PVNS of the TMJ with skull base extension is exceedingly rare. Almost all researches were a case series report and retrospective study. We aimed to present a large case series of the condition.
Contrary to other studies, our study found that the condition had male predominance (1.64:1) which was contrary to other studies [7]. The disease usually occurred in the middle-aged, although it could be seen even at the age of 18 in our case series. Pain and mass were the most common presentations; symptoms related to neurological symptoms such as hearing loss and tinnitus were not uncommon, especially in cases with serious petrosal bone erosion [8]. Trauma has been reported to be contributing factor [5], but it was not common.
Imaging examinations are necessary for the initial and differential diagnosis, and surgical evaluation [9]. Careful reading images before surgery are important for surgical planning. On MRI, the lesion generally shows low signal intensity on T2 images and “blooming” artifact on gradient-echo images due to the high hemosiderin concentration[10]. The lesion on contrast MRI presents variable levels of enhancement. CT mainly presents bony destruction of the mandibular condyles and skull base invasion and areas of central calcification. The differential diagnosis mainly consists of chondroblastoma, reparative granuloma, chondrosarcoma, giant cell tumor of bone, and synovial sarcoma. Digital subtraction angiography may show highly vascular lesions with multiple arteries blood supply and arteriovenous shunting, especially for giant tumors. It is useful to conduct preoperative embolization to decrease blood loss during operation. One of our cases was performed digital subtraction angiography embolism because the patient had a big tumor with 3.8 cm in diameters.
Pathologic evaluation is essential for a definitive diagnosis. Biopsy and postoperative pathological examination are the most common method. Because PVNS of TMJ is involved in the facial region and the tumor is relatively superficial, the pathologic examination before surgery provides valuable information for diagnosis and treatment. FNAB with the guidance of ultrasound or CT is a simple method [11]. Pathologic results reveal that the lesion is rich in giant cells and focal cartilage formation. Different kinds of cells, including mononuclear histiocytes, multinucleated giant cells, and xanthoma cells, are present in the lesion[12]. The differential diagnosis includes giant cell granuloma, giant cell tumor, and brown tumor[13]. Due to the overlapping of histopathological characteristics of these lesions, the definitive diagnosis may need additional testing tools. A brown tumor can be ruled out with normal parathyroid and calcium levels. Giant cell tumors are generally more destructive. Giant cell granulomas are more commonly bone-based. Evidence of hemosiderin deposition may also aid in the appropriate diagnosis of PVNS[7]. In our cases, one patient has undergone an FNAB test.
The most effective treatment for the condition is surgery. GTR leads to a lower recurrence rate and is associated with a longer period of relief when compared with SR. However, GTR may bring damage to occlusal function because of the destruction of the functions of the TMJ. Some scholars think that SR may be a risk factor for recurrence[14]. RT after SR for residual tumors is an adjuvant treatment [15]. The most common radiation dose is 35-60Gy in 20–30 fractions. Therefore, GTR should be conducted when it was feasible. When complete removal of the tumor could not be achieved, RT after surgery is recommended. In our series, GTR was achieved in 9 patients and 1 patient had SR for facial nerve preservation. Truly GTR included completely bony resection and soft tissue resection. We often perform extended removal for suspicious bony invasion including the greater sphenoid wing, squamous and petrous parts of the temporal bone. All our cases were extradural lesions, invasion of the periosteal layer of the dura mater was found in 10 patients. Meticulously sharp separation was performed to remove the outer layer dura for complete resection and preserve the inner layer to prevent leakage of CSF. The operation involved multi-disciplinary cooperation, including neurosurgeons, otolaryngologists, and maxillofacial surgeons. Multi-disciplinary cooperative surgeries could achieve more radical resection, reliable skull base reconstruction, and better cosmetic results.
There are two views on defect repair. Safaee et al[7] believed that it was not necessary to reconstruct the middle skull base and that autologous fat can be used to fill large defects. However, Liu et al [16] reported that repairing bony defects in the skull base can provide a satisfying contour and stable support for the brain. Unfortunately, there is a lack of enough information relating to the comparison of cases with and without reconstruction. We mainly used temporalis muscle flaps and autologous fat to reconstruct the skull base, especially for big tumors with large defects [17].
Postoperative facial palsy is a common complication after surgery. Therefore, the perseveration of the facial nerve is important. Intraoperative electrophysiological monitoring could identify and protect the facial nerve. Although facial palsy was the most common postoperative complication, it was worth that we should perform GTR because facial palsy could recovery to some extent.
In conclusion, although PVNS of the TMJ with the extension of the skull base is destructive and leads to intracranial extension, surgery can achieve an excellent outcome, especially for GTR. The most common symptom associated with this condition is pain and swelling. RT may be reserved for extensive or recurrent disease. Because of the relatively high rates of recurrence, a sufficient period for follow-up is necessary.