Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2231010977
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Theumann, N. H.
Right arrow Articles by Drapé, J.-L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Theumann, N. H.
Right arrow Articles by Drapé, J.-L.
(Radiology 2002;223:143-151.)
© RSNA, 2002

Musculoskeletal Imaging

Recurrent Glomus Tumors of Fingertips: MR Imaging Evaluation1

Nicolas H. Theumann, MD, Sophie Goettmann, MD, Dominique Le Viet, MD, Donald Resnick, MD, Christine B. Chung, MD, Jacques Bittoun, MD, PhD, Alain Chevrot, MD and Jean-Luc Drapé, MD, PhD

1 From the Department of Radiology B, CHU Cochin, AP-HP-Université Paris V, 27 rue du fg Saint-Jacques, 75014 Paris, France (N.H.T., A.C., J.L.D.); Department of Radiology, CHUV, Lausanne, Switzerland (N.H.T.); Department of Radiology, Veterans Administration Medical Center, San Diego, Calif (N.H.T., D.R., C.B.C.); Department of Dermatology, CHU Bichat, Université Paris IX, France (S.G.); Institut de la Main, Clinique Jouvenet, Paris, France (D.L.V.); and Centre Inter Etablissement de Résonance Magnétique (CIERM), Hôpital de Bicêtre, AP-HP-Université Paris XI, Le Kremlin-Bicêtre, France (J.B., J.L.D.). From the 1997 RSNA scientific assembly. Received May 29, 2001; revision requested June 26; revision received September 27; accepted October 16. Supported by the Swiss Radiological Society. Address correspondence to J.L.D. (e-mail: jean-luc.drape@cch.ap-hop-paris.fr).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To determine the magnetic resonance (MR) imaging findings in recurrent glomus tumors of the fingertips.

MATERIALS AND METHODS: Twenty-four consecutive patients with recurrent pain after previous excision of a glomus tumor of the fingertip underwent MR imaging studies and surgery. T1-weighted spin-echo MR images were obtained in each patient before and after intravenous injection of contrast material; T2-weighted spin-echo and three-dimensional gradient-recalled echo images were also obtained. MR angiography was performed in four patients. Postsurgical histopathologic analysis revealed recurrent glomus tumors in 22 patients. Signal intensity, enhancement, and margins of the scar tissue and the recurrent tumors at MR were assessed.

RESULTS: The postsurgical scars were depicted in 21 (88%) of 24 patients with all sequences but were best demonstrated on gradient-recalled echo MR images. Seven patients had undergone multiple surgical procedures and had extensive scar tissue and, in one case, a neuroma. In all patients, MR imaging revealed a nodule compatible with the diagnosis of a recurrent glomus tumor. In 13 (54%) of 24 patients, the nodule had typical features of a glomus tumor. In eight (33%) of 24 patients, the tumors had low signal intensity or isointensity compared with the nail bed on T2-weighted images. In six (25%) of 24 patients, the tumors had faint enhancement after intravenous gadolinium chelate administration. The margins of the tumors were blurred by scar tissue in nine of 24 cases.

CONCLUSION: MR imaging can aid in the evaluation of recurrent glomus tumors.

© RSNA, 2002

Index terms: Fingers and toes, MR, 436.121411, 436.121412 • Fingers and toes, neoplasms, 436.369 • Paraganglioma, 436.369


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
The glomus body is a highly specialized arteriovenous anastomosis responsible for thermoregulation. Glomus bodies consist of an afferent arteriole, a tortuous arteriovenous anastomosis, a system of collecting veins, and a neurovascular reticulum that regulates the flow of blood through the anastomosis. Glomus bodies are present in the stratum reticularis of the dermis throughout the body, but they are more numerous in the digits, the palms, and the soles of the feet. In 1812, Wood described the glomus tumor as a painful subcutaneous "tubercle" (1). Seventy-five percent of glomus tumors occur in the hand, especially in the fingertips, and particularly in the subungual area. Of all hand tumors, 1%–2% are glomus tumors (2). Multiple glomus tumors are present in 2.3% of cases (3). The average age of patients at diagnosis ranges from 30 to 50 years (4). Men are affected less frequently than are women.

Glomus tumors are characterized clinically by intense, often pulsating pain that may be spontaneous or provoked by a slight trauma. If the tumor cannot be localized clinically or on radiographs, ultrasonography (US) may be performed. US performed with a high-frequency transducer depicts tumors as small as 3 mm in diameter, particularly in the pulp of the finger (5,6). However, the curvature of the nail plate may create artifacts in the lateral nail folds (folds of skin covering the radial and ulnar margins of the nail plate), and small and flattened subungual lesions are difficult to detect with US (5,7). Magnetic resonance (MR) imaging has been shown to aid in the diagnosis of glomus tumors of the fingertip (8–13), with high-spatial-resolution MR imaging offering more information about the location and histologic type of glomus tumors (14,15). Most glomus tumors are highly vascular and show high signal intensity on T2-weighted spin-echo MR images and strong enhancement after injection of a gadolinium compound (14).

