HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Moore, S. L. Articles by Golimbu, C. Search for Related Content PubMed PubMed Citation Articles by Moore, S. L. Articles by Golimbu, C. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?

MRI of Sarcoidosis Patients with Musculoskeletal Symptoms

¹ Department of Radiology (Musculoskeletal), NYU Medical Center, 560 First Ave., New York, NY 10016.
² Department of Pulmonary and Critical Care Medicine, Mount Sinai Medical Center, New York, NY.

Received February 19, 2004; accepted after revision September 23, 2004.

 
Address correspondence to S. L. Moore (sandra.moore{at}med.nyu.edu).

Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
OBJECTIVE. Our objective was to determine MRI findings in sarcoidosis patients with musculoskeletal symptoms.

CONCLUSION. In sarcoidosis patients with musculoskeletal complaints, MRI reveals marrow and soft-tissue lesions that are occult or underestimated on radiographs. Axial and large-bone lesions may resemble osseous metastases on MRI. Most lesions detected are nonspecific in appearance, except nodular muscle lesions. MRI reveals features suggesting the diagnosis, but with standard protocols, no pathognomonic MRI features were determined.

Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Sarcoidosis is an inflammatory granulomatous disease of unknown cause that can involve the muscles, joints, and bones. Musculoskeletal involvement can be symptomatic or asymptomatic. In the radiographic literature, it has been emphasized that most skeletal lesions are seen in the small bones of the hand and feet, usually associated with skin lesions. Descriptions of MRI findings of large-bone sarcoidosis are primarily in case reports, more often of the spine [1-7] than the appendicular skeleton [8-13]. The MRI appearance of sarcoid muscle lesions has been described in retrospective studies [14, 15]. Although osteoarticular complaints are common among sarcoidosis patients, imaging often is limited to a negative radiograph, a radiograph showing a classic lacy pattern of osteolysis in the digits, or, rarely, a lytic or sclerotic large-bone lesion. Axial skeletal and long-bone involvement are not common radiographic findings [16]. Neither a bone scan nor a skeletal survey reliably predicts the location of sarcoidal bone lesions except in the small bones of the hands and feet [17].

To our knowledge, no prospective studies have been performed evaluating MRI for musculoskeletal symptoms in sarcoidosis patients. We hypothesized that MRI can improve the detection of lesions in sarcoidosis patients with musculoskeletal complaints.

Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
A prospective study was performed over a -year period with 42 consecutive patients referred for MRI by the pulmonologists at a hospital-based sarcoidosis clinic, on the basis of sarcoidosis diagnosis and unexplained musculoskeletal complaints. Musculoskeletal complaints were defined as local pain, weakness, and/or extratruncal swelling or mass of unexplained clinical cause, with no history of recent trauma. Institutional review board approval was obtained and informed consent signed.

Of 42 patients referred for MRI, in all but three cases, the diagnosis of sarcoidosis was proven before the MRI evaluation by a positive Kveim-Siltzbach test (12 patients) or fiberoptic bronchoscopy (12 patients), or biopsy of lung (one patient), lymph node (four patients), muscle (two patients), bone (one patient), skin (nine patients), or lacrimal gland (two patients); or a combination of these methods, revealing noncaseating epithelioid granulomas in the absence of known granulomagenic agents. One patient with radiographic and clinical findings of Löfgren's syndrome did not undergo biopsy. Two patients originally considered likely to have sarcoidosis on clinical grounds underwent MRI scanning but subsequent biopsy results were inconclusive for sarcoidosis and they were excluded from the study. Thus, a total of 40 patients were evaluated. The age range of the study subjects was 35-71 years, with a mean age of 49 years. The duration of disease from establishment of sarcoidosis diagnosis ranged from less than 1 year to 39 years, with an average duration of 8.2 years. In the study population, 35 patients (88%) had intrathoracic disease, 13 patients (33%) had involvement of three or more organ systems, and 15 patients (38%) had cutaneous sarcoid lesions.

