Neurosurgical forum Letters to the editor
Unilateral versus bilateral deep brain stimulation To The Editor: I read with great interest the article by Taba et al.1 (Taba HA, Wu SS, Foote KD, et al: A closer look at unilateral versus bilateral deep brain stimulation: results of the National Institutes of Health COMPARE cohort. Clinical article. J Neurosurg 113:1224–1229, December 2010). In that article 52 patients with advanced Parkinson disease (PD) were randomized to receive deep brain stimulation (DBS) to the subthalamic nucleus (STN) or the globus pallidus internus (GPi); complete data sets were available in 44 of the patients. All cases were started with unilateral implantation, and the patients were offered the choice of a contralateral implantation after 6 months based on inadequacy to address motor symptoms. In the last sentence of the Results section, the following result was mentioned: “For each 1% increase in asymmetry in the baseline off-medication UPDRS [Unified Parkinson’s Disease Rating Motor Scale]–III score, the odds of receiving bilateral DBS decreased by a factor of 0.96.” That result is congruent with the data presented in Tables 3 and 4. It is also similar to the second sentence in the Conclusions: “There was a strong association between the degree of asymmetry in each patient’s PD and the preference for unilateral DBS.” However, those correct statements are contradicted in the Abstract. The last line in the Results section of the Abstract is “For every 1% increase in asymmetry, the odds of bilateral DBS increased by 0.96,” whereas I believe it was meant to be “… decreased by 0.96.” Mohamad Khaled, M.D., M.Surg. University of Virginia Health System Charlottesville, VA Disclosure The author reports no conflict of interest. Reference 1. Taba HA, Wu SS, Foote KD, Hass CJ, Fernandez HH, Malaty IA, et al: A closer look at unilateral versus bilateral deep brain stimulation: results of the National Institutes of Health COMPARE cohort. Clinical article. J Neurosurg 113:1224–1229, 2010
Response: Dr. Khaled has correctly pointed out that in our article there is a contradiction between text in the last sentence of the Results in the paper and the corresponding text in the Results section of the Abstract. The text in the Abstract is incorrect and should read “For every 1% increase in asymmetry, the odds of bilateral DBS J Neurosurg / Volume 119 / October 2013
decreased by 0.96.” The data contained in the Results and Conclusions in the paper are correct. We thank Dr. Khaled for pointing out this error.
Houtan A. Taba, M.D. Samuel S. Wu, Ph.D. Kelly D. Foote, M.D. Michael S. Okun, M.D. University of Florida College of Medicine Gainesville, FL
Please include this information when citing this paper: published online August 9, 2013; DOI: 10.3171/2010.12.JNS102097. ©AANS, 2013
Normal pressure hydrocephalus To The Editor: I read with great interest the paper by Penn et al.8 (Penn RD, Basati S, Sweetman B, et al: Ventricle wall movements and cerebrospinal fluid flow in hydrocephalus. Clinical article. J Neurosurg 115:159–164, July 2011). As part of their study, the authors measured the CSF net flow rate through the aqueduct in 2 patients with normal pressure hydrocephalus (NPH) before and after shunt insertion and compared the findings with findings in 8 controls. I wish to discuss 2 topics of interest raised by the authors’ work: 1) Why is the amount of CSF exiting the ventricular system in the controls higher than the accepted CSF formation rate? 2) How is a high net flow of CSF into the ventricles in NPH sustained? In the 8 controls, Penn et al. found the average net flow through the aqueduct to be 1.14 ± 0.599 ml/min passing from the ventricles toward the subarachnoid space. Given that there are 1440 minutes in a day, this would equate to an apparent CSF formation rate of over 1600 ml/day. The accepted CSF formation rate is about 500 ml/ day, and there is clearly a significant difference between these 2 figures. This is not the first time this discrepancy has been noted. Oresković and Klarica7 in their review noted 4 of 6 MRI studies in which the CSF formation rate was elevated above accepted figures and found a range of between 1.5 and 3 times normal for this metric. Penn et al.8 quote a study by Kim et al.6 who found a net aqueduct stroke volume of 30.1 ml per cycle. Unfortunately, Kim et al. do not indicate the average heart rate in their study but at a heart rate of 60 BPM this would equate to a CSF formation rate of 2600 ml/day. Clearly, this anomaly is real, but to date no one has tried to explain the discrepancy. The accepted CSF formation rate has been derived from CSF infusion studies and indicator dilution studies and is actually a measure of the amount of fluid leaving the cerebrospinal cavity principally via the arachnoid granulations but also by other accessory routes (the formation rate equals the absorption rate at steady state). 1075
Neurosurgical forum So in order for 1600 ml to be entering the basal cisterns per day, but only 500 ml leaving the arachnoid granulations per day, the cortex of the brain between these 2 sites would have to be absorbing 1100 ml per day. There is no pressure gradient between the subarachnoid space and the brain to sustain such an absorption rate,9 so an energy-utilizing pump would be required. Just such a pumping mechanism appears to exist. It has long been known that large molecules such as horseradish peroxidase will rapidly pass from the subarachnoid space into the perivascular spaces and then into the interstitial spaces of the brain.4 The arterial pulsations in the penetrating arterioles passing through the pial sheaths appear to drive CSF ahead of them through the perivascular spaces, not unlike the way a train entering a tunnel will push a column of air ahead of it. Simulated arterial pulsations induce fluid movements in the perivascular spaces of approximately 1 ml per heart beat.3 This would equate to about 80 ml per day per penetrating vessel, suggesting that perivascular absorption of 1100 ml/day is not out of the question. All that would then be required to explain the high aqueduct CSF flow would be for the CSF/interstitial brain fluid to make its way to the ventricle walls and then pass back into the ventricles. Thus, although only 500 ml/day of CSF would originate from the choroid plexus, with some of this fluid coming from the leakage of fluid from brain capillaries (10%–30%),4 about 1100 ml would be added from CSF recirculation through the interstitial spaces. Therefore, 1600 ml would leave the aqueduct per day. On the way through the subarachnoid space 1100 ml would be recirculated into the brain, with 500 ml continuing on to be reabsorbed over the vertex. Penn et al. accurately depict this physiology in their Fig. 3 (left) in which arrows pass from the subarachnoid space through the brain to the ventricles, but the mechanism of the brain absorption of CSF from the subarachnoid space is not discussed. The second anomaly I wish to discuss is the finding of Penn et al. that in 2 cases of NPH the net aqueduct flow was reversed, averaging 7.2 ml/min or over 10 L per day. Other groups have found negative but much lower flows. Kim et al.6 noted a flow equating to approximately 5.4 L/ day, and Balédent et al.1 noted a flow entering the ventricles of about 1.6 L/day in hydrocephalus. The measured absorption rate of CSF into the arachnoid granulations and accessory pathways is lower in NPH than in controls, at about 0.25 ml/min or 360 ml/day.5 Obviously if 10 L/ day is leaving the subarachnoid space via the aqueduct, 360 ml/day is exiting via the arachnoid granulations, and there is no net flow from the choroid plexus through the aqueduct to replenish the subarachnoid space; then in the absence of another source of fluid, the subarachnoid space will be completely depleted. Penn et al. get around this problem in their model by suggesting that most of the excess CSF entering the ventricles passes into the brain, through the parenchyma, and exits the brain surface back into the subarachnoid space (see Fig. 2 right). This is the mirror image of the normal findings in Fig. 3 left. The problem is that again there is no pressure gradient from ventricle to subarachnoid space to propel the fluid,9 but in addition there is no obvious energy-utilizing pump available to do this either (unlike the perivascular pump, 1076