The only treatment for glomus tumors is surgical removal. The incidence of tumor recurrence after surgical excision ranges from 5% to 50% (4,16–18). The ability of MR imaging to enable detection of recurrent glomus tumors is not well described in the literature, although there are multiple descriptions of the MR imaging features of primary glomus tumors (10,14,15).

The purpose of our study was to determine the MR imaging findings in recurrent glomus tumors of the fingertips.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Patients
Twenty-four consecutive patients with recurrent pain after excision of a glomus tumor of the fingertip underwent MR imaging and surgery at the Centre Inter Etablissement de Résonance Magnétique (CIERM) or at the CHU Cochin during the past 7 years (between November 1993 and February 2001). In a single patient, the first surgical procedure did not reveal a subungual tumor. There were 23 women and one man. The mean age was 44 years (range, 22–74 years). The time of the recurrence of symptoms ranged from 3 months to 20 years after the initial surgery. The involved rays were the thumb (n = 2) and index (n = 7), middle (n = 2), ring (n = 8), and little (n = 5) fingers. Most of the patients (17 [71%] of 24) underwent surgery that was performed with a lateral (radial or ulnar) approach. Seven patients underwent surgery multiple times (range, 2–5). After our MR imaging examination, surgery was performed in all cases. Pathologic results revealed 22 recurrent glomus tumors, a neuroma, and a hemangioendothelioma.

MR Imaging
All MR imaging examinations were performed with a 1.5-T MR unit (Signa; GE Medical Systems, Milwaukee, Wis). A high-spatial-resolution module for skin imaging was used in 17 patients. A surface gradient coil was used in place of the anteroposterior gradient coil of the imaging system to increase the gradient strengths. The surface gradient coil consisted of a one-sided pair of saddle-shaped coils. The maximum gradient strength was obtained perpendicular to the finger pad and attained 42 mT/m. A voxel size of 117 x 234 x 700–3,000 µm3 was used, with a 3-cm field of view, a 256 x 128 acquisition matrix, and a 16-kHz receiver bandwidth. A surface radio-frequency coil with a 1.5-cm radius was placed at the center of the gradient coil. The more technical specifications of this coil have previously been described (19). The remaining seven patients were imaged with a commercial phased-array wrist coil. The spatial resolution was 156–195 x 312–390 x 3,000 µm3, with a 4–5-cm field of view, a 256 x 128 acquisition matrix, and a 15.6-kHz receiver bandwidth.

All patients underwent imaging in a supine or prone position with the arm extended above the head. Transverse T2-weighted spin-echo (repetition time msec/echo time msec, 2,000/80) or short inversion time inversion-recovery (repetition time msec/echo time msec/inversion time msec, 2,060/14.2/150; echo train length, six) images were obtained first. Transverse and sagittal T1-weighted spin-echo (500/20) or fast spin-echo (560/14.2; echo train length, two) sequences were performed before and after intravenous injection of 0.1 mmol/kg gadoterate meglumine (Dotarem; Guerbet, Aulnay-sous-Bois, France). Fat suppression was applied to postcontrast fast spin-echo images. Section thickness was 3 mm, with a 0.3–1.0-mm gap. Three-dimensional (3D) gradient-recalled echo (GRASS; GE Medical Systems) imaging (50/12, 40° flip angle) consisted of 28 transverse or sagittal 0.7-1.0-mm-thick contiguous sections.

MR angiography has been performed in our institution (CHU Cochin) since 1999 and was performed in four patients in this study with 3D coronal spoiled gradient-echo (FSPGR; GE Medical Systems) sequences (9.9/2.4/40 [spectral inversion of lipids was used for fat suppression]; 20° flip angle; number of signals acquired, one; 31.2-kHz bandwidth; a square field of view of 14 x 14 cm; a matrix of 256 x 256; 20 partitions of 2 mm) and elliptical k-space acquisition. Zero filling was used to interpolate the section thickness to 1 mm and the matrix to 512 in image reconstruction. Acquisition time was 1 minute 12 seconds. A power injector (Spectris; Medrad, Philadelphia, Pa) was used to administer a 2-mL test dose of gadoterate meglumine followed by a 20-mL saline flush and to determine the time to arterial peak. A precontrast series was initially performed for image subtraction. Postcontrast imaging was performed after intravenous administration of 0.2 mmol/kg gadoterate meglumine followed by a 20-mL saline flush; both substances were injected at a rate of 1 mL/sec. Two successive sequences were performed to assess the signal behavior of the tumor over time. Image processing was done by means of subtracting the precontrast data set from the postcontrast data set at a Sparc workstation (Sun Microsystems, Palo Alto, Calif). The subtracted images were used for targeted maximum intensity projections (MIPs) that "cut away" overlying veins.