The patients had chest radiographs for evaluation and staging of intrathoracic involvement (staged according to Mitchell and Scadding [18]), and laboratory analysis of serum calcium, alkaline phosphatase, and serum angiotensin-converting enzyme (ACE). Patient demographics, clinical laboratories, and Scadding stage are presented in Table 1. For all of the subjects with clinical or imaging evidence of joint disease, serology for antinuclear antibodies (ANA) and rheumatoid factor was obtained.

View larger version (114K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1 —38-year-old woman with pulmonary sarcoidosis and hand and foot deformities. Axial proton density-weighted image (TR/TE 4,016/15) shows diffuse, poorly marginated marrow infiltration of both calcanei. Patient also had MRI findings of tendinosis and tenosynovitis of ankle (not shown).

Image Evaluation
Two musculoskeletal radiologists independently reviewed the MR images. At the time of MRI review, the reviewers were aware that the patients had the clinical diagnosis of sarcoidosis, but were not aware of the laboratory or radiographic findings. For each case, the radiologists noted the presence or absence of and distribution and location of bone lesions, muscle lesions, soft-tissue masses, soft-tissue infiltration, and/or joint pathology. The two reviewers' interpretations were in complete agreement in 38/40 (95%) of the cases and, in two cases, the final agreement was based on consensus after discussion.

The criterion used for the diagnosis of marrow infiltration was intramedullary signal alteration compared with the adjacent fatty and/or hematopoietic marrow. Differentiation of bone lesions from red marrow deposition was based on morphology (e.g., multiple, discrete cannonball-like foci not characteristic of marrow replacement vs patchy areas more characteristic of red marrow), location (not characteristic, such as proximal phalanges, vs characteristic of red marrow residua, such as the proximal femurs). The lesions were described as confined to the medullary space versus with extraosseous extension, and as violating the cortex or not. Determination was made as to whether the lesions were solitary or multiple, focal or diffuse, and distinctly or vaguely marginated. Assessment for avascular necrosis (AVN) was made by established criteria [19]. Findings considered diagnostic for joint involvement on MRI included synovitis, tenosynovitis, effusions, marrow edema at articular margins, and/or subchondral cysts or erosions. Muscle abnormalities were divided into three groups: fatty replacement of muscle, edema of muscle, and masslike lesions within the muscle compartments [20]. Fatty infiltration was determined on T1-weighted pulse sequences as focal or diffuse replacement of the muscle with fat compared with other muscles. Muscle edema was determined on inversion recovery or T2-weighted images as areas of increased signal within the muscle compared with other muscles. Masslike lesions were defined as discrete regions in the subcutaneous or muscle compartments with morphology and signal intensity differing from the surrounding tissue [20]. Soft-tissue infiltration was defined as discrete noncircumferential reticulation and/or nodularity of altered signal compared with the surrounding tissue in that compartment. A subject having multiple areas scanned with positive findings on any single study was considered a positive case. Two cases with overlapping findings (small bone lesions and arthritis) were categorized by the more dominant lesion and included as a subset of bone lesions.

Radiographs of the symptomatic area were obtained for 17 patients, and for three patients with lesions of the head, CT was obtained for evaluation of the skull. All but two patients with bone lesions had radiographs or CT scans. We did not correlate radiographs with MR images for patients with soft-tissue lesions. The radiographs and CT scans (or reported interpretations for three cases without radiographs available for review) were correlated with the MRI findings by consensus by the two radiologists. All abnormal findings detected on either technique were compared for size, location, and extent and scored as equivalent, more or less conspicuous or extensive, or detected on one technique and not the other.

