J Neurosurg 121:1380–1387, 2014 ©AANS, 2014
Spinal cerebrospinal fluid 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 subdural hematoma (CSDH) in nongeriatric patients (≤ 60 years old) often remains unclear. The primary objective of this study was to identify spinal CSF leaks in young patients, after formulating 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 patient 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, myelography/fluoroscopy, and postmyelography CT. Spinal pathologies 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. Hematomas 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
C
hronic subdural hematomas (CSDHs) are generally regarded to be consequences of head trauma. The most commonly accepted pathophysiological explanation of CSDH is that mild head trauma leads to tearing of bridging veins with subsequent bleeding between the inner layer of the dura mater and the arachnoid, or between the outer and inner layers of the dura, thus creating the hematoma.5,14,28,29 Further and chronic en-
Abbreviations used in this paper: CSDH = chronic subdural hematoma; mRS = modified Rankin Scale; SIH = spontaneous intracranial hypotension; SPACE = sampling perfection with application of optimized contrasts using different flip angle evolutions.
1380
largement of the subdural hematoma is believed to be due to microbleedings from fragile walls of sprouting vessels, accumulation of fluid in this newly created space following an osmotic gradient created by blood degradation products, or a combination of these events.5,10 In older patients, mechanical forces related to brain atrophy enable formation and growth of CSDH. A history of mild head trauma in many older patients is suggestive of trauma as a causative agent. In younger patients, brain atrophy is generally not so prominent and a history of head trauma is often not present. Triggered by our experience with a young man with recurrent CSDH whose recurrence only stopped after J Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma a spinal CSF leak was sealed, we started to include the search for spinal CSF leakage into a prospective protocol for all consecutive nongeriatric CSDH patients. Our primary objectives were to identify spinal CSF leaks and spinal meningeal cysts in our cohort with the hypothesis that spinal CSF leaks are causally related to CSDH. In this paper we report our findings on the frequency of this condition. Study Design
Methods
This prospective cohort study was approved by the local ethics committee of Bern University Hospital. All consecutive patients ≤ 60 years old with CSDH verified by CT or MRI who underwent operations between September 1, 2009, and April 12, 2011, were eligible for the study. After formulation of our hypothesis that spinal CSF egress may be a cause of CSDH in younger patients, we distributed the study protocol among all staff of the Departments of Neurosurgery and Neuroradiology at several meetings and conferences that took place beginning 4 months prior to study initiation. Beginning September 1, 2009, all staff from both departments began to screen for patients who underwent operations for CSDH. Furthermore, we used the daily interdisciplinary neurosurgical-neuroradiological conferences, where all elective and emergency cases were presented, as a second screening platform. Patients operated on for a CSDH who were 60 years of age or younger were eligible for spinal imaging workup. We excluded patients who died early (< 48 hours) due to epilepsy, who presented with recurrent SDH, who were not compliant during spinal imaging, who had recent brain surgery (< 6 months), or who had a shunt system for CSF diversion. As a next step, the computerized and written log books of all operating theaters were screened for patients with CSDH who had not been included in the study. A flow diagram of the number of patients at each stage of the study is provided in Fig. 1. An idealized diagram for how imaging modalities can be used, in order of increasing invasiveness, is shown in Fig. 2.
Spinal Imaging
For MRI, a Magnetom Avanto 1.5-T or Magnetom Verio 3-T (Siemens) machine was used. The following sequences were employed: T1-weighted, T2-weighted, magnetization-prepared rapid acquisition gradient echo (MPRAGE), and sampling perfection with application of optimized contrasts using different flip angle evolutions (SPACE) with multiplanar reconstructions. In cases of intrathecal contrast application, gadopentetate dimeglumine (Magnevist, Bayer) at a standard dose of 0.5 ml dissolved in 9.5 ml CSF was administered via a lumbar puncture at the L3–5 level. Myelography was performed using a Siemens Artis Zee Monoplan System (Siemens). A maximum dosage of 30 ml of iodine contrast material was administered after lumbar puncture using an atraumatic puncture kit. Postmyelography CT was performed on a LightSpeed Ultra scanner (GE) immediately after myelography and again 4 hours later in selected cases.
J Neurosurg / Volume 121 / December 2014
Stepwise Protocol of the Search
After a review of existing imaging results or suggestions of CSF leaks, 3-T SPACE imaging of the complete spinal axis was conducted. Next, 0.5 ml of contrast medium (Magnevist) was injected intrathecally, and T1weighted MRI of the spinal axis was completed. Myelography was then performed, and finally, postmyelography CT was completed. If a CSF leak was identified, the imaging algorithm was stopped with the least invasive study. Example images from a patient are provided in Fig. 3; following the diagnosis of a CSDH (Fig. 3A), spinal imaging without contrast demonstrated extrathecal fluid accumulation (Fig. 3B). The CSF fistula was then proven by fluoroscopy (Fig. 3C) and postmyelography CT (Fig. 3D).
Classification of Spinal Imaging Findings
We classified the findings of the spinal imaging according to the likelihood of pathological CSF outflow from the intrathecal to the extrathecal space on a scale from 0 to 5 in an ascending manner (Table 1): no pathological findings (category 0), detection of spinal meningeal cysts in the sacral region (category 1), detection of spinal meningeal cysts in the cervicothoracic region (category 2), detection of extrathecal fluid accumulation on spinal MRI without intrathecal contrast application (category 3), detection of extrathecal contrast (Magnevist/iopamidol 300 [Iopamiro, Bracco]) on spinal MRI or postmyelography CT after intrathecal administration of contrast (category 4), and direct visualization of outflow of intrathecally administered contrast (iodine) from the intrathecal to the extrathecal space under fluoroscopy (category 5). The proof of an existing CSF leak (categories 4 and 5) was defined as the primary outcome.
Surgery for CSDH
Whenever possible, the patients were operated on under local anesthesia via the 2-bur-hole technique;18 drain insertion was performed if it was considered safe, after careful rinsing with large amounts of normal saline solution.26 For nonliquified space-occupying blood clots, open craniotomy under general anesthesia was the first procedure or a backup procedure. Postsurgical Care
All patients were admitted to the intermediate care unit or the intensive care unit after surgery. The drains were left in place, restricted to 10 ml/hr and 120 ml/day, for a maximum of 48 hours. Mobilization of patients was started on the day of surgery.
