J Neurosurg 121:1380–1387, 2014 ©AANS, 2014

Spinal leak as the cause of chronic subdural hematomas in nongeriatric patients

Clinical article

Jürgen Beck, M.D.,1 Jan Gralla, M.D., M.Sc.,2 Christian Fung, M.D.,1 Christian T. Ulrich, M.D.,1 Philippe Schucht, M.D.,1 Jens Fichtner, M.D.,1 Lukas Andereggen, M.D.,1 Martin Gosau, M.D.,4 Elke Hattingen, M.D.,5 Klemens Gutbrod, M.D.,3 Werner J. Z’Graggen, M.D.,1,3 Michael Reinert, M.D.,1,6 Jürg Hüsler, Ph.D.,7 Christoph Ozdoba, M.D.,2 and Andreas Raabe, M.D.1 Departments of 1Neurosurgery, 2Neuroradiology, and 3Neurology, Bern University Hospital, Bern, Switzerland; 4Department of Cranio-Maxillo-Facial Surgery, University Medical Center, Regensburg, Germany; 5Institute of Neuroradiology, University of Frankfurt, Frankfurt/Main, Germany; 6Department of Neurosurgery, Ospedale Cantonale di Lugano, Switzerland; and 7Institute of Mathematical Statistics and Actuarial Science, University of Bern, Switzerland

Object. The etiology of chronic (CSDH) in nongeriatric patients (≤ 60 years old) often re- mains unclear. The primary objective of this study was to identify spinal CSF leaks in young patients, after formulat- ing the hypothesis that spinal CSF leaks are causally related to CSDH. Methods. All consecutive patients 60 years of age or younger who underwent operations for CSDH between September 2009 and April 2011 at Bern University Hospital were included in this prospective cohort study. The pa- tient workup included an extended search for a spinal CSF leak using a systematic algorithm: MRI of the spinal axis with or without intrathecal contrast application, /, and postmyelography CT. Spinal patholo- gies were classified according to direct proof of CSF outflow from the intrathecal to the extrathecal space, presence of extrathecal fluid accumulation, presence of spinal meningeal cysts, or no pathological findings. The primary outcome was proof of a CSF leak. Results. Twenty-seven patients, with a mean age of 49.6 ± 9.2 years, underwent operations for CSDH. Hema- tomas were unilateral in 20 patients and bilateral in 7 patients. In 7 (25.9%) of 27 patients, spinal CSF leakage was proven, in 9 patients (33.3%) spinal meningeal cysts in the cervicothoracic region were found, and 3 patients (11.1%) had spinal cysts in the sacral region. The remaining 8 patients (29.6%) showed no pathological findings. Conclusions. The direct proof of spinal CSF leakage in 25.9% of patients suggests that spinal CSF leaks may be a frequent cause of nongeriatric CSDH. (http://thejns.org/doi/abs/10.3171/2014.6.JNS14550)

Key Words • chronic subdural hematoma • spinal CSF leak • geriatric • vascular disorders • meningeal cyst

hronic subdural hematomas (CSDHs) are gener- largement of the subdural hematoma is believed to be due ally regarded to be consequences of head trauma. to microbleedings from fragile walls of sprouting vessels, The most commonly accepted pathophysiological accumulation of fluid in this newly created space follow- explanationC of CSDH is that mild head trauma leads to ing an osmotic gradient created by blood degradation tearing of bridging veins with subsequent bleeding be- products, or a combination of these events.5,10 In older pa- tween the inner layer of the and the arachnoid, tients, mechanical forces related to atrophy enable or between the outer and inner layers of the dura, thus formation and growth of CSDH. A history of mild head creating the hematoma.5,14,28,29 Further and chronic en- trauma in many older patients is suggestive of trauma as a causative agent. In younger patients, brain atrophy is Abbreviations used in this paper: CSDH = chronic subdural generally not so prominent and a history of head trauma hematoma; mRS = modified Rankin Scale; SIH = spontaneous intra- is often not present. cranial hypotension; SPACE = sampling perfection with application Triggered by our experience with a young man with of optimized contrasts using different flip angle evolutions. recurrent CSDH whose recurrence only stopped after

