 - S12. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
