Guy M. McKhann II
Andrew T. Dailey, Guy M. McKhann II and Mitchel S. Berger
✓ Mutism following posterior fossa tumor resection in pediatric patients has been previously recognized, although its pathophysiology remains unclear. A review of the available literature reveals 33 individuals with this condition, with only a few adults documented in the population. All of these patients had large midline posterior fossa tumors.
To better understand the incidence and anatomical substrate of this syndrome, the authors reviewed a 7-year series of 110 children who underwent a posterior fossa tumor resection. During that time, nine (8.2%) of the 110 children exhibited mutism postoperatively. They ranged from 2.5 to 20 years of age (mean 8.1 years) and became mute within 12 to 48 hours of surgery. The period of mutism lasted from 1.5 to 12 weeks after onset: all children had difficulty coordinating their oral pharyngeal musculature as manifested by postoperative drooling and inability to swallow. Further analysis of these cases revealed that all children had splitting of the entire inferior vermis at surgery, as confirmed on postoperative magnetic resonance studies. Lower cranial nerve function was intact in all nine patients.
Current concepts of cerebellar physiology emphasize the importance of the cerebellum in learning and language. The syndrome described resembles a loss of learned activities, or an apraxia, of the oral and pharyngeal musculature. To avoid the apraxia, therefore, the inferior vermis must be preserved. For large midline tumors that extend to the aqueduct, a combined approach through the fourth ventricle and a midvermis split may be used to avoid injuring the inferior vermis.
Kirsty E. Bortnik, Guy M. McKhann II and Marla J. Hamberger
Electrical stimulation mapping (ESM) is considered the gold standard for identification of essential language cortex and is especially important in patients for whom classic language landmarks are less useful because of reorganization in response to epileptogenic or neoplastic cortex. However, little is known regarding the reliability of the procedure, particularly over extended time intervals. The authors present the case of a young man with refractory left temporal lobe epilepsy in the setting of a low-grade left temporal tumor who had undergone intraoperative language mapping at age 14 years and repeat mapping at age 25. Results from both the initial ESM and the repeat ESM 11 years later revealed a positive auditory description naming site in the same location on the superior temporal gyrus, at the anterior aspect of the tumor. This case provides support for the reliability of ESM and underscores intraindividual reliability in the location of language cortex over a prolonged period.
Timothy H. Lucas II, Guy M. McKhann II and George A. Ojemann
Object. The aim of this investigation was to address three questions in bilingualism research: 1) are multiple languages functionally separated within the bilingual brain; 2) are these languages similarly organized; and 3) does language organization in bilinguals mirror that in monolinguals?
Methods. During awake dominant-hemisphere craniotomy in each of 25 bilingual patients, the authors mapped both languages by using identical object-naming stimuli. Essential sites for primary (L1) and secondary (L2) languages were compared. Sites were photographically recorded and plotted onto an anatomically referenced grid system. Language organization in bilinguals was then compared with that in 117 monolinguals and 11 monolingual children.
Conclusions. The authors found distinct language-specific sites as well as shared sites that support both languages. The L1 and L2 representations were similar in total cortical extent but significantly different in anatomical distribution. The L2-specific sites were located exclusively in the posterior temporal and parietal regions, whereas the L1 and shared sites could be found throughout the mapped regions. Bilinguals possessed seven perisylvian language zones, in which L2 sites were significantly underrepresented when compared with the distribution of language sites in monolinguals. These L2-restricted zones overlapped the primary language areas found in monolingual children, indicating that these zones become dedicated to L1 processing. These findings support three conclusions. First, it is necessary to map both languages in bilinguals because L1 and L2 sites are functionally distinct. Second, differences exist in the organization of L1 and L2 sites, with L2-specific sites located exclusively in the posterior temporal and parietal lobes. Third, language organization comparisons in bilingual and monolingual brains demonstrate the presence of L2-restricted zones, which are dedicated to L1.
Brett E. Youngerman, Andrew K. Chan, Charles B. Mikell, Guy M. McKhann and Sameer A. Sheth
Deep brain stimulation (DBS) is an emerging treatment option for an expanding set of neurological and psychiatric diseases. Despite growing enthusiasm, the patterns and implications of this rapid adoption are largely unknown. National trends in DBS surgery performed for all indications between 2002 and 2011 are reported.
Using a national database of hospital discharges, admissions for DBS for 14 indications were identified and categorized as either FDA approved, humanitarian device exempt (HDE), or emerging. Trends over time were examined, differences were analyzed by univariate analyses, and outcomes were analyzed by hierarchical regression analyses.
Between 2002 and 2011, there were an estimated 30,490 discharges following DBS for approved indications, 1647 for HDE indications, and 2014 for emerging indications. The volume for HDE and emerging indications grew at 36.1% annually in comparison with 7.0% for approved indications. DBS for emerging indications occurred at hospitals with more neurosurgeons and neurologists locally, but not necessarily at those with the highest DBS caseloads. Patients treated for HDE and emerging indications were younger with lower comorbidity scores. HDE and emerging indications were associated with greater rates of reported complications, longer lengths of stay, and greater total costs.