which goes the other way). The flow out of the cortex would have to be swamping the perivascular flow in. The perivascular flow theoretically continues to occur into the brain even against a pressure gradient.3 The only solution available to the problem is for the cerebral cortex to become a net producer of CSF and the subependymal white matter to become a net absorber of CSF with little fluid passing across the parenchyma. As previously noted, the cortex is a small net producer of CSF (10%–30%). According to Starling’s laws, interstitial fluid exits the arteriolar side of the capillary bed because the hydrostatic pressure exceeds the oncotic pressure and returns on the venular side because the oncotic pressure exceeds the hydrostatic pressure. As the flows do not match exactly, a small net outflow of interstitial fluid adds to the CSF volume. In order for the cortex to produce up to 10 L per day, the venous pressure would need to be elevated above the interstitial pressure, and for the deep white matter to be a net absorber of 10 L/day, the venous pressure would need to be lower than the ventricular pressure. I have studied NPH for over 10 years using MRI to measure differential blood flow from the deep and superficial venous territories in the brain and have noted discrepancies entirely consistent with the suggestion that the superficial veins increase in pressure in NPH but the deep veins remain unchanged.2 Thus, unless a pump from the ventricle to cortex can be found in NPH, Penn et al.’s findings of reversal of flow in the aqueduct in NPH provide additional evidence that the real underlying physiology of NPH is altered venous pressure in the superficial brain. Finally, Penn et al. noted that the CSF dynamics returned to normal after shunt insertion, indicating that not all of the ventricular CSF exits via the shunt. Therefore, the CSF absorption abnormalities may be reversible. I have noted similar findings; shunting NPH patients increases sagittal sinus blood flow and would be consistent with a reduction in venous pressure and a reversal of the abnormal physiology.2 Thus, I believe that explaining the anomalies noted in the CSF outflow in the paper of Penn et al. appears to be important in understanding both the normal physiology of CSF flow and the physiology of NPH. Grant A. Bateman, M.B.B.S., M.D., F.R.A.N.Z.C.R. John Hunter Hospital Newcastle, Australia Disclosure The author reports no conflict of interest. References 1. Balédent O, Gondry-Jouet C, Meyer ME, De Marco G, Le Gars D, Henry-Feugeas MC, et al: Relationship between cerebrospinal fluid and blood dynamics in healthy volunteers and patients with communicating hydrocephalus. Invest Radiol 39:45–55, 2004 2. Bateman GA: The pathophysiology of idiopathic normal pressure hydrocephalus: cerebral ischemia or altered venous hemodynamics? AJNR AM J Neuroradiol 29:198–203, 2008 3. Bilston LE, Fletcher DF, Brodbelt AR, Stoodley MA: Arterial pulsation-driven cerebrospinal fluid flow in the perivascular
J Neurosurg / Volume 119 / October 2013
Neurosurgical forum space: a computational model. Comput Methods Biomech Biomed Eng 6:235–241, 2003 4. Brodbelt A, Stoodley M: CSF pathways: a review. Br J Neurosurg 21:510–520, 2007 5. Johanson CE, Duncan JA III, Klinge PM, Brinker T, Stopa EG, Silverberg GD: Multiplicity of cerebrospinal fluid functions: new challenges in health and disease. Cerebrospinal Fluid Res 14:10, 2008 6. Kim DS, Choi JU, Huh R, Yun PH, Kim DI: Quantitative assessment of cerebrospinal fluid hydrodynamics using a phasecontrast cine MR image in hydrocephalus. Childs Nerv Syst 15:461–467, 1999 7. Oresković D, Klarica M: The formation of cerebrospinal fluid: nearly a hundred years of interpretations and misinterpretations. Brain Res Rev 64:241–262, 2010 8. Penn RD, Basati S, Sweetman B, Guo X, Linninger A: Ventricle wall movements and cerebrospinal fluid flow in hydrocephalus. Clinical article. J Neurosurg 115:159–164, 2011 9. Stephensen H, Tisell M, Wikkelsö C: There is no transmantle pressure gradient in communicating or noncommunicating hy drocephalus. Neurosurgery 50:763–773, 2002
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 9, 2013; DOI: 10.3171/2011.2.JNS11216. ©AANS, 2013
Aneurysm morphology To The Editor: We are very interested in the clinical article by Lin et al.2 (Lin N, Ho A, Gross BA, et al: Differences in simple morphological variables in ruptured and unruptured middle cerebral artery aneurysms. Clinical article. J Neurosurg 117:913–919, November 2012). Ruptured aneurysms often contribute to a high mortality rate and severe morbidity. In particular, aneurysms on middle cerebral arteries (MCAs) are more prone to rupture than those in other anatomical locations.1 It is a crucial issue for clinicians to determine the treatment for unruptured MCA aneurysms before they rupture. Lin et al.2 proposed 3 new morphological parameters to investigate the differences between ruptured and unruptured MCA aneurysms. They found that aspect ratio, flow an gle, and parent-daughter angle are the 3 predominant pa rameters correlated to aneurysm rupture. Importantly, they showed that size ratio was a significant parameter (p = 0.04 < 0.05) in univariate analyses of the morphological parameters for MCA aneurysms. Conversely, when analyzed in multivariate analyses after adjustment for clinical and morphological risk factors, size ratio was not a significant parameter (p = 0.51 > 0.05) in their study. To this end, 39 unruptured MCA aneurysms were compared with 40 ruptured MCA aneurysms; the sample size in their study is relative smaller than that in other series. Ghosh et al.,1 for example, investigated the association of morphological and demographic features of intracranial aneurysms with rupture in 5138 aneurysms. They found that, among the morphological factors, a size greater than 10 mm, right sidedness, an aspect ratio greater than 1.6, deviated neck type, and multiplicity were correlated with higher rates of rupture.1 J Neurosurg / Volume 119 / October 2013
Despite the aforementioned limitations, Lin et al.’s study has provided 3 new morphological parameters of aneurysms. Further prospective, multicenter, and larger sample size–based study of morphological parameters in determining practical factors for unruptured and ruptured aneurysm is warranted.