Image Interpretation and Data Analysis
All MR images were retrospectively reviewed in consensus by two musculoskeletal radiologists (J.L.D., N.H.T.) with 14 and 5 years of experience who were blinded to the surgical findings. The institutional review boards of our institutions (CIERM, CHU Cochin) did not require their approval or patient informed consent for this study. All ill-defined signal intensity abnormalities in the surgical bed were considered to be scar tissue. Different patterns of the effects of previous surgery were recorded. These included the signal intensity behavior of the scar tissue (ie, whether its signal intensity was lower than, equivalent [isointense] to, or higher than that of the nail bed), the presence of any artifacts (few or many) on images obtained with different sequences, the presence of enhancement after intravenous injection of contrast material (recorded as no enhancement, slight enhancement, or strong enhancement compared with the nail bed), the definition of the margins of the scar (eg, not visible or ill defined), and the location of the scar (ie, transungual [on the median part of the nail bed] or lateral [on the radial or ulnar side of the nail bed]).

The different patterns of the recurrent glomus tumors were also noted. The location of each tumor in the sagittal plane was characterized as follows: the submatrical area (ie, the dermis beneath the nail matrix), the nail bed, or the hyponychium (ie, the distal portion of the nail bed). The location of each tumor in the transverse plane was characterized as follows: the medial nail bed, the lateral (radial or ulnar) nail bed, the lateral nail fold, the rima ungualum (ie, the bony ligamentous tunnel between the nail bed and the pulp), and the pulp. The margins of the tumor were classified as well defined or, if the whole circumference of the tumor could not be discerned on the MR images, as ill defined. The largest diameter of the tumor was measured in millimeters in the transverse and sagittal planes. The signal intensity of each tumor on T1- and T2-weighted spin-echo MR images compared with the signal intensity of the normal nail bed was characterized as low, equivalent (isointense), or high and as either homogeneous or heterogeneous. The appearance of the tumor after intravenous injection of contrast material was characterized as no enhancement, slight enhancement, or strong enhancement compared with the appearance of the nail bed. On MR angiograms, an early nodular enhancement in the arterial phase was considered to be diagnostic of a recurrent tumor. The results of postsurgical histologic examinations of the recurrent tumors were reviewed (S.G., D.L.V.) to note the amount of infiltration of the tumor by scar tissue.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Scar Tissue
Scar tissue in the surgical beds (appearing as areas of low signal intensity on T1-weighted images and as areas of low or high signal intensity on T2-weighted spin-echo MR images) was depicted in 21 (88%) of 24 patients and was more obvious on gradient-recalled echo MR images; artifacts were present on the gradient-recalled echo images in all 21 patients (Table 1, Fig 1). In eight (33%) of 24 patients, the gradient-echo images had many artifacts; in four of the eight patients, T1-weighted spin-echo images also had obvious artifacts. Three of these four patients had undergone two previous surgical procedures. The scars of two patients demonstrated only high signal intensity on T2-weighted spin-echo and gradient-echo images (Fig 2). The scar margins were ill defined in all 21 cases, and the scars were located most often in the lateral nail fold. The scar tissue showed no enhancement in 19 (90%) of 21 patients and slight or strong enhancement after intravenous gadolinium chelate administration in two patients (Fig 3). These two patients (patients 6 and 19) had undergone multiple previous surgical procedures with lateral and transungual approaches and had extensive scar tissue. In one patient (patient 19), however, a neuroma in the pulp was revealed that had previously been misdiagnosed as a recurrent glomus tumor (Fig 3).


View this table:
[in this window]
[in a new window]

  		TABLE 1. MR Imaging Findings in Scar Tissue

 


View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. Patient 13. Scar tissue in the nail bed 3 years after a transungual resection of a glomus tumor of the second finger. (a) Transverse T1-weighted spin-echo MR image (600/24) depicts miniscule low-signal-intensity artifacts (arrows) in the nail bed. (b) These artifacts (arrows) appear much larger on the transverse 3D gradient-recalled echo image (50/12, 40° flip angle).

 


View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. Patient 13. Scar tissue in the nail bed 3 years after a transungual resection of a glomus tumor of the second finger. (a) Transverse T1-weighted spin-echo MR image (600/24) depicts miniscule low-signal-intensity artifacts (arrows) in the nail bed. (b) These artifacts (arrows) appear much larger on the transverse 3D gradient-recalled echo image (50/12, 40° flip angle).

 


View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Patient 8. Postoperative scar tissue 1 year after a lateral excision of a glomus tumor of the thumb. Transverse T2-weighted spin-echo MR image (2,000/80) shows ill-defined scar tissue with low signal intensity in the lateral nail fold (arrowhead) and higher signal intensity in the lateral part of the nail bed (arrow).

 


View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Patient 19. Postoperative scar tissue 2 years after a fifth lateral and transungual resection of a glomus tumor of the ring finger. (a) Transverse postcontrast T1-weighted spin-echo MR image (600/24) depicts scar tissue with low signal intensity (arrowheads) in the nail fold and enhancement of fibrous tissue in the thickened nail bed (*). Note the sequela (arrow) of the dorsal cortical resection of the phalanx. (b) Transverse T1-weighted spin-echo MR image (600/24) and (c) transverse T2-weighted spin-echo MR image (2,000/80) at the level of the distal interphalangeal joint depict a 3-mm neuroma (arrowheads) along the course of the medial collateral nerve. Arrow = contralateral collateral nerve.