Biopsy was recommended for all of the patients with large-bone and soft-tissue lesions, including masses, muscle lesions, myopathy, and subcutaneous reticulation. Biopsy was not recommended for patients with hand and foot lesions or joint findings except one patient with massive tenosynovitis. The final decision for or against biopsy was determined by the clinicians. Four patients declined recommended biopsy. Biopsy evaluation of the musculoskeletal lesions showed at MRI was obtained for 17 of the subjects (49% of the patients with positive MRI findings). Biopsy findings were counted positive for sarcoidosis if reported as noncaseating granulomas in the absence of other causes of granulomatous disease. Seventeen biopsies were performed (nine of bone, four of muscle, two of masses, one of tenosynovium, and one of subcutaneous reticulation).

View larger version (168K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2A —58-year-old woman with sarcoidosis and pelvic pain. Anteroposterior radiograph of pelvis was interpreted as normal. Coronal images of pelvis and hips obtained with T1-weighting (TR/TE 550/8) in A, and proton density weighting with fat saturation in B (3,600/22) show innumerable discrete subcentimeric lesions infiltrating marrow.

View larger version (152K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2B —58-year-old woman with sarcoidosis and pelvic pain. Anteroposterior radiograph of pelvis was interpreted as normal. Coronal images of pelvis and hips obtained with T1-weighting (TR/TE 550/8) in A, and proton density weighting with fat saturation in B (3,600/22) show innumerable discrete subcentimeric lesions infiltrating marrow.

View larger version (165K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2C —58-year-old woman with sarcoidosis and pelvic pain. Anteroposterior radiograph of pelvis was interpreted as normal. Coronal detail of sacrum, coronal proton density fat-saturated (TR/TE 3,600/22) image shows innumerable high-signal discretely marginated round lesions, most 1 cm or smaller. Although patient had sarcoidosis and no known primary tumor, clinical assumption was metastases and patient underwent deep bone biopsy that revealed noncaseating granulomas.

View larger version (138K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3 —55-year-old woman with hip pain. T1-weighted axial image (TR/TE 550/9) obtained through upper pelvis. Low signal lesion in left ilium does not violate cortex (black arrow). Apparent asymmetry of iliac bones is due to positioning.

View larger version (102K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 4 —52-year-old woman with wrist "mass." T1-weighted (TR/TE 500/10) axial image performed with fat saturation after administration of IV gadolinium chelate shows tenosynovitis of extensor compartment with marked thickening and enhancement of synovium. Synovial biopsy showed noncaseating granulomas.

View larger version (83K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 5 —55-year-old woman with sarcoidosis and leg weakness. Patient was treated with steroids. Coronal T1-weighted image (TR/TE 550/8) shows fatty replacement of right gluteal and bilateral hamstring muscles. Biopsy of gastrocnemius muscle showed noncaseating granulomas.

View larger version (126K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 6A —51-year-old woman with painful palpable calf nodules. T1-weighted axial image of right calf (TR/TE 550/9) shows donut-shaped lesion, nearly isointense with muscle, with low signal intensity centrally (white arrow).

View larger version (113K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 6B —51-year-old woman with painful palpable calf nodules. T2-weighted axial image (4,550/54), same level. Lesion showed mild increase in signal intensity compared with surrounding muscle, with low signal intensity centrally but not as bright as fluid (white arrow).

View larger version (108K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 6C —51-year-old woman with painful palpable calf nodules. T1-weighted axial image (616/9) with fat saturation after IV administration of gadopentetate dimeglumine. Periphery of lesion enhances (white arrow), but central "dark star" does not.

Top Abstract Introduction Subjects and Methods Results Discussion References

Joint Abnormalities
Nine patients had joint abnormalities on MRI, all with negative serology for ANA and rheumatoid factor. Only one patient underwent synovial biopsy, which showed noncaseating granulomas (Fig. 4).

Muscle Lesions
Two patients showed fatty replacement of the thigh muscles (Fig. 5). One patient showed muscle edema in the thigh adductors, which enhanced. Gastrocnemius biopsies in these three cases showed noncaseating granulomas. The fourth patient had tender, palpable calf nodules manifesting as discrete intramuscular nodules on MRI (Figs. 6A, 6B, 6C), with an appearance characteristic of nodular sarcoidosis of muscle [14, 15] confirmed on biopsy.