I-Duo Wang Dueng-Yuan Hueng, M.D., Ph.D. Tri-Service General Hospital National Defense Medical Center Taipei, Taiwan
Disclosure The authors report no conflict of interest. References 1. Ghosh S, Dey S, Tjoumakaris S, Gonzalez F, Rosenwasser R, Pascal J, et al: Association of morphologic and demographic features of intracranial aneurysms with their rupture: a retrospective analysis. Acta Neurochir Suppl 115:275–278, 2013 2. Lin N, Ho A, Gross BA, Pieper S, Frerichs KU, Day AL, et al: Differences in simple morphological variables in ruptured and unruptured middle cerebral artery aneurysms. Clinical article. J Neurosurg 117:913–919, 2012
Response: We appreciate Dr. Hueng and colleagues’ interest in our study and their thoughtful comments. We believe that morphological parameters play an important role in the hemodynamics of an aneurysm and that the importance of each parameter is dependent on the specific location of the aneurysm. Both aspect ratio and size ratio describe the morphology of the aneurysm itself and are likely closely related, which may explain why size ratio was a significant factor in univariate but not in multivariate analysis. While our study is smaller than that of Ghosh et al.,1 it is unique in that we examined MCA aneurysms specifically and in greater detail. Despite the size limitation of our study, our results regarding aspect ratios are consistent with those of Ghosh et al. and others. Moreover, we were able to identify other significant factors not previously examined. This is likely due to the increased power of this study afforded by a more specific phenotype. We agree that future prospective, large, multicenter studies would be important in further advancing our understanding of the effects of aneurysm morphology on rupture risk. Rose Du, M.D., Ph.D. Ning Lin, M.D. Brigham and Women’s Hospital Boston, MA Harvard Medical School Boston, MA
Reference 1. Ghosh S, Dey S, Tjoumakaris S, Gonzalez F, Rosenwasser R, Pascal J, et al: Association of morphologic and demographic features of intracranial aneurysms with their rupture: a retrospective analysis. Acta Neurochir Suppl 115:275–278, 2013 Please include this information when citing this paper: published online August 2, 2013; DOI: 10.3171/2012.11.JNS121983. ©AANS, 2013
1077
Neurosurgical forum Rapid closure To The Editor: I read with interest the very significant article by Güresir et al.5 (Güresir E, Vatter H, Schuss P, et al: Rapid closure technique in decompressive craniectomy. Clinical article. J Neurosurg 114:954–960, April 2011). Decompressive craniectomy is one of the most frequently used operations in neurosurgery. Patients who receive decompressive craniectomy are critical and need urgent operations. Particularly in those patients who need bilateral craniectomies, operational duration becomes a key issue that determines whether we can save a patient’s life. In addition, encephalocele usually occurs due to brain swelling in some patients during the operation. There is no time to perform a duraplasty with insertion and watertight suturing of the dura graft in these cases. By analyzing a large series of data, the literature reveals that the rapid closure technique can significantly shorten surgical time without increasing the incidence of complications such as wound healing disturbances, CSF fistulas, epidural hematomas, or subdural hematomas. I am particularly interested in one of details of the surgical technique of rapid closure decompressive craniectomy (RCDC) noted in the paper by Güresir et al. Part of the exposed brain is covered by the loosely autogenous dura, and the remaining brain tissue is covered by Surgicel (Ethicon, Inc.). We know that Surgicel can be fully absorbed within 7–14 days (data on file; Ethicon). The brain tissue that was not covered by autogenous dura would contact subcutaneous tissue directly. I do not know why the authors did not use a synthetic dura graft to cover the exposed brain tissue. This kind of dura graft is simply curved to fit the surface of the brain, and fixed gently, and a tight suture with autogenous dura is unnecessary. This procedure would also not cost much time. The function of the synthetic dura is to separate brain tissue from subcutaneous tissue. There are two main purposes for use of synthetic dura. One purpose is to establish a clear boundary between brain tissue and subcutaneous tissue. The synthetic dura can be used as a mark to properly dissect subcutaneous tissue from brain tissue when preforming cranioplasty. What was shown in Fig. 2A in the article might be the galea layer, with the fascia and periosteum still attached to brain tissue. In particular, when dissecting temporal muscle from brain tissue, we can elevate the temporal muscle adequately if we dissect it along the synthetic dura layer. By doing this, the anatomical structure of temporal muscle could be preserved. Furthermore, the incidence of injuring brain tissue could be reduced. In the article, a study was cited that reported deep wound infection rates in a synthetic dural substitute group that were higher than in a pericranium group. However, many studies remain that do not support this opinion.1,3,4 The other purpose of using synthetic dura is reducing the interference of subcutaneous tissue with brain tissue. Some studies have revealed that a defect of dura was one risk of epilepsy after traumatic brain injury.2,6,7 I believe it would have been more appropriate to have included postoperative epilepsy as one of the complications in the study, so that we could know the morbidity of epilepsy 1078
after RCDC. We also could determine whether RCDC without dural implantation could increase the incidence of epilepsy.