 


View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Patient 19. Postoperative scar tissue 2 years after a fifth lateral and transungual resection of a glomus tumor of the ring finger. (a) Transverse postcontrast T1-weighted spin-echo MR image (600/24) depicts scar tissue with low signal intensity (arrowheads) in the nail fold and enhancement of fibrous tissue in the thickened nail bed (*). Note the sequela (arrow) of the dorsal cortical resection of the phalanx. (b) Transverse T1-weighted spin-echo MR image (600/24) and (c) transverse T2-weighted spin-echo MR image (2,000/80) at the level of the distal interphalangeal joint depict a 3-mm neuroma (arrowheads) along the course of the medial collateral nerve. Arrow = contralateral collateral nerve.

 


View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c. Patient 19. Postoperative scar tissue 2 years after a fifth lateral and transungual resection of a glomus tumor of the ring finger. (a) Transverse postcontrast T1-weighted spin-echo MR image (600/24) depicts scar tissue with low signal intensity (arrowheads) in the nail fold and enhancement of fibrous tissue in the thickened nail bed (*). Note the sequela (arrow) of the dorsal cortical resection of the phalanx. (b) Transverse T1-weighted spin-echo MR image (600/24) and (c) transverse T2-weighted spin-echo MR image (2,000/80) at the level of the distal interphalangeal joint depict a 3-mm neuroma (arrowheads) along the course of the medial collateral nerve. Arrow = contralateral collateral nerve.

 
Recurrent Glomus Tumor
In all patients, MR imaging revealed a 1.5–10.0-mm nodule compatible with the diagnosis of recurrent glomus tumor (Table 2). In 13 (54%) of 24 patients, the nodule had signal intensity features typical of a glomus tumor (ie, isointensity or high signal intensity on T1-weighted spin-echo MR images, high signal intensity on T2-weighted spin-echo MR images, and strong enhancement after intravenous injection of contrast material) (Fig 4). In eight (33%) of 24 patients, the signal intensity behavior of the tumor was atypical, with low intensity or isointensity on T2-weighted spin-echo images. In six (25%) of 24 patients, slight enhancement was also noted. These two unusual signal intensity features were both present in three (12%) of 24 patients. In these three patients, postcontrast 3D GE images were helpful in the delineation of the tumor margins (Fig 5). Most lesions had well-defined margins (15 [62%] of 24) and homogeneous signal intensity (19 [79%] of 24). However, in nine (38%) of 24 patients, the tumor had ill-defined margins; histologic examination revealed fibrous tissue infiltrating the stroma of the glomus tumor. In five (21%) of 24 patients, the signal intensity of the nodule was heterogeneous. In one of these patients (patient 3), unusual flow artifacts contributed to a heterogeneous appearance at MR, and pathologic examination revealed a vegetative hemangioendothelioma. This lesion had previously been misdiagnosed as a recurrent glomus tumor (Fig 6).


View this table:
[in this window]
[in a new window]

  		TABLE 2. MR Imaging Findings in Recurrent Glomus Tumor

 


View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. Patient 23. Typical recurrent glomus tumor of the index finger 8 years after a lateral resection. (a) Transverse short inversion time inversion-recovery MR image (2,060/14.2/150) depicts a 3-mm submatrical tumor with very high signal intensity (thin arrows). The margins of the scar tissue (arrowheads) are ill defined. (b, c) Transverse T1-weighted fast spin-echo MR images (560/14.2) obtained (b) before and (c) after injection of contrast medium show the strong enhancement of the glomus tumor (thin arrows in c) and the lack of enhancement of the lateral scar tissue (arrowheads). Note the bone erosion (thick arrow) in a-c. (d, e) Targeted MIP-reformatted images obtained at MR angiography with (d) a 25-second delayed enhanced and (e) a 1 minute 37-second delayed enhanced coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) show an early, peripheral, and homogeneous delayed enhancement of the glomus tumor. Note the two contiguous tumor nodes (arrows) beneath the nail matrix.

 


View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. Patient 23. Typical recurrent glomus tumor of the index finger 8 years after a lateral resection. (a) Transverse short inversion time inversion-recovery MR image (2,060/14.2/150) depicts a 3-mm submatrical tumor with very high signal intensity (thin arrows). The margins of the scar tissue (arrowheads) are ill defined. (b, c) Transverse T1-weighted fast spin-echo MR images (560/14.2) obtained (b) before and (c) after injection of contrast medium show the strong enhancement of the glomus tumor (thin arrows in c) and the lack of enhancement of the lateral scar tissue (arrowheads). Note the bone erosion (thick arrow) in a-c. (d, e) Targeted MIP-reformatted images obtained at MR angiography with (d) a 25-second delayed enhanced and (e) a 1 minute 37-second delayed enhanced coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) show an early, peripheral, and homogeneous delayed enhancement of the glomus tumor. Note the two contiguous tumor nodes (arrows) beneath the nail matrix.

