Abstract
Background/Aim: Tenosynovial giant cell tumor (TGCT) is a rare disease of young adults with a high number of cases going unreported. Despite an international consensus meeting of experts in June 2022, the majority of such patients repeatedly pose problems, even for experienced clinicians. This study deals with the question of how often postoperative histopathological findings are consistent with preoperative magnetic resonance imaging (MRI) findings.
Patients and Methods: In a retrospective data analysis, we investigated 137 patients at our department who had undergone synovectomy between 1991 and 2019. Inclusion criteria: positive MRI findings with evidence of T1- and T2-weighted sequence low-signal representation of the TGCT with inhomogeneous contrast medium uptake because of hemosiderin, subsequent synovectomy and complete histological report showing inconsistency with MRI findings. Because of the heterogeneity of the study group, we can only report descriptive statistics.
Results: The average age at diagnosis was 38 (range=9-73) years. Of 137 cases, 52 with complete data sets were included in the study. In 37 (71%) out of these 52 patients, MRI and histological findings were consistent. Of the 52 patients, 15 (29%) had a false-positive MRI finding for TGCT.
Conclusion: Although imaging showed pathognomonic characteristics, the diagnosis of TGCT was not confirmed histopathologically in almost 29%. We therefore recommend a preoperative biopsy, especially in case of doubt, and treatment of TGCT in designated centers.
- Tenosynovial giant cell tumor
- giant cell tumor of tendon sheath
- pigmented villonodular synovitis
- knee arthroscopy
- radical synovectomy
- knee swelling
- knee pain
Introduction
Tenosynovial giant cell tumor (TGCT) is a rare disease (~2/1000,000) in young adults, but with higher incidence rates than previously known (1-5). Despite an international consensus meeting of experts in June 2022, this disease remains difficult to diagnose and treat (3). This prompted us to carry out the study in which we processed our own data from clinical work.
In German-speaking countries, TGCT is also called pigmented villonodular synovitis (Riesenzelltumor der Sehnenscheide oder fibröses Histiozytom des Synoviums), but this is an outdated term, and in accordance with the World Health Organization, this term should no longer be used (3, 4, 6-8).
TGCT causes the synovium to thicken and possibly overgrow (2, 3, 6). In a healthy joint, the synovium produces a small amount of fluid that ‘lubricates’ the cartilage and supports mobility (9). In TGCT, the proliferation of the synovium produces extra fluid that causes swelling, pain and reduction of mobility. Symptoms are nonspecific, which is why patients often have a protracted course of disease. Differentially, it can feign a disease from the rheumatoid form group (2, 10). TGCT is a progressive disease that slowly worsens and can lead to bone damage and arthritis (4).
Two types of growth can be differentiated: a nodular form and a diffuse form. Metastasis or sarcomatous transformation is exceedingly rare. The diffuse form of TGCT tends to grow much more destructively, be more difficult to treat and have a slightly higher tendency to recur. The main site of predilection for TGCT is the knee (Figure 1 and Figure 2) being affected in 80% of patients (11-13). The hip (~15%), ankle, shoulder and elbow can also be affected. In 80% of patients, the disease is monoarticular (1, 5). Incidental findings of TGCT during arthroscopy or endoprosthesis surgery are also possible (6, 11, 14, 15). Findings include a localized brown-greenish, rather coarse, nodular lesion that grows more or less pedunculated out of the synovium, or a diffuse synovitis that, compared to reactive or rheumatoid arthritis, is less bright red and darker, tinged green by hemosiderin and sometimes yellowish by lipid deposits (11). When such observations are made intraoperatively, it is always advisable to obtain histological samples. Histologically, histiocytic cells (CD68-positive) and dendritic cells, macrophages with hemosiderin are typically seen (1, 5). At the cellular level, hemosiderin deposits and multinucleated giant cells are eponymous and pathognomonic (16).
Magnetic resonance imaging of the knee with an inhomogeneous signal because of hemosiderin deposits (arrows), proliferation of the synovium and joint effusion.