1380 J Neurosurg / Volume 121 / December 2014

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC Spinal CSF leakage and chronic subdural hematoma a spinal CSF leak was sealed, we started to include the Stepwise Protocol of the Search search for spinal CSF leakage into a prospective proto- After a review of existing imaging results or sug- col for all consecutive nongeriatric CSDH patients. Our gestions of CSF leaks, 3-T SPACE imaging of the com- primary objectives were to identify spinal CSF leaks and plete spinal axis was conducted. Next, 0.5 ml of contrast spinal meningeal cysts in our cohort with the hypothesis medium (Magnevist) was injected intrathecally, and T1- that spinal CSF leaks are causally related to CSDH. In this paper we report our findings on the frequency of this weighted MRI of the spinal axis was completed. Myelog- condition. raphy was then performed, and finally, postmyelography CT was completed. If a CSF leak was identified, the im- aging algorithm was stopped with the least invasive study. Methods Example images from a patient are provided in Fig. 3; Study Design following the diagnosis of a CSDH (Fig. 3A), spinal im- aging without contrast demonstrated extrathecal fluid ac- This prospective cohort study was approved by the cumulation (Fig. 3B). The CSF was then proven by local ethics committee of Bern University Hospital. All fluoroscopy (Fig. 3C) and postmyelography CT (Fig. 3D). consecutive patients ≤ 60 years old with CSDH verified by CT or MRI who underwent operations between Sep- Classification of Spinal Imaging Findings tember 1, 2009, and April 12, 2011, were eligible for the study. After formulation of our hypothesis that spinal We classified the findings of the spinal imaging ac- CSF egress may be a cause of CSDH in younger patients, cording to the likelihood of pathological CSF outflow we distributed the study protocol among all staff of the from the intrathecal to the extrathecal space on a scale Departments of Neurosurgery and Neuroradiology at sev- from 0 to 5 in an ascending manner (Table 1): no path- eral meetings and conferences that took place beginning ological findings (category 0), detection of spinal men- 4 months prior to study initiation. Beginning September ingeal cysts in the sacral region (category 1), detection 1, 2009, all staff from both departments began to screen of spinal meningeal cysts in the cervicothoracic region for patients who underwent operations for CSDH. Fur- (category 2), detection of extrathecal fluid accumulation thermore, we used the daily interdisciplinary neurosurgi- on spinal MRI without intrathecal contrast application cal-neuroradiological conferences, where all elective and (category 3), detection of extrathecal contrast (Magnev- emergency cases were presented, as a second screening ist/iopamidol 300 [Iopamiro, Bracco]) on spinal MRI or platform. Patients operated on for a CSDH who were 60 postmyelography CT after intrathecal administration of years of age or younger were eligible for spinal imaging contrast (category 4), and direct visualization of outflow workup. We excluded patients who died early (< 48 hours) of intrathecally administered contrast (iodine) from the due to , who presented with recurrent SDH, who intrathecal to the extrathecal space under fluoroscopy were not compliant during spinal imaging, who had recent (category 5). The proof of an existing CSF leak (catego- brain surgery (< 6 months), or who had a shunt system for ries 4 and 5) was defined as the primary outcome. CSF diversion. As a next step, the computerized and writ- ten log books of all operating theaters were screened for Surgery for CSDH patients with CSDH who had not been included in the Whenever possible, the patients were operated on un- study. A flow diagram of the number of patients at each der local anesthesia via the 2-bur-hole technique;18 drain stage of the study is provided in Fig. 1. An idealized dia- insertion was performed if it was considered safe, after gram for how imaging modalities can be used, in order of careful rinsing with large amounts of normal saline so- increasing invasiveness, is shown in Fig. 2. lution.26 For nonliquified space-occupying blood clots, Spinal Imaging open craniotomy under general anesthesia was the first procedure or a backup procedure. For MRI, a Magnetom Avanto 1.5-T or Magnetom Verio 3-T (Siemens) machine was used. The following Postsurgical Care sequences were employed: T1-weighted, T2-weighted, All patients were admitted to the intermediate care magnetization-prepared rapid acquisition gradient echo (MPRAGE), and sampling perfection with application of unit or the intensive care unit after surgery. The drains optimized contrasts using different flip angle evolutions were left in place, restricted to 10 ml/hr and 120 ml/day, (SPACE) with multiplanar reconstructions. In cases of for a maximum of 48 hours. Mobilization of patients was intrathecal contrast application, gadopentetate dimeglu- started on the day of surgery. mine (Magnevist, Bayer) at a standard dose of 0.5 ml Treatment of Spinal CSF Leak dissolved in 9.5 ml CSF was administered via a at the L3–5 level. In cases in which a spinal CSF leak was detected, the Myelography was performed using a Siemens Artis following treatment algorithm was used: 1) bed rest for Zee Monoplan System (Siemens). A maximum dosage of as long as 3 days; 2) undirected blood patch at the L3–4 30 ml of iodine contrast material was administered after level (with an option to repeat the treatment if necessary); lumbar puncture using an atraumatic puncture kit. Post- 3) directed blood patch under fluoroscopy or CT guid- myelography CT was performed on a LightSpeed Ultra ance (with an option to repeat the treatment if necessary); scanner (GE) immediately after myelography and again 4 and 4) microsurgical exposure and closure of the CSF hours later in selected cases. leak. For treatment options 3 and 4, the exact site of the