Nan-Xiang Xiong, M.D. Union Hospital Tongji Medical College Huazhong University of Science and Technology Wuhan, China Hannover Medical School Hannover, Germany Disclosure
The author reports no conflict of interest. References 1. Biroli F, Fusco M, Bani GG, Signorelli A, Esposito F, de Divitiis O, et al: Novel equine collagen-only dural substitute. Neurosurgery 62 (3 Suppl 1):273–274, 2008 2. Diaz-Arrastia R, Agostini MA, Madden CJ, Van Ness PC: Posttraumatic epilepsy: the endophenotypes of a human model of epileptogenesis. Epilepsia 50 Suppl 2:14–20, 2009 3. Esposito F, Cappabianca P, Fusco M, Cavallo LM, Bani GG, Biroli F, et al: Collagen-only biomatrix as a novel dural substitute. Examination of the efficacy, safety and outcome: clinical experience on a series of 208 patients. Clin Neurol Neurosurg 110:343–351, 2008 4. Gazzeri R, Neroni M, Alfieri A, Galarza M, Faiola A, Esposito S, et al: Transparent equine collagen biomatrix as dural repair. A prospective clinical study. Acta Neurochir (Wien) 151: 537–543, 2009 5. Güresir E, Vatter H, Schuss P, Oszvald A, Raabe A, Seifert V, et al: Rapid closure technique in decompressive craniectomy. Clinical article. J Neurosurg 114:954–960, 2011 6. Temkin NR: Risk factors for posttraumatic seizures in adults. Epilepsia 44 Suppl 10:18–20, 2003 7. Yao Y, Mao Y, Zhou L: Decompressive craniectomy for massive cerebral infarction with enlarged cruciate duraplasty. Acta Neurochir (Wien) 149:1219–1221, 2007
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 2, 2013; DOI: 10.3171/2011.8.JNS11864. ©AANS, 2013
Recurrent glioma To The Editor: We are highly interested in the clinical article by Hoover et al.1 (Hoover JM, Nwojo M, Puffer R, et al: Surgical outcomes in recurrent glioma. Clinical article. J Neurosurg 118:1224–1231, June 2013). Glioma is characterized by highly invasive, angiogenic, and frequently recurrent features. Hoover et al.1 conducted a retrospective study to investigate surgical outcomes for recurrent glioma. They showed that postoperative survival was comparatively extended but complication risk increased in those patients with glioma who underwent several cranial surgeries. The major increase in neurological complications happened between the first and second craniotomy. Nevertheless, these complications only led to a moderately increased risk of functional decline after 2 or more craniotomies. J Neurosurg / Volume 119 / October 2013
Neurosurgical forum Because this was a retrospective study, it contains limitations in the assessment of quality of life (QOL) before and after surgery. However, patient-oriented assessment and preservation of QOL is very important. Preoperative and postoperative scores for QOL could be considered in the assessment of outcomes in patients with recurrent gliomas.2 In the future, using questionnaires to investigate QOL would provide a wider spectrum of outcomes assessment from the viewpoint of not only neurosurgeons, but also of patients undergoing multiple surgeries for recurrent glioma. Chun-Pi Chang Yu-Min Chang Chuang-Yen Huang Hung-Shiang Fang Chia-Hua Lin Dueng-Yuan Hueng, M.D., Ph.D. Tri-Service General Hospital National Defense Medical Center Taipei, Taiwan
Disclosure The authors report no conflict of interest. References 1. Hoover JM, Nwojo M, Puffer R, Mandrekar J, Meyer FB, Parney IF: Surgical outcomes in recurrent glioma. Clinical article. J Neurosurg 118:1224–1231, 2013 2. Jakola AS, Unsgård G, Solheim O: Quality of life in patients with intracranial gliomas: the impact of modern image-guided surgery. Clinical article. J Neurosurg 114:1622–1630, 2011
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 9, 2013; DOI: 10.3171/2013.3.JNS13559. ©AANS, 2013
Wrong-site craniotomy To The Editor: I read with great interest the article of Cohen et al.2 (Cohen FL, Mendelsohn D, Bernstein M: Wrong-site craniotomy: analysis of 35 cases and systems for prevention. Clinical article. J Neurosurg 113:461–473, September 2010). This study demonstrated that a broad range of events and factors can cause human errors to breach patient safeguards and lead to a wrong-site craniotomy (WSC). The authors concluded that, in essence, in all of the cases analyzed, the WSCs could have been prevented with strict adherence to comprehensive and thorough protocols. I agree with the authors that preoperative checks that might have prevented WSC include the following: 1) preoperative review of the medical rec ords and examination of the patient; 2) confirming the side and site of the operation by speaking with the patient and a family member; 3) marking the surgical site at the J Neurosurg / Volume 119 / October 2013
time of site confirmation; 4) imaging verification, including confirmation of the patient’s identity and confirmation that laterality is correctly labeled; 5) the presence of the patient’s imaging studies in the operating room before and during the operation; and 6) surgical time-outs to verify the patient’s identity and the identity on the images, correct positioning of the images, the procedure to be performed, and the site marking, as well as performing detailed perioperative patient and procedure verification and preoperative site marking. Today is different from the past. I would like to mention the most dramatic wrong-site craniotomy in the neurosurgical literature. Many of these preoperative checks had been done before the deadly WSC performed on Maurice Ravel by Clovis Vincent. Ravel (1875–1937), the great impressionist-classicist composer of many popular musical works, such as Boléro, suffered from a progressive disease and died following an exploratory craniotomy.3 Although the opinion of doctors and his friends was divided, it was decided that it was better to try to do something to rescue him rather than to let him continue as he was. Maurice Ravel was admitted to the rue Boileau clinic, believing that he was just going in for a test. In December 1937 Edouard Ravel gave his permission for an operation. At the time, hydrocephalus was suspected. Before the craniotomy, Ravel’s friends did not want him to be frightened when his hair was to be cut off for the intervention, and they suggested that this was another radiologic examination. The composer was clearly aware of what was going on, saying, “Not at all: I know exactly that they will cut of my head.” At the urging of Ravel’s friends, 2 pioneering French neurosurgeons, first Thiery de Martel and subsequently Clovis Vincent, had carried our numerous examinations, including pneumoencephalography. After a long phase of discussion and persuasion by Ravel’s friends, Professor Clovis Vincent agreed to conduct a neurosurgical operation; de Martel did not consider an intervention, but Vincent argued for a craniotomy.1 Ravel underwent an exploratory craniotomy that he hoped would restore much of his health on December 19, 1937. Professor Clovis Vincent undertook the procedure to see if a tumor was the cause of Ravel’s problems. After the surgery, Ravel lapsed into coma, and on December 28, 1937, he died. Vincent had performed a right-sided craniotomy, and the tentative preoperative diagnosis had been either an intracranial expansive process such as a right temporoparietal meningioma or a chronic right hemisphere subdural hematoma (SDH) resulting from a 1932 taxi accident. My colleagues and I have hypothesized that the 1932 accident caused a minimal left hemisphere SDH,3 but because the surgery was performed on the right side, no such lesion could be appreciated. Vincent was convinced that the right hemisphere, which in his opinion was no longer compensating for deficiencies of the left, had to be inflated. “What is the worst that could happen if neurosurgeons do something?” “What is the worst that could happen if neurosurgeons do nothing?” Probably, these were questions asked of Vincent before the operation. Unfortunately, in Ravel’s day there were no mod ern neuroimaging methods, such as CT, MRI, functional MRI, SPECT, or PET. Neurosurgeons, today, can under1079
Neurosurgical forum stand the feelings of Vincent, who had limited diagnostic tools. Ayhan Kanat, M.D. Tayyip Erdogan University Medical School Rize, Turkey Disclosure The author reports no conflict of interest.
