Doo-Sik Kong

Microvascular decompression for treating hemifacial spasm: lessons learned from a prospective study of 1,174 operations

The authors critically analyzed a large series of patients with hemifacial spasm (HFS) and who underwent microvascular decompression (MVD) under a prospective protocol. We describe several “lessons learned” that are required for achieving successful surgery and proper postoperative management. The purpose of this study is to report on our experience during the previous 10xa0years with this procedure and we also discuss various related topics. From April 1997 to June 2009, over 1,200 consecutive patients underwent MVD for HFS. Among them, 1,174 patients who underwent MVD for HFS with a minimum 1xa0year follow-up were enrolled in the study. The median follow-up period was 3.5xa0years (range, 1-9.3xa0years). Based on the operative and medical records, the intraoperative findings and the postoperative outcomes were obtained and then analyzed. At the 1-year follow-up examination, 1,105 (94.1%) patients of the total 1,174 patients exhibited a “cured” state, and 69 (5.9%) patients had residual spasms. In all the patients, the major postoperative complications included transient hearing loss in 31 (2.6%), permanent hearing loss in 13 (1.1%), transient facial weakness in 86 (7.3%), permanent facial weakness in 9 (0.7%), cerebrospinal fluid leak in three (0.25%) and cerebellar infarction or hemorrhage in two (0.17%). There were no operative deaths. Microvascular decompression is a very effective, safe modality of treatment for hemifacial spasm. MVD is not sophisticated surgery, but having a basic understanding of the surgical procedures is required to achieve successful surgery.

Hemifacial spasm: neurovascular compressive patterns and surgical significance

SummaryBackground. The aim of this study was to report further investigation of neurovascular compression as a cause of hemifacial spasm (HFS) and to provide useful surgical guidelines by describing the compression patterns.Material and methods. From January 2004 to February 2006, 236 consecutive patients with HFS underwent microvascular decompression (MVD) in a single centre. Based on the operation and medical records, the intraoperative findings and post-operative outcomes were obtained and analysed.Results. We found that 95.3% of lesions had accompanying causative factors that made the neurovascular compression inevitable. Based on the contributing factors, compression patterns were categorised into six different types including: loop (n = 11: 4.6%), arachnoid (n = 66: 27.9%), perforator (n = 58: 24.6%), branch (n = 18: 7.6%), sandwich (n = 28: 11.9%), and tandem (n = 52: 22.0%). The compression patterns were significantly correlated with the compressing vessels involved. Thirty-two (86.5%) of 37 lesions where the vertebral artery was the compressing vessel involved the tandem type. Anterior inferior cerebellar artery was the compressing vessel involved in 49 (84.5%) of 58 perforator type compressions, while posterior inferior cerebellar artery was the compressing vessel involved in 8 (72.7%) of 11 loop type compressions.Conclusions. Once the compressing vessel responsible for the neurovascular compression are identified, the probable pattern of compression can be anticipated; this knowledge could facilitate the application of the appropriate operative procedures and minimise post-operative complications.

Intraoperative management to prevent cerebrospinal fluid leakage after microvascular decompression: dural closure with a “plugging muscle” method

Our objective is to present surgical techniques used for the prevention of cerebrospinal fluid leakage after microvascular decompression (MVD). From January 1996 to February 2006, microvascular decompression for hemifacial spasm or trigeminal neuralgia was performed in 678 consecutive patients. In order to achieve watertight dural closure, several pieces of muscle were interposed between the dura when the dura was sutured; the dura was stitched with the addition of muscle pieces to plug the dural defect. In cases where the mastoid air cell system was opened, bone wax was used to seal the opened surface of the cavity, and a muscle patch was applied for the secondary sealing. The cranioplasty was performed using polymethylmethacrylate (PMMA) bone cement. Only 2 (0.29%) of 678 patients, who underwent MVD followed by dural closure using several muscle pieces to plug the potential dural defect, suffered from CSF leaks. Both were treated with lumbar subarachnoid drainage; neither patient required a lumbar peritoneal shunt or a revision operation. A watertight dural closure with the addition of muscle pieces in a “plugging” fashion, along with sealing the opened surface of the mastoid cavity using bone wax and cranioplasty using bone cement, provides a simple and effective technique for the prevention of CSF leakage after MVD.

