Conclusion: Percutaneous balloon compression is a simple, safe, and effective procedure. It is a reasonable first choice for elderly patients, those who are at a high risk for posterior fossa exploration, and/or patients with recurrent trigeminal neuralgia. (p23-29)

Keywords: Balloon Compression, Gasserian Ganglion, PercutaneousProcedures, TrigeminalNeuralgia.

INTRODUCTION

During the 1950's, open trigeminal ganglion compression, through a subtemporal approach, was the most popular method for the treatment of trigeminal neuralgia. 21 In 1983, Mullan and Lichtor (18) described percutaneous compression of the trigeminal ganglion using inflatable balloon introduced through the foramen ovate under X-ray guidance. Since then more than 800 patients have been reported as been treated with this technique with good relief and minimal morbidity (Table 1).

We report here our experience with percutaneous balloon gangliolysis (PBG) for trigeminal neuralgia.

METHODS

Patient population. Thirty-three patients, ranging in age from 32 to 83 years (mean 61 years) underwent 37 PBG between November 1989 and December 1995. Ten were males and 23 females (M:F ratio 1: 2.3). The right side was more commonly involved (71%), and in 54% the neuralgia was in the distribution of the maxillary division of the trigeminal nerve. Nine patients (27%) had prior surgical treatment for trigeminal neuralgia, such as microvascular decompression (MVD), percutaneous radiofrequency rhizotomy (RFR) and/or peripheral neurectomy. One patient underwent complete resection of a tentorial meningioma. The mean follow-up was 3 years (0.5-6 years).

Patient selection. The neurologists referred all the patients after having been treated with carbamazepine, dilantin and / or baclofen, with either no response or with drug side effects. Cranial computed tomographic (CT) and/or magnetic resonance (MR) scans were usually obtained in the early patient work up, mainly to exclude a secondary cause. Thirty-one patients suffered from idiopathic trigeminal neuralgia. One patient with multiple sclerosis and one with ipsilateral tentorial meningioma presented with trigeminal neuralgia. The procedure is offered to elderly patients who are at high surgical risk, patients with recurrence or failure after other surgical treatment(s), and for patients with multiple sclerosis.

Operative technique. The procedure performed as day surgery in the majority of cases (84%), in the operating room under light general anaesthesia. Premedication with atropine and sedation given on a routine basis and with prophylactic antibiotic administered on induction. The patient positioned supine, with head extended. The procedure is performed using aseptic technique and C-arm fluoroscopy guidance. The length of the 14 F spinal needle is marked on the #3 F Fogarty balloon catheter with a stopper, and the balloon is tested. Skin puncture is performed at a point 2.5 cm from the corner of the mouth, with the 14 F spinal needle, which is then directed at a 60-degree angle toward the ipsilateral medial canthus. The needle is advanced under lateral fluoroscopy towards the petro-clival angle. Final needle position is adjusted to be at, or within, 5 mm from foramen ovate. Egress of cerebrospinal fluid (CSF) is an accurate indicator of needle tip location. The balloon catheter is then inserted and inflated with 0.75 cc of contrast material and a "pear shape" is seen on the image (Fig 1). Inflation is maintained for a mean period of 15 minutes (range 5-20) and checked on fluoroscopy intermittently to monitor for balloon rupture or displacement to the posterior fossa. Following deflation, the needle and the balloon are removed together and mild pressure is applied to the cheek for 5 minutes. The patient is usually observed for 6 hours in the recovery room and then discharged home with follow up after 4 weeks and advised that his/her medications to be tapered over a week and then discontinued.

RESULTS

Failure to cannulate foramen ovale occurred in two patients; one in a young female and the other in an elderly lady with limited neck movement due to severe spondylosis, which prevented extension. All the 31 patients who had their PBG completed successfully were initially relieved of their pain. Twenty-eight patients (90%) developed postoperative facial numbness. Postoperative mild dysesthesia occurred in 4 (13%). There were no cases of anaesthesia dolorosa or corneal anaesthesia. One patient developed postoperative diplopia secondary to 4th' cranial nerve palsy, which resolved completely within 3 months. One patient developed intraoperative asystole, during the cannulation of foramen ovale. Intravenous atropine and external massage quickly restored the patient's rhythm. Four patients (13%) developed postoperative mild ipsilateral masseter-pterygoid weakness that resolved within 3 to 6 months. One patient developed postoperative aseptic meningitis consisting of headache, fever, neck stiffness, and confusion, all occurring within 48 hours of surgery. CT scanning did not show any subarachnoid blood but CSF microscopy revealed mild pleocytosis with normal RBC's protein and sugar values. CSF cultures did not reveal any growth. This patient was treated empirically with antibiotics, and his symptoms resolved within 48 hours. The overall recurrence rate was 32% (10/31) over a mean period of 24 months (range: 1-56 months). Four patients with recurrence had a repeat PBG that again provided pain relief. Three young patients, initially declined MVD, were subsequently considered for MVD and another three underwent peripheral neurectomy.