Three patients showed soft-tissue mass lesions on MRI. One mass had MRI characteristics of a simple lipoma. Two patients showed masses with mixed signal characteristics. Biopsy revealed infarcted lipoma and noncaseating granulomas, respectively. In two cases, subcutaneous soft-tissue reticulation (more focal than typically seen with edema) was seen in the legs, decreased in signal intensity on T1, and bright on water-sensitive sequences. In one of these cases, biopsy showed noncaseating granulomas.

The associations of the MRI findings with chest radiograph stage, multiorgan and skin involvement, and biopsy findings are presented in Table 2. No statistically significant differences between subjects with or without marrow lesions on MRI were found for the parameters examined: skin lesions (p = 0.8), chest radiograph stage (p = 0.15), disease duration equal to or less than 2 years (p = 0.5), elevated alkaline phosphatase, calcium, or angiotensin-converting enzyme (p = 0.16), multiorgan disease (three or more organs) (p = 0.3), sex (p = 0.3), or race (p = 0.7). However, within the study population, those with bone lesions differed from those without bone lesions, showing a higher prevalence of stage I or II chest radiograph abnormality (76% with bone lesions, 52% all others), multiorgan involvement (41% with bone lesions, 26% all others), elevated alkaline phosphatase, ACE, and/or serum calcium (53% with bone lesions, 35% all others). The prevalence of skin lesions differed little between these groups, as 6/17 (35%) of patients with and 9/23 (39%) without bone lesions had skin lesions. There were no statistically significant differences in duration since sarcoidosis diagnosis between the two groups.

MRI/Radiographic and CT Correlation
Eleven of the 20 cases reviewed showed comparable or equivalent MRI and radiographic or CT findings. Of the remaining nine subjects, six with bone lesions on MRI had normal radiographs and three had radiographic or CT findings less conspicuous/extensive than bone lesions detected on MRI. The association of intramedullary lesions on MR images with a negative radiograph was most often noted with large-bone lesions (n = 5).

Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
This study shows the utility of MRI to detect and delineate lesions in patients with sarcoidosis with musculoskeletal signs and symptoms. Our study detected a large range of musculoskeletal lesions of biopsy-proven sarcoidal, nonsarcoidal (e.g., infarcted lipoma), and histologically undetermined cause that may be encountered in the MRI investigation of musculoskeletal complaints in sarcoidosis patients.

In our study, MRI revealed more axial skeleton and large-bone disease than had been recognized on standard films. Using standard musculoskeletal protocols, the MRI appearance of these lesions is not specific and should be differentiated from metastatic disease, multiple myeloma, lymphoma, osseous hemangioma, and disseminated granulomatous infection [3, 17]. In no large-bone lesion was there MRI evidence of cortical destruction, although cortical disruption and extraosseous extension were seen with small-bone lesions. The lack of cortical destruction with large-bone lesions might explain why these lesions were radiographically occult. We do not consider MRI necessary in most cases for the diagnosis of sarcoidal lesions of the small bones of the hand and feet (in proven sarcoidosis patients), as the radiographic finding of lacy osteolysis of the small bones of the hands and feet is virtually pathognomonic in the setting of clinical sarcoidosis.

The joint abnormalities detected on MRI reflect nonspecific arthropathy of indeterminate origin, which could include common arthritides such as degenerative disease. Whether sarcoidal arthropathy may have contributed to some of the findings is speculative. Histologic evaluation is not routinely performed to differentiate possible sarcoidal arthropathy from other causes of joint pathology, and the diagnosis of sarcoidosis arthritis is usually based on clinical presentation rather than imaging. Only in one case did joint findings at MRI (tenosynovitis) lead to biopsy. All of the patients with joint abnormalities on MRI had normal rheumatoid factor and ANA laboratory values.