Treatment of Spinal CSF Leak
In cases in which a spinal CSF leak was detected, the following treatment algorithm was used: 1) bed rest for as long as 3 days; 2) undirected blood patch at the L3–4 level (with an option to repeat the treatment if necessary); 3) directed blood patch under fluoroscopy or CT guidance (with an option to repeat the treatment if necessary); and 4) microsurgical exposure and closure of the CSF leak. For treatment options 3 and 4, the exact site of the 1381
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 symptoms, the escalation from an undirected treatment (1 or 2) to a directed treatment (3 or 4) was accelerated, including emergency neuroradiological workup to find the spinal CSF leak and to enable directed treatment. Patient Outcome
Outcome was assessed after 6 months during regularly scheduled follow-up visits in the outpatient clinics. Patient outcome was based on the modified Rankin scale (mRS) score, working status, major neurological deficits, dependency, and living situation.
Statistical Analysis
For descriptive statistics, data are expressed for continuous variables as means, standard deviations, and medians. Associations between categorical variables such as multiple recurrence, bilateral hematoma, head trauma, and CSF leak were calculated using a 2-sided Fisher exact test. Multiple logistic regression analysis was applied to
1382
consider the event of a CSF leak or of a recurrent CSDH in relation to explanatory variables by forward selection. Because the sample size and the number of events were small, only a few explanatory variables and no interactions could be investigated together. The statistical analysis was conducted using SAS Enterprise Guide (version 4.3). A p value < 0.05 was considered indicative of a significant result.
Results
Patient Characteristics and Clinical Findings
Nineteen men and 8 women, with a mean age of 49.6 ± 9.2 years (median 52.0 years), were included in the study (Table 2). Patient 1, who presented in November 2008 and whose clinical course led to the formation of our hypothesis, was included in the final analysis. The leading symptoms included headaches, neurological deficits, cranial nerve deficits, decreased level of consciousness, and seizures (Table 2). Fifteen patients (55.6%) had J Neurosurg / Volume 121 / December 2014
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 invasiveness, during the search for spinal CSF leaks and spinal meningeal cysts.
a history of head trauma, 1 patient was taking oral anticoagulants, 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 headache, for 12 of these patients the headache was not orthostatic; overall, 24 patients had no orthostatic symptoms. Characteristics and Treatment of CSDH
Chronic SDHs were unilateral in 20 patients (74.1%) and bilateral in 7 patients (25.9%; Table 2). Initial treatment was via 2-bur-hole trephination (25 cases), via 2-bur-hole trephination with conversion to craniotomy (1 case), and craniotomy (1 case). One patient showed a yellowish 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. All patients who underwent operations for multiple recurrences had a proven CSF leak (3 of 7 with a leak vs 0 of 20 without a leak; p = 0.012).
Results of Spinal Imaging
According to our classification system (Table 1), 5 patients had a spinal CSF leak proven by direct visualization of contrast material outflow (category 5) and 2 patients had a spinal CSF leak proven by extrathecal contrast material accumulation after intrathecal application (category 4), resulting in 25.9% of patients with a proven CSF leak as the primary outcome. Spinal cervicothoracic meningeal cysts (category 2) were found in 9 patients (33.3%), including patients with multiple cysts. Cysts in the sacral region only (category 1) were detected in 3 cases (11.1%). No pathology (class 0) was found in 8 patients (29.6%). The specific imaging findings and imaging modalities used per patient are listed in Tables 2 and
J Neurosurg / Volume 121 / December 2014
Fig. 3. Images from a patient with CSDH. Extrathecal fluid accumulation (arrowheads) was demonstrated by axial T2-weighted (A) and sagittal SPACE 3-T MRI (B). The spinal CSF leak was proven by direct visualization of the passage of contrast material from the intra- to the extrathecal space by anterior-posterior projection fluoroscopy (C). The leak is located in the thoracic spine (arrow) at the T10–11 level (C). The extradural accumulation of contrast material (arrow) is also demonstrated by postmyelography axial CT (D).
3. Three (43%) of the 7 patients with proven CSF leaks had a history of head trauma as compared with 15 (75%) of 20 patients without a proven leak (p = 0.175). Imaging comprised 46 spinal MRI procedures, 15 myelographies, 13 postmyelography CT scans, 1 cisternography, and 1 nuclear medicine examination. Treatment of Spinal CSF Leaks
Eight patients, including 7 with proven CSF leaks and 1 without spinal pathology but with recurrent CSDH, were treated with 13 epidural blood-patching procedures. Three patients underwent 4 spinal microsurgical operations for closure of the leak. Two of these patients needed microsurgery after repeated blood patching.
Patient Outcome
Outcome according to the mRS score at 6 months was a score of 0 for 15 patients, a score of 1 for 9 patients, and a score of 2 for 3 patients (Table 2). One of the 1383
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 1 2 3
no suggestion of CSF loss spinal meningeal cyst, sacral region spinal meningeal cyst, cervicothoracic region extrathecal fluid detection w/o intrathecal contrast extrathecal contrast accumulation after intra thecal contrast application direct visualization of contrast passage from intrathecal to extrathecal space
MRI, myelography, CT, cisternography MRI, myelography, CT MRI, myelography, CT MRI (T2 SPACE)
Gd, iodine, radionuclide Gd, iodine Gd, iodine —
no CSF fistula CSF fistula unlikely CSF fistula likely CSF fistula very likely
CT/MRI
Gd, iodine
CSF fistula proven
fluoroscopy (myelography)
iodine
CSF fistula proven
4 5
patients with an mRS score of 2 had suffered an ischemic stroke 8 years earlier with persistent hemiparesis. All 27 patients lived independently at home and 25 (92.6%) returned to their previous work status. All patients were available for follow-up. Multiple Logistic Regression
The dependence of a CSF leak event was considered with respect to the following variables: sex, trauma, orthostatic headache, oral anticoagulation/aspirin, and bilateral CSDH. Only the lack of orthostatic headache (OR = 0.030, p = 0.030) and the presence of bilateral CSDH (OR = 21.33, p = 0.017) had significant impacts on the occurrence of a CSF leak. The model assumptions and fit were checked (area under the curve = 0.836). The same model assumptions were used for the recurrence event with the same variables and also CSF leak. In this case, the presence of bilateral CSDH (OR = 17.8, p = 0.03) and lack of trauma (OR = 0.059, p = 0.023) were significant factors. The model assumptions and fit were also checked (area under the curve = 0.889). Because of the small sample size with not many events in the explanatory variables, these results should be considered with caution. Multiple recurrences were also considered, but because there were only 3 such events among the 27 patients, a multiple logistic regression did not produce an acceptable result. However, this event was dependent on CSF leak (p = 0.012, Fisher exact test) and trauma (p = 0.029, Fisher exact test).