DISCUSSION

PBG has emerged as an innovative and effective method for the treatment of trigeminal neuralgia. The procedure is fiber selective, though not division selective and patient's cooperation is not required for a functional localisation. The procedure usually performed under neuroleptic or light general anaesthesia.

Brown et al. has studied the effect of balloon compression in rabbits, and was able to show that compression selectively injures medium and large myelinated fibers more than small myelinated and unmyelinated fibers.s The ganglion cells were generally preserved. Because the corneal reflex is mediated by small fibers, PBG has a great advantage compared to other percutaneous techniques in its ability to relieve neuralgia in the distribution of ophthalmic division of the trigeminal nerve, while sparing the corneal reflex. The procedure is best suited for the elderly and medically infirm patients at greater risk with posterior fossa procedures. It can also be offered to younger informed patients on the proviso that it may require repeating the procedure, and for patients who have previously undergone a destructive procedure but do not have major dysesthesia.

Techniques and complications. Adequate positioning, with moderate neck extension is important for cannulation of foramen ovale. Technical failure in cannulation has occurred on two occasions in our series (6%). In one elderly patient with severe cervical spondylosis, adequate neck extension was not possible. The use of standard Hartel landmarks (12) and lateral fluoroscopic monitoring facilitate Meckel's cave cannulation. Faulty balloon placement is revealed by the abnormal relation to bone landmarks, and by failure of the inflated balloon to adopt the characteristic "pear" shape. Compression with an oval or irregular-shaped balloon in an appropriate location is usually followed by persistent control of pain. However atypical balloon shapes seem to be associated with early recurrence of neuralgia. The catheter may also become displaced extradurally, subdurally or in the subarachnoid space of the temporal fossa. At least 0.75 cc of contrast should be used to inflate the balloon and obtain characteristic pear shape. 7,14 Venous bleeding may occur even with a properly placed needle, and the procedure usually completed successfully. If arterial bleeding is encountered, the procedure is best terminated. The escape of CSF from the cannula indicates that the tip lies beyond the foramen margin, but does not necessarily indicate that it is in Meckel's cave. If the appropriate "pear" shape is not obtained, the balloon is deflated, and the needle withdrawn in 5 mm increments with subsequent balloon re-inflation.

As with other percutaneous techniques, autonomic changes, such as bradycardia or hypertension, may occur during needle placement and compression by the balloon. This phenomenon has been observed more frequently with PBG, probably because of the larger needle required for cannulation. Brown and Preul reported bradycardia and hypotension at a frequency of 75% and 50% respectively, and they recommended premeditation with atropine or prophylactic external pacemaker. 7 Dominguez et al. recently reported a technique to overcome autonomic changes by injecting lidocaine into Meckel's cave before compression. 8 In that series two patients developed transient bilateral mydriasis within a few minutes of lidocaine injection. In our series, one patient developed a brief asystole during foramen ovule cannulation. This was reversed immediately with atropine and external cardiac massage. We have noticed a slight decrease in heart rate in some patients usually at needle cannulation, but not during the balloon compression phase. We recommended that atropine was given before placing the needle, when the baseline heart rate is less than 50 / minute or if bradycardia developed.

Injury to the second through sixth cranial nerves has been reported as a complication of percutaneous treatment of trigeminal neuralgia in addition to intracranial haematoma, subarachnoid haemorrhage, abscess, and stroke. 20 Four carotid artery injuries have been reported (19,20) one associated with PBG. 13 Carotid artery injury may result from a misguided, particularly far posterior or medial needle placement. In up to 4% of cases, there is fusion of the foramen ovule and foramen lacerum, the primitive foramen lacerum medius, which may predispose to vascular injury. I9 We emphasize that the tip of the needle should not enter the intracranial cavity to minimize the likelihood of entering an intracranial vessel.

