2025 年 12 巻 p. 415-420
Marginal zone B cell lymphoma, which can sometimes resemble a meningioma, originates in the dura mater. Reports of marginal zone B cell lymphoma complicated by intracranial hemorrhage are rare. There have been five reported cases of dural marginal zone B cell lymphoma complicated by acute subdural hematoma. To our knowledge, this is the first reported case of dural marginal zone B cell lymphoma diagnosed at the onset of a non-traumatic acute epidural hematoma. A 74-year-old man presented to our hospital with complaints of right hemiparesis without a history of head injury. He was diagnosed with left acute epidural hematoma based on the results of a computed tomography scan of the head; however, his serum soluble interleukin-2 receptor level was elevated (3,520 U/mL), and whole-body contrast-enhanced computed tomography revealed intra-abdominal and bilateral inguinal lymphadenopathy. Contrast-enhanced magnetic resonance imaging of the head revealed a well-enhanced, thickened dura mater with an acute epidural hematoma. The patient underwent a biopsy of the thickened dura mater and a hematoma removal, which on histopathology revealed dural marginal zone B cell lymphoma, as did a biopsy of an inguinal lymph node. After chemotherapy, the thickened dura mater shrank, and the marginal zone B cell lymphoma lesions showed remission. However, the mechanism underlying acute epidural hematoma in the dural marginal zone B cell lymphoma remains unclear. This case report provides new insights into dural marginal zone B cell lymphoma as a cause of non-traumatic acute epidural hematoma.
Non-traumatic acute epidural hematoma (AEDH), primarily caused by coagulopathy (40% of cases) and tumors (mainly cancer; 14% of cases), is a rare occurrence.1-6) Dural marginal zone B cell lymphoma (MZL) arises from the dura mater7) and accounts for 7% of B cell lymphomas. MZL is primarily associated with the gastric mucosa5) and is classified into three types based on anatomical location: extranodal or mucosa-associated lymphoid tissue (MALT), nodal, and splenic. The histopathological and immunohistochemical features of the three types are similar;8) therefore, MZL of the central nervous system is a type of MALT lymphoma. The site of origin of dural MZL varies, and neuroradiological findings often resemble those seen with meningiomas.7) Although there are a few reported cases of dural MZL causing intracranial hematoma,9-13) each of these cases has been complicated by acute subdural hematoma. We describe herein the rare case of non-traumatic AEDH caused by dural MZL.
A 74-year-old man presented to our hospital with a left temporal headache that began a few days before his presentation. He subsequently began to develop right hemiparesis and motor aphasia and was referred to our hospital with impaired consciousness (Glasgow Coma Scale score, E4V4M5). He had no history of head trauma, and blood tests showed no evidence of coagulopathy or thrombocytopenia. Computed tomography (CT) of the head revealed a left-sided AEDH and no bone destruction (Fig. 1A and B), although at the time of admission, the cause of the AEDH was unclear and the midline shift was mild. Upon sudden clinical deterioration, preparations for emergency surgery were made, and a whole-body CT was performed before the operation. Whole-body contrast-enhanced CT showed intra-abdominal and bilateral inguinal lymphadenopathy (Fig. 1C), suggestive of lymphoma. Additionally, the patient's serum soluble interleukin-2 receptor (sIL-2R) levels were elevated (3,520 U/mL). Given the absence of symptom progression and only mild impaired consciousness, an additional contrast-enhanced magnetic resonance imaging (MRI) of the head was performed. It revealed a well-enhanced dura mater coexisting with the AEDH (Fig. 1D). On T2-weighted images (Fig. 1E) and diffusion-weighted imaging (Fig. 1F), the hematoma was primarily composed of acute-phase components. Lymphoma was suspected as the potential cause of the AEDH with dural thickening; therefore, we performed a dural biopsy and maximal possible hematoma evacuation through a small craniotomy under local anesthesia. Intraoperatively, the epidural hematoma consisted of both soft and firm components, consistent with an acute hematoma. The outer layer of the dura mater was edematous, thickened, and exhibited a mildly yellowish discoloration. The inner layer of the dura also appeared mildly edematous, although the surface appeared nearly normal upon gross inspection. The cortical surface and arachnoid membrane maintained their normal gross appearance.
Initial computed tomography (CT) scan of the brain shows a left epidural lesion, suspected to be an epidural hematoma, meningioma, or another type of epidural tumor with no evidence of bone destruction (A, B); CT image of the inguinal region shows bilateral inguinal lymphadenopathy (white arrows) (C); contrast-enhanced magnetic resonance imaging (MRI) scan of the head shows thickening of the dura mater (D); T2-weighted imaging on MRI, the center of the hematoma showed high signal intensity, while the periphery appeared hypointense (E); DWI imaging on MRI, the hematoma exhibited a high signal intensity at the center with a low signal intensity in the surrounding area (F).
