Synovial hemangioma, a rare benign tumor that occurs most frequently in the knee in children and young adults, has four histological subtypes: Venous, arteriovenous, cavernous and capillary hemangiomas. Since the clinical presentation and radiological findings of synovial hemangioma are non-specific, there is frequently a long period between the onset and the diagnosis. The cases of nine patients, pathologically diagnosed with synovial hemangioma and surgically treated, were retrospectively analyzed. All nine patients had persistent knee pain. In addition, three patients also had a swollen knee with intra-articular hemorrhage. Plain radiography revealed intra-articular phleboliths in two patients. In seven patients, T1-weighted magnetic resonance imaging showed low signal intensity with small signal voids. On T2-weighted imaging, all patients showed high signal intensity containing small signal voids. All patients underwent surgical excision; there was no postoperative recurrence after the final operation, and the knee pain had disappeared at the final follow-up. From the pathological findings, the diagnoses were venous hemangioma, cavernous hemangioma and capillary hemangioma (three patients each).
Synovial hemangioma is a rare benign tumor that occurs most frequently in the knee in children and young adults (
After obtaining institutional review board approval, we identified the nine patients (five males and four females) who were each diagnose with synovial hemangioma of a knee and treated at Fukushima Medical University Hospital (Fukushima, Japan) or Fukushima Red Cross Hospital (Fukushima, Japan) during the period between January 1998 and December 2021. The patients' median age at surgery was 22 years (range 1-43 yrs). We retrospectively reviewed the patients' clinical presentations, radiological findings, surgical procedures, postoperative courses, and pathological findings.
On plain radiographs, intra-articular calcification consistent with phleboliths was observed in two patients. T1-weighted magnetic resonance imaging (MRI) showed low signal intensity with and without small signal voids in six patients and two patients, respectively, and high signal intensity without signal voids in the other patient (Patient #7). In all nine patients, T2-weighted imaging (WI) showed high signal intensity containing small signal voids and intra-tumoral septum comparable to joint fluid and fat. In the present cases, the synovial hemangiomas showed a small honeycomb pattern with a thin septum and a lobulated pattern. T2*-WI showed high intensity, containing low intensity area that could be post-hemorrhagic changes. Gadolinium enhancement was performed for four patients, with heterogeneous staining of the tumor (
The surgery was performed without a preoperative biopsy in all nine patients. The duration from the patient's initial visit to the surgery ranged from 1 to 9 months. Open resection (n=2 patients), arthroscopic resection (n=3), and conversion to open resection after arthroscopy (n=4) were performed. In one of the two patients who underwent an open resection (Patient #6) and in one of the patients who underwent arthroscopic resection (Patient #4), additional resections for local recurrence were performed.
The average postoperative follow-up period from the final surgery was 17.2 months, with the longest follow-up period being 60 months. After the final surgery, none of the patients developed postoperative recurrence, and their knee pain had disappeared at the final follow-up (
Based on the pathological findings of surgical specimens, the diagnoses were venous hemangioma (
A 14-year-old boy presented with a 1-month history of continuous left knee pain. Plain radiographs of his left knee joint showed an intra-articular calcification consistent with phlebolith (
Synovial hemangioma has accounted for 0.07% of all soft tissue tumors and 0.78% of resected hemangiomas (
The clinical findings of synovial hemangioma are similar to those of D-TSGCT and hemophilic arthropathy. Both may cause sudden and non-traumatic intra-articular hemorrhage, and a delayed diagnosis can thus lead to osteoarthritis and joint function disuse (
Plain radiography is usually the first radiological examination conducted in patients who are suspected to have knee pain and/or intra-articular hemorrhage. Plain radiography of a synovial hemangioma typically shows no abnormality, but periosteal reactions, osteolysis, osteopenia, intra-articular phleboliths, osteoarthritis, or soft tissue swelling may be observed (
Compared to muscle tissue, an MRI evaluation of a synovial hemangioma reveals low to iso-intensity on T1-WI and high intensity on T2-WI, but no characteristic findings are seen on fat-suppression imaging (
The size of the intra-tumoral vascular lumen differs depending on the subtype of hemangioma. In a larger vascular lumen, the blood flow becomes slower, and the contrast will be partial. It was reported that large venous synovial hemangiomas showed high signal intensity on T1-WI due to slow blood flow (