The optimal compression time with the PBG technique is not known, and no prospective randomized trial has addressed this issue. The literature review did riot reveal any direct correlation between the compression time and the incidence of dysesthesias. Lichtor and Mullan (I4) in their pioneering series of 100 patients initially used 5-7 minutes of compression (range, 0.5-15 minutes). They observed an increased incidence of dysesthesia and numbness, with longer compression time, and after the sixth procedure, they began using only one minute of compression. However, in most series, the balloon was inflated for 4.5-7 minutes.

Transient ipsilateral masseter-pterigoid muscle weakness is more frequent in patients treated with PBG than with other percutaneous techniques. It varies from mild to severe, with reported incidence in the literature around 10%. For patients with bilateral trigeminal neuralgia, the second procedure should be staged after complete recovery of the weakness.

Literature review. Tekkok and Brown (24) analyzed the data from 9 published series (2,3,6,9,10.11,15,17,18) of PBG published before 1995. The success rate of PBG procedure was 92.1% in more than 500 patients. The recurrence rates vary but average 25% (range, 9.7- 54%) at 3 years. The mortality rate was 0.1% (one patient). 22 Masseter-pterigoid muscle weakness occurred in approximately 10% of cases. The rates of mildmoderate and severe dysesthesia were 6.4% and 2°/> respectively. Mild-moderate and severe hypesthesia occurred in 57% and 1.6% respectively. Only one case (0.1%) of anaesthesia dolorosa has been reported, and the risk of corneal anaesthesia from balloon compression is negligible (0.1%).

Comparison with other procedures. Taha and 'I'ew recently reported their opinion in a retrospective selective analysis of the literature relative to the management of trigeminal neuralgia. 23 They compared their experience in RFR with larger patient series of RFR, PBG, glycerol rhizotomy (GR), MVD, and partial trigeminal rhizotomy (P'I'R). Pain relief after PBG was 93% in comparison to 98% for MVD and RFR. The recurrence rate of PBG was 20% in 2 years, and 20% in 9 years for MVD, 15% in 5 years for RFR, and 54% in 4 years for GR. Facial numbness was highest among I'TR and RFR techniques, 100% and 98% respectively, and 72% with PBG. Dysesthesia and anaesthesia dolorosa was most frequently seen after RFR compared to MVD, PBG, and GR. Trigeminal motor dysfunction was high with PBG (66%), and corneal anaesthesia and keratitis was not reported in PBG. Radiosurgery of the trigeminal root entry zone, recently evolved as a noninvasive treatment option in some cases, but was not included in their review. This review is not comprehensive and not a meta-analysis, and no prospective randomized trials have been reported in the literature.

Conclusion. The PBG procedure is easy to perform requires a short anaesthetic, and a brief period of hospitalisation. It is well tolerated by patients who describe it as a "totally pain free experience". Morbidity is minimal, and the recurrence rate is not significantly higher than other procedures for trigeminal neuralgia.

Figure 1 — Intraoperative lateral radiograph of inflated balloon in the Meckel's cave forming a characteristic "pear" shape.

Table 1 - Results of the reported PBG in the literature

Authors/Year No. of patients Initial success/% Recurrence/% Follow-up/ yr.

Bricolo & Dalla Ore 1983 [3] Esposito et al 1985 [9] Fiume et al 1985 [10] Belber & Rak 1987[2] Meglio et al 1987 [17] McIntosh 1987 [16] Fraiolio et al 1989 [11] Lobato et al 1990 [15] Lichtor & Mullan 1990 [14] Abdennabi & Amzer 1991 [1] Brown et al 1996 [4] Current study Total 51 50 10 33 47 48 159 144 100 20 141 33 836 100 64 100 100 100 85 90 94 100 92 92 94 93 20 NS 30 24 55 27 9.8 9.7 28 14 26 32 23 <2 <1 <1 0.5-7 (2.8) 2 0.08-3.5 3.5 0.4-4.5(2) 1-10 0.5-4.5(2) 0.16-10(2) 0.5-6 (3) -2

