SPG RFA should be performed by highly trained physicians with experience with such procedures under fluoroscopy. These physicians typically are fellowship trained and board-certified in pain medicine with backgrounds in anesthesiology, physical medicine, and rehabilitation (PM&R), neurology, emergency medicine, family medicine, and psychiatry. Highly trained supporting staff should include a radiology technician, a qualified nurse to assist with equipment, a certified registered nurse anesthetist and patient support.[10]

Patients need to be informed about the procedure in detail in a separate clinic visit before the procedure. Details should include procedural steps, benefits of the procedure, contraindications and complications of the procedure, and alternative therapies. All patients must be off anticoagulants for the appropriate time.  In more complex situations, patients should be advised to visit their PCP to bridge to short-acting anticoagulants.

Procedural clearance should be obtained in patients with cardiovascular diseases. The patient should be advised not to eat 6-8 hours prior to the appointment. Diabetic patients require dose adjustment for insulin and oral diabetic medicine. Primary care doctors can help with this adjustment. Accurate medication reconciliation including allergies should be done prior to the procedure. Activities like driving or operating machinery should be avoided for at least 24 hours after the procedure. Detailed consent is necessary for the physician who will perform or assist the procedure. Consent should mention the name of the patient, identification (medical record number, age, gender). The procedure site should be marked accurately. A technician should make sure the equipment is in proper working order. A time out should be taken when the patient is on the procedure table in the presence of all the personnel to ascertain the correct patient is getting the correct procedure in the correct location. The enactment of thorough aseptic precautions and a sterile procedure tray are standard procedure.[10][11][12]

The SPG can is approachable via the intranasal, infrazygomatic, and suprazygomatic routes but radiofrequency ablation techniques are typically done via the infrazygomatic approach.

Fluoroscopic approach: Narouze et al. described the fluoroscopic guided infrazygomatic approach for an SPG RFA. For this technique, the patient is positioned in the supine position and obtain a true lateral fluoroscopic image, which can be accomplished by rotating the patient’s head or the C-arm fluoroscope in a manner that superimposes rami of the mandibles on one another and creates an “inverted vase” appearance of the pterygopalatine fossa. Mark the skin entry site just anterior to the mandible and inferior to the zygomatic arch. Raise a skin wheel at this skin entry point using 1% lidocaine anesthetic solution. Use a blunt RFA needle with 22-gauge diameter, 10-cm length, with a slightly bent 5-mm active tip. Insert the RFA needle at the skin site and advance it superiorly and medially towards pterygopalatine fossa under the real-time fluoroscopy guidance. Confirm the proper direction of the needle using an anteroposterior view. Obtain an AP view to confirm the needle tip at the lateral to the lateral nasal wall. Change the needle direction to cephalad and anterior if you come across the lateral pterygoid plate. This change in needle direction will help the curved needle tip to reach to the pterygopalatine fossa. Apply 50-Hz sensory stimulation first to produce deep paresthesia behind the root of the nose at <0.5 V. Inject 0.2-0.5 ml contrast agent using real-time fluoroscopy to prior to lesioning to rule out intranasal or intravascular spread. Anesthetize the region using 0.5 ml lidocaine 2% solution.  Carry out two radiofrequency lesions in this region. Use the temperature at 80°C for 60 seconds each. Radiofrequency ablation is an extremely painful procedure. Inject 0.5 mL of bupivacaine 0.5% along with 5 mg of triamcinolone after lesioning to prevent neuritis after the procedure.[10]

CBCT approach: Loomba et al. described another performing to perform the SPG RFA. They used cone beam computed tomography (CBCT) to lesion SPG. All the steps are similar to the technique described by Narouze et al. except a trained interventional radiologist confirms the location of the tip of the radiofrequency cannula after the needle tip walks off the lateral pterygoid plate. The advantage of CBCT over bi-plane fluoroscopy is CBCT offers improved anatomic visualization and allows accurate needle tip placement. Also, CBCT is relatively less expensive than other procedures and has less radiation exposure.[11]

Complications of SPG RFA are mostly associated to the needle insertion and may include infection, internal bleeding or epistaxis, injury to the maxillary artery and nerve and its branches (greater and lesser petrosal nerves), hemodynamic instability. Sometimes it results in numbness and/or dysesthesia of the upper teeth, hard palate, or pharynx, a decrease in lacrimation, and nasal mucus production. SPG RFA often causes reflex sinus bradycardia, and so close monitoring of vital signs is a requirement for early diagnosis of this complication.

During the procedure, if the patient feels liquid in the back of the throat, the RFA needle should be withdrawn slightly and retested. If proper stimulation is achieved, local anesthetic should be injected to anesthetize the area; if the patient does not report a liquid feeling in the back of his throat the radiofrequency ablation should be completed, and the patient should then be monitored for any symptoms such as bleeding or infection.

Sphenopalatine ganglion RFA is relatively quick, cost-effective and safe procedure with a low complication rate that has demonstrated that it is a useful method to treat chronic cluster headaches. Salgado-Lopez et al. from Spain published their results on the safety and efficacy of SPG RFA. They followed 37 SPG RFA patients for a mean follow-up of 68.1 months (range, 15-148 months). Five of their patients (13.5%) experienced complete clinical relief of both pain and parasympathetic symptoms, 21 patients (56.8%) had the partial and transient relief, and 11 patients (29.7%) did not improve. They documented no complications in any of these patients. The authors claim this is the largest number of patients and the longest follow-up period published in the literature. There is a prospective, multicenter, randomized, controlled, blinded-endpoint trial to test the safety and efficacy of SPG pulsed radiofrequency treatment for cluster headache underway in China. The sample size of this study is 80 patients and follow up is three years. This study aims to investigate whether the SPG pulsed radiofrequency ablation method is superior to a traditional SPG nerve block. It will also provide us with the evidence that the SPG pulsed RFA treatment can provide a safe, effective and minimally invasive alternative for patients with cluster headaches with failed medical management.[13][14]

The sphenopalatine ganglion is a potentially favorable target for treating cluster headache using blocks, RFAs, or neurostimulation. Additionally, SPG RFA has great evidence associated with its use in a few other conditions as mentioned in the indications above. However, most of these studies were with small sample size, shorter duration of follow-ups and without reproductions. It warrants further controlled studies to produce consistent results and magnify these prior findings.