References 1. Baeck E: The terminal illness and last composition of Maurice Ravel, in Bogousslavsky J, Boller F (eds): Neurological Disorders in Famous Artists. Basel: Karger, 2005, pp 132–140 2. Cohen FL, Mendelsohn D, Bernstein M: Wrong-site craniotomy: analysis of 35 cases and systems for prevention. Clinical article. J Neurosurg 113:461–473, 2010 3. Kanat A, Kayaci S, Yazar U, Yilmaz A: What makes Maurice Ravel’s deadly craniotomy interesting? Concerns of one of the most famous craniotomies in history. Acta Neurochir (Wien) 152:737–742, 2010
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 9, 2013; DOI: 10.3171/2010.10.JNS101577. ©AANS, 2013
Deep brain stimulation To The Editor: I read with interest the article by Snellings et al.6 (Snellings A, Sagher O, Anderson DJ, et al: Identification of the subthalamic nucleus in deep brain stimulation surgery with a novel wavelet-derived measure of neural background activity. Clinical article. J Neurosurg 111:767–774, October 2009), but I found that there are several issues that need clarification. The title and the Methods section are misleading. Therefore I think that this article needs commentary. In fact, it does not describe a “wavelet-derived measure of neural background activity” by the measure of the background activity after larger amplitude spike removal using the wavelet transform. This is an important distinction because, as I have demonstrated many years ago,4 the wavelet transform may help to characterize local variations in time and frequency patterns of the neural activity. Such patterns are very difficult to find by standard methods of averaging in the time or in the frequency domain such as, for example, the PSTH (post- or peri-stimulus time histograms) or the power spectra analysis with Fast Fourier Transform. My analysis demonstrated wavelet application as a mathematical “microscope” that helps to extract different transient neural responses, and until now it was not applied to the neural activity of the subthalamic nucleus (STN). Another point is that in describing their method, the 1080
authors are not very precise. From Methods: “The signal energy concentration properties of the wavelet transform allow us to identify and separate large amplitude neural spikes from background noise, then measure the root mean square amplitude of that background and output the result as a single identification index.” They identified large spikes and probably removed them, but the question is what wavelet did they use and how did they replace spikes? There are several possibilities for spike replacement that might influence the results of their analysis: 1) by using a signal with 0 amplitude (see Novak et al.2), 2) by a signal with the mean value of the noise (see Novak et al.3), or 3) by the spline function (as in Przybyszewski et al.5), among others. Also, the authors compared the root mean square (RMS) amplitude of the signal (a measure that is very sensitive to noise) in the frequency range 300–5000 Hz with a power spectra density value in a different frequency range (between 400 Hz and 2 kHz). We know from our previous2 and recent3 works that activity in the STN is highly variable in different frequency ranges. It was recently demonstrated that the RMS amplitude of the signal including large spikes may be good measure for the STN,1 but the power spectrum of the background activity after large spike removal seems to be a better measure.2,3 Andrzej W. Przybyszewski, Ph.D. University of Massachusetts Medical School Worcester, MA Polish-Japanese Institute of Information Technology Warsaw, Poland Disclosure Dr. Przybyszewski was partly supported by NCN 2011/03/B/ ST6/03816 and 2892/B/T02/2011/40.
References 1. Miyagi Y, Okamoto T, Morioka T, Tobimatsu S, Nakanishi Y, Aihara K, et al: Spectral analysis of field potential recordings by deep brain stimulation electrode for localization of subthalamic nucleus in patients with Parkinson’s disease. Stereotact Funct Neurosurg 87:211–218, 2009 2. Novak P, Daniluk S, Ellias SA, Nazzaro JM: Detection of the subthalamic nucleus in microelectrographic recordings in Parkinson disease using the high-frequency (> 500 Hz) neuronal background. Technical note. J Neurosurg 106:175–179, 2007 3. Novak P, Przybyszewski AW, Barborica A, Ravin P, Margolin L, Pilitsis JG: Localization of the subthalamic nucleus in Parkinson disease using multiunit activity. J Neurol Sci 310: 44–49, 2011 4. Przybyszewski AW: Analysis of the oscillatory patterns in the central nervous system with the wavelet method. J Neurosci Methods 38:247–257, 1991 5. Przybyszewski AW, Lankheet MJM, van de Grind WA: On the complex dynamics of intracellular ganglion cell light responses in the cat retina. Biol Cybern 74:299–308, 1996 6. Snellings A, Sagher O, Anderson DJ, Aldridge JW: Identification of the subthalamic nucleus in deep brain stimulation surgery with a novel wavelet-derived measure of neural background activity. Clinical article. J Neurosurg 111:767–774, 2009
J Neurosurg / Volume 119 / October 2013
Neurosurgical forum Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 9, 2013; DOI: 10.3171/2011.2.JNS101964. ©AANS, 2013
Intracranial hypotension To The Editor: I read with great interest the article by Vogel et al.1 (Vogel TW, Dlouhy BJ, Howard MA III: Use of confirmatory imaging studies to illustrate adequate treatment of cerebrospinal fluid leak in spontaneous intracranial hypotension. Technical note. J Neurosurg 113:955–960, November 2010). The authors reported an interesting case of spontaneous intracranial hypotension that was complicated by paradoxical herniation following a craniectomy. They concluded that the treatment of spontaneous intracranial hypotension sometimes necessitates multiple epidural blood patches and that radionuclide imaging is useful for confirming the presence of any CSF leaks and also the efficacy of the epidural blood patches. I agree with their conclusions, but I have one question about their radionuclide images. They described their radionuclide images as “single-photon emission CT of the lumbar and thoracic spine.” Their radionuclide cisternography images in Figs. 3 and 4 appear to be ordinary 2D projection images. There is no detailed description of the image acquisition methods. Single-photon emission CT produces tomographic images. Brain single-photon emission CT is a routine imaging method; however, it appears to be difficult to obtain high-quality images by means of spinal single-photon emission CT. If their radionuclide images are indeed single-photon emission CT images, then the authors should provide image orientation and a detailed description of the image acquisition methods. In addition, there is no description regarding the relationship between the lumbar puncture for radionuclide injection and the findings of the radionuclide images. Yasushi Shibata, M.D. University of Tsukuba Ibaraki, Japan