Discussion
With this prospective search for spinal pathologies in a consecutive neurosurgical series of nongeriatric patients with CSDH, we found an unexpectedly high rate of spinal CSF leaks (25.9%). Although spinal CSF leaks have been described as a potential cause of CSDH in rare cases, spinal CSF leakage has not been regarded as a frequent diagnosis and is therefore not included in a standard workup. Our findings suggest that investigation of spinal CSF leakage should be a mandatory part of a dedicated workup. In addition to the 25.9% of patients with directly proven CSF leaks, 33.3% of patients were found to have spinal cervicothoracic meningeal cysts. These cysts have been proposed as the cause or likely cause for spinal CSF outflow.20 1384
Cerebrospinal Fluid Leakage as a Cause of CSDHs
Spinal outflow of CSF may lead to sagging of the brain,19 which may tear congested bridging veins and thus enable formation of a CSDH. A causal relationship between CSF outflow and CSDH is supported by the formation of CSDH as a complication of lumbar puncture,6 spontaneous intracranial hypotension (SIH),19,21 neurosurgical shunting procedures,11,15 spinal anesthesia,1 spinal microdiscectomy with a dural tear,2 or any neurosurgical procedure associated with excessive CSF outflow.11 Until now, the clinical relevance and incidence of spinal CSF outflow for nongeriatric patients with CSDH has not been documented and is very likely underestimated. The prevailing diagnostics focus on iatrogenic mechanisms (such as spinal tap or CSF shunts) leading to CSF loss, whereas an occult, idiopathic, or acquired form of spinal CSF leakage that leads to CSDH is currently not regarded as a relevant differential diagnosis and is thus not routinely investigated. To our knowledge, a systematic study of spinal CSF leakage in patients with CSDH has never been performed or reported. Because SIH is a possible diagnosis leading to CSDH, we specifically inquired about orthostatic symptoms at admission, and again during outpatient follow-up visits, including questioning of relatives. This led to identification of 3 patients (11%) with some orthostatic component. The majority of subdural fluid collections in SIH reported 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 hematomas.3,23 Both disorders, CSDH and SIH, occur with the same incidence of about 5 per 100,000 people.16,19 Thus, it is extremely unlikely that a simultaneous occurrence of these disorders would happen by chance in 25.9% of a group of patients. The presence of a spinal CSF leak might lead to different clinical phenotypes (SIH or CSDH), which raises pathophysiological considerations. Whether there are different kinds of spinal CSF leaks with specific anatomical or functional (flow) characteristics, with some leading to SIH and others to CSDH, or whether the clinical manifestations change over time or represent a spectrum of a disease, is unknown. Relevance of Spinal Meningeal Cysts
The incidence and natural history of spinal meningeJ Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma TABLE 2: Patient characteristics Variable age (yrs) mean ± SD median males (n = 19) mean age ± SD (yrs) median age females (n = 8) mean age ± SD (yrs) median age leading symptoms (multiple) headache somnolence coma seizure mydriasis hemiparesis ataxia apraxia perseveration paresthesia cranial nerve deficit cognitive decline delirium location of CSDH rt lt bilat category of spinal CSF leak 0 1 2 3 4 5 warfarin therapy aspirin mRS score at 6 mos 0 1 2
Value 49.6 ± 9.2 52.0 48.1 ± 9.7 51.0 53.1 ± 7.1 55.5 15 4 1 2 1 3 2 2 1 1 2 1 1 11 9 7 8 3 9 0 2 5 1 4 15 9 3
al cysts are not exactly known. Small perineural cysts in the region of the sacrum appear to have a benign course.9 Spinal meningeal cysts have been regarded as the source of CSF outflow,3,22 and Schievink and his group suggested that these are, together with imaging and clinical features, a diagnostic criterion for spinal CSF leaks.20 Spinal CSF leaks in combination with anticoagulation treatment were recently suggested as a cause of CSDH, particularly in elderly patients.22 Interestingly, all 3 reported patients in J Neurosurg / Volume 121 / December 2014
that study had spinal cervicothoracic cysts (33.3%, comparable to the current cohort) but none had a proven CSF leak. To take this into consideration, our classification differentiates proven CSF outflow (categories 4 and 5) from the mere presence of spinal cysts (categories 1 and 2). Patient Profile and Risk Factors
The reported incidence of recurrent CSDH varies from 3% to 39%,4,17,24,27,30 with reports of second or multiple recurrence ranging from 2% to 9%.4,7,18,24–26 Our series showed a high rate of first (33.3%), second (11.1%), and third recurrences (3.7%), but no reference group of nongeriatric patients with CSDH is available for comparison. Multiple recurrences occurred significantly more often in patients with proven CSF leaks (categories 4 and 5) as compared with patients without such leaks. What makes the multiple recurrences noteworthy is the fact that in 4 of the 7 current cases, the recurrence of the CSDH only stopped after the CSF leak was sealed. This finding supports the hypothesis of a causal relationship between CSF outflow and CSDH. Other researchers have reported that patients with recurrent CSDH have a lower incidence of head trauma than patients without recurrent CSDH,13 as was also the case in the current series. In addition, bilateral CSDH and young age appear to involve a higher risk of recurrence.8,12,26,27 The current report describes an association of bilateral hematomas, orthostatic headaches, and multiple CSDH recurrences with a proven CSF leak. Based on our current observations, we suggest that criteria such as young age, early or multiple CSDH recurrence, continuing or unexplained clinical deterioration, orthostatic headaches, or the presence of bilateral CSDH should initiate the search for spinal CSF leaks.
Conclusions
Spinal imaging results are challenging the pathogenetic concept of CSDH in nongeriatric patients. Our results support the hypothesis that spinal CSF leakage is a frequent cause of CSDH in young patients, which could influence the diagnostic and therapeutic management of these patients. Potential spinal CSF leaks should be methodically investigated in young patients and in patients with certain risk factors. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Beck, Raabe. Acquisition of data: Beck, Gralla, Fung, Ulrich, Schucht, Fichtner, Andereggen, Gutbrod, Z’Graggen, Reinert, Ozdoba, Raabe. Analysis and interpretation of data: Beck, Gralla, Fung, Ulrich, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, Raabe. Drafting the article: Beck, Fung, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, Raabe. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Beck. Statistical analysis: Hüsler, Beck. Study supervision: Beck.
1385
J. Beck et al. TABLE 3: Neuroradiological workup of all patients studied* Cranial
Spinal
Case No.
