Abstract
Background
Soft-tissue chondroma (STC) is a rare benign soft tissue tumor that arises primarily in acral extra-skeletal locations. Occasionally, STCs may arise in more proximal non-acral locations, accompanied by non-classic features that label them as indeterminate lesions and pose diagnostic challenge for both radiologists and pathologists alike.
Purpose
To explicate the potential of diagnostic imaging in the identification and characterization of appendicular non-acral STCs with emphasis on their morphologic magnetic resonance imaging (MRI) enhancement.
Material and Methods
Our clinical database records were searched for patients with histologically proven primary soft-tissue chondroid lesions over a five-year period. Two musculoskeletal (MSK) trained radiologists evaluated the imaging studies and an MSK pathologist revised the pathological findings.
Results
The study included six cases of appendicular non-acral STCs (mean age = 40.5 years). The mean size of the tumors was 5.6 cm, with four localized to the knee region, one in the thigh, and one in the sternoclavicular region. All cases showed high signal intensity matrix with low-signal intensity septa on T2-weighted MRI and post-contrast marginal/septal enhancement. The lesions were lobulated and lacked host tissue reaction except for one showing subjacent mild soft-tissue edema. Histologically, the cases lacked overt features of malignancy although one was originally misdiagnosed as chondrosarcoma.
Conclusion
Non-acral STCs are benign cartilaginous tumors that may pose a diagnostic challenge, both radiologically and pathologically. Collaborative imaging and pathologic workup is needed for better characterization of non-aggression of these lesions, and to avoid diagnostic pitfalls and unnecessary radical resections.
Keywords
Introduction
Soft-tissue chondroma (STC) is a rare benign soft-tissue tumor, composed of hyaline cartilage, that arises primarily in extra-skeletal locations (1,2). Radiologically and pathologically, STCs may resemble other soft tissue chondroid masses such as periosteal chondroma, synovial chondromatosis, soft-tissue chondrosarcoma, and metaplastic chondroid lesions (such as myositis ossificans and chondrolipoma) (2). However, conventional STCs are readily recognized by their small size (often less than 3 cm) and peri-articular acral location, typically in the hands and feet (1,3–6). They are rarely found in other parts of the body (7–10). STCs are usually treated with conservative local excision sparing the patient unnecessary radical surgery and decreasing morbidity (11,12), with local recurrences reported in 10–18% of cases (4–6).
Occasionally, STCs may arise in more proximal non-acral locations in the appendicular skeleton, accompanied by non-classic imaging features and a relatively larger size compared to their acral counterparts. Radiologically, these tumors are problematic posing diagnostic concerns for more aggressive tumors that warrant aggressive radical surgery, and accordingly they are often labeled as indeterminate lesions (11–14). Similarly, distinguishing benign from malignant cartilaginous tumors at extra-osseous sites may present a diagnostic challenge to the pathologist because of the frequent occurrence of atypical cellular features in STCs (15). Therefore, accurate diagnosis requires close radiological–pathological correlation.
Herein we present a series of six primary soft-tissue chondroid tumors that fulfilled the criteria for non-acral STC, and explicate the potential of diagnostic imaging in the identification and characterization of these lesions with emphasis on their morphologic MRI features.
Material and Methods
Study design and research ethics
The study was conducted at Al-Razi Orthopedic Hospital, Kuwait. Study approval and patient consent were waived by our institutional review board owing to its retrospective nature.
Study population inclusion and exclusion criteria
We searched our clinical database records for patients with histologically proven primary soft-tissue chondroid tumors over a five-year period from May 2011 to April 2016. Inclusion criteria included: (i) any case of extra-osseous chondroid mass that was located exclusively in the soft tissues of extremities, or related to a non-acral joint (juxta-, peri-, or intra-articular); and (ii) these cases should have available MRI studies and a pathologic confirmation of its chondroid nature. Exclusion criteria were: (i) heterogeneous lesions with metaplastic/secondary chondroid elements histologically such as lipoma with metaplastic cartilage, tumoral calcinosis, synovial sarcoma, and myositis ossificans containing cartilage; (ii) radiological evidence of connection to underlying bone or presence of an underlying skeletal cartilaginous mass; (iii) cases with no available MR study; (iv) cases without confirmatory tissue diagnosis or with inconclusive biopsy results; and (v) recurrent tumors with previous surgical excision of a bone or soft-tissue tumor from same site.
Demographic data, relevant clinical presentation, symptoms durations, and management strategies were also recorded.
Imaging studies acquisition and review
All MR studies of the study population were retrieved from our hospital Picture Archiving and Communication System (PACS). Auxiliary radiographic and multidetector computed tomography (MDCT) studies relevant to the studied region were also reclaimed.