J Neurosurg / Volume 121 / December 2014 1381

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC J. Beck et al.

Fig. 1. Flow chart showing the number of patients at each stage of this prospective cohort study. Two hundred twenty-one patients were screened and 27 patients underwent spinal imaging and were included in the final analysis.

CSF leak had to be known. Depending on the severity of consider the event of a CSF leak or of a recurrent CSDH symptoms, the escalation from an undirected treatment (1 in relation to explanatory variables by forward selection. or 2) to a directed treatment (3 or 4) was accelerated, in- Because the sample size and the number of events were cluding emergency neuroradiological workup to find the small, only a few explanatory variables and no interac- spinal CSF leak and to enable directed treatment. tions could be investigated together. The statistical analy- sis was conducted using SAS Enterprise Guide (version Patient Outcome 4.3). A p value < 0.05 was considered indicative of a sig- Outcome was assessed after 6 months during regu- nificant result. larly scheduled follow-up visits in the outpatient clinics. Patient outcome was based on the modified Rankin scale Results (mRS) score, working status, major neurological deficits, dependency, and living situation. Patient Characteristics and Clinical Findings Nineteen men and 8 women, with a mean age of Statistical Analysis 49.6 ± 9.2 years (median 52.0 years), were included in For descriptive statistics, data are expressed for con- the study (Table 2). Patient 1, who presented in November tinuous variables as means, standard deviations, and me- 2008 and whose clinical course led to the formation of dians. Associations between categorical variables such our hypothesis, was included in the final analysis. The as multiple recurrence, bilateral hematoma, head trauma, leading symptoms included , neurological defi- and CSF leak were calculated using a 2-sided Fisher exact cits, cranial nerve deficits, decreased level of conscious- test. Multiple logistic regression analysis was applied to ness, and (Table 2). Fifteen patients (55.6%) had

1382 J Neurosurg / Volume 121 / December 2014

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC Spinal CSF leakage and chronic subdural hematoma

