Selective Percutaneous Controlled Radiofrequency Thermocoagulation of the Gasserian Ganglion To Control Facial Pain Due to Medication-Related Osteonecrosis of the Jaw

Introduction

Bisphosphonates are potent inhibitors of bone resorption. As a result, they are widely used to prevent skeletal-related events in patients with osteoporosis, solid cancers with bone metastasis, and multiple myeloma, and to manage hypercalcemia of malignancy. In contrast, Marx ¹ was the first to report, in 2003, that some patients taking bisphosphonates developed serious osteonecrosis of the jaw. This was followed by a similar report, ² and this condition was called “bisphosphonate-related osteonecrosis of the jaw,” a serious complication of bisphosphonate treatment. Recently, development of similar jaw osteonecrosis was also reported in patients being treated with denosumab^3,4 and bevacizumab. ⁵

The Position Paper 2014 Update ⁶ of the American Association of Oral and Maxillofacial Surgeons adopted the name “medication-related osteonecrosis of the jaw” (MRONJ) for osteonecrosis cases associated with antiresorptive and antiangiogenic agents. Patients may be considered to have MRONJ if all of the following are present: (1) current or previous treatment with antiresorptive or antiangiogenic agents; (2) exposed bone or bone that can be probed through an intraoral or extraoral fistula in the maxillofacial region that has persisted for longer than eight weeks; and (3) no history of radiation therapy to the jaws or obvious metastatic disease to the jaws.

The underlying diseases of MRONJ were reported to be mostly cancers, including multiple myeloma in 31.2%, breast cancer in 29.5%, prostate cancer in 9.6%, other cancers in 7.6%, and metastasis in 2.5%, compared with osteoporosis in 19.6%. ⁷ Common symptoms of MRONJ include bone exposure, pain, swelling, paresthesia, suppuration, soft tissue ulceration, intraoral fistula, extraoral fistula, and loosening of teeth. ⁸ The pathophysiology of MRONJ remains unclear, and appropriate approaches for its prevention and treatment have not been established. ⁹ Since treatment of MRONJ is extremely difficult, ¹⁰ it is a major, distressing, long-term complication. ¹¹

A 68-year-old woman with facial pain associated with MRONJ is presented. This is the first case report of selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion to control the pain due to MRONJ.

Case History

A 68-year-old woman with breast cancer had recurrence in the left trapezius muscle 14 years after left mastectomy and was started on chemotherapy. One year later, bone scintigraphy detected metastasis in the left scapula, and intravenous bisphosphonate (zoledronate) was started to prevent skeletal-related events. In the same year, a computed tomography (CT) scan detected left pleural metastasis, and oxycodone was prescribed for cancer pain in the left chest. A good analgesic effect was obtained with extended-release oxycodone at 70 mg/day and naproxen at 300 mg/day. One year after starting the intravenous zoledronate, the patient began to complain of swelling and pain in the left mandibular molar gingiva. A dentist performed conservative treatment consisting of antimicrobial therapy and local irrigation, but the swelling and pain were treatment resistant and progressive. Three months later, exposed bone was noted in the same left mandibular molar region (Fig. 1). CT showed left mandibular osteomyelitis. Since the patient had been administered zoledronate at 4 mg every four weeks for six months, and then denosumab at 120 mg every three weeks for eight months, she was diagnosed as having MRONJ due to antiresorptive agents. Even after denosumab was discontinued after diagnosis of MRONJ, the patient's medical condition continued to worsen, with an extraoral fistula. In addition, hypoesthesia and numbness developed in the left chin and lower lip. The patient rated the severity of continuous pain in the left jaw as 7 out of 10 on the Numerical Pain Rating Scale. The pain was especially severe when eating, and this led to reduced oral intake. Because the patient also complained of severe pain at the time of local irrigation, adequate local care could not be performed, and infections of the mandible and the gingiva secondary to the MRONJ became uncontrollable. She was very distressed due to left jaw pain. To treat the pain, immediate-release oxycodone, 10 mg, and loxoprofen sodium, 60 mg, were added to her treatment regimen, but they had no effect. We recommended her to take neuropathic agents, carbamazepine or amitriptyline, but she refused to take them that might worsen her sleepiness due to extended-release oxycodone at 70 mg/day.

OPEN IN VIEWER

FIG. 1.

Exposed necrotic bone extending beyond the region of alveolar bone, resulting in extraoral fistula.

Consultations among the Breast Surgery Department, Dentistry, and Oral Surgery Department, and our department resulted in the following conclusions regarding the patient's condition. (1) The prognosis for breast cancer was six months to about one year. (2) The MRONJ was treatment resistant and progressive, and alleviation was unlikely within the patient's life expectancy. (3) Surgical resection would be invasive and thus not indicated, and as conservative treatment, it was necessary to continue local irrigation despite the accompanying severe pain. (4) Analgesics would not afford satisfactory pain relief. Finally, (5) the patient's quality of life (QOL) was severely reduced by the MRONJ. The patient's pain due to the MRONJ was nociceptive pain and neuropathic pain in the distribution of the third branch of the trigeminal nerve, and it was resistant to analgesics. For these reasons, it was decided to investigate the usefulness of selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion to control that pain. The patient's informed consent to try this therapy was obtained after she was given an explanation that there was a possibility of actually aggravating the hypoesthesia and numbness in the left chin and lower lip, a possibility of expanding hypoesthesia and numbness in the distribution of the third branch of the trigeminal nerve, and a possibility of paralysis of the masseter muscle. CT showed no lesions at the planned needle insertion site or the needle track. The procedure was performed under fluoroscopic control. The patient was lightly sedated by intravenous propofol. The needle was inserted at a point 2.5 cm lateral to the left corner of the mouth. It was advanced toward the foramen ovale and from there into Meckel's cavity. Sensory stimulation was performed, and paresthesia was confirmed to be felt in the third division selectively at 0.5 mA. After administration of 0.2 mL of 2% lidocaine, a radiofrequency lesion was produced by 90°C for 180 seconds (Fig. 2). There were no serious complications, including meningitis and paralysis of the masseter muscle, during or after the procedure.