The MRI studies were acquired on a 1.5-T unit (Signa HDe MR system; GE Healthcare, Milwaukee, WI, USA). All MRI examinations included a combination of both anatomic images (fast-spin echo T1-weighted [T1W], TR/TE = 300–400/9–14 ms) and water-sensitive images (either frequency-selective fat-saturated [FS] fast-spin echo T2-weighted [T2W] [TR/TE = 4000–4500/80–100 ms], proton density [PD; TR/TE = 2500–3000/35–40 ms], and/or STIR [TR/TE = 3500–5000/50 ms; inversion time = 150 ms] images) in orthogonal planes, which depict the lesion well. Post-contrast T1W images (TR/TE = 300–400/9–14 ms) with and/or without fat suppression were available in five of the study cases.
Two radiologists, specialized in musculoskeletal imaging, evaluated the imaging studies in consensus. On MR, the topography of the lesion (widest dimension, margins of the lesion, matrix texture, presence or absence of breaking down) was defined. Moreover, adjacent soft-tissue edema and definition of surrounding soft-tissue margins were recorded. The signal intensity of the lesions on T1W and water-sensitive images were characterized in comparison to adjacent skeletal muscles, while MR contrast enhancement patterns were visually evaluated.
In addition, plain radiographs and/or CT, when available, were assessed for presence or absence of osteo-articular continuity or involvement, osseous adaptive changes, periosteal reaction, presence/absence of tissue mineralization and its pattern.
Lesions biopsy
Five patients were biopsied, using a 16–18-G automated Tru-cut needle, under imaging guidance (two ultrasound [US]-assisted while the other three were CT-guided). One patient had excision biopsy as the tumor was initially removed en bloc.
Generally, our department standard practice, for image-guided procedures, is to collect between three and six core-biopsy specimens per procedure according to the lesion’s size and criteria as well as the in-charge interventionalists review. A subjective assessment of sample adequacy was made by visual inspection at the time of biopsy due to unavailability of an on-site cytopathologist at our institution.
Histopathological review
Histopathology reports and glass slides were reviewed by a pathologist specialized in musculoskeletal pathology. The cases were reviewed for the following histopathological features: quality of matrix (hyaline cartilage, chondromyxoid); cellular features (degree of nuclear atypia, binucleation, mitosis); calcifications; and evidence of invasion into adjacent structures and/or bone that would exclude the diagnosis of a benign chondroma and favor a more aggressive soft-tissue neoplasm. The pathological features were evaluated in view of the imaging findings.
Results
Clinical findings and lesions topography
Six cases met the criteria of STC; their clinical and imaging features are summarized in Suppl. Table 1. The patients comprised three women and three men (mean age = 40.5 years, age range = 26–58 years). The main clinical presentation was that of a slow-growing painless mass and/or locking of a joint/limitation of movement. The symptoms had been present for 12–66 months prior to presentation (average = 41 months).
The tumor size was in the range of 2.6–10.2 cm at presentation (mean = 5.6 cm). Four tumors were located in the knee region, one in the upper left thigh inferor and medial to the femoral head capping the lesser trochanter, and one ventral to the right sterno-clavicular joint. Of the four knee tumors, one was para-articular and three were intra-articular as follows: one in the inferio-medial joint recess encroaching on the Hoffa’s pad, one in the superior-lateral recess and one obliterating the supra-patellar pouch.
Radiographic and CT imaging findings
Conventional radiographs were available in 4/6 patients, three of which also had a pre-biopsy non-enhanced MDCT. All four cases showed at least focal matrix mineralization ranging from a central calcific fleck only visible on CT (Fig. 1a) but not appreciable on the corresponding radiograph, to flocculent calcifications on radiographs that appeared as central nodular calcifications on CT (Fig. 1b), to the classic rings and arcs calcifications, to dense ossifications (Fig. 1c). No adjacent periosteal reactions and/or bone remodeling were identified on available radiographs and/or CT of these cases.
(a) Suprasternal STC adjacent to the right sterno-clavicular joint (SCJ) from case 6. Axial CT scan, in soft-tissue window, shows a hypodense round mass splaying the distal fibers of right sternomastoid muscle with a central fleck of calcification, (b) Para-articular mass of right knee from case 1. Axial CT scans in soft tissue window setting revealing a hypodense para-articular mass with central nodular calcifications. (c) A suprapatellar intra-articular STC of the knee from case 4. Lateral radiograph of the left knee shows densely calcified mass overlying the distal femoral shaft as well as superior patellar pole and obliterating the patello-femoral space.