Fig. 2. Flow chart for spinal imaging. The diagram shows how we believe that imaging modalities can be used, in order of increasing inva- siveness, during the search for spinal CSF leaks and spinal meningeal cysts. a history of head trauma, 1 patient was taking oral antico- agulants, and 4 patients were taking aspirin. Orthostatic headaches prevailed in 1 patient and were retrospectively documented in 2 additional cases. Thus, although the chief complaint of 15 patients was , for 12 of these patients the headache was not orthostatic; overall, 24 patients had no orthostatic symptoms. Fig. 3. Images from a patient with CSDH. Extrathecal fluid accu- Characteristics and Treatment of CSDH mulation (arrowheads) was demonstrated by axial T2-weighted (A) and sagittal SPACE 3-T MRI (B). The spinal CSF leak was proven by direct Chronic SDHs were unilateral in 20 patients (74.1%) visualization of the passage of contrast material from the intra- to the and bilateral in 7 patients (25.9%; Table 2). Initial treat- extrathecal space by anterior-posterior projection fluoroscopy (C). The ment was via 2-bur-hole trephination (25 cases), via leak is located in the thoracic spine (arrow) at the T10–11 level (C). 2-bur-hole trephination with conversion to craniotomy (1 The extradural accumulation of contrast material (arrow) is also demon- case), and craniotomy (1 case). One patient showed a yel- strated by postmyelography axial CT (D). lowish subdural fluid collection (hygroma, 3.7%), and 26 (96.3%) showed typical hematomas. Ten patients (37.0%) had recurrent CSDH treated with a total of 14 procedures. 3. Three (43%) of the 7 patients with proven CSF leaks All patients who underwent operations for multiple recur- had a history of head trauma as compared with 15 (75%) rences had a proven CSF leak (3 of 7 with a leak vs 0 of of 20 patients without a proven leak (p = 0.175). Imaging 20 without a leak; p = 0.012). comprised 46 spinal MRI procedures, 15 myelographies, 13 postmyelography CT scans, 1 cisternography, and 1 Results of Spinal Imaging nuclear medicine examination. According to our classification system (Table 1), Treatment of Spinal CSF Leaks 5 patients had a spinal CSF leak proven by direct visu- alization of contrast material outflow (category 5) and Eight patients, including 7 with proven CSF leaks 2 patients had a spinal CSF leak proven by extrathecal and 1 without spinal pathology but with recurrent CSDH, contrast material accumulation after intrathecal applica- were treated with 13 epidural blood-patching procedures. tion (category 4), resulting in 25.9% of patients with a Three patients underwent 4 spinal microsurgical opera- proven CSF leak as the primary outcome. Spinal cervi- tions for closure of the leak. Two of these patients needed cothoracic meningeal cysts (category 2) were found in 9 microsurgery after repeated blood patching. patients (33.3%), including patients with multiple cysts. Cysts in the sacral region only (category 1) were detected Patient Outcome in 3 cases (11.1%). No pathology (class 0) was found in 8 Outcome according to the mRS score at 6 months patients (29.6%). The specific imaging findings and imag- was a score of 0 for 15 patients, a score of 1 for 9 pa- ing modalities used per patient are listed in Tables 2 and tients, and a score of 2 for 3 patients (Table 2). One of the

J Neurosurg / Volume 121 / December 2014 1383

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC J. Beck et al.

TABLE 1: Classification of imaging findings and the likelihood of spinal CSF leakage