CT
MRI
MRI w/o Contrast
1† 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
11 6 5 1 4 4 4
2 1
2 3 1
4 1 2 3 1 1 2 1 8 1 1 3 1 1 6 6 1 2
1 6
1 3 1 3
1 1 5 1 1 1 2 2 1 1
MRI & CM (iv) MRI & CM (it) Myelography 1
3 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 2 1 1 2 1 1
2
Postmyelography CT
3 1
2 1
1 1
3
2
1
1
1
1
1
1
1
1
1
1 1 2
1 1 2
2
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 1. Amorim JA, Remígio DS, Damázio Filho O, de Barros MA, Carvalho VN, Valença MM: Intracranial subdural hematoma post-spinal anesthesia: report of two cases and review of 33 cases in the literature. Rev Bras Anestesiol 60:620–629, 344–349, 2010 2. Beier AD, Soo TM, Claybrooks R: Subdural hematoma after microdiscectomy: a case report and review of the literature. Spine J 9:e9–e12, 2009 3. Cohen-Gadol AA, Mokri B, Piepgras DG, Meyer FB, Atkinson JL: Surgical anatomy of dural defects in spontaneous spinal cerebrospinal fluid leaks. Neurosurgery 58 (4 Suppl 2): ONS-238–ONS-245, 2006 4. Forster MT, Mathé AK, Senft C, Scharrer I, Seifert V, Gerlach R: The influence of preoperative anticoagulation on outcome and quality of life after surgical treatment of chronic subdural hematoma. J Clin Neurosci 17:975–979, 2010 5. Gardner WJ: Traumatic subdural hematoma: with particular reference to the latent interval. Arch Neurol Psychiatry 27:847–858, 1932 6. Gaucher DJ Jr, Perez JA Jr: Subdural hematoma following lumbar puncture. Arch Intern Med 162:1904–1905, 2002 7. Ishihara H, Ishihara S, Kohyama S, Yamane F, Ogawa M, Sato
1386
A, et al: Experience in endovascular treatment of recurrent chronic subdural hematoma. Interv Neuroradiol 13 Suppl 1: 141–144, 2007 8. Ko BS, Lee JK, Seo BR, Moon SJ, Kim JH, Kim SH: Clinical analysis of risk factors related to recurrent chronic subdural hematoma. J Korean Neurosurg Soc 43:11–15, 2008 9. Langdown AJ, Grundy JR, Birch NC: The clinical relevance of Tarlov cysts. J Spinal Disord Tech 18:29–33, 2005 10. Markwalder TM: Chronic subdural hematomas: a review. J Neurosurg 54:637–645, 1981 11. Mori K, Maeda M: Risk factors for the occurrence of chronic subdural haematomas after neurosurgical procedures. Acta Neurochir (Wien) 145:533–540, 2003 12. Mori K, Maeda M: Surgical treatment of chronic subdural hematoma in 500 consecutive cases: clinical characteristics, surgical outcome, complications, and recurrence rate. Neurol Med Chir (Tokyo) 41:371–381, 2001 13. Oishi M, Toyama M, Tamatani S, Kitazawa T, Saito M: Clinical factors of recurrent chronic subdural hematoma. Neurol Med Chir (Tokyo) 41:382–386, 2001 14. Putnam TJ, Cushing H: Chronic subdural hematoma: its pathology, its relation to pachymeningitis hemorrhagica and its surgical treatment. Arch Surg 11:329–393, 1925
J Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma 15. Samuelson S, Long DM, Chou SN: Subdural hematoma as a complication of shunting procedures for normal pressure hydrocephalus. J Neurosurg 37:548–551, 1972 16. Santarius T, Hutchinson PJ: Chronic subdural haematoma: time to rationalize treatment? Br J Neurosurg 18:328–332, 2004 17. Santarius T, Kirkpatrick PJ, Ganesan D, Chia HL, Jalloh I, Smielewski P, et al: Use of drains versus no drains after burr-hole evacuation of chronic subdural haematoma: a randomised controlled trial. Lancet 374:1067–1073, 2009 18. Santarius T, Kirkpatrick PJ, Kolias AG, Hutchinson PJ: Working toward rational and evidence-based treatment of chronic subdural hematoma. Clin Neurosurg 57:112–122, 2010 19. Schievink WI: Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. JAMA 295:2286–2296, 2006 20. Schievink WI, Maya MM, Louy C, Moser FG, Tourje J: Diagnostic criteria for spontaneous spinal CSF leaks and intracranial hypotension. AJNR Am J Neuroradiol 29:853–856, 2008 21. Schievink WI, Maya MM, Moser FG, Tourje J: Spectrum of subdural fluid collections in spontaneous intracranial hypotension. 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 hematomas in elderly patients on anticoagulation. Report of 3 cases. J Neurosurg 112:295–299, 2010 23. Sencakova D, Mokri B, McClelland RL: The efficacy of epidural blood patch in spontaneous CSF leaks. Neurology 57:1921–1923, 2001 24. Smely C, Madlinger A, Scheremet R: Chronic subdural haematoma–a comparison of two different treatment modalities. Acta Neurochir (Wien) 139:818–826, 1997
J Neurosurg / Volume 121 / December 2014
25. Stroobandt G, Fransen P, Thauvoy C, Menard E: Pathogenetic factors in chronic subdural haematoma and causes of recurrence after drainage. Acta Neurochir (Wien) 137:6–14, 1995 26. Tahsim-Oglou Y, Beseoglu K, Hänggi D, Stummer W, Steiger HJ: Factors predicting recurrence of chronic subdural haematoma: the influence of intraoperative irrigation and low-molecular-weight heparin thromboprophylaxis. Acta Neurochir (Wien) 154:1063–1068, 2012 27. Torihashi K, Sadamasa N, Yoshida K, Narumi O, Chin M, Yamagata S: Independent predictors for recurrence of chronic subdural hematoma: a review of 343 consecutive surgical cases. Neurosurgery 63:1125–1129, 2008 28. Trotter W: Chronic subdural haemorrhage of traumatic origin, and its relation to pachymeningitis haemorrhagica interna. Br J Surg 2:271–291, 1914 29. Virchow R: Das haematom der dura mater. Verh Phys Med Ges Würzburg 7:134–142, 1857 30. Weigel R, Schmiedek P, Krauss JK: Outcome of contemporary surgery for chronic subdural haematoma: evidence based review. J Neurol Neurosurg Psychiatry 74:937–943, 2003
Manuscript submitted September 26, 2013. Accepted June 9, 2014. Please include this information when citing this paper: published online July 18, 2014; DOI: 10.3171/2014.6.JNS14550. Address correspondence to: Jürgen Beck, M.D., Department of Neurosurgery, Bern University Hospital, Freiburgstrasse 10, Bern 3010, Switzerland. email: [email protected].