Intraoperative view of the knee with a Trickey surgical approach. A demarcable, nodular form of tenosynovial giant cell tumor was present.
Magnetic resonance imaging (MRI) remains the diagnostic gold standard, showing partly homogeneous, partly inhomogeneous signal behavior, depending on the extent of hemosiderin deposition and the collagen content. TGCT shows as a low signal in T1- and T2-weighted sequences, with inhomogeneous contrast medium uptake. In T2-weighted fast low-angle shot (FLASH) 3D/gradiant recalled echo (GRE) sequences, there are typical susceptibility artifacts due to hemosiderin deposition, as can be seen in Figure 1 with (extensive) joint effusion. This can help to differentiate between a localized and diffuse form. It is recommended that a conventional X-ray be performed to assess possible destruction of the bone and to exclude calcifications. Plain X-ray plays a minor role, but periarticular soft-tissue swelling can be observed due to the increased density caused by iron deposits. There may be erosive changes at the osseocartilaginous junction (1, 5). It is important to note that bone density remains the same, in contrast to osteoarthritis.
The disease therefore does not have a typical presentation, imaging is not always clear, so that after a prolonged period of suffering, only histological confirmation remains for definitive diagnosis. In the following, we examine the mismatch between preoperative MRI findings and postoperative histological reports.
Patients and Methods
The study protocol was approved by the local Ethics Committee (EK Nr: 1142/2018) Applying a retrospective study design, we investigated all patients diagnosed and treated at the Department of Orthopedics and Traumatology, Medical University of Innsbruck, Austria, for TGCT between 1991 and 2019. For our investigation, we included all patients with a positive MRI finding for TGCT (with no further differentiation between nodular and diffuse types) with typical hemosiderin deposits and surgical treatment followed by a histological result inconsistent with the preoperative MRI finding. A positive MRI finding and a consistent postoperative histological examination regarding TGCT was defined as an exclusion criterion.
In accordance with the ICD-10 coding, we defined the diagnostic key to be used to search the hospital’s internal patient information system. Additionally, we undertook cross-searches by reviewing medical histories, written MRI results and surgical reports. The MRI was reviewed retrospectively by radiologists and a specialist for orthopedic oncology. We defined complete documentation as a preoperative medical report in which not only was the code M12.2-M12.28 used, but TGCT was also mentioned by name as the diagnosis, in combination with a positive MRI report and consecutive surgical removal with histological processing of the obtained tissue, and a clinical follow-up examination. Further details of case selection are shown in Figure 3. We report our results as descriptive statistics that were analyzed with Excel® (version 16.92; Microsoft Corporation, Redmond, WA, USA).
Results
In the database, 137 cases were identified with ICD-10 code M12.2-M12.28 (“treated with synovectomy” between 1991-2019). The hospital’s internal documentation was studied for all 137 of these patients including: medical history, clinical examination, radiological findings, surgical reports if available, histopathological reports, and clinical follow-up treatment. Of these, 69 patients were excluded because the diagnosis of TGCT was never mentioned or the MRI reviewed retrospectively by radiologists and specialists for orthopedic oncology did not show the typical MRI findings mentioned above. Of the remaining 68 patients, 16 were excluded because the documentation could not be fully reconstructed; 52 patients with complete documentation remained. Thirty-seven patients with true-positive results were excluded. This means that they had a medical assessment, received a diagnosis made from positive MRI findings, and the TGCT diagnosis was mentioned by name in the doctor’s letter and was confirmed by the histopathological report.
The 52 patients who fell under the treatment algorithm for a TGCT included 22 (42%) male and 30 (58%) female patients. Their mean age at diagnosis was 38 (range=9-73) years. Overall, MRI findings and histological results were consistent in approximately 71% of cases (n=37, 71.16%). Open surgical repair (Figure 2 and Figure 4) was the most frequently chosen treatment procedure, in 31 (59.61%) cases regardless of the histopathological finding.