The pathological diagnosis of synovial hemangioma is classified into the venous, arteriovenous, cavernous, and capillary subtypes (
An important differential diagnosis for synovial hemangioma is D-TSGCT. Because of these two tumors' similarity in age of onset and clinical presentation, D-TSGCT can be difficult to distinguish in clinical practice. We noted that all nine of the present patients' cases showed high signal intensity on T2-WI, similar to circumferential fatty tissue and joint fluid. D-TSGCTs usually show low-signal changes on T2-WI due to hemosiderin deposition (
In the present series, diffusion-weighted imaging (DWI) was performed in one of the capillary subtype cases. The DWI showed low signal intensity in the tumor and the high apparent diffusion coefficient (ADC) value of 2,184/mm2/sec, excluding the area with high signal intensity that was thought to be post-hemorrhage changes (
Despite their similar clinical findings, synovial hemangiomas and D-TSGCT have different postoperative outcomes. Although two of the nine patients in our present series had postoperative recurrence because of an uncomplete resection, synovial hemangiomas have been reported to have a postoperative recurrence rate of 3.4% and a symptom persistence rate of 11.4% (
A limitation of the present study is that the imaging conditions used for the MRI examinations were not standardized, and DWI and ADC values were measured in only one case. When we encounter patients with intra-articular tumors including synovial hemangioma and D-TSGCT in the future, we will obtain MRI findings under uniform conditions for the purpose of increasing the number of cases for an additional review.
Not applicable.
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
TS and MH participated in the design of study, interpreted clinical data and drafted the manuscript. YK and HY participated in the design of the clinical part of this study and provided clinical data. OH analyzed and contributed to the interpretation of the radiological findings. SY analyzed and contributed to the interpretation of the pathological findings. SK participated in the design of study and supervised the project. MH and SK confirm the authenticity of all the raw data. All authors read and approved the final manuscript.
This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study was approved by the Ethical Review Committee of Fukushima Medical University (approval no. 2022-36).
This study is retrospective study. Patients were not required to give informed consent to the study because the analysis used anonymous clinical data that were obtained after each patient agreed to treatment by written consent. We applied Opt-out method to obtain consent on this study. The Opt-out was approved by the Ethical Review Committee of Fukushima Medical University.
The authors declare that they have no competing interests.
Venous synovial hemangioma (patient #4). On magnetic resonance imaging, (A) T1-WI (sagittal view) showed low-signal intensity without a signal void and (B) T2-WI (sagittal view) showed high signal intensity containing small signal void (arrow). (C) Gadolinium-enhanced T1-weighted fat-suppression imaging (sagittal view) showed heterogeneous staining of the tumor. Pathologically, the vascular smooth muscle is thickened and vascular lumen is dilated: (D) low-power field (magnification, x20), H-E staining; (E) high-power field (magnification, x200), H-E staining. Slow blood flow is prone to forming thrombus and phleboliths. H-E, hematoxylin and eosin.
Capillary hemangioma (patient #9). (A) On magnetic resonance imaging, T2-weighed fat-suppression imaging showed high signal intensity containing small signal void (arrow). (B) Gadolinium enhancement showed heterogeneous staining of the tumor. (C) DWI showed low-signal intensity in the tumor and (D) high apparent diffusion coefficient (ADC) value of 2,184/mm2/sec. Microscopic findings show small vascular lumen, which was less likely to form thrombus or phleboliths, and this tumor was difficult to diagnose without microscopic confirmation: (E) low-power field (magnification, x20), H-E staining; (F) high-power field, (magnification, x200), H-E staining. H-E, hematoxylin and eosin.
Representative case (patient #3, a 14-year-old boy). (A) Intra-articular calcification consistent with phlebolith (arrow). (B) Magnetic resonance imaging demonstrating a mass with high signal intensity containing a small signal void on T2-weighted images (sagittal plane). (C) Intraoperative gross findings revealing a dark red tumor covered by the synovium on the surface of lateral condyle of the femur. (D) Microscopic findings of the surgical specimen showing expanded blood lumens lacking vascular smooth muscle (high-power field, magnification x200). (E) Formation of thromboses and phleboliths caused by slow blood flow was easily identified (arrows) (low-power field, magnification x20).