1387
Spinal cerebrospinal fluid 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 subdural hematoma (CSDH) in nongeriatric patients (≤ 60 years old) often remains unclear. The primary objective of this study was to identify spinal CSF leaks in young patients, after formulating 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 patient 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, myelography/fluoroscopy, and postmyelography CT. Spinal pathologies 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. Hematomas 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
C
hronic subdural hematomas (CSDHs) are generally regarded to be consequences of head trauma. The most commonly accepted pathophysiological explanation of CSDH is that mild head trauma leads to tearing of bridging veins with subsequent bleeding between the inner layer of the dura mater and the arachnoid, or between the outer and inner layers of the dura, thus creating the hematoma.5,14,28,29 Further and chronic en-
Abbreviations used in this paper: CSDH = chronic subdural hematoma; mRS = modified Rankin Scale; SIH = spontaneous intracranial hypotension; SPACE = sampling perfection with application of optimized contrasts using different flip angle evolutions.
1380
largement of the subdural hematoma is believed to be due to microbleedings from fragile walls of sprouting vessels, accumulation of fluid in this newly created space following an osmotic gradient created by blood degradation products, or a combination of these events.5,10 In older patients, mechanical forces related to brain atrophy enable formation and growth of CSDH. A history of mild head trauma in many older patients is suggestive of trauma as a causative agent. In younger patients, brain atrophy is generally not so prominent and a history of head trauma is often not present. Triggered by our experience with a young man with recurrent CSDH whose recurrence only stopped after J Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma a spinal CSF leak was sealed, we started to include the search for spinal CSF leakage into a prospective protocol for all consecutive nongeriatric CSDH patients. Our primary objectives were to identify spinal CSF leaks and spinal meningeal cysts in our cohort with the hypothesis that spinal CSF leaks are causally related to CSDH. In this paper we report our findings on the frequency of this condition. Study Design
Methods
This prospective cohort study was approved by the local ethics committee of Bern University Hospital. All consecutive patients ≤ 60 years old with CSDH verified by CT or MRI who underwent operations between September 1, 2009, and April 12, 2011, were eligible for the study. After formulation of our hypothesis that spinal CSF egress may be a cause of CSDH in younger patients, we distributed the study protocol among all staff of the Departments of Neurosurgery and Neuroradiology at several meetings and conferences that took place beginning 4 months prior to study initiation. Beginning September 1, 2009, all staff from both departments began to screen for patients who underwent operations for CSDH. Furthermore, we used the daily interdisciplinary neurosurgical-neuroradiological conferences, where all elective and emergency cases were presented, as a second screening platform. Patients operated on for a CSDH who were 60 years of age or younger were eligible for spinal imaging workup. We excluded patients who died early (< 48 hours) due to epilepsy, who presented with recurrent SDH, who were not compliant during spinal imaging, who had recent brain surgery (< 6 months), or who had a shunt system for CSF diversion. As a next step, the computerized and written log books of all operating theaters were screened for patients with CSDH who had not been included in the study. A flow diagram of the number of patients at each stage of the study is provided in Fig. 1. An idealized diagram for how imaging modalities can be used, in order of increasing invasiveness, is shown in Fig. 2.
Spinal Imaging
For MRI, a Magnetom Avanto 1.5-T or Magnetom Verio 3-T (Siemens) machine was used. The following sequences were employed: T1-weighted, T2-weighted, magnetization-prepared rapid acquisition gradient echo (MPRAGE), and sampling perfection with application of optimized contrasts using different flip angle evolutions (SPACE) with multiplanar reconstructions. In cases of intrathecal contrast application, gadopentetate dimeglumine (Magnevist, Bayer) at a standard dose of 0.5 ml dissolved in 9.5 ml CSF was administered via a lumbar puncture at the L3–5 level. Myelography was performed using a Siemens Artis Zee Monoplan System (Siemens). A maximum dosage of 30 ml of iodine contrast material was administered after lumbar puncture using an atraumatic puncture kit. Postmyelography CT was performed on a LightSpeed Ultra scanner (GE) immediately after myelography and again 4 hours later in selected cases.
J Neurosurg / Volume 121 / December 2014
Stepwise Protocol of the Search
After a review of existing imaging results or suggestions of CSF leaks, 3-T SPACE imaging of the complete spinal axis was conducted. Next, 0.5 ml of contrast medium (Magnevist) was injected intrathecally, and T1weighted MRI of the spinal axis was completed. Myelography was then performed, and finally, postmyelography CT was completed. If a CSF leak was identified, the imaging algorithm was stopped with the least invasive study. Example images from a patient are provided in Fig. 3; following the diagnosis of a CSDH (Fig. 3A), spinal imaging without contrast demonstrated extrathecal fluid accumulation (Fig. 3B). The CSF fistula was then proven by fluoroscopy (Fig. 3C) and postmyelography CT (Fig. 3D).
Classification of Spinal Imaging Findings
We classified the findings of the spinal imaging according to the likelihood of pathological CSF outflow from the intrathecal to the extrathecal space on a scale from 0 to 5 in an ascending manner (Table 1): no pathological findings (category 0), detection of spinal meningeal cysts in the sacral region (category 1), detection of spinal meningeal cysts in the cervicothoracic region (category 2), detection of extrathecal fluid accumulation on spinal MRI without intrathecal contrast application (category 3), detection of extrathecal contrast (Magnevist/iopamidol 300 [Iopamiro, Bracco]) on spinal MRI or postmyelography CT after intrathecal administration of contrast (category 4), and direct visualization of outflow of intrathecally administered contrast (iodine) from the intrathecal to the extrathecal space under fluoroscopy (category 5). The proof of an existing CSF leak (categories 4 and 5) was defined as the primary outcome.
Surgery for CSDH
Whenever possible, the patients were operated on under local anesthesia via the 2-bur-hole technique;18 drain insertion was performed if it was considered safe, after careful rinsing with large amounts of normal saline solution.26 For nonliquified space-occupying blood clots, open craniotomy under general anesthesia was the first procedure or a backup procedure. Postsurgical Care
All patients were admitted to the intermediate care unit or the intensive care unit after surgery. The drains were left in place, restricted to 10 ml/hr and 120 ml/day, for a maximum of 48 hours. Mobilization of patients was started on the day of surgery.