Category Imaging Finding Imaging Modality Intrathecal Contrast Material Interpretation 0 no suggestion of CSF loss MRI, myelography, CT, cisternography Gd, iodine, radionuclide no CSF fistula 1 spinal meningeal cyst, sacral region MRI, myelography, CT Gd, iodine CSF fistula unlikely 2 spinal meningeal cyst, cervicothoracic region MRI, myelography, CT Gd, iodine CSF fistula likely 3 extrathecal fluid detection w/o intrathecal MRI (T2 SPACE) — CSF fistula very likely contrast 4 extrathecal contrast accumulation after intra- CT/MRI Gd, iodine CSF fistula proven thecal contrast application 5 direct visualization of contrast passage from fluoroscopy (myelography) iodine CSF fistula proven intrathecal to extrathecal space patients with an mRS score of 2 had suffered an ische- Cerebrospinal Fluid Leakage as a Cause of CSDHs mic 8 years earlier with persistent hemiparesis. All Spinal outflow of CSF may lead to sagging of the 27 patients lived independently at home and 25 (92.6%) brain,19 which may tear congested bridging veins and returned to their previous work status. All patients were thus enable formation of a CSDH. A causal relationship available for follow-up. between CSF outflow and CSDH is supported by the for- 6 Multiple Logistic Regression mation of CSDH as a complication of lumbar puncture, spontaneous intracranial hypotension (SIH),19,21 neurosur- The dependence of a CSF leak event was considered gical shunting procedures,11,15 spinal anesthesia,1 spinal with respect to the following variables: sex, trauma, or- microdiscectomy with a dural tear,2 or any neurosurgical thostatic headache, oral anticoagulation/aspirin, and bilat- procedure associated with excessive CSF outflow.11 eral CSDH. Only the lack of (OR = Until now, the clinical relevance and incidence of spi- 0.030, p = 0.030) and the presence of bilateral CSDH (OR nal CSF outflow for nongeriatric patients with CSDH has = 21.33, p = 0.017) had significant impacts on the occur- not been documented and is very likely underestimated. rence of a CSF leak. The model assumptions and fit were The prevailing diagnostics focus on iatrogenic mecha- checked (area under the curve = 0.836). The same model nisms (such as spinal tap or CSF shunts) leading to CSF assumptions were used for the recurrence event with the loss, whereas an occult, idiopathic, or acquired form of same variables and also CSF leak. In this case, the pres- spinal CSF leakage that leads to CSDH is currently not ence of bilateral CSDH (OR = 17.8, p = 0.03) and lack of regarded as a relevant differential diagnosis and is thus trauma (OR = 0.059, p = 0.023) were significant factors. not routinely investigated. To our knowledge, a systematic The model assumptions and fit were also checked (area un- study of spinal CSF leakage in patients with CSDH has der the curve = 0.889). Because of the small sample size never been performed or reported. with not many events in the explanatory variables, these Because SIH is a possible diagnosis leading to CSDH, results should be considered with caution. Multiple recur- we specifically inquired about orthostatic symptoms at rences were also considered, but because there were only admission, and again during outpatient follow-up visits, 3 such events among the 27 patients, a multiple logistic including questioning of relatives. This led to identifica- regression did not produce an acceptable result. However, tion of 3 patients (11%) with some orthostatic component. this event was dependent on CSF leak (p = 0.012, Fisher The majority of subdural fluid collections in SIH report- exact test) and trauma (p = 0.029, Fisher exact test). ed in the literature are hygromas (which do not usually require evacuation); in contrast, in our series there were only 4% hygromas and the remaining 96% were hemato- mas.3,23 Both disorders, CSDH and SIH, occur with the Discussion same incidence of about 5 per 100,000 people.16,19 Thus, With this prospective search for spinal pathologies it is extremely unlikely that a simultaneous occurrence in a consecutive neurosurgical series of nongeriatric pa- of these disorders would happen by chance in 25.9% of a tients with CSDH, we found an unexpectedly high rate group of patients. of spinal CSF leaks (25.9%). Although spinal CSF leaks The presence of a spinal CSF leak might lead to dif- have been described as a potential cause of CSDH in rare ferent clinical phenotypes (SIH or CSDH), which raises cases, spinal CSF leakage has not been regarded as a fre- pathophysiological considerations. Whether there are dif- quent diagnosis and is therefore not included in a standard ferent kinds of spinal CSF leaks with specific anatomical workup. Our findings suggest that investigation of spinal or functional (flow) characteristics, with some leading to CSF leakage should be a mandatory part of a dedicated SIH and others to CSDH, or whether the clinical mani- workup. In addition to the 25.9% of patients with directly festations change over time or represent a spectrum of a proven CSF leaks, 33.3% of patients were found to have disease, is unknown. spinal cervicothoracic meningeal cysts. These cysts have been proposed as the cause or likely cause for spinal CSF Relevance of Spinal Meningeal Cysts outflow. 20 The incidence and natural history of spinal meninge-

1384 J Neurosurg / Volume 121 / December 2014

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC Spinal CSF leakage and chronic subdural hematoma

