- Email: [email protected]
Chronic subdural hematomas—causes of morbidity and mortality
Surgical Neurology 67 (2007) 367 – 373 www.surgicalneurology-online.com
Vascular
Chronic subdural hematomas—causes of morbidity and mortalityB Ramnarayan Ramachandran, MCha,4, Thimmappa Hegde, MChb a
Department of Neurosurgery, Dr SM CSI Medical College Hospital, Karakonam 695504, India b Narayana Institute of Neurosciences, Bangalore 560099, India Received 27 February 2006; accepted 28 July 2006
Abstract
Objective: Chronic subdural hematoma is a very common condition seen usually in the later stages of life. Treatment, although apparently simple, is associated with some morbidity and mortality with a potential for recurrence. This is especially important as the average life span of humanity increases all over the world. Methods: A retrospective analysis of 647 cases of chronic subdural hematomas treated in NIMHANS was done. The aim of this study was to determine the factors responsible for the morbidity, mortality, and recurrence in patients with chronic subdural hematomas. Results: The patients were mostly men in the sixth and seventh decade of life and presented with headache, cognitive decline, or focal deficits. CT scan was done in all cases. 94% of the patients underwent evacuation of the hematoma. There was a mortality rate of 5%, and there was a recurrence of 21%. Statistical analysis for factors both for mortality and morbidity were done. Conclusions: It was seen that statistically significant factors determining mortality were age, GCS at presentation, and associated illnesses like cardiac and renal failure. The statistically significant factors for recurrence are the presence of a thick subdural membrane visualized during surgery and brain, remaining at a depth at the end of evacuation of hematoma. Use of a subdural drain significantly reduced recurrences. The chronicity of the hematoma was not a factor determining mortality or morbidity. Drains were seen to significantly reduce recurrence in younger patients with better GCS and when the subdural membrane is seen at surgery and the brain remains at a depth at the end of hematoma evacuation. D 2007 Elsevier Inc. All rights reserved.
Keywords:
Chronic subdural hematoma; Elderly; Morbidity; Mortality
Abbreviations: Contrast CT, Contrast-enhanced computed tomography; CT, Computed tomography; GCS, Glasgow coma score; GOS, Glasgow outcome score; NIMHANS, National Institute of Mental Health and NeuroSciences, Government of India. B This study was done at the Department of Neurosurgery, The National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, India. There were no financial relationships involved for this study and there is no conflict of interests. This is purely a retrospective case note study and does not involve any human beings or animals directly. The corresponding author has the right to grant on behalf of all authors and, does grant on behalf of all authors, an exclusive license (or nonexclusive for government employees) on a worldwide basis to permit this article (if accepted) to be published in Surgical Neurology and any other products and sublicenses that use and exploit all subsidiary rights, as set out in our license. 4 Corresponding author. Tel.: +91 479 2469469; fax; +91 479 2468709. E-mail addresses: [email protected] (R. Ramachandran)8 [email protected] (T. Hegde). 0090-3019/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.07.022
1. Introduction Head injury sequelae with posttraumatic hematomas of various types form a sizeable percentage of admissions of any neurosurgical center [6]. Scotti [30] has defined subdural hematoma radiologically based on the density in the CT scan and defined chronic subdural hematomas as hypodense to parenchyma and presenting 21 days after trauma. It is a problem related to age. Various surgical treatments described like burr holes and twist drills to craniotomies, are not always effective in managing this condition. Despite the apparent simplicity of the management of this condition, mortality and morbidity does occur, as also recurrence. The aim of this study was to analyze the causes of morbidity, mortality, and recurrence in chronic subdural hematomas.
368
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
2. Material and methods A retrospective analysis of the records of all patients of chronic subdural hematomas admitted to NIMHANS over a 10-year period (1989-1999) was done. Cases were identified by the coding in the medical records. The details recorded were the age and sex of the patients; any associated comorbidity such as hypertension, cardiac or renal disease, and clinical features at presentation, including the postresuscitation GCS; and the radiology and operative procedure. Outcome was assessed on the seventh postoperative day. A total of 665 patients were treated for chronic subdural hematoma during this time period. However, only 647 case notes were available for analysis. Most of the patients were referred from local hospitals on clinical suspicion. They were admitted, stabilized, clinically assessed, and then sent for plain and contrast CT scan for diagnosis. Patients with chronic subdural hematoma were then managed based on clinical and radiological findings and the operative findings. They were observed after definitive procedure for 24 hours and then referred back to the secondary level hospitals. These patients were then called back on the 7th postoperative day for suture removal (and also clinical assessment). The patients were kept for follow-up for at least up to 1 year. v 2 and proportion tests were done appropriately to find out the statistical significance. As the patients were referred from other smaller hospitals on clinical suspicion, a plain and contrast CT was done in all cases. There were patients with other radiological findings such as tumors and vascular accidents, but all those cases were not included in this study. 3. Results There were 534 male and 113 female patients. The age groups ranged from the first to the ninth decade, most being in the fifth and sixth decade of life. The clinical features at presentation were varied with most having altered behavior and headache. Examination revealed that all patients had some amount of cognitive decline or limb weakness (Table 1). The postresuscitation GCS was then recorded. Of the patients, 86% had a postresuscitation GCS of 9 or higher (Table 1).
Fig. 1. Findings in the CT scan taken at time of admission.
From the radiology (Fig. 1), it was seen that almost all cases had mass effect and midline shift. An enhancing inner subdural membrane was seen in 11% of the CT scans. Associated lesions noted included diffuse cerebral edema, resolving contusions, bony fractures, or small pneumocephalus. The surgical procedures undertaken in patients with radiologically confirmed chronic subdural hematoma were burr holes or twist drill evacuation and craniotomy in 5 cases (Fig. 2). Most of the patients were treated operatively. Forty patients were not operated because either the patients refused surgery or were medically unfit, or the hematoma was not producing sufficient mass effect to merit surgery or was thought of not being the cause of the patients’ complaint. Of the patients, 6 underwent twist drill evacuation of hematoma as the primary method of evacuation. Of these, 4 presented in an acute stage with metabolic problems, and 2 were considered unfit for any major procedure. Repeat scans in this group had shown reduction in mass effect. As mentioned earlier, the patients were clinically reassessed on the seventh postoperative day and during follow-up (Figs. 3 and 4). The neurologic status including
Table 1 Clinical features at presentation (n = 647) Altered behavior Headache Cognitive disturbances History of trauma Hemiparesis Neurologic deficits Seizures Postresuscitation GCS 13-15 9-12 3-8
428 386 370 360 277 247 45
66% 60% 57% 56% 43% 38% 7%
400 155 92
62% 24% 14%
The table shows clinical features including the postresuscitation GCS.