DWI: diffusion-weighted imaging
Hematoxylin and eosin staining revealed the infiltration of small-to-medium-sized lymphocytes into the dura (Fig. 2A and B), the immunohistochemical examination of which indicated cluster of differentiation (CD) 20-positivity (Fig. 2C), CD-3-negativity (Fig. 2D), weak CD-10-positivity (Fig. 2E), and a Ki-67 index less than 5% (Fig. 2F). These findings led to a diagnosis of dural MZL, a low-grade B cell lymphoma. Additionally, an inguinal lymph node biopsy revealed the same diagnosis.
Hematoxylin and eosin (H&E) staining (×40) shows lymphocyte infiltration into the dura mater (A); H&E staining (×200) reveals neovascularization in the outer dura mater (B); lymphocytes are positive for cluster of differentiation (CD) 20 (C) and negative for CD3 (×200) (D); CD10 shows weak positivity (×200) (E); and Ki-67 index < 5% (×200) (F); Vascular endothelial growth factor staining was barely detectable (×200) (G); CD34 staining revealed neovascularization, with a CD34 microvessel density of 58/mm2 (×200) (H); Elastica van Gieson staining demonstrates lymphocyte infiltration into newly formed blood vessels with the loss of collagen fibers in the vascular walls (white triangle), black triangles indicate blood vessels with preserved collagen fibers (×200) (I).
Additional staining was performed on the intracranial biopsy specimens using vascular endothelial growth factor (VEGF) stains (Fig. 2G), CD34 (Fig. 2H), and elastica van Gieson (EVG) stains (Fig. 2I). VEGF expression was minimal, while intra-tumor staining showed positive staining for CD34. EVG staining revealed positivity in the neovascular walls and suggested infiltration of lymphoma cells into some of the vascular walls.
Postoperatively, the patient's right hemiparesis and motor aphasia improved. He subsequently underwent six courses of chemotherapy with bendamustine and rituximab, and the dural lesion and lymph node swelling were resolved on follow-up imaging over the following year (Fig. 3A and B).
Postoperative CT confirms the removal of the hematoma (A); Contrast-enhanced MRI at 6 months postoperatively shows a reduction in dural thickening (B).
CT: computed tomography; MRI: magnetic resonance imaging
We have described herein the case of MZL resulting in an AEDH, where the thickened dura mater was enhanced on MRI. Inguinal lymph node involvement was detected using whole-body CT, and elevated sIL-2R levels suggested lymph node metastasis. The pathological findings of both lesions contributed to the administration of additional chemotherapy.
Dural MZL is a low-grade lymphoma that arises from the dura mater and lacks lymphoid tissue; however, the etiology of MZL in the dura mater remains unclear.4,5,7,14-16) Dural MZL is difficult to diagnose based on neuroimaging alone; therefore, blood tests and whole-body CT are important for guiding the diagnosis, as in our case.7,16,17) The final diagnosis should be confirmed by the biopsy of the dura mater. Given the high radiosensitivity of dural MZL, treatment is typically surgical resection and/or radiotherapy. Systemic chemotherapy is usually unnecessary in localized cases; however, gamma-knife surgery is an effective option for recurrent lesions, highlighting the importance of local disease control.7)
A total of six cases (including this case) of dural MZL presenting with intracranial hemorrhage have been reported (Table 1). The mean patient age was 65 years, and excluding this case, all of the patients were females, and all cases were pathologically diagnosed with MZL. The site of occurrence was most frequently the frontal region, and other than this case, the hemorrhages were all located in the subdural space. Treatment included chemotherapy (33% of cases) and radiotherapy (67% of cases), both of which resulted in favorable outcomes. However, there is no consensus regarding the most effective treatment for dural MZL.7) The two-year overall survival rate was reported to be 80%, while the two-year disease-free survival rate was 59% in primary dural MZL. Patient prognosis worsens in the presence of systemic metastasis;15) therefore, our case requires careful long-term follow-up.