Nodular and diffuse forms of tenosynovial giant cell tumor. On the left, the intraoperative tissue of a nodular form can be well delineated; on the right, that of a diffuse form cannot be well delineated.
The remaining 15 patients for whom there was a mismatch between preoperative MRI findings and postoperative histological reports (28.84%) constitute our study group. For this group, MRI showed a TGCT, but the tissue samples obtained intraoperatively when processed histologically did not confirm TGCT. This group included nine females and six males. Their mean age at diagnosis was 41 (range=17-72) years. Histological findings were: nonspecific synovitis, joint chondromatosis, ganglion, hemangioma, of psoriatic arthritis.
We confirm symptoms are nonspecific, include swelling, pain (acute and chronic), and limitation of range of motion. About 50% of patients reportedly have persistent symptoms after treatment (3, 4, 8, 17, 18). Due to our dependence on non-standardized documentation, we cannot specify the exact duration of pain. The subjective perception of pain was described as ranging from “acute” to “years”. However, the trend saw the complaints persist for at least several months. Recurrence was observed in 12 (32%) patients. These occurred on average after 2.5 (range=1-7) years. The knee was most frequently affected, in 27 (72%) cases, a finger and an elbow in one, an ankle in four, and the hip joint in two cases.
Discussion
TGCT is a rare disease, and for this reason, there are few studies on this topic, and even fewer prospective studies (3, 11-13, 19-21). Therefore, the consensus meeting recommends treatment and follow-up in designated centers (3).
Symptoms of patients suffering from TGCT are nonspecific, including swelling, pain (acute and chronic), limitation of range of motion, as we can confirm (3, 4, 8, 17). The symptom duration and intensity in our study ranged from “acute” to “long-term” and from “mild to severe”. About 50% of patients continue to have persistent symptoms after treatment (18); some patients may also be asymptomatic. Diffuse joint pain and painless joint swelling are usually reported as already existing for months at initial manifestation (11, 16, 19, 22, 23). In terms of symptoms, the clinician’s experience is more likely to play a role than specific symptoms. We can confirm this observation. Because of the diffuse symptoms, the diagnosis is often made late. Due to our retrospective data analysis, we cannot give an exact time between the onset of symptoms and diagnosis.
In our collective, the knee joint was most often affected (72%). In the nodular form, symptoms of entrapment or blockage may simulate a meniscal lesion. Recurrent joint effusions are also possible. Thus, in the case of the above-mentioned diverse symptoms, MRI tomographic clarification should be performed (8, 11, 22, 24). MRI remains the undisputed gold standard, but a plain X-ray should always be performed to rule out bone erosion. The localized type presents with a single mass with a capsule and small low intensity foci in T1 and T2 sequences representing faint hemosiderin deposits. In T2 and short-tau-inversion-recovery-sequence (STIR), the signal is heterogenous due to differing amounts of hemosiderin, fibrous, cellular, fat, fluid elements (25, 26). The signal characteristics are: T1 low-to-intermediate signal, T2 variable and heterogeneous, low-to-intermediate signal. High signal areas can occur due to joint fluid or inflammation of synovium. In STIR, the signal is variable and heterogeneous; on GRE, it is low and may demonstrate susceptibility artefacts (25, 26). The findings in the case of the diffuse type is an extensive synovial proliferation, which may be mass-like, without lobulated or ill-defined margins along with low T1/T2 signal areas due to hemosiderin deposits in combination with or without joint erosions (27). The T2 and STIR signal is variable due to hemosiderin deposits, fluid, fat, fibrous and cellular elements. The other signal characteristics are the same as in the nodular type (25-27).
In the present study, the MR images were screened for the above-mentioned artifacts by both an orthopedic surgeon and a radiologist. The study includes a total of 52 patients whose records were analyzed according to scientific standards. The actual study group used to answer the study question contained only 15 patients. Although we were looking for false-positive values, this should be seen in a positive context. The high loss of data shows the importance of conscientious and accurate medical documentation: 69 patients were excluded because the diagnosis of TGCT was never mentioned or the MRI reviewed retrospectively by radiologists and specialists for orthopedic oncology did not show the typical MRI findings of TGCT mentioned above. These 69 patients make up almost 50% of the original 137. Another approximately 11% should probably be excluded because of inaccurate documentation. In total, almost 61% of the original cohort was lost.