Fig. 2. Operative procedures undertaken in these patients.
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
369
Table 2 Comparison of age with GOS (n = 607) Age
Total
Good
Modera
Severe
Vegetat
Dead
b 40 40-60 N 60
117 287 203
86 (74%) 227 (79%) 134 (66%)
12 (10%) 30 (11%) 19 (9%)
9 (8%) 9 (3%) 20 (10%)
5 (4%) 7 (2%) 12 (6%)
5(4%) 14 (5%) 18 (9%)
Comparing the age with the outcome showed that age was significant above 60 years ( P = .003).
Fig. 3. Glasgow outcome score on the seventh day.
the GOS was done at the time of this assessment. Follow-up in this study was up to 1 year, and 6% of cases were lost for follow-up. Most patients (89%) had made a very good recovery by the seventh postoperative day (Fig. 3). Although 14 patients were not seen on the seventh postoperative day, the clinical details were obtained from the secondary level hospitals, and they were subsequently seen in the outpatient clinic at later dates. This information was then analyzed. Sixteen patients had deteriorated postoperatively; 8 had recurrence of the clot and required reexploration, and 3 patients had large pneumocephalus, which was aspirated. In 5 patients, the deterioration was due to extracranial causes. With appropriate management, all these patients had good improvement. During follow-up, it was seen that 83% of patients were independent (based on GOS) (Fig. 4) and 4% had died due to unrelated causes. Comparing the age with the GOS, it was seen that younger patients had a much better outcome (Table 2). Age as a factor was statistically significant when the patients of 60 years or below were compared with those older than 60 years ( P = .003). Five hundred fifteen patients had a postresuscitation GCS of above 8, of which 83% had a good GOS. However this figure dropped to 35% when the postresuscitation GCS was 8 or below (Table 3). This finding was also statistically significant ( P = .000).
The mortality in this series was 5%. Of this, 28 patients (75%) had initial post resuscitation GCS 8 or below. When the information about GCS 8 or below was further broken down, it was seen that when the GCS was 3, the mortality was 70%, and with GCS of 4, it was 50%. The mortality of patients who presented in GCS of 8 was only 6% (Fig. 5). When the factors for mortality were analyzed, it was seen that most of the patients died due to herniation syndromes and raised intracranial pressure (Fig. 6). Preexisting comorbidities such as myocardial and renal problems also were the cause of death in about 30% of patients. Herniation was diagnosed clinically by features such as pupillary changes and loss of eye movements. Recurrence was seen in 136 patients (21%). This was identified when the patient presented with reappearance of clinical symptoms and confirmed by CT scan. During analysis, it was found that the factors for recurrence, such as presence of enhancing inner subdural membrane, membrane being visualized at surgery, and the brain remaining at a depth at end of surgery, were statistically significant ( P = .000) (Table 4). In the first few years of this study, a subdural drain was not used. However, after a change of policy at NIMHANS in 1994, a subdural drain was used if the brain was seen through the burr holes to be at a depth from surface at the end of the saline washout. A subdural drain was used in 27% of our patients, and this was retained for 48 hours postoperatively. After this the incidence, recurrence was reduced from 30% to 4% (Table 5). This value also was found to be significant statistically ( P = .000). There was another interesting fact noticed during this study. Biopsy of the inner membrane was done in 28 patients in whom the subdural membrane was visualized. The recurrence rate in these patients was 100%. An attempt was made to make a table comparing the various parameters such as age, sex, GCS, and radiological and operative findings and correlate this fact with recurrence and use of drain. It can be seen that use of drain prevented Table 3 Comparison of GCS with GOS (n = 607)
Fig. 4. Glasgow outcome score during the follow-up period of one year.
GCS
Total
Good
Modera
Severe
Vegetat
Dead
N8 V8 N 60
515 92 203
427 (83%) 33 (35%) 134 (66%)
46 (9%) 13 (15%) 19 (9%)
21 (4%) 10 (11%) 20 (10%)
12 (2%) 8 (9%) 12 (6%)
9 (2%) 28 (30%) 18 (9%)
Comparing the GCS with the outcome showed that GCS above 8 was significant ( P = .003).
370
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373 Table 4 Factors for recurrence (n = 607) Enhancing subdural membrane in CT Subdural membrane at surgery Brain remaining at a depth
Total
Recurrence
72 339 403
35 (49%) 72 (21%) 95 (24%)
Table shows the factors for recurrence that are statistically significant ( P = .000).
Fig. 5. Comparison between the GCS and mortality.
recurrence when the age of the patient was less than 60 years, patient was male, postresuscitation GCS was higher, the subdural membrane was visualized at surgery, and the brain remained at a depth at the end of surgery (Table 6). All these parameters are statistically significant. Sex may not be important because the sex ratio in this study was unequal. So, drain may significantly reduce recurrence in younger patients with better GCS, when the subdural membrane is seen at surgery, and the brain remains at a depth at the end of hematoma evacuation. 4. Discussion This discussion focuses on a retrospective analysis of 647 patients of chronic subdural hematoma over a 10-year period (1989-1999).The results suggest that the factors for morbidity in chronic subdural hematoma are age, low GCS, and comorbid factors such as myocardial or renal diseases, and recurrences can be reduced by the use of drains. In this study a higher incidence in men in the sixth and seventh decades was noted. This agrees with the series by Markwalder [20] Cameron [4], Robinson [28] and Kaste [15]. Svein and Gelety [31] had an average age of 63 years in their series. Fogelholm [9] and Markwalder had high incidence of chronic subdural hematomas in the sixth and seventh decade.
The common presenting features of patients in this series were with abnormalities in behavior (66%), nonspecific headache (60%), and focal neurologic deficits in the form of hemiparesis (44%). Cameron [4], in his series, had limb weakness in 40%, cognitive disturbance in 30%, and headache in 20%. Of the patients of Kaste et al [15], 72% presented with headache, and 48% had limb weakness, whereas in Dronfield’s series [7], cognitive disturbance was seen in 100%. Mckissock and Loud [23] had limb weakness as the most common presenting symptom. A history of trauma was obtained in 56% of patients. Among these, 281 (78%) had a history of trivial trauma. Trivial trauma was taken when, after trauma to the head, there was no loss of consciousness with no vomiting or seizures and without any posttraumatic sequelae. This compares with the studies of Mckissock and Loud [23], Fogelholm [9] and Richter et al [27] with an average of 52% to 75% incidence of trauma in their series [13,14]. In this study, 62% of the patients were conscious and had GCS 13 to 15, and another 24% were above GCS 9. Cognitive disturbances were seen in 59% and limb weakness in 44%. Papilloedema was seen in 50% of patients. In Dronfield’s series [7], common signs were cognitive dysfunction and limb weakness. Mckissock [23] also had similar findings. Kaste et al [15] had papilloedema as the most common sign. All patients underwent a non–contrast- and a contrastenhanced CT scan [2]. A contrast CT scan was asked for because of reasons mentioned earlier and also because the subdural membrane was better visualized with contrast [16]. Forbes [10,11] had also advocated contrast CT scan in all cases, especially if clinically suspected. Kim [18] showed the usefulness of contrast CT in isodense subdural hematomas. In a recent study from Japan, Nakaguchi et al [26] have argued that contrast CT may be useful for diagnosing the extent of enlargement of the inner membrane, predicting the risk of local brain damage, and evaluating the stage in natural history and can be predictive of recurrence [22,24]. Many operative methods have been described [1,3,5,12, 21,25,28]. Svein and Gelety [31], Robinson [28], and Table 5 Use of subdural drain (n = 607) With drain Without drain
Fig. 6. Cause in patients who died.
