Ko, Kai-Hsiung MD*; Hsu, Yi-Chih MD*; Lee, Herng-Sheng MD†; Lee, Chian-Her MD‡; Huang, Guo-Shu MD*
Tophaceous gout usually develops in the chronic phase of the disease process and is typically diagnosed using clinical history, laboratory studies, and characteristic radiographic findings.1 The knee is the third most commonly involved large joint, after the foot and ankle. In cases that affect the knee, the patient may experience a variety of symptoms, including local pain and swelling, soft-tissue mass, and limitation of motion, which mimic internal derangement, inflammatory, or neoplastic processes.2,3 Thus, preoperative diagnosis from radiographs and physical examination is sometimes difficult, particularly in the absence of known hyperuricemia or previous episodes of gouty arthritis. In these circumstances, magnetic resonance imaging (MRI) may be performed.4 However, the MRI findings of gout have been described as being nonspecific and must be differentiated from those of pigmented villonodular synovitis (PVNS) or from the synovitis related to various conditions of the inflammatory process, localized nodular synovitis, and amyloidosis.2,5,6 The purpose of this study was to reevaluate the characteristic MRI features of tophaceous gout and identify morphologic patterns and common deposit locations in the knee, to differentiate gout from those diseases.
MATERIALS AND METHODS
The cases were retrieved retrospectively using a computer-based database program that searched for patients who underwent MRI of the knee in our hospital during a 5-year period by using the following search words: synovitis and intra- or extra-articular mass (solid or cystic mass). We identified 30 patients (32 knees) who had proven tophaceous gout. Medical records were also reviewed and the patients' demographic and clinical features are provided in Table 1.
Of the 30 patients, 22 had a known history of gouty arthritis before the imaging studies were performed, 15 of whom suffered from chronic tophaceous gout. The remaining 8 patients had not received a diagnosis of gout until the MRI studies were performed. The diagnosis of gouty tophi was confirmed through the analysis of histopathological specimens (14 patients) or the identification of needle-shaped crystals in the joint fluid that exhibited negative birefringence under polarized-light microscopy (16 patients).
MRI was performed on 1.5-T scanners (Vista, Picker, Cleveland, OH; Signa, General Electric Medical Systems, Milwaukee, WI) with dedicated extremity coils. The examinations consisted principally a combination of spin-echo T1-weighted (repetition time [TR]/echo time [TE], 450–650/15–20), spin-echo proton density-weighted (1800–2200/20–30) or fast spin-echo proton density-weighted (TR range/TE, 3000–3600/20), spin-echo T2-weighted (TR range/TE range, 1800–2200/80–90) or fast spin-echo T2-weighted (2500–3300/55–80), and fat-suppressed fast spin-echo T2-weighted (3000–3600/55–80) sequences. Various combinations of sequences and imaging planes were used. In 6 patients, T1-weighted (TR range/TE range, 450–650/15–20) spin-echo coronal and axial images of 6 knees were acquired after the intravenous administration of gadolinium (0.1 mmol/kg of body weight). The field of view varied between 14 and 16 cm, the slice thickness ranged from 3 to 5 mm, and the interslice gap was from 0 to 1 mm. The number of acquisitions was either 1 or 2. The imaging matrix ranged from 192 × 256 to 256 × 256.
All MRI studies were reviewed simultaneously by 2 musculoskeletal radiologists who had 18 years (G.S.H.) and 5 years (Y.C.H.) of experience and were aware that all patients had proven tophaceous gout. In cases where there was discordance in interpretation, a conclusion was reached by consensus. Each tophaceous lesion was evaluated for location, morphology, signal intensity, associated findings of bone erosion and intraosseous involvement, absence or presence of synovitis, and bone-marrow edema. Synovitis was defined as the presence of an abnormal thickening of the synovial lining of the joint that was distinct from the lesion. The signal intensity of the lesion was determined by comparison with that of skeletal muscle. The degree of enhancement of the lesion on fat-suppressed T1-weighted fast spin-echo imaging was quantified (none, mild, or moderate). To complete the assessment, we also evaluated the integrity of the articular cartilage, menisci, and ligaments. The changes of articular cartilage were graded into 4 categories using the MRI classification described by Recht et al7: grade 1, internal signal intensity alteration only; grade 2, a defect of the cartilage of less than 50%; grade 3, a defect of the cartilage of 50% to 99%; and grade 4, full-thickness defects that expose the bone. The meniscal abnormalities were graded based on the classification described by Crues et al8: grade 1, globular intrasubstance signal without articular communication; grade 2, linear intrasubstance signal without articular communication; and grade 3, meniscal tear. All ligamentous abnormalities were graded according to the classification described by House et al9: grade 1, increased edema within the ligament or periligamentous edema with the ligament intact; grade 2, partial tear; and grade 3, complete tear.