MRI findings
All six cases had MRI studies available with 5/6 cases also having contrast-enhanced T1W images with or without fat suppression. All tumors had well defined margins with a more or less lobulated contour (Figs. 2–4). They were completely separable from adjacent osteo-articular structures and showed no surrounding soft-tissue edema except for one case where peri-lesional fat stranding was noted (Fig. 3a and b). The non-calcified matrices of all tumors exhibited: (i) a homogenous texture with no detected areas of breaking down; (ii) isointense signal intensity on T1W images; and (iii) predominantly high-signal intensity on water-sensitive sequences, compared to adjacent skeletal muscles, with linear low-signal bands of the capsules and septae (Figs. 2–4). Remarkable peripheral enhancement, with or without septal enhancement, was identified following intravenous gadolinium infusion in 4/5 cases with available post-contrast MRI series. The supra-patellar lesion showed characteristic segmental peripheral and patchy enhancement of its non-ossified tumor due to heavy ossification (Fig. 1c).
A mass in the Hoffa’s fat pad of left knee from case 2. MRI; sagittal (a) T1W image and (b) FS PD-weighted image the lesion exhibits isointense signal on (a) and hyperintensity on (b). (c) Post-contrast FS T1W axial image displaying marginal enhancement (white arrows). (d) Microscopic examination of the needle core biopsy reveals hyaline cartilage featuring bland chondrocytes in lacunae, consistent with a well-differentiated cartilaginous neoplasm (H&E stain, ×20 magnification). An intra-articular mass of left knee from case 3. Coronal MRI (a) T1W and (b) FS PD images show a well-defined soft-tissue mass filling the supero-lateral synovial recess under the ilio-tibial tract (white arrowheads in (a) and (b)). The mass exhibits homogenous isointensity to skeletal muscles on (a) and hyperintensity on (b). Note the adjacent peri-lesional fat stranding on T1W image and hyperintensity on PD images inferring adjacent edema small (small black arrows in (a) and (b)). Post-contrast FS T1W images in (c) coronal and (d) axial planes showing marginal enhancement (small white arrows in (c) and (d)). Of note, no reactive cortical changes are observed. A suprasternal mass from case 6. (a) Non-contrast MRI axial T1W and (b) coronal FS T2W, showing the mass superior to the right sternoclavicular joint (white stars in (b) and (c)), stretching and splaying the distal sternomastoid insertion (small white arrows in (a) and (b)) and bulges into the overlying subcutaneous tissue. The mass emits isointense low signal on (a) and predominantly hyperintense on (b). (c) Post-contrast coronal FS T1W image, following IV gadolinium administrations, the mass exhibited remarkable marginal enhancement (white arrowheads). (d) Gross resection specimen shows a circumscribed mass adherent to the outer surface of the sternoclavicular joint without infiltrating into the bone or adjacent structures (yellow arrow indicates the bone). (e) Microscopically, the lesion is composed of chondromyxoid matrix with foci of hyaline cartilage, consistent with a well-differentiated cartilaginous neoplasm (H&E stain, ×20 magnification).


Pathological findings
The pathological findings are displayed in Suppl. Table 2. Four of six cases had needle core biopsy material only for diagnosis, one had biopsy followed by resection specimen (case 6, Fig. 4d), and one a resection specimen only. Histologically, 5/6 cases were composed of hyaline cartilaginous tissue while one case had extensive chondromyxoid matrix reminiscent to that seen in myxoid chondrosarcoma (Fig. 4e). Mild to moderate chondrocyte atypia was noted in 4/6 cases and binucleation in one case. No histological evidence of invasion into adjacent tissues was noted in any of the cases. Calcifications were identified in three cases and heavy ossification in one case. Originally, none of the tumors were firmly diagnosed as STC mainly because of limited biopsy material combined with unusual sites, relatively large tumor sizes and lack of radiology–pathology correlation. Case 6 was confused with myxoid chondrosarcoma because of the extensive myxoid matrix (Fig. 4e).
Discussion
STCs are uncommon slow-growing tumors of the soft tissues characterized by their small size (mean < 3 cm in diameter) (6) and predilection to appendicular acral locations (4,5), with the hands being more common than feet (5,6,16). Thus, it is understandable that such tumors often pose a diagnostic challenge to the radiologist and pathologist alike when they are encountered as large masses in unusual proximal sites. Indeed, the relatively larger tumor size in the current series (range = 2.6–10.2 cm, mean = 5.6 cm) contributed to the diagnostic dilemma. Similar to previous reports (12,17,18), our cases showed a predilection for the lower limb, specifically the knee region. One tumor was in close proximity to the sterno-clavicular joint but completely separable from the joint cavity and the clavicular bone. The only two earlier reports of clavicle-related chondromas described in the English literature were periosteal variants at the lateral and medial ends of the clavicle, respectively (19,20). Our cases were in the third to sixth decades of life (age range = 26–58 years, mean age = 40.5 years) with no sex predilection, in agreement with previous data (4,6).