No.
Recurrence
165 442
6 (4%) 130 (30%)
Use of drain was statistically significant ( P = .000).
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373 Table 6 Comparison of recurrence and drain with variable parameters
Age b 40 40-60 N 60 Sex Male Female GCS 3-8 9-12 13-15 Enhancing membrane Subdural at surgery Brain at depth
Recurrence with drain
Recurrence without drain
Total
Recurrence
Total
Recurrence
33 77 55
0 (0%) 2 (3%) 4 (7%)
84 210 148
20 (24%) 76 (36%) 34 (23%)
3.08 5.67 2.55
.005 .001 .01
136 29
5 (4%) 1 (3%)
398 44
118 (29%) 12 (27%)
6.21 2.60
.000 .01
23 40 102 20
1 2 4 3
69 105 268 52
12 38 80 9
(17%) (36%) (30%) (17%)
1.56 3.76 5.32 0.24
NS .004 .000 NS
(4%) (4%) (4%) (15%)
Z
P
86
5 (6%)
102
49 (45%)
6.06
.000
103
6 (6%)
125
68 (54%)
7.79
.000
Age less than 60 years, higher GCS, and operative findings were statistically significant. NS indicates not significant.
Markwalder et al [21] advocated the use of burr holes only for removing the hematomas. In 1997, Sambasivan [29] had described a new method of a temporal craniectomy and showed a very low recurrence in that series of 2300 cases. Hamilton et al [12] compared burr hole evacuation and craniotomy and found no difference between the procedures. Camel and Gruble [3] had achieved an 86% excellent outcome by using twist drill craniostomy with continuous catheter drainage, whereas Williams et al [35] concluded that burr hole evacuation was superior to twist drill evacuation of chronic subdural hematoma, with respect to clinical outcomes and complications. Mohamed [25] argued for craniotomy and outer membranectomy with minimal incidence of recollection, morbidity, and mortality. Because of these uncertainties, Dunn [8,9] suggested a well-designed and adequately sized clinical trial to provide patients with a rational basis for the treatments offered. In this study, 92% of patients underwent burr hole evacuation, and 1% had twist drill drainage as the primary procedure. In this series there was a mortality of 6%. It was seen that age and post resuscitation GCS were 2 significant factors. However, the comorbidities such as renal and cardiac illnesses caused death in more than 30% [38]. Villagrasa [33] showed that a low GCS at admission had a higher mortality. This was echoed by Roselle also. Zumkeller et al [39] concluded from their work that the outcome and mortality were determined by age, preexisting illnesses, and complicating illnesses. There was a recurrence in 136 patients in this series. It was seen that a very well-enhancing subdural membrane in contrast CT as well as visualizing this membrane during surgery were statistically significant. Many studies have considered the importance of this membrane in recurrence. Yamashima et al [36] noticed formation of gap junctions in the endothelium of the subdural membrane. They concluded
371
that the most important factors for the development of chronic subdural hematoma existed in the vessels of the capsule, which show a marked proliferation potential and a fragile nature. Yamashima and Yamomoto [37,38] also described the pathology and the pathophysiology of the inner membrane in determining location of the hematoma. Takahashi et al [32] noticed difference between the collagen arrangement between inner and outer membrane. In a study in dogs, Watanabe et al [34] postulated that a peculiar type of fibrin was essential for capsule formation of the enlarging hematoma. Killeffer [17] combined the technique of radioactive red cell labeling and chemical analysis of the subdural fluid with observations of the ultrastructure of the normal meninges and outer neomembrane. They postulated that the outer membrane evolves as a reaction to cleavage of the dural border cell layer of the meninges. Interaction between the components of the outer membrane and the chronic sudural fluid contributes to the growth of the outer neomembrane and recurrent hemorrhage. All these studies may also explain the 100% recurrence in this series when the membrane was biopsied. Many authors have advocated the importance of the brain to expand after hematoma evacuation and fill up the subdural space as a factor preventing recurrence. Lalonde advocated intrathecal saline during surgery to cause expansion of the brain [5]. Robinson [28] reported excellent results in 87 of 121 patients with this method. However, this result could not be duplicated. In our study, there was a recurrence rate of 24% when the brain remained at a depth at end of surgery. This correlated with the finding of Robinson [28], Watanabe et al [34], and Yamashima et al [36]. The use of a subdural drain has been a subject of repeated studies. In his study in 1985, Markwalder et al [21] did not see any long-term benefit of subdural drain but noticed steadily progressive clinical improvement in early postoperative period (24 hours) when a drain was used. Lind et al [19] reported lower rates of repeated surgery if subdural drains are placed after evacuation of a subdural hematoma via a burr hole. This study also had similar findings. When subdural drains were not used, there was a recurrence rate of 30%, but with the use of drains, recurrence was 4%. The finding that drains prevent or reduce recurrence when age of the patient was less than 60 years, post resuscitation GCS was higher, the subdural membrane was visualized at surgery, and the brain remained at a depth at the end of surgery is a very novel one. It is suggested that a drain be used when these criteria are present. However, further studies may be necessary before these are accepted as indications for drainage. There is one self criticism in this study. The nonexpansion of the brain could not be defined or quantified exactly because this is a retrospective study. Only the surgeons’ visual impression has been taken for analysis. A definite relationship between the nonexpansion of the brain and recurrence has been shown by other studies [5,27,33,35], but none have quantified the depth.
372
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
5. Conclusions 1.
2.
3.
4.
The factors determining mortality were age, GCS at presentation, and associated illnesses such as cardiac and renal problems. The factors responsible for recurrence were visualizing a thick subdural membrane in the contrast CT as well as during surgery and brain remaining at a depth during surgery. Manipulating the membrane can cause recurrence (except in nonexpansile armorized brains of children with sticky chronic subdural hematomas). The use of a subdural drain significantly reduced recurrence.