Disclosure The author reports no conflict of interest. Reference 1. Vogel TW, Dlouhy BJ, Howard MA III: Use of confirmatory imaging studies to illustrate adequate treatment of cerebrospinal fluid leak in spontaneous intracranial hypotension. Technical note. J Neurosurg 113:955–960, 2010
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 2, 2013; DOI: 10.3171/2011.2.JNS102007. ©AANS, 2013
J Neurosurg / Volume 119 / October 2013
Trigeminal neuralgia To The Editor: We read with great interest the article by Bahgat et al.2 (Bahgat D, Ray DK, Raslan AM, et al: Trigeminal neuralgia in young adults. Clinical article. J Neurosurg 114:1306–1311, May 2011). The authors reviewed the clinical materials and treatment outcomes in 7 patients younger than 25 years who suffered from trigeminal neuralgia. In their research, they encountered a significant finding about the etiology of trigeminal neuralgia in young adults, which is no obvious vessel compression, only venous compression in the surgical and imaging findings. We strongly agree with this point of view. Trigeminal neuralgia in patients under the age of 25 years is really rare. The clinical features and treatment are very specific and should be approached with caution. In the Discussion of their article, they addressed the question of how they manage cases in which preoperative MR images demonstrate no compression. Here we would like to elaborate on this subject. Compared with other treatment methods, such as Gamma Knife surgery (GKS), percutaneous radiofrequency retrogasserian rhizotomy, and percutaneous retrogasserian glycerol rhizolysis, microvascular decompression (MVD) is still known as the first choice for trigeminal neuralgia.3 Previous researches have indicated that patients with no vessel compression who underwent MVD still gained a good result. Baechli and Gratzl1 reported that 5 patients who had no visible vascular compression were treated with traditional MVD one or two times. All patients were finally pain free. It should be pointed out that the mean patient age was 65.6 years. The opinion that partial sectioning of the trigeminal nerve may be considered in patients who have negative explorations during an MVD was put forth by Liu and Apfelbaum9 and Delitala et al.4 Bahgat et al. also discussed the effectiveness of partial sensory rhizotomy for such patients in their article.2 Actually, we disagree with the opinion that trigeminal nerve rhizotomy should be performed in young adults. Many reports have revealed that the clinical results are not satisfactory given the higher rate of recurrence and side effects.6,8,14 In addition, numbness is inevitable after partial sensory rhizotomy. Patients would be uncomfortable with facial numbness. Younger patients especially are more self-conscious of persistent numbness than older patients, which affects the quality of life as well. In cases in which resection involves the ophthalmic branch of the trigeminal nerve, denervation of the cornea would result in decreased corneal reflex in the affected side. This may weaken the self-protection tendency of the cornea to external stimuli. Younger patients would endure longer periods in which there is less of the self-protective tendency, so there is a higher risk of keratohelcosis in younger patients than in older patients. Consequently, we prefer to perform GKS in young adults in whom preoperative MR images demonstrate no compression. In Bahgat and colleagues’ research, 2 cases did not achieve satisfactory clinical results after GKS, although we do not know the details about the dose and plan of the GKS. Several researches have revealed a good outcome after GKS.11,12 Sheehan et al.13 found that there 1081
Neurosurgical forum is no significant difference in pain relief between those with and those without vascular compression following GKS. In most Gamma Knife institutes, only one 4-mm isocenter is applied to target the root entry zone of the trigeminal nerve. However, at our institute, a maximal dose of 80 Gy is used with two 4-mm collimators. The 50% isodose line not only involves the trigeminal nerve root entry zone, but also the whole cisternal portion of the trigeminal nerve. In our experience, 2 young adults (23 and 26 years old, respectively) achieved good outcomes after treatment with GKS. It is a pity that there has been no statistical study to explore the difference in outcome between a single-isocenter group and a double-isocenter group thus far. Despite the ineffectiveness of therapy and the recurrence in some of the cases treated with GKS, long-term observation of repeat GKS for trigeminal neuralgia has shown good pain relief in a large number of patients.5–7 Facial numbness and weakness of the corneal reflex also happened in some patients, but the incidences were lower than that for partial sensory rhizotomy.10 Altogether, we consider GKS as a good choice for trigeminal neuralgia in young adults. Nan-Xiang Xiong, M.D.1,2 Hong-Yang Zhao, M.D.1 1 Union Hospital Tongji Medical College Huazhong University of Science and Technology Wuhan, China 2 Hannover Medical School Hannover, Germany Disclosure The authors report no conflict of interest. References 1. Baechli H, Gratzl O: Microvascular decompression in trigeminal neuralgia with no vascular compression. Eur Surg Res 39:51–57, 2007 2. Bahgat D, Ray DK, Raslan AM, McCartney S, Burchiel KJ: Trigeminal neuralgia in young adults. Clinical article. J Neurosurg 114:1306–1311, 2011 3. Broggi G, Ferroli P, Franzini A, Galosi L: The role of surgery in the treatment of typical and atypical facial pain. Neurol Sci Suppl 2:s95–s100, 2005 4. Delitala A, Brunori A, Chiappetta F: Microsurgical posterior fossa exploration for trigeminal neuralgia: a study on 48 cases. Minim Invasive Neurosurg 44:152–156, 2001 5. Dvorak T, Finn A, Price LL, Mignano JE, Fitzek MM, Wu JK, et al: Retreatment of trigeminal neuralgia with Gamma Knife radiosurgery: is there an appropriate cumulative dose? Clinical article. J Neurosurg 111:359–364, 2009 6. Gellner V, Kurschel S, Kreil W, Holl EM, Ofner-Kopeinig P, Unger F: Recurrent trigeminal neuralgia: long term outcome of repeat gamma knife radiosurgery. J Neurol Neurosurg Psychiatry 79:1405–1407, 2008 7. Kimball BY, Sorenson JM, Cunningham D: Repeat Gamma Knife surgery for trigeminal neuralgia: long-term results. Clinical article. J Neurosurg 113:178–183, 2010 8. Laghmari M, El Ouahabi A, Arkha Y, Derraz S, El Khamlichi A: Are the destructive neurosurgical techniques as effective as microvascular decompression in the management of trigeminal neuralgia? Surg Neurol 68:505–512, 2007