Frequency and prognosis of delayed facial palsy after microvascular decompression for hemifacial spasm.

SummaryBackground. Microvascular decompression (MVD) for hemifacial spasm (HFS) provides a long-term cure rate. Delayed facial palsy (DFP) is not an unusual complication, but it has only been sporadically described in the literature. The purpose of this report is to evaluate the incidence of delayed facial palsy after MVD and its clinical course and final results.n Methods. From January, 1998 to April, 2004, 410 patients underwent microvascular decompression for hemifacial spasm at our Institute. During this time, 21 patients (5.4%) developed delayed facial weakness; eighteen of them were given steroid medication and they were followed up in the out-patient clinic.n Findings. Twenty-one patients developed DFP after microvascular decompression an incidence of 5.4%. There were seventeen women (81.0%) among the 21 patients with DFP who were included in this study. In twenty of them, the symptoms of HFS improved completely after the operation, but the spasm remained with one of them. The onset of palsy occurred between postoperative day 7 and 23 (average: 12.1 days). The palsy was at least Grade II or worse on the House-Brackmann (HB) scale. The time to recovery averaged 5.7 weeks (range: 25 days–17 weeks); 20 patients improved to complete recovery and 1 patient remained with minimal weakness, as Grade II on the HB scale, at the follow-up examination.n Conclusion. Our findings demonstrated that the incidence of DFP was not so low as has been reported the literature, and it did not have any striking predisposing factors. Even though the degree of facial palsy was variable, almost all patients exhibited a complete recovery without any further special treatment. The etiology of DFP and its association with herpes infection should be further clarified.

Microvascular decompression for hemifacial spasm: long-term outcome and prognostic factors, with emphasis on delayed cure.

The postoperative course of microvascular decompression (MVD) for hemifacial spasm (HFS) is variable, and the optimal time for assessing the results is unclear. From April 1997 to October 2007, MVD for HFS was performed in 801 patients. Patients were divided into two groups (cured or failed) according to subjective patient assessments over a 3-year period. We analyzed patient characteristics and surgical findings to determine prognostic factors. Medical records were analyzed retrospectively over the 3-year follow-up period. Of the 801 patients who underwent surgery, 743 (92.8xa0%) appeared to be cured, 70 (8.7xa0%) had residual or recurrent spasms more than 1xa0year after surgery, 11 (1.3xa0%) had gradual improvement over 3xa0years, and 1 (0.1xa0%) had delayed improvement more than 3xa0years after surgery. Fifty-eight patients (7.2xa0%) had residual or recurrent spasms more than 3xa0years after surgery, of which 19 (2.4xa0%) had recurrence after initial relief. The mean time to spasm recurrence was 18.9xa0months. Intraoperative resolution of the lateral spread response (LSR) after decompression (pu2009=u20090.048) and severe indentation (pu2009=u20090.038) were significant predictors of good long-term outcome after MVD for HFS. In our series, 70 patients (8.7xa0%) had residual or recurrent spasms more than 1xa0year after surgery, of which 12 (17.1xa0%) improved gradually after 1xa0year. If the surgeon can confirm intraoperative resolution of the LSR and severe indentation, reoperation can be delayed until 3xa0years after MVD.