The serum urate levels of patients varied and ranged from 5.2 to 11.3 mg/dL (mean, 7.9 mg/dL) at the time of MRI studies. Five patents who had normal urate level (<7 mg/dL) had undergone recent urate lowering treatment with uricosuric agents. All patients presented with local pain and swelling, limitation of motion, or soft-tissue mass for a duration of 1 to 16 months (mean, 7.3 months). The medical records of these patients were reviewed to investigate associated comorbidities (Table 1). We found hypertension in 9 patients, diabetes mellitus in 5, obesity in 4, and chronic renal failure in 2 patients. Seventeen patients exhibited involvement of other joints, which included the foot (16 patients), ankle (13 patients), elbow (5 patients), and wrist (3 patients).
Table 2 summarizes the MRI findings for all patients. Five of 6 patients who had extra-articular tophaceous lesions also had intra-articular lesions, whereas 24 patients had only intra-articular tophaceous deposits, without extra-articular involvement. Tophaceous deposits were identified at multiple locations in the knees in 24 patients (26 knees). The remaining 6 patients (6 knees) had an isolated tophaceous mass. The tophaceous masses were commonly located in the infrapatellar fat pad and anterior joint recess, which were observed in 28 (87%) knees; the space adjacent to the lateral rim of the lateral femoral condyle, observed in 25 (78%) knees; and the intercondylar fossae, observed in 22 (69%) knees (Fig. 1).
Morphologically, the tophaceous lesions revealed amorphous crystalline-like masses in 27 knees, with low to intermediate signal intensity on T1-weighted images and variable signal intensity on T2-weighted images, ranging from low to heterogeneously high signals. Contrast-enhanced T1-weighted images of 6 knees in 6 patients showed mild-to-moderate heterogeneous enhancement or peripheral enhancement of the tophaceous masses (Figs. 2, 3). Another pattern presenting as linear tissue of crystalline-like deposits of low signal intensity on T2-weighted images was noted in 6 knees (Fig. 4). Cystic lesions of the bursa around the knee were identified in 3 knees (Fig. 5).
Associated bone erosions (Table 3) were seen in tophaceous lesions at 23 locations, including the lateral rim of the lateral femoral condyle (n = 8), the roof of the intercondylar notch (n = 7), the tibial eminence (n = 5), the medial rim of the tibial plateau (n = 2), and the lateral rim of the tibial plateau (n = 1). Intraosseous tophi were seen in 4 lesions, including in the tibial plateau (n = 3) and the patella (n = 1) (Fig. 6). A thickened synovium with a low-to-intermediate T2 signal was identified in 28 affected knees. Bone-marrow edema was only observed in 8 affected knees and was usually associated with intraosseous tophi and bone erosion (Fig. 6).
Of the 16 knees with chondral lesions, 6, 4, 2, and 4 knees had grade 1, 2, 3, and 4 defects of the cartilage, respectively. The abnormalities of lateral meniscus were identified in 13 knees with grade 1 (n = 6), 2 (n = 4), and 3 (n = 3) meniscal changes. The abnormalities of medial meniscus were evident in 17 knees with grade 1 (n = 7), 2 (n = 6), and 3 (n = 4) meniscal changes. With regard to the ligamentous abnormalities, grade 1 change of the cruciate ligaments and collateral ligaments were present in 9 and 8 knees, respectively. The increased signal inside the cruciate ligaments might result from myxoid change, intraligamentous ganglion formation, or tophaceous deposition. Otherwise, no partial or complete ligament tear was identified.