OPEN IN VIEWER

FIG. 2.

Position of the radiofrequency needle in the anteroposterior oblique view. The needle tip is placed in the Gasserian ganglion through the foramen ovale.

After the procedure, the continuous pain in the left jaw, the pain felt when eating, and the pain at the time of local irrigation all disappeared. Adequate local irrigation could be performed because it was no longer painful. There was no change in the MRONJ-derived hypoesthesia and numbness in the left chin and lower lip. Hypoesthesia in the left parotid gland region and the left external auditory canal, which was in the distribution of the third branch of the trigeminal nerve, developed anew, but the patient was able to tolerate them. Oral intake also improved, and the patient's satisfaction was very high. The conservative therapy was continued, and after one month, the skin fistula closed. There was no worsening of the MRONJ, such as recurrence of pain, infections, and expansion of the bone exposure, from the time of selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion to eight months later, when the patient died at home due to progression of the breast cancer.

Discussion

The pain due to MRONJ is pain that is due to cancer treatment. In cancer patients exposed to zoledronate, the incidence of MRONJ ranged from 0.7% to 6.7%, and with exposure to denosumab, the range was 0.7% to 1.9%, ⁶ indicating that MRONJ is by no means a rare complication of antiresorptive agents treatment. As in cases of chemotherapy-related pain syndromes and radiation-related pain syndromes, the pain due to MRONJ should be recognized as pain due to cancer treatment.

Treatment objectives for patients with an established diagnosis of MRONJ are to eliminate pain, control infection of soft and hard tissues, and minimize the progression or occurrence of bone necrosis. However, for elimination of pain, there are no concrete methods other than antimicrobial agents for their anti-inflammatory effects and surgical resection of necrotic bone. Relief of symptoms by controlling pain and managing infection would be the treatment goal in terms of maintaining the QOL of oncology patients. In this patient, the MRONJ was treatment resistant and progressive, and analgesia through medication was difficult. For maintaining the patient's QOL, it was decided for selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion, since if successful it would have an immediate analgesic effect and pain relief would be long lasting.

Radiofrequency thermocoagulation of the Gasserian ganglion is a common technique to treat trigeminal neuralgia^12,13 and was first described by Sweet and Wepsic in 1974. ¹⁴ Moreover, it was effective for pain in the trigeminal nerve distribution due to malignant tumors such as tongue cancer, ¹⁵ for which pain relief by analgesics was inadequate. Mendelsohn et al. ¹⁵ recommended the procedure because it is quick and affords immediate benefit with infrequent complications in patients with intractable facial pain, and it is, therefore, ideally suited for the palliative care setting. This procedure uses a neurostimulator, and by adjusting the position of the needle tip, it is possible to selectively block each of the branches of the trigeminal nerve. ¹⁶ The median time to recurrence of pain was 24 months when treating trigeminal neuralgia by this procedure, ¹⁷ and its effectiveness is thought to be superior to that with alcohol block ¹⁸ or pulsed radiofrequency. ¹⁹ The most prevalent complication of the procedure is sensory loss in the treated division. ²⁰ Kanpolat et al. ¹⁶ reported a decreased corneal reflex (5.7%), weakness and paralysis of the masseter muscle (4.1%), dysesthesia (1%), anesthesia dolorosa (0.8%), keratitis (0.6%), and temporary paralyses of the third and fourth cranial nerves (0.8%) as complications. Fraioli et al. ²¹ reported unwanted first and second division hypoanesthesia (1.3%), paresthesia requiring transient medical treatment (3.8%), masseter dysfunction (3.8%), transient sixth nerve palsy (0.6%), orbital hematoma (0.6%), and cheek hematoma (5.7%) as complications after selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion for isolated third division trigeminal neuralgia. In a previous study ¹⁸ of the treatment of trigeminal neuralgia by this procedure, 10% of patients replied that although they had obtained pain relief, their life was considerably affected by complications from the procedure. This is a point that should be noted. Because this procedure is neurodestructive and can cause sensory loss and muscle weakness, it is important to determine whether to proceed with it based on sufficient consideration of its likely benefits and disadvantages to the patient.

In this patient, there were risks that selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion might cause sensory loss anew in the distribution of the third branch of the trigeminal nerve and paralysis of the masseter muscle. However, alleviation of the MRONJ was unlikely while she was alive, and it was judged that long-term pain relief brought about by the procedure would be a boon to the patient. The patient herself agreed with this, and the procedure was thus performed. As a result, pain relief was achieved with minimal complications, pain-free adequate local irrigation could be performed, and the management of the MRONJ was improved. Up to the time of death, there was no recurrence of pain or worsening of the MRONJ, and the selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion had been useful for improving the patient's QOL.

Conclusion

We propose that selective percutaneous controlled radiofrequency thermocoagulation of the Gasserian ganglion will be further investigated as a procedure for pain management in patients whose QOL is seriously impacted by treatment-resistant MRONJ and the accompanying analgesic-resistant pain.