Cerebral arteriovenous malformations and seizures: differential impact on the time to seizure-free state according to the treatment modalities

BackgroundTo determine the prognostic factors for the incidence and the outcome of seizure in patients with cerebral arteriovenous malformation (AVM) and to identify the time to seizure-free state according to the treatment modalities.Material and methodsBetween 1995 and 2008, the multidisciplinary team at our institution treated 399 patients with cerebral AVMs. Treatment consisted of surgical resection, radiosurgery, and embolization, either alone or in combination. The median follow-up period was 6.0xa0years (range, 3.0–16.2xa0years). Eighty-six patients (21.5xa0%) experienced seizures before treatment. We investigated the variables associated with seizure incidence and seizure outcome and analyzed the outcomes of seizure among each treatment modality.ResultsAfter treatment, 60 (70xa0%) patients were seizure-free. Compared with 313 patients who did not experience seizures, we found that younger age (≤ 35xa0years), sizeu2009≥u20093xa0cm, and location of temporal lobe were associated with seizures (pu2009<u20090.05). Short seizure history, accompanying intracerebral hemorrhage, generalized tonic-clonic type seizure, deep-seated or infratentorial AVM, complete obliteration of AVM, and a favorable neurological outcome at 12xa0months were closely associated with Engel Class I outcomes (pu2009<u20090.05). Seizure-free outcomes after microsurgery, radiosurgery, or embolization were 78xa0%, 66xa0%, and 50xa0%, respectively. The overall annual bleeding rate was 1.0xa0% and 2.2xa0% in microsurgery-treated and radiosurgery-treated AVMs, respectively. In the surgery group, the median time to seizure-free status was 1.1xa0months (95xa0% CI, 0.7–1.2xa0months), whereas the radiosurgery group and embolization-alone group showed 20.5xa0months (95xa0% CI, 18.3–23.8xa0months), and 8.1xa0months (95xa0% CI, 6.0–13.5xa0months), respectively.ConclusionsA multidisciplinary team approach for cerebral AVMs achieved satisfactory seizure control results. Microsurgery led to the highest percentage of seizure-free outcomes and had the lowest annual bleeding rate, whereas radiosurgery had a higher bleeding rate. Median time to seizure-free status in surgically treated patients was shorter than in patients who underwent radiosurgical or endovascular treatment.

The patterns and risk factors of hearing loss following microvascular decompression for hemifacial spasm

ObjectiveThe aim of this study was to reveal the risk factors including intraoperative brain stem auditory evoked potential (BAEP) changes and to define parameter and warning values of BAEP beyond which the probability of hearing impairment rises significantly.MethodsFrom April 1997 to February 2009, 1,156 patients underwent microvascular decompression (MVD) for hemifacial spasm (HFS) and their medical records and audiologic data. The intraoperative BAEP monitoring was performed in all operations during surgery from the time of administration of general anesthesia until the time of skin closure. Pure tone audiometry (PTA) and Speech Discrimination Score (SDS) were performed on all patients before and after surgery for categorizing the patterns of hearing loss. There were 825 females and 331 males with a mean age of 48.7xa0years (range 17–75xa0years). The mean symptom duration was 67.8xa0months (range 1–420xa0months).ResultsAt the 1-year follow-up examination, 1,091 (94.4%) patients of the total 1,156 patients exhibited a cured state, and 65 (5.6%) patients had residual spasms. Hearing loss occurred in 46 patients (3.9%). In 26 patients, PTA was decreased more than 15xa0dB with a proportional decrease of the SDS. In 10 patients, poor SDS without hearing loss occurred. Total deafness was developed in 10 patients. A higher incidence of BAEP change and a poor recovery especially amplitude in wave V during surgery was observed in patients with poor SDS (eight patients) and total deafness (seven patients) (pu2009=u20090.000). Reduction of amplitude more than 50% in wave V was a strong indicator for a worse outcome of the hearing capacity. The difference in other risk factors according to hearing loss pattern was not statistically significant (pu2009>u20090.05). Only female was significant (pu2009=u20090.005).ConclusionsThe intraoperative BAEP change and a poorer recovery, especially reduction of amplitude more than 50% in wave V, was a strong indicator for a worse outcome of the hearing capacity. Vigilant intraoperative monitoring of the BAEP and adequate steps for recovery of the BAEP change could prevent hearing loss after MVD for HFS.

Feasibility of neuroendoscopic biopsy of pediatric brain tumors.