TABLE 2: Patient characteristics that study had spinal cervicothoracic cysts (33.3%, com- parable to the current cohort) but none had a proven CSF Variable Value leak. To take this into consideration, our classification dif- age (yrs) ferentiates proven CSF outflow (categories 4 and 5) from the mere presence of spinal cysts (categories 1 and 2). mean ± SD 49.6 ± 9.2 median 52.0 Patient Profile and Risk Factors males (n = 19) mean age ± SD (yrs) 48.1 ± 9.7 The reported incidence of recurrent CSDH varies 4,17,24,27,30 median age 51.0 from 3% to 39%, with reports of second or mul- tiple recurrence ranging from 2% to 9%.4,7,18,24 –26 Our series females (n = 8) showed a high rate of first (33.3%), second (11.1%), and third mean age ± SD (yrs) 53.1 ± 7.1 recurrences (3.7%), but no reference group of nongeriatric median age 55.5 patients with CSDH is available for comparison. Multiple leading symptoms (multiple) recurrences occurred significantly more often in patients headache 15 with proven CSF leaks (categories 4 and 5) as compared somnolence 4 with patients without such leaks. What makes the multiple recurrences noteworthy is the fact that in 4 of the 7 current 1 cases, the recurrence of the CSDH only stopped after the 2 CSF leak was sealed. This finding supports the hypothesis mydriasis 1 of a causal relationship between CSF outflow and CSDH. hemiparesis 3 Other researchers have reported that patients with ataxia 2 recurrent CSDH have a lower incidence of head trauma 13 apraxia 2 than patients without recurrent CSDH, as was also the case in the current series. In addition, bilateral CSDH perseveration 1 and young age appear to involve a higher risk of recur- paresthesia 1 rence.8,12,26,27 The current report describes an association cranial nerve deficit 2 of bilateral hematomas, orthostatic headaches, and mul- cognitive decline 1 tiple CSDH recurrences with a proven CSF leak. Based delirium 1 on our current observations, we suggest that criteria such location of CSDH as young age, early or multiple CSDH recurrence, con- tinuing or unexplained clinical deterioration, orthostatic rt 11 headaches, or the presence of bilateral CSDH should ini- lt 9 tiate the search for spinal CSF leaks. bilat 7 category of spinal CSF leak 0 8 Conclusions 1 3 Spinal imaging results are challenging the pathoge- 2 9 netic concept of CSDH in nongeriatric patients. Our re- 3 0 sults support the hypothesis that spinal CSF leakage is a 4 2 frequent cause of CSDH in young patients, which could 5 5 influence the diagnostic and therapeutic management of warfarin therapy 1 these patients. Potential spinal CSF leaks should be me- thodically investigated in young patients and in patients aspirin 4 with certain risk factors. mRS score at 6 mos 0 15 1 9 Disclosure 2 3 The authors report no conflict of interest concerning the mate- rials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript prepara- al cysts are not exactly known. Small perineural cysts in tion include the following. Conception and design: Beck, Raabe. the region of the sacrum appear to have a benign course.9 Acquisition of data: Beck, Gralla, Fung, Ulrich, Schucht, Fichtner, Spinal meningeal cysts have been regarded as the source Andereggen, Gutbrod, Z’Graggen, Reinert, Ozdoba, Raabe. of CSF outflow,3,22 and Schievink and his group suggested Analysis and interpretation of data: Beck, Gralla, Fung, Ulrich, that these are, together with imaging and clinical features, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, Raabe. Drafting the 20 article: Beck, Fung, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, a diagnostic criterion for spinal CSF leaks. Spinal CSF Raabe. Critically revising the article: all authors. Reviewed submit- leaks in combination with anticoagulation treatment were ted version of manuscript: all authors. Approved the final version of recently suggested as a cause of CSDH, particularly in the manuscript on behalf of all authors: Beck. Statistical analysis: elderly patients.22 Interestingly, all 3 reported patients in Hüsler, Beck. Study supervision: Beck.