1082
9. Liu JK, Apfelbaum RI: Treatment of trigeminal neuralgia. Neurosurg Clin N Am 15:319–334, 2004 10. Matsuda S, Nagano O, Serizawa T, Higuchi Y, Ono J: Trigeminal nerve dysfunction after Gamma Knife surgery for trigeminal neuralgia: a detailed analysis. Clinical article. J Neurosurg 113:184–190, 2010 11. Pan HC, Sheehan J, Huang CF, Sheu ML, Yang DY, Chiu WT: Quality-of-life outcomes after Gamma Knife surgery for trigeminal neuralgia. Clinical article. J Neurosurg 113:191– 198, 2010 12. Riesenburger RI, Hwang SW, Schirmer CM, Zerris V, Wu JK, Mahn K, et al: Outcomes following single-treatment Gamma Knife surgery for trigeminal neuralgia with a minimum 3-year follow-up. Clinical article. J Neurosurg 112:766–771, 2010 13. Sheehan JP, Ray DK, Monteith S, Yen CP, Lesnick J, Kersh R, et al: Gamma Knife radiosurgery for trigeminal neuralgia: the impact of magnetic resonance imaging-detected vascular impingement of the affected nerve. Clinical article. J Neurosurg 113:53–58, 2010 14. Zakrzewska JM, Lopez BC, Kim SE, Coakham HB: Patient reports of satisfaction after microvascular decompression and partial sensory rhizotomy for trigeminal neuralgia. Neurosurgery 56:1304–1312, 2005
Response: No response was received from the authors of the original article. Please include this information when citing this paper: published online August 16, 2013; DOI: 10.3171/2012.12.JNS11965. ©AANS, 2013
Magnetic resonance imaging and thermal injury To The Editor: We read with interest the recent case report by Tanaka and coauthors3 (Tanaka R, Yumoto T, Shiba N, et al: Overheated and melted intracranial pressure transducer as cause of thermal brain injury during magnetic resonance imaging. Case report. J Neurosurg 117:1100–1109, December 2012). In the report, they describe a case in which a patient sustained a thermal brain injury due to overheating of an intraparenchymal intracranial pressure (ICP) transducer (Codman MicroSensor; Johnson & Johnson, Codman & Shurtleff, Inc.) during MRI within a 1.5-T Magnetom Vision (Siemens AG). Analysis of the cause of the incident revealed that the manufacturer’s MRI safety recommendations had not been followed. Specifically, the transducer had been positioned in a straight line during imaging instead of in the coiled configuration advised by the manufacturer. This resulted in heating of the wire within the transducer due to resonant radiofrequency (RF) waves produced by the “antenna effect.” Our institution has extensive experience with imaging acute traumatic brain injury patients, many with Codman MicroSensors in situ, as part of our research program.2,4 An article published in this journal describes the extensive safety testing we performed using a gel phantom with a Codman MicroSensor in situ within a 3-T Magnetom Total Imaging Matrix (TIM) Trio (Siemens Medical Solutions).2 We reported that extreme and J Neurosurg / Volume 119 / October 2013
Neurosurgical forum rapid heating to 64°C was noted when the transducer wire was not coiled and was placed straight within the scanner when using the transmit body coil and receive-only head coil. Following this heating, the probe no longer functioned. These findings were consistent with the phenomenon of resonance, and further testing demonstrated that the probe had a distinct resonant response. Coiling some of the transducer wire outside of the receive-only head coil reduced the current generated and so stopped the thermogenesis. Coiling had no impact on ICP transducer performance. Subsequently we found that fixing the coiled transducer wire to the head coil instead of to the patient’s head reduces the impact of any image artifacts that are produced by the coiled wire. We concluded that the Codman ICP MicroSensor is MRI conditional. We emphasized, however, that the findings were specific to our particular imaging system and caution should be used in applying them to other systems. Since the publication of this article we have added another 3-T scanner to our imaging center (Magnetom Verio 3T, Siemens Medical Solutions), and we have confirmed that the same configuration of the ICP transducer and cabling can be safely used. The accompanying illustration (Fig. 1) shows the configuration of the Codman MicroSensor system that is used within our center. Audit of our practice over the past 12 months demonstrated that we completed 59 imaging sessions in patients undergoing ICP monitoring within a Siemens 3-T scanner. During this time the Codman MicroSensor ceased to work following imaging in one patient. Analysis of the cause of this incident revealed that sensor malfunction was unrelated to the MR study, and was mechanical in origin: the sensor wire had been bent and fractured during transfer back to the intensive care unit, resulting in probe malfunction. The patient had a cranial access device in place, so the ICP sensor was removed and immediately replaced. There was no resultant harm.
To conclude, our experimental work and practical experience show that the Codman ICP MicroSensor is MR conditional. It may be safely used during MRI if the manufacturer’s safety instructions are followed for a 1.5-T system1 or if proper safety testing is performed prior to use in a 3-T system. However, as the manufacturer states in its instructions and as Tanaka et al. found in practice, failure to follow the published guidelines can result in serious injury to the patient.3 In addition, our experience suggests that siting the coiled transducer wire away from the patient’s head and attaching it to the imaging coil may retain the RF attenuation needed for safety, while reducing artifact from susceptibility effects.
Eleanor L. Carter, F.R.C.A., M.A. Virginia F. J. Newcombe, M.R.C.P., Ph.D. Robert C. Hawkes, Ph.D. Jonathan P. Coles, F.R.C.A., Ph.D. University of Cambridge Addenbrooke’s Hospital Cambridge, United Kingdom Disclosure
The authors report no conflict of interest. References 1. DePuy Synthes: Addendum Sheet: CODMAN® MicroSensor Products. (http://www.depuy.com/sites/default/files/products/ files/Microsensor%20IFU.pdf) [Accessed July 25, 2013] 2. Newcombe VFJ, Hawkes RC, Harding SG, Willcox R, Brock S, Hutchinson PJ, et al: Potential heating caused by intraparenchymal intracranial pressure transducers in a 3-tesla magnetic resonance imaging system using a body radiofrequency resonator: assessment of the Codman MicroSensor Transducer. Technical note. J Neurosurg 109:159–164, 2008 3. Tanaka R, Yumoto T, Shiba N, Okawa M, Yasuhara T, Ichikawa T, et al: Overheated and melted intracranial pressure transduc-
Fig. 1. Schematic diagram demonstrating the ICP transducer configuration used during MRI in our institution.