PurposeThe purpose of this study was to investigate the potential value of neuroendoscopic biopsies in pediatric patients with peri- or intraventricular tumors.MethodsFrom 2001 to 2008, 49 pediatric patients (mean age, 12.16xa0years) with tumors located in the intraventricular or paraventricular areas underwent neuroendoscopic biopsy, with or without simultaneous endoscopic third ventriculostomy. Neuroendoscopic biopsies were performed to verify the histological diagnosis of neoplasms and to establish pathological diagnoses necessary for planning appropriate treatment strategies.ResultsIn 45 of 49 patients (91.8%) neuroendoscopic biopsy specimens were appropriate for diagnosis and revealed 27 germinomas, 11 astrocytomas, and one ependymoma, etc. The tumor location included the pineal gland (nu2009=u200928), thalamus (nu2009=u20097), intraventricle (nu2009=u20093), hypothalamus (nu2009=u20093), suprasellar area (nu2009=u20092), and diffuse multifocal area (nu2009=u20093). In two patients (4.1%) biopsy specimens were informative but not diagnostic. Tumor tissue specimens were undiagnostic in two patients (4.1%). There were eight transient morbidities, including four EOM limitations, two central DI, one EVD infection, and one CSF leakage. One patient experienced postoperative tumor bleeding requiring emergent operation. There was no case of operative mortality.ConclusionNeuroendoscopic biopsy can be considered as the first choice for tissue sampling of periventricular and intraventricular tumors with acceptable risks.

Microsurgical resection of deep-seated lesions using transparent tubular retractor: pediatric case series

BackgroundTo facilitate effective resection of deep-seated brain lesions without causing significant trauma to the overlying cortex, the authors used a transparent plastic tubular retractor to approach these lesions.MethodsBetween July 2009 and January 2011, we used an 11-mm diameter transparent plastic tubular retractor in combination with a frameless stereotactic navigation system to remove 18 deep lesions.ResultsGross total resection of the lesions was achieved in 14 of 18 patients, and subtotal removal occurred in four patients. Effective resection of lesions was achieved in all patients through small size craniotomy window and small cortical incision. The histopathologic diagnosis was established in all 18 patients: 3 hematomas, 3 cavernous angioma, 7 low-grade glioma, 2 dysembryoplastic neuroepithelial tumor, 1 choroid plexus papilloma, 1 abscess, and 1 meningioma.ConclusionMicrosurgery using a transparent tubular retractor guided by a neuronavigation system facilitated accurate and effective removal of these deep-seated brain lesions.

Challenging reconstructive techniques for skull base defect following endoscopic endonasal approaches

ObjectiveWe assessed the outcomes of various reconstructive methods for skull base defect after endoscopic endonasal approaches (EEA) depending on the degree of intraoperative cerebrospinal fluid (CSF) leaks.MethodsBetween Jan. 2008 and Sep. 2009, 122 consecutive patients underwent 124 EEA for sellar and extra-sellar lesions. Intraoperative CSF leaks were classified as grade 0, no intraoperative CSF leak; grade 1, low output; and grade 2, high-output based on the degree of CSF leakage and size of opening in the arachnoid membrane (<5 or ≥5xa0mm).ResultsPostoperative CSF leaks or meningitis occurred in 13 of 124 cases (10.5%). In 77 patients with grade 0, there was no postoperative CSF leak. Among 20 patients with grade 1 CSF leaks, four patients developed meningitis or postoperative CSF leak. Postoperative CSF leaks occurred in nine of 26 patients (34.6%) with grade 2 leaks. Comparison of reconstructive methods revealed that gasket-seal method provided better control of CSF leaks than free-fat graft in patients with grade 2 leaks (11.8% vs. 66.7%, pu2009=u20090.028). However, in grades 0 and 1, we found no difference among the various reconstructive methods.ConclusionThe selection of reconstructive methods for skull base defects should be determined by the degree of CSF leaks. Although grade 0 or 1 leak requires relatively conservative management such as simple closure or free-tissue grafting, a more aggressive reconstructive technique is required to prevent postoperative complication in grade 2 CSF leak.