Gout is a clinical disorder resulting from urate crystal deposition. Tophi usually form in the chronic phase of gouty arthritis and may appear at intra- or extra-articular locations or within the subcutaneous tissue. Hench reported that the interval from the first gouty attack to the beginning of chronic arthritis or visible tophi is 3 to 42 years, with an average of 11.6 years.10 Subcutaneous or extra-articular tophi are usually a late manifestation and often occur together with intra-articular tophaceous deposition. Intra-articular tophi may develop very early because crystal shedding is assumed to pathogenetically precipitate acute gouty attacks.11,12 In our study, extra-articular tophi (6/30) were relatively rarer than intra-articular tophi and in most patients (5/6), extra-articular tophi were associated with intra-articular lesions. A clinical diagnosis can be established easily on the basis of recurrent gouty attacks with diffuse multiple subcutaneous tophi, evidence of hyperuricemia, and characteristic radiographic findings revealing juxta-articular soft-tissue masses, sharply defined erosions, and overhanging margins of bone.1,13 However, when intra-articular tophi occur with the absence of subcutaneous tophi and without known hyperuricemia or previous episodes of gouty arthritis, this condition may be underestimated by plain radiography. Computed tomography best evaluates bone changes and intralesional calcification,5,11 whereas MRI is superior in assessing abnormalities of the bone and soft tissue, synovial membrane thickness, and inflammatory changes.5,14–16
MRI of tophaceous gout in the knee has been described in few reports and case-series studies.2,3,5 The findings are usually nonspecific because tophi have a wide spectrum of signal intensity characteristics, which reflect their variable compositions and relative proportions of protein, fibrous tissue, crystals, and hemosiderin.2,6 Most lesions are isointense relative to muscle on T1-weighted images. On T2-weighted images, most lesions show low-to-intermediate heterogeneous signal intensities,3,5 although lesions may display high signal intensities in the presence of high protein or inflammation with local edema.2 Heterogeneous peripheral or homogeneous enhancement is frequently observed in contrast-enhanced images; these enhanced patterns may result from hypervascular granulation tissue surrounding the tophus or hypervascularized inflammatory tissue within the tophi.5,6,17 Although the MRI findings in our study are consistent with those of previous studies, they are not pathognomonic features of tophaceous gout and may deceptively mimic those of inflammatory arthritis, PVNS, amyloidosis, or soft-tissue tumor.
As with signal intensities and contrast-enhanced patterns, the MRI morphologic patterns of gouty tophi in the knee may offer some information for a differential diagnosis. The classified patterns are rarely described in the published data. In our study, 3 morphologic patterns were observed. Amorphous crystalline-like tophaceous masses (n = 27) were the most common pattern and reflect the traditional idea of how soft-tissue urate deposition occurs. The second pattern of tophaceous gout reveals cystic masses of the bursae (n = 3) without evidence of tophaceous lesions in the knee, which arise from urate crystals deposited in the synovial membrane, with synovial inflammation and reactive proliferation leading to effusion. This pattern of the tophaceous gout is difficult to diagnose and must be proven by histopathologic analysis. The last pattern of tophaceous gout shows linear tissue of crystalline-like deposits (n = 6) floating in the joint effusion. To the best of our knowledge, this pattern has not been previously reported and may be a relatively specific finding, which helped us to differentiate gout from other entities. These 3 patterns of tophaceous gout may represent different stages of the disease. A longitudinal follow-up study is required to establish how these patterns of deposition evolve through various stages of gouty tophi.
Tophi can involve any location around the knee, such as the synovium, tendon, cartilage, or intraosseous locations,18 but they seem to be preferentially deposited in several common locations, including the infrapatellar fat pad, anterior recess, intercondylar fossae, and intercondylar roof, which have been mentioned in some reports.3,5 Our series confirmed these findings. Apart from the aforementioned locations, the space adjacent to lateral rim of the femoral condyle(popliteal groove) is also the common location for tophaceous deposition. Bony erosions also frequently occur at these common sites of tophus deposition. The underlying pathologic mechanism of bone erosion in chronic gout remains unclear but several possible explanations have been suggested: (1) local pressure develops from adjacent tophaceous masses;1 (2) intra-articular monosodium urate crystal deposition induces various degrees of synovitis with similarities to chronic rheumatic arthritis, causing erosive synovial arthropathies19; (3) locally produced enzymes within the tophus degrade the bone and cartilage matrix and contribute to the development of erosion20,21; and (4) osteoclast activation within the tophus plays a potential role in bone erosion.19 Our results support the first 2 theories because the tophaceous masses were associated with bone erosions in our patients, often located at stress-bearing points of the knee joint, such as the origin and insertion sites of the cruciate ligaments and femoral condyles, and were also associated with varying degrees of synovitis. Four intraosseous tophi were also found and most of them (3/4) were on the tibial plateau. The pathogenetic mechanism may involve the invagination of the tophi through the focal bone erosions or a pressure defect over the tibial attachment site of the cruciate ligament, rather than by the direct intraosseous deposition of tophaceous gout. This finding is also consistent with the conclusion of Dalbeth et al, who demonstrated a close relationship between bone erosion and the presence and size of the intraosseous tophus.19
The differential diagnosis of tophaceous gout of the knee should include chronic rheumatoid arthritis, PVNS, localized nodular synovitis, and amyloidosis. Although any of these diseases occur at any location in the knee and may be associated with erosion, none has demonstrated a preference for these locations. The presence of amorphous or linear hypointense lesions on T2-weighted images in the aforementioned common locations, associated with bone erosions and an inhomogeneous enhancement pattern after gadolinium administration, allowed us to suggest tophaceous gout in most of our patients. Chronic rheumatoid arthritis usually presents with marked synovial proliferation and synovitis, with tissue debris or rice bodies, which are less common in tophaceous gout. PVNS is almost invariably monoarticular and usually presents as a profoundly dark T2 signal, reflecting the proliferating synovium together with the paramagnetic effect of hemosiderin, which is seen less often in tophaceous gout.22 Although the appearance of amyloidosis on MRI may simulate that of gout, in our experience amyloid deposition infrequently appears as a linear crystalline-like deposit in the joint effusion.23 Clinical information, laboratory data, and characteristic MRI patterns should allow us to make these diagnoses correctly. However, percutaneous needle aspiration may be necessary in some equivocal cases.