Treatment of Spinal CSF Leak
In cases in which a spinal CSF leak was detected, the following treatment algorithm was used: 1) bed rest for as long as 3 days; 2) undirected blood patch at the L3–4 level (with an option to repeat the treatment if necessary); 3) directed blood patch under fluoroscopy or CT guidance (with an option to repeat the treatment if necessary); and 4) microsurgical exposure and closure of the CSF leak. For treatment options 3 and 4, the exact site of the 1381
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 symptoms, the escalation from an undirected treatment (1 or 2) to a directed treatment (3 or 4) was accelerated, including emergency neuroradiological workup to find the spinal CSF leak and to enable directed treatment. Patient Outcome
Outcome was assessed after 6 months during regularly scheduled follow-up visits in the outpatient clinics. Patient outcome was based on the modified Rankin scale (mRS) score, working status, major neurological deficits, dependency, and living situation.
Statistical Analysis
For descriptive statistics, data are expressed for continuous variables as means, standard deviations, and medians. Associations between categorical variables such as multiple recurrence, bilateral hematoma, head trauma, and CSF leak were calculated using a 2-sided Fisher exact test. Multiple logistic regression analysis was applied to
1382
consider the event of a CSF leak or of a recurrent CSDH in relation to explanatory variables by forward selection. Because the sample size and the number of events were small, only a few explanatory variables and no interactions could be investigated together. The statistical analysis was conducted using SAS Enterprise Guide (version 4.3). A p value < 0.05 was considered indicative of a significant result.
Results
Patient Characteristics and Clinical Findings
Nineteen men and 8 women, with a mean age of 49.6 ± 9.2 years (median 52.0 years), were included in the study (Table 2). Patient 1, who presented in November 2008 and whose clinical course led to the formation of our hypothesis, was included in the final analysis. The leading symptoms included headaches, neurological deficits, cranial nerve deficits, decreased level of consciousness, and seizures (Table 2). Fifteen patients (55.6%) had J Neurosurg / Volume 121 / December 2014
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 invasiveness, during the search for spinal CSF leaks and spinal meningeal cysts.
a history of head trauma, 1 patient was taking oral anticoagulants, 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 headache, for 12 of these patients the headache was not orthostatic; overall, 24 patients had no orthostatic symptoms. Characteristics and Treatment of CSDH
Chronic SDHs were unilateral in 20 patients (74.1%) and bilateral in 7 patients (25.9%; Table 2). Initial treatment was via 2-bur-hole trephination (25 cases), via 2-bur-hole trephination with conversion to craniotomy (1 case), and craniotomy (1 case). One patient showed a yellowish 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. All patients who underwent operations for multiple recurrences had a proven CSF leak (3 of 7 with a leak vs 0 of 20 without a leak; p = 0.012).
Results of Spinal Imaging
According to our classification system (Table 1), 5 patients had a spinal CSF leak proven by direct visualization of contrast material outflow (category 5) and 2 patients had a spinal CSF leak proven by extrathecal contrast material accumulation after intrathecal application (category 4), resulting in 25.9% of patients with a proven CSF leak as the primary outcome. Spinal cervicothoracic meningeal cysts (category 2) were found in 9 patients (33.3%), including patients with multiple cysts. Cysts in the sacral region only (category 1) were detected in 3 cases (11.1%). No pathology (class 0) was found in 8 patients (29.6%). The specific imaging findings and imaging modalities used per patient are listed in Tables 2 and
J Neurosurg / Volume 121 / December 2014
Fig. 3. Images from a patient with CSDH. Extrathecal fluid accumulation (arrowheads) was demonstrated by axial T2-weighted (A) and sagittal SPACE 3-T MRI (B). The spinal CSF leak was proven by direct visualization of the passage of contrast material from the intra- to the extrathecal space by anterior-posterior projection fluoroscopy (C). The leak is located in the thoracic spine (arrow) at the T10–11 level (C). The extradural accumulation of contrast material (arrow) is also demonstrated by postmyelography axial CT (D).
3. Three (43%) of the 7 patients with proven CSF leaks had a history of head trauma as compared with 15 (75%) of 20 patients without a proven leak (p = 0.175). Imaging comprised 46 spinal MRI procedures, 15 myelographies, 13 postmyelography CT scans, 1 cisternography, and 1 nuclear medicine examination. Treatment of Spinal CSF Leaks
Eight patients, including 7 with proven CSF leaks and 1 without spinal pathology but with recurrent CSDH, were treated with 13 epidural blood-patching procedures. Three patients underwent 4 spinal microsurgical operations for closure of the leak. Two of these patients needed microsurgery after repeated blood patching.
Patient Outcome
Outcome according to the mRS score at 6 months was a score of 0 for 15 patients, a score of 1 for 9 patients, and a score of 2 for 3 patients (Table 2). One of the 1383
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 1 2 3
no suggestion of CSF loss spinal meningeal cyst, sacral region spinal meningeal cyst, cervicothoracic region extrathecal fluid detection w/o intrathecal contrast extrathecal contrast accumulation after intra thecal contrast application direct visualization of contrast passage from intrathecal to extrathecal space
MRI, myelography, CT, cisternography MRI, myelography, CT MRI, myelography, CT MRI (T2 SPACE)
Gd, iodine, radionuclide Gd, iodine Gd, iodine —
no CSF fistula CSF fistula unlikely CSF fistula likely CSF fistula very likely
CT/MRI
Gd, iodine
CSF fistula proven
fluoroscopy (myelography)
iodine
CSF fistula proven
4 5
patients with an mRS score of 2 had suffered an ischemic stroke 8 years earlier with persistent hemiparesis. All 27 patients lived independently at home and 25 (92.6%) returned to their previous work status. All patients were available for follow-up. Multiple Logistic Regression
The dependence of a CSF leak event was considered with respect to the following variables: sex, trauma, orthostatic headache, oral anticoagulation/aspirin, and bilateral CSDH. Only the lack of orthostatic headache (OR = 0.030, p = 0.030) and the presence of bilateral CSDH (OR = 21.33, p = 0.017) had significant impacts on the occurrence of a CSF leak. The model assumptions and fit were checked (area under the curve = 0.836). The same model assumptions were used for the recurrence event with the same variables and also CSF leak. In this case, the presence of bilateral CSDH (OR = 17.8, p = 0.03) and lack of trauma (OR = 0.059, p = 0.023) were significant factors. The model assumptions and fit were also checked (area under the curve = 0.889). Because of the small sample size with not many events in the explanatory variables, these results should be considered with caution. Multiple recurrences were also considered, but because there were only 3 such events among the 27 patients, a multiple logistic regression did not produce an acceptable result. However, this event was dependent on CSF leak (p = 0.012, Fisher exact test) and trauma (p = 0.029, Fisher exact test).