REFERENCES

1.	Abdennebi B and Amzar Y. Traitement de la neuralgie trigeminale essentielle par compression gasserienne par ballonet. Neurochirrurgie, 1991, 37: 115-118.
2.	Belber CJ and Rak RA. Balloon compression rhizolysis in the surgical management of trigeminal neuralgia. Neurosurgery, 1987, 20: 908-913.
3.	Bricolo A and Dalla Ore G. Percutaneous microcompression of the gasserian ganglion for trigeminal neuralgia. Acta Neurochir (alien), 1983, 69: 102 (ahstr.).
4.	Brown JA, Chittum CJ, Sabol D, Gouda JJ. Percutaneous balloon compression of the trigeminal nerve for treatment of trigeminal neuralgia. Neurosurg Focus, 1996, 1:4; 1-8.
5.	Brown JA, Hoeflinger B, Long PB, Gunning WT, Rhoades R, Bennett-Clarke CA, Chiaia NL, Weaver MT. Axon and ganglion cell injury in rabbits after percutancous trigeminal balloon compression. Neurosurgery, 1996, 38: 993- 1004.
6.	Brown JA, McDaniel MD, Weaver- MT. I'crcutaneous trigeminal nerve compression fir trcatnwnt (>f t rigenrinal neuralgia. Results in 5C p,uicnW Nrur(murgcry, 1993, 32: 570-573.
7.	Brown JA and Preul MC,.trigeminal ganglion compression for trigeminal neuralgia: experience in 22 patients and review of the literature. J Neurosurg, 1989, 70: 900-904.
8.	Dominguez J, Lobato RD, Rivas JJ, Gargallo MC, Castells V, Gozalo A, Sarabia R. Changes in systemic blood pressure and cardiac rhythm induced by therapeutic compression of the trigeminal ganglion. Neurosurgery, 1994, 34: 422- 428.
9.	Esposito S, Delitalia A, Bruni P, Hernandez R, Callovini GM. Therapeutic protocol in the treatment of trigeminal neuralgia. Appl Neurophysiol, 1985, 48: 271-273.
10.	Flume D, Scarda G, Natali G, Della Valle G. La microcompressione percutanea del ganglio di gasser. Una nuova terapia per la neuralgia del trigemino. Riv Neurol, 1985, 5: 387-391.
11.	Fraioli B, Esposito V, Guidetti B, Crucco G, Manfredi M. Treatment of trigeminal neuralgia by thermocoagulation, glycerolization and percutaneous compression of the gasserian ganglion and/or gasserian rootlets: long-term results and therapeutic protocol. Neurosurgery, 1989, 24: 239-245.
12.	Hartel F.Die Behandlung der Trigeminusneuralgie nut intrakraniellen Alkoholeinspritzungen. Dtsche 22. Z Chir, 1914, 126: 429-552.
13.	Kuether TA, O'Neill OR, Nesbit GM, Barnwell SL. Direct carotid cavernous fistula after trigeminal balloon microcompression gangliolysis. Case report. Neurosurgery, 1996, 39: 853-836.
14.	Lichtor and Mullan. A 10-year follow-up review of percutaneous microcompression of the trigeminal ganglion. J Neurosurg1990, 72: 49- 54.
15.	Lobato RE, Rivas JJ, Sarabia R, Lamas E. Percutaneous microcompression of the gasserian ganglion for trigeminal neuralgia. J Neurosurg, 1990, 72: 546-553.
16.	McIntosh JW. Micro-compression for trigeminal trigeminal neuralgia. J Neurol Neurosurg Psychiatry, 1987, 50: 645-646.
17.	Meglio M, Cioni B, d'Annunzio V. Percutaneous microcompression of the gasserian ganglion: personal experience. Acta Neurochir Suppl, 1987, 39: 142-143.
18.	Mullan S and Lichtor T. Percutaneous microcompression of the trigeminal ganglion for trigeminal neuralgia. J Neurosurg, 1983, 95: 10071012.
19.	Revuelta R, Nathal E, Balderrama J, Tello A, Zenteno M. External carotid artery fistula due to microcompression of the gasserian ganglion for relief of trigeminal neuralgia. Case report. Neurosurgery, 1993, 78: 499-500.
20.	Sekhar LN, Heros RC, Kerber CW. Carotid cavernous fistula following percutaneous retrogasserian procedures. Report of two cases. J Neurosurg, 1979, 51: 700-706.
21.	Shelden Ch, Pudenz RH, Freshwater DB, Crue BL. Compression rather than decompression for trigeminal neuralgia. J Neurosurg, 1955, 12: 123126.
22.	Spaziante R, Cappabianca P, Peca C, de Divitiis E. Subarachnoid hemorrhage and normal pressure hydrocephalus: fatal complication of percutaneous microcompression of the Gasserian ganglion. Neurosurgery, 1988, 22: 148.
23.	Taha JM and Tew JM. Comparison of surgical treatments for trigeminal neuralgia: revaluation of radiofrequency rhizotomy. Neurosurgery, 1996, 38: 865-871.
24	Tekkok IH and Brown JA. The neurosurgical management of trigeminal neuralgia. Neurosurg Q, 1996, 6: 89-107.