DBS for HDE and emerging indications underwent rapid growth in the last decade, and it is not exclusively the most experienced DBS practitioners leading the charge to treat the newest indications. Surgeons may be selecting younger and healthier patients for their early experiences. Differences in reported complication rates warrant further attention and additional costs should be anticipated as surgeons gain experience with new patient populations and targets.
Guy M. McKhann II, Andrew W. McEvoy, Robert E. Gross and Stephan Chabardes
Toru Fukuhara, Guy M. McKhann II, Paul Santiago, Joseph M. Eskridge, John D. Loeser and H. Richard Winn
✓ The authors describe a patient with right-sided central pain resulting from a left parietal arteriovenous malformation (AVM). The AVM was treated with staged embolization and stereotactic radiosurgery, and its obliteration was documented on follow-up angiographic studies. Surprisingly, the patient noted complete resolution of her pain syndrome after embolization, which is an extremely rare result. Central pain and its proposed mechanisms are discussed.
Robert A. McGovern, John P. Sheehy, Brad E. Zacharia, Andrew K. Chan, Blair Ford and Guy M. McKhann II
Early work on deep brain stimulation (DBS) surgery, when procedures were mostly carried out in a small number of high-volume centers, demonstrated a relationship between surgical volume and procedural safety. However, over the past decade, DBS has become more widely available in the community rather than solely at academic medical centers. The authors examined the Nationwide Inpatient Sample (NIS) to study the safety of DBS surgery for Parkinson disease (PD) in association with this change in practice patterns.
The NIS is a stratified sample of 20% of all patient discharges from nonfederal hospitals in the United States. The authors identified patients with a primary diagnosis of PD (332.0) and a primary procedure code for implantation/replacement of intracranial neurostimulator leads (02.93) who underwent surgery between 2002 and 2009. They analyzed outcomes using univariate and hierarchical, logistic regression analyses.
The total number of DBS cases remained stable from 2002 through 2009. Despite older and sicker patients undergoing DBS, procedural safety (rates of non-home discharges, complications) remained stable. Patients at low-volume hospitals were virtually indistinguishable from those at high-volume hospitals, except that patients at low-volume hospitals had slightly higher comorbidity scores (0.90 vs 0.75, p < 0.01). Complications, non-home discharges, length of hospital stay, and mortality rates did not significantly differ between low- and high-volume hospitals when accounting for hospital-related variables (caseload, teaching status, location).
Prior investigations have demonstrated a robust volume-outcome relationship for a variety of surgical procedures. However, the present study supports safety of DBS at smaller-volume centers. Prospective studies are required to determine whether low-volume centers and higher-volume centers have similar DBS efficacy, a critical factor in determining whether DBS is comparable between centers.
Guy M. McKhann II, Julie Schoenfeld-McNeill, Donald E. Born, Michael M. Haglund and George A. Ojemann
Object. Among the variety of surgical procedures that are performed for the treatment of medically refractory mesial temporal lobe epilepsy (TLE), no consensus exists as to how much of the hippocampus should be removed. Whether all patients require a maximal hippocampal resection has not yet been determined.
Methods. At the University of Washington, all TLE operations are performed in a tailored fashion, guided by electrocorticography (ECoG). The amount of hippocampal resection is determined intraoperatively by the extent of interictal epileptiform abnormalities on ECoG recorded from that structure, resulting in a hippocampal resection that is individualized for each patient. Using this approach, the authors prospectively observed 140 consecutive patients who underwent surgery for mesial TLE with pathological diagnoses of either mesial temporal sclerosis with neuronal loss (MTS group) or mild gliosis without neuronal loss (non-MTS group) to determine whether the extent of hippocampal resection correlates with outcome when a tailored approach is used. Additionally, the authors analyzed whether the presence of residual interictal epileptiform activity on ECoG following mesial temporal resection predicts poorer seizure control.
With at least 18 months of clinical follow up, 67% of the 140 patients were seizure free or had only a single postoperative seizure. There was no correlation between the size of the hippocampal resection and seizure control in the group as a whole or when stratified by pathological subtype. Using an intraoperatively tailored strategy, individuals with a larger hippocampal resection (> 2.5 cm) were not more likely to have seizure-free outcomes than patients with smaller resections (p = 0.9). Additionally, both MTS and non-MTS patients, in whom postoperative ECoG detected residual epileptiform hippocampal (but not cortical or parahippocampal) interictal activity following surgical resection, had significantly worse seizure outcomes (p = 0.01 in the MTS group; p = 0.002 in the non-MTS group).
Conclusions. Intraoperative hippocampal ECoG can predict how much hippocampus should be removed to maximize seizure-free outcome, allowing for sparing of possibly functionally important hippocampus.