Discussion
With this prospective search for spinal pathologies in a consecutive neurosurgical series of nongeriatric patients with CSDH, we found an unexpectedly high rate of spinal CSF leaks (25.9%). Although spinal CSF leaks have been described as a potential cause of CSDH in rare cases, spinal CSF leakage has not been regarded as a frequent diagnosis and is therefore not included in a standard workup. Our findings suggest that investigation of spinal CSF leakage should be a mandatory part of a dedicated workup. In addition to the 25.9% of patients with directly proven CSF leaks, 33.3% of patients were found to have spinal cervicothoracic meningeal cysts. These cysts have been proposed as the cause or likely cause for spinal CSF outflow.20 1384
Cerebrospinal Fluid Leakage as a Cause of CSDHs
Spinal outflow of CSF may lead to sagging of the brain,19 which may tear congested bridging veins and thus enable formation of a CSDH. A causal relationship between CSF outflow and CSDH is supported by the formation of CSDH as a complication of lumbar puncture,6 spontaneous intracranial hypotension (SIH),19,21 neurosurgical shunting procedures,11,15 spinal anesthesia,1 spinal microdiscectomy with a dural tear,2 or any neurosurgical procedure associated with excessive CSF outflow.11 Until now, the clinical relevance and incidence of spinal CSF outflow for nongeriatric patients with CSDH has not been documented and is very likely underestimated. The prevailing diagnostics focus on iatrogenic mechanisms (such as spinal tap or CSF shunts) leading to CSF loss, whereas an occult, idiopathic, or acquired form of spinal CSF leakage that leads to CSDH is currently not regarded as a relevant differential diagnosis and is thus not routinely investigated. To our knowledge, a systematic study of spinal CSF leakage in patients with CSDH has never been performed or reported. Because SIH is a possible diagnosis leading to CSDH, we specifically inquired about orthostatic symptoms at admission, and again during outpatient follow-up visits, including questioning of relatives. This led to identification of 3 patients (11%) with some orthostatic component. The majority of subdural fluid collections in SIH reported 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 hematomas.3,23 Both disorders, CSDH and SIH, occur with the same incidence of about 5 per 100,000 people.16,19 Thus, it is extremely unlikely that a simultaneous occurrence of these disorders would happen by chance in 25.9% of a group of patients. The presence of a spinal CSF leak might lead to different clinical phenotypes (SIH or CSDH), which raises pathophysiological considerations. Whether there are different kinds of spinal CSF leaks with specific anatomical or functional (flow) characteristics, with some leading to SIH and others to CSDH, or whether the clinical manifestations change over time or represent a spectrum of a disease, is unknown. Relevance of Spinal Meningeal Cysts
The incidence and natural history of spinal meningeJ Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma TABLE 2: Patient characteristics Variable age (yrs) mean ± SD median males (n = 19) mean age ± SD (yrs) median age females (n = 8) mean age ± SD (yrs) median age leading symptoms (multiple) headache somnolence coma seizure mydriasis hemiparesis ataxia apraxia perseveration paresthesia cranial nerve deficit cognitive decline delirium location of CSDH rt lt bilat category of spinal CSF leak 0 1 2 3 4 5 warfarin therapy aspirin mRS score at 6 mos 0 1 2
Value 49.6 ± 9.2 52.0 48.1 ± 9.7 51.0 53.1 ± 7.1 55.5 15 4 1 2 1 3 2 2 1 1 2 1 1 11 9 7 8 3 9 0 2 5 1 4 15 9 3
al cysts are not exactly known. Small perineural cysts in the region of the sacrum appear to have a benign course.9 Spinal meningeal cysts have been regarded as the source of CSF outflow,3,22 and Schievink and his group suggested that these are, together with imaging and clinical features, a diagnostic criterion for spinal CSF leaks.20 Spinal CSF leaks in combination with anticoagulation treatment were recently suggested as a cause of CSDH, particularly in elderly patients.22 Interestingly, all 3 reported patients in J Neurosurg / Volume 121 / December 2014
that study had spinal cervicothoracic cysts (33.3%, comparable to the current cohort) but none had a proven CSF leak. To take this into consideration, our classification differentiates proven CSF outflow (categories 4 and 5) from the mere presence of spinal cysts (categories 1 and 2). Patient Profile and Risk Factors
The reported incidence of recurrent CSDH varies from 3% to 39%,4,17,24,27,30 with reports of second or multiple recurrence ranging from 2% to 9%.4,7,18,24–26 Our series showed a high rate of first (33.3%), second (11.1%), and third recurrences (3.7%), but no reference group of nongeriatric patients with CSDH is available for comparison. Multiple recurrences occurred significantly more often in patients with proven CSF leaks (categories 4 and 5) as compared with patients without such leaks. What makes the multiple recurrences noteworthy is the fact that in 4 of the 7 current cases, the recurrence of the CSDH only stopped after the CSF leak was sealed. This finding supports the hypothesis of a causal relationship between CSF outflow and CSDH. Other researchers have reported that patients with recurrent CSDH have a lower incidence of head trauma than patients without recurrent CSDH,13 as was also the case in the current series. In addition, bilateral CSDH and young age appear to involve a higher risk of recurrence.8,12,26,27 The current report describes an association of bilateral hematomas, orthostatic headaches, and multiple CSDH recurrences with a proven CSF leak. Based on our current observations, we suggest that criteria such as young age, early or multiple CSDH recurrence, continuing or unexplained clinical deterioration, orthostatic headaches, or the presence of bilateral CSDH should initiate the search for spinal CSF leaks.
Conclusions
Spinal imaging results are challenging the pathogenetic concept of CSDH in nongeriatric patients. Our results support the hypothesis that spinal CSF leakage is a frequent cause of CSDH in young patients, which could influence the diagnostic and therapeutic management of these patients. Potential spinal CSF leaks should be methodically investigated in young patients and in patients with certain risk factors. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Beck, Raabe. Acquisition of data: Beck, Gralla, Fung, Ulrich, Schucht, Fichtner, Andereggen, Gutbrod, Z’Graggen, Reinert, Ozdoba, Raabe. Analysis and interpretation of data: Beck, Gralla, Fung, Ulrich, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, Raabe. Drafting the article: Beck, Fung, Gosau, Hattingen, Z’Graggen, Hüsler, Ozdoba, Raabe. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Beck. Statistical analysis: Hüsler, Beck. Study supervision: Beck.
1385
J. Beck et al. TABLE 3: Neuroradiological workup of all patients studied* Cranial
Spinal
Case No.