The MRI finding of fatty atrophy of the muscles can be seen with sarcoidosis myopathy and other muscle disorders including corticosteroid myopathy, residua of trauma, chronic disuse, and/or chronic denervation [20]. Differentiation of sarcoidosis myopathy from other myopathies requires correlation with electromyography, which shows greater muscle irritability with sarcoidal myopathy than steroid myopathy. Creatine kinase levels are often elevated in muscle sarcoidosis but not corticosteroid myopathy. MRI does not differentiate between these causes but can be used to assess the degree of atrophy and guide the selection of a muscle biopsy site, obviating a biopsy of fatty-replaced muscle.

The MRI findings of musculoskeletal sarcoidosis lesions may reveal systemic involvement and a greater burden of granulomas than had been previously considered, which may have therapeutic and prognostic significance. When musculoskeletal manifestations are detected, MRI can also guide biopsy in patients with suspected but unproven sarcoidosis and can be used to follow response to treatment.

The higher prevalence of multiorgan disease and chest radiograph abnormality corresponding to Scadding stage I or II among sarcoidosis patients with bone lesions on MRI suggests a correlation of bone lesions with lymphadenopathy and total body granulomatous burden. Although cutaneous sarcoidosis lesions have been reported as associated with skeletal lesions [21], we did not find a statistically significant correlation in our study population.

Our study has several recognized limitations. It did not include a comparison group of sarcoidosis patients without osteoarticular symptoms nor patients without sarcoidosis. The clinicians referred only patients with musculoskeletal symptoms, which probably led to the relatively large percentage of cases with positive MRI findings. Thus, the frequency of lesion type detected cannot be generalized to the sarcoidosis population at large. Fewer than half the subjects underwent biopsy evaluation of the musculoskeletal abnormality, and the potential sarcoidal cause of the nonbiopsied abnormalities detected at MRI was not determined. The comparison of radiograph and MRI findings was limited by the unavailability of correlative radiographs for some cases. We scanned only the area of signs and/or symptoms; further work is needed to assess the extent of involvement in asymptomatic regions in this patient population. Some of the patients had background changes of red marrow replacement in the pelvis and spine, which made distinguishing their marrow lesions more difficult. Dedicated sequences such as opposed phase gradient-echo sequences might be more accurate for distinguishing hematopoietic from pathologic marrow.

In evaluating sarcoidosis patients with musculoskeletal symptoms, MRI detects radiographically silent disease in the soft tissues and large bones. With the standard technique, in most cases, no pathognomonic imaging findings allow differentiation from other pathologies, including neoplasm.

The clinician should consider MRI for the evaluation of sarcoidosis patients with unexplained osteoarticular complaints if standard radiographs are negative. Radiologists should include sarcoidosis in the differential diagnosis of musculoskeletal disease detected at MRI in the appropriate clinical setting and should be alerted that large-bone and axial skeleton sarcoidosis lesions encountered on MRI might resemble metastatic lesions.