J Neurosurg / Volume 121 / December 2014 1385

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC J. Beck et al.

TABLE 3: Neuroradiological workup of all patients studied*

Cranial Spinal Case No. CT MRI MRI w/o Contrast MRI & CM (iv) MRI & CM (it) Myelography Postmyelography CT 1† 11 2 2 1 2 3 2 2 6 1 3 1 1 3 5 1 4 1 1 1 5 4 6 3 1 3 2 6 4 1 7 4 1 8 1 1 9 4 3 1 1 1 1 10 1 1 11 2 1 1 1 1 12 3 3 1 13 1 1 14 1 1 15 2 1 16 1 1 17 8 1 1 2 1 1 18 1 1 1 19 1 5 2 1 1 20 1 1 21 3 1 1 22 1 1 2 2 23 1 2 1 24 6 1 25 6 2 2 1 1 26 1 1 1 2 1 1 27 2 1 1 2 2

* CM (it) = intrathecal application of contrast media; CM (iv) = intravenous application of contrast media. † Case 1 was the only patient to also undergo 1 cisternography and 1 nuclear medicine examination.

References A, et al: Experience in endovascular treatment of recurrent chronic subdural hematoma. Interv Neuroradiol 13 Suppl 1: 1. Amorim JA, Remígio DS, Damázio Filho O, de Barros MA, 141–144, 2007 Carvalho VN, Valença MM: Intracranial subdural hematoma 8. Ko BS, Lee JK, Seo BR, Moon SJ, Kim JH, Kim SH: Clinical post-spinal anesthesia: report of two cases and review of 33 cas- analysis of risk factors related to recurrent chronic subdural es in the literature. Rev Bras Anestesiol 60:620–629, 344–349, hematoma. J Korean Neurosurg Soc 43:11–15, 2008 2010 9. Langdown AJ, Grundy JR, Birch NC: The clinical relevance 2. Beier AD, Soo TM, Claybrooks R: Subdural hematoma after of Tarlov cysts. J Spinal Disord Tech 18:29–33, 2005 microdiscectomy: a case report and review of the literature. 10. Markwalder TM: Chronic subdural hematomas: a review. J Spine J 9:e9–e12, 2009 Neurosurg 54:637–645, 1981 3. Cohen-Gadol AA, Mokri B, Piepgras DG, Meyer FB, Atkin- 11. Mori K, Maeda M: Risk factors for the occurrence of chronic son JL: Surgical anatomy of dural defects in spontaneous spi- subdural haematomas after neurosurgical procedures. Acta nal cerebrospinal fluid leaks. Neurosurgery 58 (4 Suppl 2): Neurochir (Wien) 145:533–540, 2003 ONS-238–ONS-245, 2006 4. Forster MT, Mathé AK, Senft C, Scharrer I, Seifert V, Gerlach 12. Mori K, Maeda M: Surgical treatment of chronic subdural R: The influence of preoperative anticoagulation on outcome hematoma in 500 consecutive cases: clinical characteristics, and quality of life after surgical treatment of chronic subdural surgical outcome, complications, and recurrence rate. Neurol hematoma. J Clin Neurosci 17:975–979, 2010 Med Chir (Tokyo) 41:371–381, 2001 5. Gardner WJ: Traumatic subdural hematoma: with particu- 13. Oishi M, Toyama M, Tamatani S, Kitazawa T, Saito M: Clini- lar reference to the latent interval. Arch Neurol Psychiatry cal factors of recurrent chronic subdural hematoma. Neurol 27:847–858, 1932 Med Chir (Tokyo) 41:382–386, 2001 6. Gaucher DJ Jr, Perez JA Jr: Subdural hematoma following 14. Putnam TJ, Cushing H: Chronic subdural hematoma: its pa- lumbar puncture. Arch Intern Med 162:1904–1905, 2002 thology, its relation to pachymeningitis hemorrhagica and its 7. Ishihara H, Ishihara S, Kohyama S, Yamane F, Ogawa M, Sato surgical treatment. Arch Surg 11:329–393, 1925