CT
MRI
MRI w/o Contrast
1† 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
11 6 5 1 4 4 4
2 1
2 3 1
4 1 2 3 1 1 2 1 8 1 1 3 1 1 6 6 1 2
1 6
1 3 1 3
1 1 5 1 1 1 2 2 1 1
MRI & CM (iv) MRI & CM (it) Myelography 1
3 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 2 1 1 2 1 1
2
Postmyelography CT
3 1
2 1
1 1
3
2
1
1
1
1
1
1
1
1
1
1 1 2
1 1 2
2
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 1. Amorim JA, Remígio DS, Damázio Filho O, de Barros MA, Carvalho VN, Valença MM: Intracranial subdural hematoma post-spinal anesthesia: report of two cases and review of 33 cases in the literature. Rev Bras Anestesiol 60:620–629, 344–349, 2010 2. Beier AD, Soo TM, Claybrooks R: Subdural hematoma after microdiscectomy: a case report and review of the literature. Spine J 9:e9–e12, 2009 3. Cohen-Gadol AA, Mokri B, Piepgras DG, Meyer FB, Atkinson JL: Surgical anatomy of dural defects in spontaneous spinal cerebrospinal fluid leaks. Neurosurgery 58 (4 Suppl 2): ONS-238–ONS-245, 2006 4. Forster MT, Mathé AK, Senft C, Scharrer I, Seifert V, Gerlach R: The influence of preoperative anticoagulation on outcome and quality of life after surgical treatment of chronic subdural hematoma. J Clin Neurosci 17:975–979, 2010 5. Gardner WJ: Traumatic subdural hematoma: with particular reference to the latent interval. Arch Neurol Psychiatry 27:847–858, 1932 6. Gaucher DJ Jr, Perez JA Jr: Subdural hematoma following lumbar puncture. Arch Intern Med 162:1904–1905, 2002 7. Ishihara H, Ishihara S, Kohyama S, Yamane F, Ogawa M, Sato
1386
A, et al: Experience in endovascular treatment of recurrent chronic subdural hematoma. Interv Neuroradiol 13 Suppl 1: 141–144, 2007 8. Ko BS, Lee JK, Seo BR, Moon SJ, Kim JH, Kim SH: Clinical analysis of risk factors related to recurrent chronic subdural hematoma. J Korean Neurosurg Soc 43:11–15, 2008 9. Langdown AJ, Grundy JR, Birch NC: The clinical relevance of Tarlov cysts. J Spinal Disord Tech 18:29–33, 2005 10. Markwalder TM: Chronic subdural hematomas: a review. J Neurosurg 54:637–645, 1981 11. Mori K, Maeda M: Risk factors for the occurrence of chronic subdural haematomas after neurosurgical procedures. Acta Neurochir (Wien) 145:533–540, 2003 12. Mori K, Maeda M: Surgical treatment of chronic subdural hematoma in 500 consecutive cases: clinical characteristics, surgical outcome, complications, and recurrence rate. Neurol Med Chir (Tokyo) 41:371–381, 2001 13. Oishi M, Toyama M, Tamatani S, Kitazawa T, Saito M: Clinical factors of recurrent chronic subdural hematoma. Neurol Med Chir (Tokyo) 41:382–386, 2001 14. Putnam TJ, Cushing H: Chronic subdural hematoma: its pathology, its relation to pachymeningitis hemorrhagica and its surgical treatment. Arch Surg 11:329–393, 1925
J Neurosurg / Volume 121 / December 2014
Spinal CSF leakage and chronic subdural hematoma 15. Samuelson S, Long DM, Chou SN: Subdural hematoma as a complication of shunting procedures for normal pressure hydrocephalus. J Neurosurg 37:548–551, 1972 16. Santarius T, Hutchinson PJ: Chronic subdural haematoma: time to rationalize treatment? Br J Neurosurg 18:328–332, 2004 17. Santarius T, Kirkpatrick PJ, Ganesan D, Chia HL, Jalloh I, Smielewski P, et al: Use of drains versus no drains after burr-hole evacuation of chronic subdural haematoma: a randomised controlled trial. Lancet 374:1067–1073, 2009 18. Santarius T, Kirkpatrick PJ, Kolias AG, Hutchinson PJ: Working toward rational and evidence-based treatment of chronic subdural hematoma. Clin Neurosurg 57:112–122, 2010 19. Schievink WI: Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. JAMA 295:2286–2296, 2006 20. Schievink WI, Maya MM, Louy C, Moser FG, Tourje J: Diagnostic criteria for spontaneous spinal CSF leaks and intracranial hypotension. AJNR Am J Neuroradiol 29:853–856, 2008 21. Schievink WI, Maya MM, Moser FG, Tourje J: Spectrum of subdural fluid collections in spontaneous intracranial hypotension. 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 hematomas in elderly patients on anticoagulation. Report of 3 cases. J Neurosurg 112:295–299, 2010 23. Sencakova D, Mokri B, McClelland RL: The efficacy of epidural blood patch in spontaneous CSF leaks. Neurology 57:1921–1923, 2001 24. Smely C, Madlinger A, Scheremet R: Chronic subdural haematoma–a comparison of two different treatment modalities. Acta Neurochir (Wien) 139:818–826, 1997
J Neurosurg / Volume 121 / December 2014
25. Stroobandt G, Fransen P, Thauvoy C, Menard E: Pathogenetic factors in chronic subdural haematoma and causes of recurrence after drainage. Acta Neurochir (Wien) 137:6–14, 1995 26. Tahsim-Oglou Y, Beseoglu K, Hänggi D, Stummer W, Steiger HJ: Factors predicting recurrence of chronic subdural haematoma: the influence of intraoperative irrigation and low-molecular-weight heparin thromboprophylaxis. Acta Neurochir (Wien) 154:1063–1068, 2012 27. Torihashi K, Sadamasa N, Yoshida K, Narumi O, Chin M, Yamagata S: Independent predictors for recurrence of chronic subdural hematoma: a review of 343 consecutive surgical cases. Neurosurgery 63:1125–1129, 2008 28. Trotter W: Chronic subdural haemorrhage of traumatic origin, and its relation to pachymeningitis haemorrhagica interna. Br J Surg 2:271–291, 1914 29. Virchow R: Das haematom der dura mater. Verh Phys Med Ges Würzburg 7:134–142, 1857 30. Weigel R, Schmiedek P, Krauss JK: Outcome of contemporary surgery for chronic subdural haematoma: evidence based review. J Neurol Neurosurg Psychiatry 74:937–943, 2003
Manuscript submitted September 26, 2013. Accepted June 9, 2014. Please include this information when citing this paper: published online July 18, 2014; DOI: 10.3171/2014.6.JNS14550. Address correspondence to: Jürgen Beck, M.D., Department of Neurosurgery, Bern University Hospital, Freiburgstrasse 10, Bern 3010, Switzerland. email: [email protected].
1387