J Neurosurg / Volume 119 / October 2013
1083
Neurosurgical forum er as cause of thermal brain injury during magnetic resonance imaging. Case report. J Neurosurg 117:1100–1109, 2012 4. Williams EJ, Bunch CS, Carpenter TA, Downey SP, Kendall IV, Czosnyka M, et al: Magnetic resonance imaging compatibility testing of intracranial pressure probes. Technical note. J Neurosurg 91:706–709, 1999
Response: We would like to thank Dr. Carter and colleagues for their remarkably insightful and thoughtful letter. We appreciate the support that their letter lends to our conclusion that strict adherence to the manufacturer’s guidelines is very important for preventing serious complications in patients with ICP monitors undergoing MRI examinations. The number of reported thermal injuries associated with MRI procedures exceeds 160, and it is possible that many other unreported incidents have occurred. Knopp et al.3 reported third-degree burns in the medial calves of a male patient with unusual anatomy (after resection and radiation therapy for a liposarcoma) during conventional MRI scanning on a 1.5-T system that was operated within safe limits and without any external conductor. Davis et al.1 conducted experiments using non-uniform cylindrical phantoms made from a mixture of agar and saline and demonstrated that the RF fields of MR scanners can cause focal heating if the exposed object is non-uniform. Actually, the majority of burns associated with MRI have occurred where electrically conductive materials were attached to the skin. These burns are considered to be caused by electromagnetic induction in the loops due to RF radiation. It is therefore commonly believed that heating can be reduced by ensuring that the cables are not looped and that the electrodes are kept as close together as practical. However, as we described in our report, heating can be induced by RF fields used in MRI by 3 mechanisms. Especially when performing MRI for a patient with an implanted device, as overheating can cause serious complications, we need to pay close attention to many factors, including the configuration of a device. We speculated that the major mechanism of the incident in the case in our report was the antenna effect, one of electromagnetic induction heating by resonant RF waves, because only the tip of transducer was melted and scorched by the marked and rapid temperature rise and the total length of the ICP transducer was close to onefourth of the RF wavelength for a 1.5-T, 63.5-MHz MRI scanner. In that case, 3 aspects of the imaging procedure did not follow the manufacturer’s guidelines (configuration of the transducer in a specific geometry; using a 1.5-T MR system with a transmit/receive RF body coil, transmit body coil/receive-only head coil, or transmit/receive head coil; and limiting the specific absorption rate (SAR) to 1.0 W/kg). However, the precise mechanism by which the Codman MicroSensor may have caused the thermal brain injury in each violation of the above guidelines remains unclear. Experiments (including in vivo experiments) that duplicate the thermal heating in the ICP transducer would be of great interest and would help to elucidate the mechanisms by which RF heating causes injury. As mentioned by Dr. Carter and colleagues, they performed excellent experiments using a gel phantom with a Codman MicroSensor in situ within a 3-T Magnetom Total Imaging Matrix Trio.4 They showed that excessive 1084
heating to 64°C within 30 seconds occurred at the tip of the straight ICP probe when disconnected from the monitoring system and positioned where the RF fields would be greatest in the bore of the magnet. Other clinically unacceptable heating was observed with the ICP wire straight but with the connection at various locations within the bore. Temperature change within clinically acceptable limits was observed with about half of the 1-m transducer wire coiled into 4 loops and secured next to the head coil regardless of whether the probe was connected or not. No heating was observed when a transmit-and-receive head coil was used. These results of their study provide us with very valuable information, specifically concerning the configuration of the transducer, and were also useful for analysis of the cause of the incident in our case.4,6 Moreover, the fact that their method of coiling the transducer outside the head coil and fixing it to the head coil instead of to the patient’s head reduced the current and stopped the thermogenesis and their successful practical experiences using this method in 59 imaging sessions with a Siemens 3-T scanner will be very important and useful, because of the increasing use of 3-T MRI. Many factors should be taken into consideration when performing MRI scanning on a patient with an implanted device, including lead construction, the static field strength of the system, the type of RF coil, the orientation and position of the lead, the length and extension of the lead relative to the transmitter coil and the anatomical region imaged, and the amount of RF energy delivered (SAR). Furthermore, our general preconceptions that loop formation must be avoided at all times during MRI scanning have been shown to be incorrect. Even slight deviations from the guidelines have resulted in catastrophic complications.2,5 All of these facts imply that we need an expert in electromagnetic induction when we deviate from the guidelines. Generally, as experts with specialized knowledge of electromagnetic induction are not normally available at most hospitals, it seems to be difficult for us to make a correct change concerning the configuration of the manufacturer’s safety guidelines by ourselves. Moreover, new recommendations made by the manufacturer, as well as theory, excellent experimental work, and practical experience, are required for us to perform a new method of coiling the transducer outside the head coil and fixing it to the head coil in clinical cases. In summary, the manufacturer’s safety information and recommendations should be followed strictly. At tempting to generalize the data across MR system platforms, whether within or across manufacturers, is in ap propriate under the present circumstances. An old Japanese proverb, “A person who scalded his tongue with hot soup blows on cold meat dishes” (similar to the English proverb, “A scalded dog fears cold water”) may express our current situation exactly; nevertheless, we should adhere to safety guidelines and expect the manufacturer to continue revising the guidelines for greater safety or endeavor to develop devices that are completely compatible with the MR environment. Reiichiro Tanaka, M.D. Institute of Neuroscience and Orthopedics Okayama Kyokuto Hospital Okayama, Japan
J Neurosurg / Volume 119 / October 2013
Neurosurgical forum References 1. Davis PL, Shang C, Talagala L, Pasculle AW: Magnetic resonance imaging can cause focal heating in a nonuniform phantom. IEEE Trans Biomed Eng 40:1324–1327, 1993 2. Henderson JM, Tkach J, Phillips M, Baker K, Shellock FG, Rezai AR: Permanent neurological deficit related to magnetic resonance imaging in a patient with implanted deep brain stimulation electrodes for Parkinson’s disease. Neurosurgery 57:E1063, 2005 3. Knopp MV, Essig M, Debus J, Zabel HJ, Kaick GV: Unusual burns of the low extremities caused by a closed conducting loop in a patient at imaging. Radiology 200:572–575, 1996 4. Newcombe VFJ, Hawkes RC, Harding SG, Willcox R, Brock S, Hutchinson PJ, et al: Potential heating caused by intraparenchymal intracranial pressure transducers in a 3-tesla magnetic resonance imaging system using a body frequency resonator:
J Neurosurg / Volume 119 / October 2013
assessment of the Codman MicroSensor Transducer. Technical note. J Neurosurg 109:159–164, 2008 5. Spiegel J, Fuss G, Backens M, Reith W, Magnus T, Becker G, et al: Transient dystonia following magnetic resonance imaging in a patient with deep brain stimulation electrodes for the treatment of Parkinson’s disease. Case report. J Neurosurg 99:772–774, 2003 6. Williams EJ, Bunch CS, Carpenter TA, Downey SP, Kendall IV, Czosnyka M, et al: Magnetic resonance imaging compatibility testing of intracranial pressure probes. Technical note. J Neurosurg 91:706–709, 1999 Please include this information when citing this paper: published online August 23, 2013; DOI: 10.3171/2012.11.JNS122090. ©AANS, 2013
1085