 


View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c. Patient 23. Typical recurrent glomus tumor of the index finger 8 years after a lateral resection. (a) Transverse short inversion time inversion-recovery MR image (2,060/14.2/150) depicts a 3-mm submatrical tumor with very high signal intensity (thin arrows). The margins of the scar tissue (arrowheads) are ill defined. (b, c) Transverse T1-weighted fast spin-echo MR images (560/14.2) obtained (b) before and (c) after injection of contrast medium show the strong enhancement of the glomus tumor (thin arrows in c) and the lack of enhancement of the lateral scar tissue (arrowheads). Note the bone erosion (thick arrow) in a-c. (d, e) Targeted MIP-reformatted images obtained at MR angiography with (d) a 25-second delayed enhanced and (e) a 1 minute 37-second delayed enhanced coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) show an early, peripheral, and homogeneous delayed enhancement of the glomus tumor. Note the two contiguous tumor nodes (arrows) beneath the nail matrix.

 


View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4d. Patient 23. Typical recurrent glomus tumor of the index finger 8 years after a lateral resection. (a) Transverse short inversion time inversion-recovery MR image (2,060/14.2/150) depicts a 3-mm submatrical tumor with very high signal intensity (thin arrows). The margins of the scar tissue (arrowheads) are ill defined. (b, c) Transverse T1-weighted fast spin-echo MR images (560/14.2) obtained (b) before and (c) after injection of contrast medium show the strong enhancement of the glomus tumor (thin arrows in c) and the lack of enhancement of the lateral scar tissue (arrowheads). Note the bone erosion (thick arrow) in a-c. (d, e) Targeted MIP-reformatted images obtained at MR angiography with (d) a 25-second delayed enhanced and (e) a 1 minute 37-second delayed enhanced coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) show an early, peripheral, and homogeneous delayed enhancement of the glomus tumor. Note the two contiguous tumor nodes (arrows) beneath the nail matrix.

 


View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4e. Patient 23. Typical recurrent glomus tumor of the index finger 8 years after a lateral resection. (a) Transverse short inversion time inversion-recovery MR image (2,060/14.2/150) depicts a 3-mm submatrical tumor with very high signal intensity (thin arrows). The margins of the scar tissue (arrowheads) are ill defined. (b, c) Transverse T1-weighted fast spin-echo MR images (560/14.2) obtained (b) before and (c) after injection of contrast medium show the strong enhancement of the glomus tumor (thin arrows in c) and the lack of enhancement of the lateral scar tissue (arrowheads). Note the bone erosion (thick arrow) in a-c. (d, e) Targeted MIP-reformatted images obtained at MR angiography with (d) a 25-second delayed enhanced and (e) a 1 minute 37-second delayed enhanced coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) show an early, peripheral, and homogeneous delayed enhancement of the glomus tumor. Note the two contiguous tumor nodes (arrows) beneath the nail matrix.

 


View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Patient 17. Atypical recurrent glomus tumor of the middle finger after a second transungual resection 6 months previously. (a) Transverse T2-weighted spin-echo MR image (2,000/80) and (b) postcontrast T1-weighted spin-echo MR image (600/24) show a bone erosion (arrow) of the dorsal cortex of the phalanx without an overlying well-defined mass. (c, d) Postcontrast (c) sagittal T1-weighted spin-echo image (600/24) and (d) transverse 3D gradient-recalled echo image (50/12, 40° flip angle) better delineate a slightly enhanced 2-mm glomus tumor (arrow in c, arrows in d) in the distal portion of the nail bed.

 


View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Patient 17. Atypical recurrent glomus tumor of the middle finger after a second transungual resection 6 months previously. (a) Transverse T2-weighted spin-echo MR image (2,000/80) and (b) postcontrast T1-weighted spin-echo MR image (600/24) show a bone erosion (arrow) of the dorsal cortex of the phalanx without an overlying well-defined mass. (c, d) Postcontrast (c) sagittal T1-weighted spin-echo image (600/24) and (d) transverse 3D gradient-recalled echo image (50/12, 40° flip angle) better delineate a slightly enhanced 2-mm glomus tumor (arrow in c, arrows in d) in the distal portion of the nail bed.

 


View larger version (102K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. Patient 17. Atypical recurrent glomus tumor of the middle finger after a second transungual resection 6 months previously. (a) Transverse T2-weighted spin-echo MR image (2,000/80) and (b) postcontrast T1-weighted spin-echo MR image (600/24) show a bone erosion (arrow) of the dorsal cortex of the phalanx without an overlying well-defined mass. (c, d) Postcontrast (c) sagittal T1-weighted spin-echo image (600/24) and (d) transverse 3D gradient-recalled echo image (50/12, 40° flip angle) better delineate a slightly enhanced 2-mm glomus tumor (arrow in c, arrows in d) in the distal portion of the nail bed.

 


View larger version (101K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d. Patient 17. Atypical recurrent glomus tumor of the middle finger after a second transungual resection 6 months previously. (a) Transverse T2-weighted spin-echo MR image (2,000/80) and (b) postcontrast T1-weighted spin-echo MR image (600/24) show a bone erosion (arrow) of the dorsal cortex of the phalanx without an overlying well-defined mass. (c, d) Postcontrast (c) sagittal T1-weighted spin-echo image (600/24) and (d) transverse 3D gradient-recalled echo image (50/12, 40° flip angle) better delineate a slightly enhanced 2-mm glomus tumor (arrow in c, arrows in d) in the distal portion of the nail bed.