If a mismatch between preoperative MRI findings and the postoperative histological report is calculated, the rate is approximately 29%. How can such a high false-positive rate occur? Therefore, we began to question our own diagnostic algorithm.
The indication should be made after discussion in a designated expert meeting (3, 8, 17, 21). A paper recently published by the community of experts (3) gives a treatment algorithm which we would like to summarize in part. Various authors agree that contrast MRI (using gadolinium, for example) is an indispensable gold standard (3, 4, 8, 17, 21). The prevailing expert opinion is that when “D-TGCT (diffuse form of TGCT) is suspected, a biopsy is recommended. A biopsy can be avoided if the radiological evaluation in an expert center shows strong evidence of TGCT and resection is planned” (3). The pathological diagnosis is then confirmed based on the surgical specimen. Regarding our own results, we would like to go a step further and recommend prior biopsy in all ambiguous cases and resection biopsy only if the tumor is definitely resectable. If the image morphology is unclear or access to the tumor is difficult and it is uncertain whether it can be safely resected, a biopsy should be planned. Based on our own results, we would like to recommend that these specific cases be biopsied before synovectomy. We recognize that accurately predicting these patients is very difficult.
The expert recommendation for both nodular and diffuse forms is to assess the surgical potential for improvement (2, 3). Consideration should be given to the location of the tumor, i.e. its surgical accessibility or its location (intra-/extraarticular), the associated possibility of local recurrence and the patient’s individual risk factors; although the tumor is benign in origin, it has a high mitotic rate (1, 5, 28). We had a recurrence rate of approximately 32%. In the case of surgery, our recommendation is to perform a marginal resection to treat the nodular form and an extensive synovectomy for the diffuse form (3). For the nodular form, after complete marginal resection, low local relapse rates are expected. We agree with literature, which reports a slightly lower local relapse rate after open versus arthroscopic surgery (13% vs. 20%) (3). In 59.61% of our patients, we chose open surgical repair as the treatment procedure. Extra-articular nodular TGCT requires macroscopic complete resection (3, 29). Incomplete synovectomy is associated with a 44-55% recurrence rate (3, 12). Therefore, treatment at designated centers is recommended to reduce the risk of incomplete synovectomy. Early diagnosis, surgical therapy, and adjuvant measures, if necessary, are essential to reduce the severity of the disease and the recurrence rate (30, 31). If TGCT is malignant, it should be treated as a soft-tissue sarcoma (3). Our experience is in line with the latter statements. Debulking is controversially discussed in literature and by experts – it can also lead to reactivation of the tumor with subsequent growth (3, 32). We would like to point out that this is a potential source of failure, especially in joint replacement. Perioperative systemic therapy in terms of local recurrence rates and morbidity reduction has been the subject of some scientific studies (3, 33). Systemic therapy might be an option in patients with frequent recurrences because authors seem to agree that once a recurrence occurs, the risk of a further recurrence is higher (21, 34-36). For treatment of the diffuse form, a higher recurrence rate and more perioperative complications are described (3, 34).
On the basis of the published literature, we have moved away from radiosynoviorthesis (yttrium-90) (3, 37). Yttrium-90 (β-emitter) is administered at our Nuclear Medicine Department approximately 6 weeks after marginal removal of the synovium, but it is now used only in selected cases (8, 19, 38, 39). Other therapeutic options, which we cannot comment on, but which are mentioned are immunomodulators: all TGCTs express colony-stimulating factor 1 (CSF1). “CSF1R pathway inhibitors have shown activity in TGCT with substantial tumor shrinkage, symptomatic and functional improvement, and long-term disease control (3, 32, 40-43).