References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Bushara KO, Petermann G, Waclawik AJ, Brown WD, Schutta HS. Sarcoidosis of the spinal cord with extensive vertebral involvement: a case report. Comput Med Imaging Graph1995; 19:443 -446[CrossRef][Medline]
2. Fisher AJ, Gilula LA, Kyriakos M, Holzaepfel. MR imaging changes of lumbar vertebral sarcoidosis. AJR1999; 173:354 -356[Free Full Text]
3. Ginsberg LE, Williams DW, Stanton D. MRI of vertebral sarcoidosis. J Comput Assist Tomogr1993; 17:158 -159[Medline]
4. Golzarian J, Matos C, Golstein M, Stallenberg B, Depierreux M, Struyven J. Case report: osteosclerotic sarcoidosis of spine and pelvis: plain film and magnetic resonance imaging findings. Br J Radiol 1994;67:401 -404[Abstract/Free Full Text]
5. Kenney CM III, Goldstein ST. MRI of sarcoidosis spondylolisthesis. J Comput Assist Tomogr1992; 16:660 -662[Medline]
6. Poyanli A, Poyanli O, Sencer S, Akan K, Sayrak H, Acunas B. Vertebral sarcoidosis: imaging findings. Eur Radiol2000; 10:92 -94[CrossRef][Medline]
7. Sundaram M, Place H, Shaffer WO, Martin DS. Progressive destructive vertebral sarcoid leading to surgical fusion. Skeletal Radiol 1999;28:717 -722[CrossRef][Medline]
8. Beasley EW III, Peterman SB, Hertzler GL. An unusual form of tibial sarcoidosis. AJR1987; 149:754 -756[Free Full Text]
9. Blacksin MF, Acello AN, Kowalc J, Lyons MM. Osseous sarcoidosis of the foot: detection by MR imaging. AJR1994; 163:1444 -1445[Free Full Text]
10. Blank NM, Steininger H, Kalden JR, Burkhardt H. Symptomatic bone lesion of the tibia head as onset manifestation of sarcoid. J Rheumatol 1999;26:936 -937[Medline]
11. Ferrozzi F, Folegnani F, Delnevo D, Campani R. Tibial sarcoidosis. Radiologic and magnetic resonance aspects. Radiol Med (Torino) 1997;94:675 -677[Medline]
12. Powers TA, Limbird TJ. MRI of the knee in sarcoidosis: synovial and marrow involvement. J Comput Assist Tomogr1994; 18:313 -314[Medline]
13. Rayner CK, Burnet SP, McNeil JD. Osseous sarcoidosis—a magnetic resonance imaging diagnosis. Clin Exp Rheumatol 2002;20:546 -548[Medline]
14. Otake S, Ishigaki T. Muscular sarcoidosis. Semin Musculoskelet Radiol 2001;5:167 -170[CrossRef][Medline]
15. Otake S. Sarcoidosis involving skeletal muscle: imaging findings and relative value of imaging procedures. AJR1994; 162:369 -375[Abstract/Free Full Text]
16. Resnick D. Sarcoidosis. In: Resnick D, ed. Diagnosis of bone and joint disorders, 4th ed. Philadelphia, PA: Saunders,2002 : 4771-4791
17. Milman N, Lund JO, Graudal N, Enevoldsen H, Evald T, Norgard P. Diagnostic value of routine radioisotope bone scanning in a series of 63 patients with pulmonary sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 2000;17:67 -70[Medline]
18. Mitchell DN, Scadding JG. Sarcoidosis. Am Rev Respir Dis 1974;110:774 -802[Medline]
19. Mitchell DG, Rao VM, Dalinka MK, et al. Femoral head avascular necrosis: correlation of MR imaging, radiographic staging, radionuclide imaging, and clinical findings. Radiology1987; 162:709 -715[Abstract/Free Full Text]
20. May DA, Disler DG, Jones EA, Balkissoon AA, Manaster BJ. Abnormal signal intensity in skeletal muscle at MR imaging: patterns, pearls, and pitfalls. RadioGraphics2000; 20[Spec No]:S295 -S315. Review[Abstract/Free Full Text]
21. Funk SD, Kremer JM. Cutaneous and musculoskeletal features, diagnostic modalities, and immunopathology in sarcoidosis. Curr Opin Rheumatol 1994;6:78 -81[Medline]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. B. Prabhakar, C. B. Rabinowitz, and F. S. Chew Imaging of sarcoidosis: self-assessment module. Am. J. Roentgenol., March 1, 2008; 190(3 Suppl): S7 - 10. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Moore, S. L. Articles by Golimbu, C. Search for Related Content PubMed PubMed Citation Articles by Moore, S. L. Articles by Golimbu, C. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?