 


View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Patient 3. Recurrence of a vegetative hemangioendothelioma of the index finger after a lateral resection 5 years previously. Results of previous histologic examination had indicated the mass was a glomus tumor. Sagittal postcontrast T1-weighted spin-echo MR image (600/24) depicts a 9-mm well-defined mass (arrowheads) in the pulp. The tumor is enhanced but has an unusual central flow void artifact (arrow).

 
The 3D postcontrast MR angiograms in four patients showed marked enhancement of the nodule, an appearance that is highly suggestive of a recurrent glomus tumor. This "tumor blush" was more obvious on late images obtained with a mixed arterial and venous timing and may be helpful in characterizing multiple nodules (Fig 4). In one of the four patients (patient 24), only MR angiography revealed a small, 2-mm recurrent tumor beneath the nail matrix, while T1- and T2-weighted images showed a nodule in the scar tissue with signal intensity equivalent to that of the nail bed (Fig 7). Thirteen (54%) of these 24 patients had evidence of tumor recurrence during the year following surgery; the other 11 patients had evidence of tumor recurrence after 1 year.



View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a. Patient 24. Recurrent glomus tumor of the ring finger after a lateral resection 5 years previously. (a) The glomus tumor was not visible on the transverse T1-weighted fast spin-echo MR images (500/14.7) after injection of contrast medium. (b) Only the targeted MIP-reformatted MR angiogram from the postcontrast coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) highlights the 2-mm glomus tumor (arrow) beneath the posterolateral horn of the nail matrix.

 


View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b. Patient 24. Recurrent glomus tumor of the ring finger after a lateral resection 5 years previously. (a) The glomus tumor was not visible on the transverse T1-weighted fast spin-echo MR images (500/14.7) after injection of contrast medium. (b) Only the targeted MIP-reformatted MR angiogram from the postcontrast coronal 3D gradient-echo sequence (9.9/2.4/40, 20° flip angle) highlights the 2-mm glomus tumor (arrow) beneath the posterolateral horn of the nail matrix.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Recurrent glomus tumors occurred in 27% of the 94 glomus tumors that were evaluated at our institutions (CIERM, CHU Cochin) during the past 7 years. The previously reported incidence of glomus tumor recurrence after surgical excision has ranged from 5% to 50% (4,16–18).

The only treatment of primary glomus tumor is surgical removal. The surgical approach depends on an accurate evaluation of the tumor location. Small tumors may be removed by punching a 6-mm hole into the nail plate, incising the nail bed, and enucleating the lesion. The small nail disk is placed in its original location as a physiologic dressing. Larger tumors may be treated after removal of the proximal half of the nail plate. Lateral tumors are removed through an L-shaped incision parallel to and 4–6 mm toward the volar side of the lateral nail fold. The nail bed is carefully dissected from the bone until the tumor is reached and extirpated. Extirpation is usually curative, although the pain may take several weeks to disappear (20).

Clinically, the correct diagnosis of tumor recurrence may be difficult to establish. Other complications that may occur after surgery, including scar formation and possible development of a traumatic neuroma, make diagnosis difficult. In relation to the interval between the recurrence and the primary operation, recurrences are classified as early (<=1 year) or delayed (>1 year) (21). Early recurrences can be attributed to incomplete excision (4,16) or to the presence of a second tumor that was not previously diagnosed and excised during the initial operation (22,23). Delayed "recurrence" is possibly due to the development of a new glomus tumor near the excision site (17,18,24,25). In our series, the number of patients with early recurrence was nearly equal to the number of patients with delayed recurrence, with a slight predominance of early recurrences (54%). Most of the patients with early recurrence underwent surgery with a lateral approach, which offers a narrower view of the tumor bed compared with a transungual approach (26). However, this technique is preferred because it minimizes the risk of postoperative nail dystrophy.

In a postoperative context, investigations in addition to clinical examinations appear to be essential for the accurate delineation of the extent of these lesions. Radiographs can depict glomus tumors (4), but they have low diagnostic sensitivity for this tumor and do not allow differentiation between recurrent and primary bone erosions. The low percentage of positive radiographs is due to the small size of most lesions. US performed with a high-frequency transducer can depict tumors as small as 3 mm in diameter, but the limits of small and flattened subungual lesions are difficult to detect (5–7). Arteriography (27), thermography (28), and scintigraphy, which have been advocated by some authors, are no longer indicated because of their invasiveness or their poor sensitivity.