The cartilaginous nature of the current cases was suspected based on the high signal intensity of the matrix and low signal intensity of the intervening septa on T2W MRI, followed by peripheral and septal vascular enhancement on post-contrast series. All STCs in our series were well-defined and lobulated lacking invasion into adjacent tissues and host tissue reaction except for one case that showed mild peri-lesional soft-tissue edema on MR. Although the presence of surrounding soft-tissue edema may suggest a more aggressive neoplastic lesion (21–23), the edema in our case may be explained by the mass effect exerted by the tumor on the fat surrounding the superior-lateral recess of the knee causing venous backpressure. All cases lacked adjacent bone remodeling on radiography and CT. Other radiological clues that pointed to the benign nature of the lesions in the current series were signal homogeneity on T1W images and absence of tissue heterogeneity on T2W images apart from that caused by calcified matrix. This is comparable to the results of previous reports (21–23) that addressed the use of different MR morphologic and signal criteria to differentiate benign and malignant soft-tissue tumors, especially those in the extremities. Post-contrast MR studies in our series revealed a characteristic peripheral and septal pattern of enhancement in all lesions, also in agreement with previous studies (16,24–26). This pattern is uncommon in malignant cartilaginous tumors which usually show intense and homogenous patterns of enhancement (16,22–24).
Due to their lobular growth pattern, cartilaginous tumors exhibit a characteristic arciform pattern of calcification, recapitulating the final stages of endochondral ossification (27). In up to 70% of STCs, various patterns of calcifications have been described on radiographic and/or CT studies, including arc and ring patterns, calcium flecks, and granular and diffuse patterns (26,28). Our cases also showed a spectrum of mineralization from a central calcium fleck to diffuse dense ossification. These areas of calcification showed loss of signal on all MR pulse sequences due to paucity of free precision water molecules. Detection of these calcification patterns on ancillary radiographic and CT imaging is invaluable in recognizing cartilaginous lesions within soft tissues and predicting their histology (29).
Histologically, STCs are composed of lobules of mature hyaline cartilage separated by fibrous septa and surrounded by a fibrous pseudo-capsule (1,24,27), a picture that parallels what is seen on MRI and explains the peripheral/septal enhancement post contrast. However, this pattern is not specific for benign chondromas (24,30) and one needs to exclude cytological atypia and tumor permeation beyond the capsule. Although all of our cases were proven to be of chondroid nature histologically, none was firmly diagnosed as chondroma at the outset. The main limiting factor was the scanty material available for pathological assessment, making it very difficult to establish a definite diagnosis without radiological guidance. Indeed, our suprasternal tumor was originally diagnosed as myxoid chondrosarcoma on core-needle biopsy based on the large tumor size, unusual location, and extensive myxoid matrix. Re-evaluation of the pathology slides, in view of the radiological features, revealed lack of significant cytological atypia, necrosis, mitosis, and permeation into adjacent tissues and the final diagnosis of STC was rendered in consensus.
From imaging perspectives, STCs should be differentiated from other appendicular soft-tissue masses, especially those with calcifications, such as myositis ossificans, depositional entities; like tumoral calcinosis, gout, and calcium pyrophosphate deposition disease (CPPD); and benign neoplasms as synovial chondromatosis. Further malignant entities such as synovial sarcoma, pleomorphic sarcoma (aka malignant fibrous histiocytoma [MFH]), extraskeletal chondrosarcoma, and/or osteosarcoma have to be considered while interpreting extraskeletal soft-tissue masses.
Myositis ossificans is commonly characterized by its zonal pattern on radiography and CT with recall of traumatic precedents (5). Depositional diseases tend to have peri-articular amorphous or bizarre-shaped calcifications with relatively poorly defined margins and an attenuation value of about 160 HU (31). Synovial chondromatosis usually exhibits multiple calcific nodules, with classic ring and arc patterns, related to a large joint and/or bursa (5,6).
Synovial sarcoma prevails in younger age group, occurs in peri-articular locations and 30% of the tumors show calcification, which is more dense and peripheral (32). On the other side, pleomorphic sarcoma is more common in older age groups with lesser calcification tendency (33). Extra-skeletal chondrosarcomas are larger masses with discontinuous cartilaginous calcific patterns. It lacks both lobular growth pattern and vascular septal pattern of enhancement seen in STCs (34). Juxtacortical osteosarcomas are related to bone surface and usually exhibits a cloudy pattern of amorphous calcification mirroring osteogenesis in the tumor without enchondral ossification (35).
We acknowledge that our study is limited by the small number of cases included, but this is inevitable due to the rarity of these non-acral STCs. Another limitation is the lack of assessment of quantitative enhancement parameters due to the retrospective nature of the study, although the validity and clinical applicability of these parameters are still controversial (36).
In conclusion, non-acral STCs are benign cartilaginous tumors that may pose a diagnostic challenge radiologically and pathologically. Collaborative imaging and pathologic workup is needed for better characterization of non-aggression of these lesions, to avoid diagnostic pitfalls and unnecessary radical resections.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