References [1] Aoki N. Percutaneous tapping in the treatment of chronic subdural haematomas in adults. Neurol Res 1987;9:19. [2] Bergstrom M, Ericsson K, Levander B, et al. Computed Tomography of cranial subdural and epidural haematomas. J Comput Assist Tomogr 1977;1:449. [3] Camel M, Grubb RL. Treatment of chronic subdural haematomas by twist drill craniostomy with continuous drainage. J Neurosurg 1986; 65:183. [4] Cameron MM. Chronic subdural haematomas: review of 114 cases. J Neurol Neurosurg Psychiatry 1978;41:834. [5] Chee CP. Exteriorization of subdural pocket for chronic recurrent subdural haematoma. Neurosurgery 1988;22:780. [6] Cooper PR. In: Cooper PR, editor. Head injury. 3rd ed. Williams & Wilkins; 1993. p. 285 - 97. [7] Dronfield MW, Mead GM, Langman MJ. Survival and death from subdural haematoma on medical wards. Postgrad Med J 1997;53:57. [8] Dunn LT. Surgery for chronic subdural haematoma: is there an evidence base? J Neurol Neurosurg Psychiatry 2003;74:842. [9] Fogelholm R. Chronic subdural haematoma: influence of patients age on symptoms signs and thickness of hematoma. J Neurosurg 1975;42:43. [10] Forbes GS. CT scan in the evolution of chronic subdural haematoma. Radiology 1978;126:143. [11] Goodell CL, Mealey Jr J. Pathogenesis of chronic subdural hematomas. Experimental studies. Arch Neurol 1963;8:429. [12] Hamilton MG, Frizzell JB, Traumer BI. Chronic subdural haematomas. The role of craniotomy re-evaluated. Neurosurgery 1993;33:67. [13] Handa J, Handa H, Nakono Y. Rim enhancement in computed tomography with chronic subdural haematomas. Surg Neurol 1979; 11:217. [14] Kalyanaraman S, Ramamurthy B, et al. Chronic subdural haematoma. Neurol India 1970;18:25. [15] Kaste M, Waltimo O, Haiskanen O. Chronic bilateral subdural haematoma in adults. Acta Neurochir 1979;48:231. [16] Kawakami Y, Chikama M, Tanuja T. Coagulation and fibrinolysis in chronic subdural haematoma. Neurosurgery 1989;25:25. [17] Killeffer JA, Killeffer FA, Schochet SS. The outer neomembrane of chronic subdural haematoma. Neurosurg Clin N Am 2000;11:407. [18] Kim KS. Computed tomography in isodense chronic subdural haematoma. Radiology 1978;128:71. [19] Lind CR, Lind CJ, Mee EW. Reduction in the number of repeated operations of subacute and chronic subdural haematomas by placement of subdural drains. J Neurosurg 2003;99:44. [20] Markwalder TM. Chronic subdural haematomas a review. J Neurosurg 1981;54:637.
[21] Markwalder TM, Stemsepe KF, Rohner M, et al. The course of subdural haematomas after burr hole craniostomy and closed system drainage. J Neurosurg 1981;55:390. [22] Markwalder TM, Seiler RW. Chronic subdural haematomas: to drain or not to drain. Neurosurgery 1985;16:185. [23] Mckissock W, Loud MS, Richardson A, et al. Subdural haematoma a review of 389 cases. Lancet 1960;1:1365. [24] Mellergrad P, Wisten O. Operations and re-operations for chronic subdural haematomas during a 25 year period in a well defined population. Acta Neurochir 1996;138:708. [25] Mohamed EE. Chronic subdural haematoma treated by craniotomy, durectomy, outer membranectomy and subgaleal suction drainage. Personal experience in 39 patients. Br J Neurosurg 2003;17:244. [26] Nakaguchi H, Yoshimasu N, Tanishima T. Relationship between the natural history of chronic subdural hematoma and enhancement of the inner membrane on post-contrast CT scan. No Shinkei Geka 2003; 31(2):157 - 64 [Article in Japanese]. [27] Richter HP, Klein HJ, Schafer M. Chronic Subdural haematomas treated by enlarged burrhole craniostomy and closed System drainage. Retrospective study of 120 patients. Acta Neurochir 1984; 71:179. [28] Robinson RG. Chronic subdural haematoma, surgical management in 133 patients. J Neurosurg 1984;61:263. [29] Sambasivan M. An overview of chronic subdural haematoma. Experience with 2300 cases. Surg Neurol 1997;47:418. [30] Scotti G, Terbrugge K, Melamcom D, et al. Evaluation of age of subdural haematomas in various clinical stages by CT scan. J Neurosurg 1977;47:311. [31] Svein HJ, Gelety JE. On the surgical management of encapsulated subdural haematoma. A comparison of results of membranectomy and simple evacuation. J Neurosurg 1964;21:172. [32] Takahashi Y, Okura A, Yoshimura G, et al. Ultra structure of collagen fibers in the outer membrane of recurrent chronic subdural haematoma. Neurol Med Chir 1996;36:627 [Japanese]. [33] Villagrasa J, Prat R, Diaz JF. Analysis of prognostic factors in adults with chronic subdural haematomas. Neurologia 1998;13:120 [Spanish]. [34] Watanabe S, Shimada WE, Newman P, et al. The role of craniectomy in the treatment of chronic subdural haematomas in experimental animals. J Neurosurg 1972;37:552. [35] Williams GR, Baskaya MK, Menendez J, Polin R, Willis B, Nanda A. Burr-hole versus twist-drill drainage for the evacuation of chronic subdural haematoma: a comparison of clinical results. J Clin Neurosci 2001;8:551. [36] Yamashima T, Yamomoto S, Friede RL. The role of endothelial gap junctions in the enlargement of chronic subdural haemotomas. J Neurosurg 1983;59:298. [37] Yamashima T, Yamomoto S. How do vessels proliferate in the capsule of a chronic subdural haematoma. Neurosurgery 1984; 15:672. [38] Yamashima T. The inner membrane of chronic subdural haematomas: pathology and pathophysiology. Neurosurg Clin N Am 2000;11:413. [39] Zumkeller M, Hollerhage HG, Dietz H. Treatment outcome in patients with chronic subdural haematoma with reference to age and concurrent internal diseases. Wien Med Wochenscher 1997;147:55.