Our study has some limitations. First, the small number of patients selected may have made it difficult to identify all the MRI features of tophaceous gout of the knee in this study. Second, this study retrospectively assessed the MRI features of tophaceous gout in the knees of patients with pathohistologically proven disease. Therefore, not all patients included in the study underwent the same protocol or received contrast reagent at the time of the MRI study. Furthermore, our study mainly demonstrates the morphologic patterns and common locations of tophaceous gout in the knee, with no comparison with those of PVNS or amyloidosis patients. Such a comparison should be made in a future study.
1.Rubenstein J, Pritzker KPH. Crystal-associated arthropathies. Am J Roentgenol
2.Yu JS, Chung C, Recht M, et al. MR imaging of tophaceous gout. Am J Roentgenol
3.Yu KH, Lien LC, Ho HH. Limited knee joint range of motion due to invisible gouty tophi. Rheumatology (Oxford)
4.Kobayashi S, Taniguchi A, Okamoto H, et al. A case of isolated gouty tophus in a patient without a previous history of gouty arthritis. J Clin Rheumatol
5.Chen CK, Yeh LR, Pan HB, et al. Intra-articular gouty tophi of the knee: CT and MR imaging in 12 patients. Skeletal Radiol
6.Popp JD, Bidgood WD Jr, Edwards NL. Magnetic resonance imaging of tophaceous gout in the hands and wrists. Semin Arthritis Rheum
7.Recht MP, Piraino DW, Paletta GA, et al. Accuracy of fat-suppressed three-dimensional spoiled gradient-echo FLASH MR imaging in the detection of patellofemoral articular cartilage abnormalities. Radiology
8.Crues JV, Mink J, Levy TL, et al. Meniscal tears of the knee: accuracy of MR imaging. Radiology
9.House CV, Connell DA, Saifuddin A. Posteromedial corner injuries of the knee. Clin Radiol
10.Hench PS. The diagnosis of gout and gouty arthritis. J Lab Clin Med
11.Gerster JC, Landry M, Duvoisin B, et al. Computed tomography of the knee joint as an indicator of intraarticular tophi in gout. Arthritis Rheum
12.Yu KH. Intraarticular tophi in a joint without previous gouty attack. J Rheumatol
13.Dalbeth N, McQueen FM. Use of imaging to evaluate gout and other crystal deposition disorders. Curr Opin Rheumatol
14.Gentili A. Advanced imaging of gout. Curr Rheumatol Rep
15.Chen CK, Chung CB, Yeh L, et al. Carpal tunnel syndrome caused by tophaceous gout: CT and MR imaging features in 20 patients. Am J Roentgenol
16.Schumacher HR, Becker MA, Edwards NL, et al. Magnetic resonance imaging in the quantitative assessment of gouty tophi. Int J Clin Pract
17.Hsu CY, Shih TT, Huang KM, et al. Tophaceous gout of the spine: MR imaging features. Clin Radiol
18.Recht MP, Seragini F, Kramer J, et al. Isolated or dominant lesions of the patella in gout: a report of seven patients. Skeletal Radiol
19.Dalbeth N, Clark B, Gregory K, et al. Mechanisms of bone erosion in gout: a quantitative analysis using plain radiography and computed tomography. Ann Rheum Dis
20.Hasselbacher P, McMillan RM, Vater CA, et al. Stimulation of secretion of collagenase and prostaglandin E2 by synovial fibroblasts in response to crystals of monosodium urate monohydrate: a model for joint destruction in gout. Trans Assoc Am Physicians
21.Liu R, Liote F, Rose DM, et al. Proline-rich tyrosine kinase 2 and Src kinase signaling transduce monosodium urate crystal-induced nitric oxide production and matrix metalloproteinase 3 expression in chondrocytes. Arthritis Rheum
22.Murphey MD, Rhee JH, Lewis RB, et al. Pigmented villonodular synovitis: radiologic-pathologic correlation. Radiographics
23.Cobby MJ, Adler RS, Swartz R, et al. Dialysis-related amyloid arthropathy: MR findings in four patients. Am J Roentgenol
© 2010 Lippincott Williams & Wilkins, Inc.