1386 J Neurosurg / Volume 121 / December 2014

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC Spinal CSF leakage and chronic subdural hematoma

15. Samuelson S, Long DM, Chou SN: Subdural hematoma as a 25. Stroobandt G, Fransen P, Thauvoy C, Menard E: Pathogenetic complication of shunting procedures for normal pressure hy- factors in chronic subdural haematoma and causes of recur- drocephalus. J Neurosurg 37:548–551, 1972 rence after drainage. Acta Neurochir (Wien) 137:6–14, 1995 16. Santarius T, Hutchinson PJ: Chronic subdural haematoma: time 26. Tahsim-Oglou Y, Beseoglu K, Hänggi D, Stummer W, Steiger to rationalize treatment? Br J Neurosurg 18:328–332, 2004 HJ: Factors predicting recurrence of chronic subdural haema- 17. Santarius T, Kirkpatrick PJ, Ganesan D, Chia HL, Jalloh toma: the influence of intraoperative irrigation and low-mo- I, Smielewski P, et al: Use of drains versus no drains after lecular-weight heparin thromboprophylaxis. Acta Neurochir burr-hole evacuation of chronic subdural haematoma: a ran- (Wien) 154:1063–1068, 2012 domised controlled trial. Lancet 374:1067–1073, 2009 27. Torihashi K, Sadamasa N, Yoshida K, Narumi O, Chin M, Ya- 18. Santarius T, Kirkpatrick PJ, Kolias AG, Hutchinson PJ: Work- magata S: Independent predictors for recurrence of chronic ing toward rational and evidence-based treatment of chronic subdural hematoma: a review of 343 consecutive surgical subdural hematoma. Clin Neurosurg 57:112–122, 2010 cases. Neurosurgery 63:1125–1129, 2008 19. Schievink WI: Spontaneous spinal cerebrospinal fluid leaks 28. Trotter W: Chronic subdural haemorrhage of traumatic origin, and intracranial hypotension. JAMA 295:2286–2296, 2006 and its relation to pachymeningitis haemorrhagica interna. Br 20. Schievink WI, Maya MM, Louy C, Moser FG, Tourje J: Di- J Surg 2:271–291, 1914 agnostic criteria for spontaneous spinal CSF leaks and intra- 29. Virchow R: Das haematom der dura mater. Verh Phys Med cranial hypotension. AJNR Am J Neuroradiol 29:853–856, Ges Würzburg 7:134–142, 1857 2008 30. Weigel R, Schmiedek P, Krauss JK: Outcome of contempo- 21. Schievink WI, Maya MM, Moser FG, Tourje J: Spectrum of rary surgery for chronic subdural haematoma: evidence based subdural fluid collections in spontaneous intracranial hypo- review. J Neurol Neurosurg Psychiatry 74:937–943, 2003 tension. J Neurosurg 103:608–613, 2005 22. Schievink WI, Maya MM, Pikul BK, Louy C: Spontaneous spinal cerebrospinal fluid leaks as the cause of subdural he- matomas in elderly patients on anticoagulation. Report of 3 cases. J Neurosurg 112:295–299, 2010 Manuscript submitted September 26, 2013. 23. Sencakova D, Mokri B, McClelland RL: The efficacy of Accepted June 9, 2014. epidural blood patch in spontaneous CSF leaks. Please include this information when citing this paper: pub- 57:1921–1923, 2001 lished online July 18, 2014; DOI: 10.3171/2014.6.JNS14550. 24. Smely C, Madlinger A, Scheremet R: Chronic subdural hae- Address correspondence to: Jürgen Beck, M.D., Department of matoma–a comparison of two different treatment modalities. Neurosurgery, Bern University Hospital, Freiburgstrasse 10, Bern Acta Neurochir (Wien) 139:818–826, 1997 3010, Switzerland. email: [email protected].

J Neurosurg / Volume 121 / December 2014 1387

Unauthenticated | Downloaded 10/05/21 08:02 AM UTC