MZL Complicated with Intracranial Hemorrhage
| Author, Year | Age (years) | Sex | Symptoms | Site of intracranial lesion | Site of hematoma | Other lymphoid swelling | Ki-67 index | Treatment | Outcome |
|---|---|---|---|---|---|---|---|---|---|
| EDH: epidural hematoma; F: female; RT: radiation therapy; Lt: left; M: male; MZL: marginal zone B cell lymphoma; Rt: right; SDH: subdural hematoma; WBRT: whole-brain radiation therapy | |||||||||
| Goetz et al.,9 2002 | 64 | F | Headache, Lt. hemiparesis. | Rt. Frontoparietal | SDH | No description | 6.90% | WBRT | No recurrence at 3 months |
| Gocmen et al.,10 2010 | 45 | F | Seizures, speech disturbances | Lt. Frontotemporal | SDH | Nothing | 5%–10% | Chemotherapy | No recurrence at last follow-up |
| Jesionek-Kupnicka et al.,11 2013 | 60 | F | Rt. upper limb weakness, facial cramping | Rt. Parietal | SDH | Nothing | 10% | RT | No recurrence at last follow-up |
| Thibodeau et al.,12 2022 | 70 | F | Loss of consciousness | Lt. Frontal | SDH | Nothing | 5% | RT | No recurrence at 1 year |
| Neeley et al.,13 2020 | 77 | F | Slurred speech and transient lt. facial droop | Rt. Frontal | SDH | Nothing | No description | RT | No recurrence at 3 months |
| Present case | 74 | M | Rt. hemiparesis, motor aphasias | Lt. Frontoparietal | EDH | Abdominal, bilateral inguinal | 5% | Chemotherapy | No recurrence at 6 months |
Although intracranial lymphoma is rarely associated with hemorrhage, in this case, we performed a pathological examination for angiogenesis. VEGF staining was performed in this case, which showed minimal expression, and no association between intracranial hemorrhage and VEGF was suggested. In contrast, staining for CD34, a marker of intra-tumoral angiogenesis, revealed high expression, indicating neovascularization within the tumor in the dura mater. Additionally, elastica van Gieson staining revealed collagen fibers in the microvessels, while areas where lymphoma cells had infiltrated the blood vessels showed a loss of collagen fibers. These findings are suggestive of vascular wall destruction, which may have caused the bleeding observed in this case. Whether a patient develops an epidural or subdural hematoma may depend on whether neovascularization is observed on the epidural side or the subdural side, as suggested by the pathological findings in our patient. However, no previous reports have discussed the presence or distribution of neovascularization in similar patients, and this remains a matter of speculation.
Non-traumatic AEDH is often caused by coagulopathy or vascular malformations; it can also be associated with tumors, particularly metastatic tumors.1,6) Therefore, when non-traumatic AEDH is identified, it is crucial to assess the underlying factors. Among the reported cases, these tumors include meningiomas and metastases from primary cancers such as hepatocellular carcinoma and lung cancer.6) In addition, our case suggests that lymphoma also has the potential to metastasize to the dura. Even in situations requiring emergency surgery, a dural biopsy should be considered. As few reports have provided pathological insights into patients with MZL with intracranial hematomas, the pathological understanding of this disease remains limited, highlighting the need for further research and data collection.
On a related note, a previous report described a patient with diffuse large B cell lymphoma in the epidural space extending through the skull to the overlying skin, although no discussion of pathological progression was provided, limiting comparison with our patient.18) While MZL (MALT lymphoma) can also arise in the skin, contiguous spread across anatomical barriers, such as from the dura to the skin, is extremely rare. This may reflect differences in biological behavior and progression patterns between the two diseases. In our patient, the findings suggested that the MZL had undergone distant metastasis. However, there was no evidence of direct invasion into adjacent organs. Furthermore, the mechanisms underlying MZL development in the dura or other extranodal sites lacking lymphoid tissue remain unclear, highlighting the need for further investigation.
To the best of our knowledge, this is the first reported case of dural MZL complicated by AEDH. Although a handful of cases of MZL with intracranial hemorrhage have been previously reported, few included pathological investigations, and none involved AEDH. In this case, lymphoma cells infiltrating the microvessels were thought to play a role in the hemorrhage. Therefore, when non-traumatic AEDH is encountered, as in this case, a dural biopsy is crucial for determining the underlying cause.
The authors thank Editage (www.editage.jp) for the English language polishing.
KM: Drafted the manuscript, composed the figures and tables, and contributed to manuscript writing. KF, YO, JK, TM: Reviewed and corrected the manuscript. KT: Performed pathological examinations and interpreted pathological findings. TF: Interpreted pathological findings. MM: Reviewed and corrected the manuscript.
All procedures were performed in accordance with the ethical standards of the institution involved and the 1964 Declaration of Helsinki and its later amendments.
The patient provided informed consent to the publication of his case report and the related images.
All authors have no conflict of interest.