Patients mostly become ill between 30 and 50 years of age; with a mean age of 38 (range=9-73) years at diagnosis of our patients, we can confirm that TCGT is a disease of younger adults (11). About equal numbers of men and women are affected, with slightly more women, as also confirmed by our results (mean age=38 years; 42% male and 58% female). From literature and according to the International Consensus Meeting, it became clear from a review of cases that without a uniform therapeutic regimen, the expertise and precautions of the reviewing radiologist and the experience and surgical skills of the surgeon play a critical role and influence both the diagnosis and the therapeutic algorithm more than evidence-based studies currently do (3, 8, 22, 24, 31, 41).
The value of radiotherapy for the treatment of TGCT is largely unclear. Randomized studies do not exist. In larger retrospective studies, doses of 30-50 Gy were used, which resulted in relatively good disease control (1, 5). Nevertheless, even large sarcoma centers rarely use external radiation therapy due to the rather benign nature of the disease and the mostly young age of the patients. Intra-articular radiotherapy (radiosynoviorthesis) is a nuclear medicine treatment that is primarily used for chronic inflammatory diseases of the synovium (1, 5). As TGCT often grows in the form of nodules and generally does not respect tissue boundaries, radiosynoviorthesis per se is not radical enough. A number of case series describe the use of radiosynoviorthesis for TGCT (1, 5). Unfortunately, there is a lack of randomized studies that prove the value of this treatment. Radiosynoviorthesis can be used after partial synovectomy in the diffuse form. In any case, radiosynoviorthesis is pointless in the case of diffuse extra-articular involvement (1, 3, 5).
TGCT is a rare disease, therefore there are only a few studies and hardly any prospective studies available. This makes it difficult to compare results. To be regarded as a limitation here too, we had only 52 cases. Almost 61% of the original cohort was lost. Therefore, there is the recommendation to treat TGCT in centers (3, 17). Firstly, to obtain a better understanding of the symptoms and imaging via clear diagnostic chain. Secondly, to reduce risk factors in diagnostics, such as a biopsy, but above all to document and process the knowledge gained scientifically.
Conclusion
Our findings show a false-positive MRI rate of approximately 29%, for which reason, we recommend performing a biopsy and confirming the diagnosis in all ambiguous cases. MRI with contrast medium is the diagnostic gold standard, while surgical rehabilitation including follow-up examinations is the therapeutic gold standard. We conclude that patients with TGCT should be treated at centers with appropriate expertise.
We are convinced that for a complete diagnosis, in addition to MRI with its pathognomonic low-signal representation of TGCT with inhomogeneous contrast medium uptake on T1- and T2- weighted sequences, each case should be discussed at an interdisciplinary meeting. Should there be any uncertainties surrounding the diagnosis, we recommend a biopsy to confirm it or, if possible, an excision biopsy en bloc, R0.
Acknowledgements
The Authors would like to express their personal thanks to Markus Süβ and Mary Margreiter, as they have played a key role in shaping the progress of the study and have always supported us. The Authors express their appreciation of the contribution of NÖ Landesgesundheitsagentur, legal entity of University Hospitals in Lower Austria, for providing the organizational framework to conduct this research. The Authors would also like to acknowledge support by Open Access Publishing Fund of Karl Landsteiner University of Health Sciences, Krems, Austria.
Footnotes
Authors’ Contributions
Dietmar Dammer conceived the study, obtained ethics application and observed the course of the study. Johannes Neugebauer collected the data together with Markus Süβ, and wrote the article together with Phillip Blum. Gianpaolo Leone and Marko Bergovec together with Dietmar Dammerer gave their clinical input and surgical experience. Alexander Keiler, Markus Neubauer and Patrick Reinbacher were proofreaders.
Conflicts of Interest
The Authors declare that they have no known competing financial interests or personal relationships that could appear to influence the work reported in this article.
- Received September 4, 2024.
- Revision received January 15, 2025.
- Accepted January 20, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.