The ability of MR imaging to allow detection of a recurrent glomus tumor is not well described in the literature compared with the multiple descriptions of MR imaging features in primary glomus tumor (10,14,15). In our series, MR imaging allowed depiction of glomus tumors as small as 1.5 mm in diameter. The differentiation of a glomus tumor from a surgical scar is important and was possible in most cases with MR imaging. The amount of scar tissue and its enhancement were not related to the postsurgical delay but to the number of previous surgeries. The presence of a neuroma in one patient after multiple excisions demonstrated the complexity of distinguishing the recurrence of a glomus tumor from postsurgical complications. Artifacts in the surgical bed were depicted mainly on gradient-recalled echo images. These artifacts may be useful in the accurate determination of the location of the surgical bed but may blur the limits of a recurrent tumor. Fibrous scars were never misdiagnosed as glomus tumors; their margins were ill defined in all cases, and they showed most often low signal intensity with all sequences. MR angiography did not reveal marked enhancement of the scar tissue, even on delayed-acquisition images. The diagnostic difficulty is related to the different patterns of tumor recurrence compared with primary glomus tumor. Recurrent tumors were ill defined in 38% of patients, had low signal intensity or isointensity on T2-weighted spin-echo MR images in 33% of patients, and had faint or no enhancement in 25% of patients. When these three features are concomitant, as was the case in 12% of the patients in our series, it is difficult to distinguish the MR imaging findings of scar tissue from those of recurrent tumors.

MR angiography may be helpful in the diagnosis of recurrent glomus tumors. MR angiography, which may be combined with conventional MR imaging, affords the advantage of arteriography as a noninvasive technique for establishing the diagnosis of glomus tumor. MR angiography helps confirm the established arteriographic characteristics of glomus tumor by revealing strong enhancement in the arterial phase (29). Delayed images confirm the tumor blush, which increases in size. In one patient in our series, only MR angiograms were able to highlight a tiny 2-mm recurrent tumor beneath the nail matrix that was not visible with MR imaging alone. Therefore, MR angiography is a useful adjunct to standard MR imaging sequences in the diagnosis of a recurrent glomus tumor and the establishment of its precise location (30).

The information obtained at MR imaging can help rectify an incorrect diagnosis, as occurred in one patient in our series. The patient underwent surgery for a suspected glomus tumor, which was thought to have recurred 5 years later. The MR patterns of the tumor were very unusual for a glomus tumor, with vascular pouches and huge flow artifacts. These findings suggested a vascular malformation, which was confirmed at pathologic examination to be a hemangioendothelioma, a tumor in which the value of extensive resection remains debated (31).

In conclusion, MR imaging allows the detection of recurrent glomus tumors. They can be confused with scar tissue. MR angiography is a promising technical aid, but evaluation of a larger series of patients is necessary to prove the utility of this method.


   FOOTNOTES
 
Abbreviations: CIERM = Centre Inter Etablissement de Résonance Magnétique, MIP = maximum intensity projection, 3D = three-dimensional