Commentary This is a retrospective analysis of 647 cases of chronic subdural hematoma treated in one institution, apparently by different surgeons. The information is collected and analyzed quite well, and the results are reported in different sections according to the age and GCS at presentation as a
Vascular
Chronic subdural hematomas—causes of morbidity and mortalityB Ramnarayan Ramachandran, MCha,4, Thimmappa Hegde, MChb a
Department of Neurosurgery, Dr SM CSI Medical College Hospital, Karakonam 695504, India b Narayana Institute of Neurosciences, Bangalore 560099, India Received 27 February 2006; accepted 28 July 2006
Abstract
Objective: Chronic subdural hematoma is a very common condition seen usually in the later stages of life. Treatment, although apparently simple, is associated with some morbidity and mortality with a potential for recurrence. This is especially important as the average life span of humanity increases all over the world. Methods: A retrospective analysis of 647 cases of chronic subdural hematomas treated in NIMHANS was done. The aim of this study was to determine the factors responsible for the morbidity, mortality, and recurrence in patients with chronic subdural hematomas. Results: The patients were mostly men in the sixth and seventh decade of life and presented with headache, cognitive decline, or focal deficits. CT scan was done in all cases. 94% of the patients underwent evacuation of the hematoma. There was a mortality rate of 5%, and there was a recurrence of 21%. Statistical analysis for factors both for mortality and morbidity were done. Conclusions: It was seen that statistically significant factors determining mortality were age, GCS at presentation, and associated illnesses like cardiac and renal failure. The statistically significant factors for recurrence are the presence of a thick subdural membrane visualized during surgery and brain, remaining at a depth at the end of evacuation of hematoma. Use of a subdural drain significantly reduced recurrences. The chronicity of the hematoma was not a factor determining mortality or morbidity. Drains were seen to significantly reduce recurrence in younger patients with better GCS and when the subdural membrane is seen at surgery and the brain remains at a depth at the end of hematoma evacuation. D 2007 Elsevier Inc. All rights reserved.
Keywords:
Chronic subdural hematoma; Elderly; Morbidity; Mortality
Abbreviations: Contrast CT, Contrast-enhanced computed tomography; CT, Computed tomography; GCS, Glasgow coma score; GOS, Glasgow outcome score; NIMHANS, National Institute of Mental Health and NeuroSciences, Government of India. B This study was done at the Department of Neurosurgery, The National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, India. There were no financial relationships involved for this study and there is no conflict of interests. This is purely a retrospective case note study and does not involve any human beings or animals directly. The corresponding author has the right to grant on behalf of all authors and, does grant on behalf of all authors, an exclusive license (or nonexclusive for government employees) on a worldwide basis to permit this article (if accepted) to be published in Surgical Neurology and any other products and sublicenses that use and exploit all subsidiary rights, as set out in our license. 4 Corresponding author. Tel.: +91 479 2469469; fax; +91 479 2468709. E-mail addresses: [email protected] (R. Ramachandran)8 [email protected] (T. Hegde). 0090-3019/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.07.022
1. Introduction Head injury sequelae with posttraumatic hematomas of various types form a sizeable percentage of admissions of any neurosurgical center [6]. Scotti [30] has defined subdural hematoma radiologically based on the density in the CT scan and defined chronic subdural hematomas as hypodense to parenchyma and presenting 21 days after trauma. It is a problem related to age. Various surgical treatments described like burr holes and twist drills to craniotomies, are not always effective in managing this condition. Despite the apparent simplicity of the management of this condition, mortality and morbidity does occur, as also recurrence. The aim of this study was to analyze the causes of morbidity, mortality, and recurrence in chronic subdural hematomas.
368
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
2. Material and methods A retrospective analysis of the records of all patients of chronic subdural hematomas admitted to NIMHANS over a 10-year period (1989-1999) was done. Cases were identified by the coding in the medical records. The details recorded were the age and sex of the patients; any associated comorbidity such as hypertension, cardiac or renal disease, and clinical features at presentation, including the postresuscitation GCS; and the radiology and operative procedure. Outcome was assessed on the seventh postoperative day. A total of 665 patients were treated for chronic subdural hematoma during this time period. However, only 647 case notes were available for analysis. Most of the patients were referred from local hospitals on clinical suspicion. They were admitted, stabilized, clinically assessed, and then sent for plain and contrast CT scan for diagnosis. Patients with chronic subdural hematoma were then managed based on clinical and radiological findings and the operative findings. They were observed after definitive procedure for 24 hours and then referred back to the secondary level hospitals. These patients were then called back on the 7th postoperative day for suture removal (and also clinical assessment). The patients were kept for follow-up for at least up to 1 year. v 2 and proportion tests were done appropriately to find out the statistical significance. As the patients were referred from other smaller hospitals on clinical suspicion, a plain and contrast CT was done in all cases. There were patients with other radiological findings such as tumors and vascular accidents, but all those cases were not included in this study. 3. Results There were 534 male and 113 female patients. The age groups ranged from the first to the ninth decade, most being in the fifth and sixth decade of life. The clinical features at presentation were varied with most having altered behavior and headache. Examination revealed that all patients had some amount of cognitive decline or limb weakness (Table 1). The postresuscitation GCS was then recorded. Of the patients, 86% had a postresuscitation GCS of 9 or higher (Table 1).
Fig. 1. Findings in the CT scan taken at time of admission.
From the radiology (Fig. 1), it was seen that almost all cases had mass effect and midline shift. An enhancing inner subdural membrane was seen in 11% of the CT scans. Associated lesions noted included diffuse cerebral edema, resolving contusions, bony fractures, or small pneumocephalus. The surgical procedures undertaken in patients with radiologically confirmed chronic subdural hematoma were burr holes or twist drill evacuation and craniotomy in 5 cases (Fig. 2). Most of the patients were treated operatively. Forty patients were not operated because either the patients refused surgery or were medically unfit, or the hematoma was not producing sufficient mass effect to merit surgery or was thought of not being the cause of the patients’ complaint. Of the patients, 6 underwent twist drill evacuation of hematoma as the primary method of evacuation. Of these, 4 presented in an acute stage with metabolic problems, and 2 were considered unfit for any major procedure. Repeat scans in this group had shown reduction in mass effect. As mentioned earlier, the patients were clinically reassessed on the seventh postoperative day and during follow-up (Figs. 3 and 4). The neurologic status including
Table 1 Clinical features at presentation (n = 647) Altered behavior Headache Cognitive disturbances History of trauma Hemiparesis Neurologic deficits Seizures Postresuscitation GCS 13-15 9-12 3-8
428 386 370 360 277 247 45
66% 60% 57% 56% 43% 38% 7%
400 155 92
62% 24% 14%
The table shows clinical features including the postresuscitation GCS.
Fig. 2. Operative procedures undertaken in these patients.
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
369
Table 2 Comparison of age with GOS (n = 607) Age
Total
Good
Modera
Severe
Vegetat
Dead
b 40 40-60 N 60
117 287 203
86 (74%) 227 (79%) 134 (66%)
12 (10%) 30 (11%) 19 (9%)
9 (8%) 9 (3%) 20 (10%)
5 (4%) 7 (2%) 12 (6%)
5(4%) 14 (5%) 18 (9%)
Comparing the age with the outcome showed that age was significant above 60 years ( P = .003).
Fig. 3. Glasgow outcome score on the seventh day.
the GOS was done at the time of this assessment. Follow-up in this study was up to 1 year, and 6% of cases were lost for follow-up. Most patients (89%) had made a very good recovery by the seventh postoperative day (Fig. 3). Although 14 patients were not seen on the seventh postoperative day, the clinical details were obtained from the secondary level hospitals, and they were subsequently seen in the outpatient clinic at later dates. This information was then analyzed. Sixteen patients had deteriorated postoperatively; 8 had recurrence of the clot and required reexploration, and 3 patients had large pneumocephalus, which was aspirated. In 5 patients, the deterioration was due to extracranial causes. With appropriate management, all these patients had good improvement. During follow-up, it was seen that 83% of patients were independent (based on GOS) (Fig. 4) and 4% had died due to unrelated causes. Comparing the age with the GOS, it was seen that younger patients had a much better outcome (Table 2). Age as a factor was statistically significant when the patients of 60 years or below were compared with those older than 60 years ( P = .003). Five hundred fifteen patients had a postresuscitation GCS of above 8, of which 83% had a good GOS. However this figure dropped to 35% when the postresuscitation GCS was 8 or below (Table 3). This finding was also statistically significant ( P = .000).
The mortality in this series was 5%. Of this, 28 patients (75%) had initial post resuscitation GCS 8 or below. When the information about GCS 8 or below was further broken down, it was seen that when the GCS was 3, the mortality was 70%, and with GCS of 4, it was 50%. The mortality of patients who presented in GCS of 8 was only 6% (Fig. 5). When the factors for mortality were analyzed, it was seen that most of the patients died due to herniation syndromes and raised intracranial pressure (Fig. 6). Preexisting comorbidities such as myocardial and renal problems also were the cause of death in about 30% of patients. Herniation was diagnosed clinically by features such as pupillary changes and loss of eye movements. Recurrence was seen in 136 patients (21%). This was identified when the patient presented with reappearance of clinical symptoms and confirmed by CT scan. During analysis, it was found that the factors for recurrence, such as presence of enhancing inner subdural membrane, membrane being visualized at surgery, and the brain remaining at a depth at end of surgery, were statistically significant ( P = .000) (Table 4). In the first few years of this study, a subdural drain was not used. However, after a change of policy at NIMHANS in 1994, a subdural drain was used if the brain was seen through the burr holes to be at a depth from surface at the end of the saline washout. A subdural drain was used in 27% of our patients, and this was retained for 48 hours postoperatively. After this the incidence, recurrence was reduced from 30% to 4% (Table 5). This value also was found to be significant statistically ( P = .000). There was another interesting fact noticed during this study. Biopsy of the inner membrane was done in 28 patients in whom the subdural membrane was visualized. The recurrence rate in these patients was 100%. An attempt was made to make a table comparing the various parameters such as age, sex, GCS, and radiological and operative findings and correlate this fact with recurrence and use of drain. It can be seen that use of drain prevented Table 3 Comparison of GCS with GOS (n = 607)
Fig. 4. Glasgow outcome score during the follow-up period of one year.
GCS
Total
Good
Modera
Severe
Vegetat
Dead
N8 V8 N 60
515 92 203
427 (83%) 33 (35%) 134 (66%)
46 (9%) 13 (15%) 19 (9%)
21 (4%) 10 (11%) 20 (10%)
12 (2%) 8 (9%) 12 (6%)
9 (2%) 28 (30%) 18 (9%)
Comparing the GCS with the outcome showed that GCS above 8 was significant ( P = .003).
370
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373 Table 4 Factors for recurrence (n = 607) Enhancing subdural membrane in CT Subdural membrane at surgery Brain remaining at a depth
Total
Recurrence
72 339 403
35 (49%) 72 (21%) 95 (24%)
Table shows the factors for recurrence that are statistically significant ( P = .000).
Fig. 5. Comparison between the GCS and mortality.
recurrence when the age of the patient was less than 60 years, patient was male, postresuscitation GCS was higher, the subdural membrane was visualized at surgery, and the brain remained at a depth at the end of surgery (Table 6). All these parameters are statistically significant. Sex may not be important because the sex ratio in this study was unequal. So, drain may significantly reduce recurrence in younger patients with better GCS, when the subdural membrane is seen at surgery, and the brain remains at a depth at the end of hematoma evacuation. 4. Discussion This discussion focuses on a retrospective analysis of 647 patients of chronic subdural hematoma over a 10-year period (1989-1999).The results suggest that the factors for morbidity in chronic subdural hematoma are age, low GCS, and comorbid factors such as myocardial or renal diseases, and recurrences can be reduced by the use of drains. In this study a higher incidence in men in the sixth and seventh decades was noted. This agrees with the series by Markwalder [20] Cameron [4], Robinson [28] and Kaste [15]. Svein and Gelety [31] had an average age of 63 years in their series. Fogelholm [9] and Markwalder had high incidence of chronic subdural hematomas in the sixth and seventh decade.
The common presenting features of patients in this series were with abnormalities in behavior (66%), nonspecific headache (60%), and focal neurologic deficits in the form of hemiparesis (44%). Cameron [4], in his series, had limb weakness in 40%, cognitive disturbance in 30%, and headache in 20%. Of the patients of Kaste et al [15], 72% presented with headache, and 48% had limb weakness, whereas in Dronfield’s series [7], cognitive disturbance was seen in 100%. Mckissock and Loud [23] had limb weakness as the most common presenting symptom. A history of trauma was obtained in 56% of patients. Among these, 281 (78%) had a history of trivial trauma. Trivial trauma was taken when, after trauma to the head, there was no loss of consciousness with no vomiting or seizures and without any posttraumatic sequelae. This compares with the studies of Mckissock and Loud [23], Fogelholm [9] and Richter et al [27] with an average of 52% to 75% incidence of trauma in their series [13,14]. In this study, 62% of the patients were conscious and had GCS 13 to 15, and another 24% were above GCS 9. Cognitive disturbances were seen in 59% and limb weakness in 44%. Papilloedema was seen in 50% of patients. In Dronfield’s series [7], common signs were cognitive dysfunction and limb weakness. Mckissock [23] also had similar findings. Kaste et al [15] had papilloedema as the most common sign. All patients underwent a non–contrast- and a contrastenhanced CT scan [2]. A contrast CT scan was asked for because of reasons mentioned earlier and also because the subdural membrane was better visualized with contrast [16]. Forbes [10,11] had also advocated contrast CT scan in all cases, especially if clinically suspected. Kim [18] showed the usefulness of contrast CT in isodense subdural hematomas. In a recent study from Japan, Nakaguchi et al [26] have argued that contrast CT may be useful for diagnosing the extent of enlargement of the inner membrane, predicting the risk of local brain damage, and evaluating the stage in natural history and can be predictive of recurrence [22,24]. Many operative methods have been described [1,3,5,12, 21,25,28]. Svein and Gelety [31], Robinson [28], and Table 5 Use of subdural drain (n = 607) With drain Without drain
Fig. 6. Cause in patients who died.
No.
Recurrence
165 442
6 (4%) 130 (30%)
Use of drain was statistically significant ( P = .000).
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373 Table 6 Comparison of recurrence and drain with variable parameters
Age b 40 40-60 N 60 Sex Male Female GCS 3-8 9-12 13-15 Enhancing membrane Subdural at surgery Brain at depth
Recurrence with drain
Recurrence without drain
Total
Recurrence
Total
Recurrence
33 77 55
0 (0%) 2 (3%) 4 (7%)
84 210 148
20 (24%) 76 (36%) 34 (23%)
3.08 5.67 2.55
.005 .001 .01
136 29
5 (4%) 1 (3%)
398 44
118 (29%) 12 (27%)
6.21 2.60
.000 .01
23 40 102 20
1 2 4 3
69 105 268 52
12 38 80 9
(17%) (36%) (30%) (17%)
1.56 3.76 5.32 0.24
NS .004 .000 NS
(4%) (4%) (4%) (15%)
Z
P
86
5 (6%)
102
49 (45%)
6.06
.000
103
6 (6%)
125
68 (54%)
7.79
.000
Age less than 60 years, higher GCS, and operative findings were statistically significant. NS indicates not significant.
Markwalder et al [21] advocated the use of burr holes only for removing the hematomas. In 1997, Sambasivan [29] had described a new method of a temporal craniectomy and showed a very low recurrence in that series of 2300 cases. Hamilton et al [12] compared burr hole evacuation and craniotomy and found no difference between the procedures. Camel and Gruble [3] had achieved an 86% excellent outcome by using twist drill craniostomy with continuous catheter drainage, whereas Williams et al [35] concluded that burr hole evacuation was superior to twist drill evacuation of chronic subdural hematoma, with respect to clinical outcomes and complications. Mohamed [25] argued for craniotomy and outer membranectomy with minimal incidence of recollection, morbidity, and mortality. Because of these uncertainties, Dunn [8,9] suggested a well-designed and adequately sized clinical trial to provide patients with a rational basis for the treatments offered. In this study, 92% of patients underwent burr hole evacuation, and 1% had twist drill drainage as the primary procedure. In this series there was a mortality of 6%. It was seen that age and post resuscitation GCS were 2 significant factors. However, the comorbidities such as renal and cardiac illnesses caused death in more than 30% [38]. Villagrasa [33] showed that a low GCS at admission had a higher mortality. This was echoed by Roselle also. Zumkeller et al [39] concluded from their work that the outcome and mortality were determined by age, preexisting illnesses, and complicating illnesses. There was a recurrence in 136 patients in this series. It was seen that a very well-enhancing subdural membrane in contrast CT as well as visualizing this membrane during surgery were statistically significant. Many studies have considered the importance of this membrane in recurrence. Yamashima et al [36] noticed formation of gap junctions in the endothelium of the subdural membrane. They concluded
371
that the most important factors for the development of chronic subdural hematoma existed in the vessels of the capsule, which show a marked proliferation potential and a fragile nature. Yamashima and Yamomoto [37,38] also described the pathology and the pathophysiology of the inner membrane in determining location of the hematoma. Takahashi et al [32] noticed difference between the collagen arrangement between inner and outer membrane. In a study in dogs, Watanabe et al [34] postulated that a peculiar type of fibrin was essential for capsule formation of the enlarging hematoma. Killeffer [17] combined the technique of radioactive red cell labeling and chemical analysis of the subdural fluid with observations of the ultrastructure of the normal meninges and outer neomembrane. They postulated that the outer membrane evolves as a reaction to cleavage of the dural border cell layer of the meninges. Interaction between the components of the outer membrane and the chronic sudural fluid contributes to the growth of the outer neomembrane and recurrent hemorrhage. All these studies may also explain the 100% recurrence in this series when the membrane was biopsied. Many authors have advocated the importance of the brain to expand after hematoma evacuation and fill up the subdural space as a factor preventing recurrence. Lalonde advocated intrathecal saline during surgery to cause expansion of the brain [5]. Robinson [28] reported excellent results in 87 of 121 patients with this method. However, this result could not be duplicated. In our study, there was a recurrence rate of 24% when the brain remained at a depth at end of surgery. This correlated with the finding of Robinson [28], Watanabe et al [34], and Yamashima et al [36]. The use of a subdural drain has been a subject of repeated studies. In his study in 1985, Markwalder et al [21] did not see any long-term benefit of subdural drain but noticed steadily progressive clinical improvement in early postoperative period (24 hours) when a drain was used. Lind et al [19] reported lower rates of repeated surgery if subdural drains are placed after evacuation of a subdural hematoma via a burr hole. This study also had similar findings. When subdural drains were not used, there was a recurrence rate of 30%, but with the use of drains, recurrence was 4%. The finding that drains prevent or reduce recurrence when age of the patient was less than 60 years, post resuscitation GCS was higher, the subdural membrane was visualized at surgery, and the brain remained at a depth at the end of surgery is a very novel one. It is suggested that a drain be used when these criteria are present. However, further studies may be necessary before these are accepted as indications for drainage. There is one self criticism in this study. The nonexpansion of the brain could not be defined or quantified exactly because this is a retrospective study. Only the surgeons’ visual impression has been taken for analysis. A definite relationship between the nonexpansion of the brain and recurrence has been shown by other studies [5,27,33,35], but none have quantified the depth.
372
R. Ramachandran, T. Hegde / Surgical Neurology 67 (2007) 367–373
5. Conclusions 1.
2.
3.
4.
The factors determining mortality were age, GCS at presentation, and associated illnesses such as cardiac and renal problems. The factors responsible for recurrence were visualizing a thick subdural membrane in the contrast CT as well as during surgery and brain remaining at a depth during surgery. Manipulating the membrane can cause recurrence (except in nonexpansile armorized brains of children with sticky chronic subdural hematomas). The use of a subdural drain significantly reduced recurrence.
References [1] Aoki N. Percutaneous tapping in the treatment of chronic subdural haematomas in adults. Neurol Res 1987;9:19. [2] Bergstrom M, Ericsson K, Levander B, et al. Computed Tomography of cranial subdural and epidural haematomas. J Comput Assist Tomogr 1977;1:449. [3] Camel M, Grubb RL. Treatment of chronic subdural haematomas by twist drill craniostomy with continuous drainage. J Neurosurg 1986; 65:183. [4] Cameron MM. Chronic subdural haematomas: review of 114 cases. J Neurol Neurosurg Psychiatry 1978;41:834. [5] Chee CP. Exteriorization of subdural pocket for chronic recurrent subdural haematoma. Neurosurgery 1988;22:780. [6] Cooper PR. In: Cooper PR, editor. Head injury. 3rd ed. Williams & Wilkins; 1993. p. 285 - 97. [7] Dronfield MW, Mead GM, Langman MJ. Survival and death from subdural haematoma on medical wards. Postgrad Med J 1997;53:57. [8] Dunn LT. Surgery for chronic subdural haematoma: is there an evidence base? J Neurol Neurosurg Psychiatry 2003;74:842. [9] Fogelholm R. Chronic subdural haematoma: influence of patients age on symptoms signs and thickness of hematoma. J Neurosurg 1975;42:43. [10] Forbes GS. CT scan in the evolution of chronic subdural haematoma. Radiology 1978;126:143. [11] Goodell CL, Mealey Jr J. Pathogenesis of chronic subdural hematomas. Experimental studies. Arch Neurol 1963;8:429. [12] Hamilton MG, Frizzell JB, Traumer BI. Chronic subdural haematomas. The role of craniotomy re-evaluated. Neurosurgery 1993;33:67. [13] Handa J, Handa H, Nakono Y. Rim enhancement in computed tomography with chronic subdural haematomas. Surg Neurol 1979; 11:217. [14] Kalyanaraman S, Ramamurthy B, et al. Chronic subdural haematoma. Neurol India 1970;18:25. [15] Kaste M, Waltimo O, Haiskanen O. Chronic bilateral subdural haematoma in adults. Acta Neurochir 1979;48:231. [16] Kawakami Y, Chikama M, Tanuja T. Coagulation and fibrinolysis in chronic subdural haematoma. Neurosurgery 1989;25:25. [17] Killeffer JA, Killeffer FA, Schochet SS. The outer neomembrane of chronic subdural haematoma. Neurosurg Clin N Am 2000;11:407. [18] Kim KS. Computed tomography in isodense chronic subdural haematoma. Radiology 1978;128:71. [19] Lind CR, Lind CJ, Mee EW. Reduction in the number of repeated operations of subacute and chronic subdural haematomas by placement of subdural drains. J Neurosurg 2003;99:44. [20] Markwalder TM. Chronic subdural haematomas a review. J Neurosurg 1981;54:637.
[21] Markwalder TM, Stemsepe KF, Rohner M, et al. The course of subdural haematomas after burr hole craniostomy and closed system drainage. J Neurosurg 1981;55:390. [22] Markwalder TM, Seiler RW. Chronic subdural haematomas: to drain or not to drain. Neurosurgery 1985;16:185. [23] Mckissock W, Loud MS, Richardson A, et al. Subdural haematoma a review of 389 cases. Lancet 1960;1:1365. [24] Mellergrad P, Wisten O. Operations and re-operations for chronic subdural haematomas during a 25 year period in a well defined population. Acta Neurochir 1996;138:708. [25] Mohamed EE. Chronic subdural haematoma treated by craniotomy, durectomy, outer membranectomy and subgaleal suction drainage. Personal experience in 39 patients. Br J Neurosurg 2003;17:244. [26] Nakaguchi H, Yoshimasu N, Tanishima T. Relationship between the natural history of chronic subdural hematoma and enhancement of the inner membrane on post-contrast CT scan. No Shinkei Geka 2003; 31(2):157 - 64 [Article in Japanese]. [27] Richter HP, Klein HJ, Schafer M. Chronic Subdural haematomas treated by enlarged burrhole craniostomy and closed System drainage. Retrospective study of 120 patients. Acta Neurochir 1984; 71:179. [28] Robinson RG. Chronic subdural haematoma, surgical management in 133 patients. J Neurosurg 1984;61:263. [29] Sambasivan M. An overview of chronic subdural haematoma. Experience with 2300 cases. Surg Neurol 1997;47:418. [30] Scotti G, Terbrugge K, Melamcom D, et al. Evaluation of age of subdural haematomas in various clinical stages by CT scan. J Neurosurg 1977;47:311. [31] Svein HJ, Gelety JE. On the surgical management of encapsulated subdural haematoma. A comparison of results of membranectomy and simple evacuation. J Neurosurg 1964;21:172. [32] Takahashi Y, Okura A, Yoshimura G, et al. Ultra structure of collagen fibers in the outer membrane of recurrent chronic subdural haematoma. Neurol Med Chir 1996;36:627 [Japanese]. [33] Villagrasa J, Prat R, Diaz JF. Analysis of prognostic factors in adults with chronic subdural haematomas. Neurologia 1998;13:120 [Spanish]. [34] Watanabe S, Shimada WE, Newman P, et al. The role of craniectomy in the treatment of chronic subdural haematomas in experimental animals. J Neurosurg 1972;37:552. [35] Williams GR, Baskaya MK, Menendez J, Polin R, Willis B, Nanda A. Burr-hole versus twist-drill drainage for the evacuation of chronic subdural haematoma: a comparison of clinical results. J Clin Neurosci 2001;8:551. [36] Yamashima T, Yamomoto S, Friede RL. The role of endothelial gap junctions in the enlargement of chronic subdural haemotomas. J Neurosurg 1983;59:298. [37] Yamashima T, Yamomoto S. How do vessels proliferate in the capsule of a chronic subdural haematoma. Neurosurgery 1984; 15:672. [38] Yamashima T. The inner membrane of chronic subdural haematomas: pathology and pathophysiology. Neurosurg Clin N Am 2000;11:413. [39] Zumkeller M, Hollerhage HG, Dietz H. Treatment outcome in patients with chronic subdural haematoma with reference to age and concurrent internal diseases. Wien Med Wochenscher 1997;147:55.
Commentary This is a retrospective analysis of 647 cases of chronic subdural hematoma treated in one institution, apparently by different surgeons. The information is collected and analyzed quite well, and the results are reported in different sections according to the age and GCS at presentation as a