Patients' clinical characteristics.
Patient no. | Age, years | Sex | Symptom | Intra-articular hemorrhage | Symptom duration (until first visit) | Affected side | Location |
---|---|---|---|---|---|---|---|
1 | 1 | F | Pain, swelling | + | 3 months | Right | Infrapatella and suprapatellar fat-pad |
2 | 11 | F | Pain | - | 3 months | Right | Suprapatellar bursa |
3 | 14 | M | Pain | - | 1 month | Left | In front of the medial condyle of femur |
4 | 17 | F | Pain | - | 0.5 months | Left | Suprapatellar bursa |
5 | 22 | M | Pain, Swelling | + | 1 day | Right | Infra patella fat-pad |
6 | 24 | M | Pain | - | 10 years | Left | Inter condylar fossa, infra patellar fat-pad and posterolateral capsule |
7 | 28 | M | Pain | - | 1 year | Right | Suprapatella fat-pad |
8 | 34 | F | Pain | - | 6 months | Left | In front of the medial condyle of femur |
9 | 43 | M | Pain, swelling | + | 1 months | Left | Infrapatellar fat-pad |
F, female; M, male.
Radiological characteristics of the tumor.
Patient no. | Size, cm | Plain radiography | T1-weighted image | T2*-weighted image | Gadolinium- enhanced, T1- weighted fat- suppression image | DWI ADC-mapping |
---|---|---|---|---|---|---|
1 | 3.7x2.4x1.9 | Phlebolith | Low intensity | High intensity, containing low intensity area of hemorrhage | - | - |
2 | 3.0x2.8x1.5 | Normal | Low intensity, containing small signal voids | - | - | - |
3 | 7.2x4.1x0.9 | Cortical erosion, phlebolith | Low intensity, containing small signal voids | - | Heterogeneous enhancement | - |
4 | 3.1x2.1x1.1 | Normal | Low intensity | - | Heterogeneous enhancement | - |
5 | 1.5x0.8 | Normal | Low intensity, containing small signal voids | - | - | - |
6 | Diffuse/multiple | Normal | Low intensity, containing small signal voids | - | - | - |
7 | 3.8x3.3x1.1 | Normal | High intensity, containing small signal voids | - | Heterogeneous enhancement | - |
8 | 4x3.5x1.1 | Normal | Low intensity, containing small signal voids | - | - | - |
9 | 4.7x5.9x2.9 | Normal | Low intensity, containing small signal voids | High intensity, containing low intensity area of hemorrhage | Heterogeneous enhancement | ADC value 2184 |
For all nine patients, the T2-weighted imaging result was high intensity, containing small signal voids. DWI, diffusion-weighted image; ADC, apparent diffusion coefficient.
Surgical procedure, pathological diagnoses and oncological prognosis.
Patient no. | Duration from initial visit to surgery, months | Procedure | Pathological sub-type | Follow-up period, months | Prognosis |
---|---|---|---|---|---|
1 | 45 | Open resection after arthroscopy | Cavernous hemangioma | 11 | NED |
2 | 5 | Arthroscopic resection | Capillary hemangioma | 12 | NED |
3 | 9 | Open resection | Cavernous hemangioma | 1 | NED |
4 | 2 | 1st surgery: Arthroscopic resection 2nd surgery: Open resection | Venous hemangioma | 9 | NED |
5 | 3 | Arthroscopic resection | Capillary hemangioma | 6 | NED |
6 | 1 | 1st surgery: Open resection 2nd surgery: Arthroscopic resection | Venous hemangioma | 41 | NED |
7 | 3 | Open resection after arthroscopy | Cavernous hemangioma | 13 | NED |
8 | 7 | Open resection | Venous hemangioma | 2 | NED |
9 | 2 | Open resection after arthroscopy | Capillary hemangioma | 60 | NED |
aNED after the resection of local recurrence. NED, no evidence of disease.