Author contributions: Guarantors of integrity of entire study, J.L.D., A.C.; study concepts, J.B., A.C., J.L.D.; study design, D.L.V., C.B.C., D.R.; literature research, N.H.T.; clinical studies, J.C.D., D.L.V., S.G.; data acquisition, J.C.D., D.L.V., S.G.; data analysis/interpretation, N.H.T., S.G.; manuscript preparation, N.H.T., S.G., C.C.; manuscript definition of intellectual content, J.B., D.R., D.L.V., A.C., J.L.D.; manuscript editing, N.H.T., J.L.D.; manuscript revision/review, C.B.C., S.G., N.H.T., J.L.D., D.R.; manuscript final version approval, J.L.D., A.C., J.B., D.L.V.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Wood W. On painful subcutaneous tubercle. Edinb Med J 1812; 8:283-291.
  2. Rettig AC, Strickland JW. Glomus tumor of the digits. J Hand Surg Am 1977; 2:261-265.[Medline]
  3. Heys SD, Brittenden J, Atkinson P, Eremin O. Glomus tumour: an analysis of 43 patients and review of the literature. Br J Surg 1992; 79:345-347.[Medline]
  4. Carroll RE, Berman AT. Glomus tumors of the hand: review of the literature and report on twenty-eight cases. J Bone Joint Surg Am 1972; 54:691-703.[Abstract/Free Full Text]
  5. Fornage BD. Glomus tumors in the fingers: diagnosis with US. Radiology 1988; 167:183-185.[Abstract/Free Full Text]
  6. Bianchi S, Martinoli C, Abdelwahab IF. High-frequency ultrasound examination of the wrist and hand. Skeletal Radiol 1999; 28:121-129.[CrossRef][Medline]
  7. Ogino T, Ohnishi N. Ultrasonography of a subungual glomus tumour. J Hand Surg Br 1993; 18:746-747.[CrossRef][Medline]
  8. Holzberg M. Glomus tumor of the nail: a "red herring" clarified by magnetic resonance imaging. Arch Dermatol 1992; 128:160-162.[Abstract/Free Full Text]
  9. Matloub HS, Muoneke VN, Prevel CD, Sanger JR, Yousif NJ. Glomus tumor imaging: use of MRI for localization of occult lesions. J Hand Surg Am 1992; 17:472-475.[Medline]
  10. Kneeland JB, Middleton WD, Matloub HS, Jesmanowicz A, Froncisz W, Hyde JS. High resolution MR imaging of glomus tumor. J Comput Assist Tomogr 1987; 11:351-352.[Medline]
  11. Jablon M, Horowitz A, Bernstein DA. Magnetic resonance imaging of a glomus tumor of the fingertip. J Hand Surg Am 1990; 15:507-509.[Medline]
  12. Schneider LH, Bachow TB. Magnetic resonance imaging of a glomus tumor: a case report. Orthop Rev 1991; 20:255-256.[Medline]
  13. Hou SM, Shih TT, Lin MC. Magnetic resonance imaging of an obscure glomus tumour in the fingertip. J Hand Surg Br 1993; 18:482-483.[CrossRef][Medline]
  14. Drape JL, Idy-Peretti I, Goettmann S, et al. Subungual glomus tumors: evaluation with MR imaging. Radiology 1995; 195:507-515.[Abstract/Free Full Text]
  15. Drape JL, Idy-Peretti I, Goettmann S, Guerin-Surville H, Bittoun J. Standard and high resolution magnetic resonance imaging of glomus tumors of toes and fingertips. J Am Acad Dermatol 1996; 35:550-555.[CrossRef][Medline]
  16. Brenner P, Krause-Bergmann A, Wittig K. Solitary glomus tumors of the hand: a clinico-pathologic evaluation study. Langenbecks Arch Chir 1995; 380:321-326[German].[Medline]
  17. Carlstedt T, Lugnegard H. Glomus tumor in the hand: a clinical and morphological study. Acta Orthop Scand 1983; 54:296-302.[Medline]
  18. Van Geertruyden J, Lorea P, Goldschmidt D, et al. Glomus tumours of the hand: a retrospective study of 51 cases. J Hand Surg Br 1996; 21:257-260.[CrossRef][Medline]
  19. Bittoun J, Saint-Jalmes H, Querleux BG, et al. In vivo high-resolution MR imaging of the skin in a whole-body system at 1.5 T. Radiology 1990; 176:457-460.[Abstract/Free Full Text]
  20. Jepson RP, Harris JD. Glomus tumours. Med J Aust 1970; 2:452-454.[Medline]
  21. Dailiana ZH, Drape JL, Le Viet D. A glomus tumour with four recurrences. J Hand Surg Br 1999; 24:131-132.[CrossRef][Medline]
  22. Noor MA, Masbah O, Lumpur K. Synchronous glomus tumors in a distal digit: a case report. J Hand Surg Am 1997; 22:508-510.[Medline]
  23. Gandon F, Legaillard P, Brueton R, Le Viet D, Foucher G. Forty-eight glomus tumours of the hand: retrospective study and four-year follow-up. Ann Chir Main Memb Super 1992; 11:401-405.[CrossRef][Medline]
  24. Mansat M, Bonnevialle P, Gay R, Durroux R. Glomus tumors of the hand: presentation of fourteen cases. Ann Chir Main 1985; 4:43-50.[CrossRef][Medline]
  25. Cornell SJ. Multiple glomus tumours in one digit. Hand 1981; 13:301-302.[CrossRef][Medline]
  26. Heim U, Hanggi W. Subungual glomus tumors: value of the direct dorsal approach. Ann Chir Main 1985; 4:51-54.[CrossRef][Medline]
  27. Mercier R, Bresson P, Viallet JF, Vanneuville G. Value of arteriography in sub-ungual glomus tumors. J Radiol Electrol Med Nucl 1970; 51:303-304[French].[Medline]
  28. Corrado EM, Passaretti U, Messore L, Lanza F. Thermographic diagnosis of glomus tumour. Hand 1982; 14:21-24.[CrossRef][Medline]
  29. Boudghene FP, Gouny P, Tassart M, Callard P, Le Breton C, Vayssairat M. Subungual glomus tumor: combined use of MRI and three-dimensional contrast MR angiography. J Magn Reson Imaging 1998; 8:1326-1328.[Medline]
  30. Berquist TH, Ehman RL, King BF, Hodgman CG, Ilstrup DM. Value of MR imaging in differentiating benign from malignant soft-tissue masses: study of 95 lesions. AJR Am J Roentgenol 1990; 155:1251-1255.[Abstract/Free Full Text]
  31. Bourekas EC, Cohen ML, Kamen CS. Malignant hemangioendothelioma (angiosarcoma) of the skull: plain film, CT and MR appearance. AJNR Am J Neuroradiol 1996; 17:1946-1948.[Abstract]



This article has been cited by other articles:


Home page
 J Hand Surg Eur VolHome page
M. M. AL-QATTAN, A. AL-NAMLA, A. AL-THUNAYAN, F. AL-SUBHI, and A. F. EL-SHAYEB
Magnetic Resonance Imaging in the Diagnosis of Glomus Tumours of the Hand
J Hand Surg Eur Vol., October 1, 2005; 30(5): 535 - 540.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Theumann, N. H.
Right arrow Articles by Drapé, J.-L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Theumann, N. H.
Right arrow Articles by Drapé, J.-L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE