Chapter 137 Surgical Management of Chronic Subdural Hematoma in Adults
Epidemiology
Chronic subdural hematoma (CSDH) is one of the commonest neurosurgical disorders and is especially prevalent among the elderly. Its incidence is about five per 100,000 per year in the general population but is higher for those aged 70 years and older (58 per 100,000 per year).1 Because the proportion of people aged 65 years and older is expected to double worldwide between 2000 and 2030, a substantial rise in incidence is expected.1 CSDH has a strong male preponderance, with a male-to-female ratio approximately 3:1.2 Age distributions vary among published series depending on the populations from which they were derived. Similarly to most published series,3–5 our published cohort of 215 patients had a mean age of 76.8 years, ranging from 36 to 95 years.2
Pathogenesis
CSDH arises in the dural border cell (DBC) layer, a loose cellular layer devoid of intercellular collagen and tight junctions, located between two firm meninges: the dura mater wired with abundant intercellular collagen on one side and the arachnoid matter with cells anchored to a basal membrane and clamped together with tight junctions on the other (Fig. 137-1).6 Traversing veins are being increasingly stretched by the shrinking brain until only a minor additional force is sufficient to cause rupture through stretching or shearing.6 The extravasated blood dissects the DBC layer, creating a subdural cavity. Similarly, a traumatic tear of the arachnoid mater can cause a hygroma, which can later transform into a CSDH.4,7,8 Whatever the initial mechanism, a failure of complex reparatory processes to heal the injured tissues results in CSDH. In contrast to young and healthy persons in whom healing is often successfully completed, CSDHs are more likely to arise in elderly patients with atrophic brain, impaired coagulation, repeated falls, and conditions resulting in intracranial hypotension, such as ventriculoperitoneal shunt. However, rarer causes of a CSDH such as vascular malformations, arachnoid cysts, and neoplasms have to be borne in mind.9–11 CSDH is characterized by the formation of neomembranes (the inner, visceral, membrane is relatively less vascular and usually thinner than the outer, parietal, membrane) with growth of leaky neocapillaries and enzymatic hyperfibrinolysis.12
Figure 137-1 Schematic representation of the ultrastructure of the meninges (from Haines et al6). The dura mater is composed of fibroblasts and a large amount of collagen. The arachnoid barrier cells are supported by a basement membrane and bound together by numerous tight junctions. The dural border cells layer is formed by flattened fibroblasts, with no tight junctions and no intercellular collagen. It is therefore a relatively loose layer positioned between firm dura mater and arachnoid. The subdural space is a potential space that can form within the dural border cell layer. SAS, subarachnoid space.
Clinical Features
Patients present with a variety of complaints, usually following a subacute or chronic course. In our published cohort the most common symptoms in order of decreasing frequency were gait disturbance and falls, mental deterioration, hemiparesis, headache, speech impairment, and drowsiness or coma.2 A history of head injury was established in 119 of 196 cases (61%). On admission, 86 (42%) patients had a Glasgow Coma Score (GCS) of 15, 102 (50%) had scores of 9 to 14, and 14 (7%) had scores less than 8. Limb weakness was recorded in 111 of 184 (63%), and dysphasia was found in 62 of 191 (33%).
Radiologic Evaluation
Most CSDHs can be diagnosed with a computed tomography (CT) or magnetic resonance imaging (MRI) scan. CT is currently the investigation of choice in CSDH. The density of a subdural hematoma depends on the interval(s) between the bleeding episode(s) and the time of examination.13 A hyperdense, crescent-shaped collection lying between the brain and the inner skull table represents an acute subdural hematoma (60 to 80 Hounsfield units). A subacute subdural hematoma can be nearly isodense with the adjacent cortex (30 Hounsfield units). This phase occurs 1 to 3 weeks after the acute event.14 Isodense hematomas may be difficult to differentiate from normal brain tissue, and MRI may be helpful in establishing the diagnosis.14 Certain features on CT such as medial displacement of the gray–white matter interface and a failure of cortical sulci to reach the inner skull table can point toward the presence of an isodense subdural hematoma.14 CSDHs more than 3 weeks old are encapsulated, crescent-shaped, hypodense collections (0 to 25 Hounsfield units). As the capsule of a CSDH undergoes neoangiogenesis, it enhances with contrast administration. A mixed-density pattern or a fluid level indicate rebleeding into a preexisting CSDH or hygroma.13
Nomura and colleagues15 divided subdural hematomas into five types based on CT appearances: hyperdense, isodense, hypodense, mixed-density, and layering types. The authors demonstrated that the layering type is characterized by a high hyperfibrinolytic activity. The mixed-density type has a lower hyperfibrinolytic activity than the layering type, and the hypodense hematoma has the lowest fibrinolytic activity. Nakaguchi and colleagues16 grouped subdural hematomas into four types: homogeneous-density (hyperdense, hypodense, isodense), laminar (known as mixed-density), separated (known as layering) including gradation subtype, and trabecular (Fig. 137-2). They showed that the separated type of CSDH had a higher recurrence rate (36%) compared to the laminar (19%) and homogeneous (15%) types. Stanisic and colleagues17 confirmed the high recurrence rate of the separated (29%) and laminar (27%) types.
Figure 137-2 Computed tomography scans demonstrating classification of chronic subdural hematomas according to their internal architecture (as proposed by Nakaguchi et al16): A, Homogeneous hypodense. B, Homogeneous isodense. C, Laminar. D, Gradation. E. Separated. F, Trabecular.
MRI of CSDH can better define the multicompartmental nature of certain CSDHs.18 The MRI signal intensity changes for a subdural hematoma follow the pattern of changes of an intracranial hematoma.13 Tsutsumi and colleagues19 attempted to classify CSDHs into five groups according to the intensity of their appearance on T1-weighted MRI: high, mixed high/isointense, isointense, mixed isointense/low, and low. The authors could not demonstrate a correlation between CT classifications and T1-weighted MRI appearances. However, they showed that the recurrence rate of CSDHs that exhibited homogeneous high intensity on T1-weighted images was significantly lower (3.4%) than that in the non–high-intensity groups (11.6%). Tanikawa and colleagues20 found that subdural hematomas divided into multiple layers by an intrahematomal membrane (as demonstrated on T2-weighted MRI) are more effectively treated with craniotomy than burr-hole craniostomy. However, the need for division of membranous septae remains controversial (see the section on craniotomy), and the role of imaging in selecting a surgical technique has not been sufficiently studied.
Treatment
Conservative management is usually reserved for moribund patients with significant comorbidities or asymptomatic patients with small collections.21 Most surgeons would agree that surgical evacuation of a CSDH is indicated for symptomatic patients or collections exerting significant mass effect. Preoperatively, iatrogenic anticoagulation and coagulopathies need to be reversed. With respect to the use of prophylactic antiepileptic medications, there is still not a sufficient evidence base to support their routine use. The many surgical techniques used in the treatment of CSDH can be broadly classified into the following three categories: twist-drill craniostomy (diameter less than 5 mm), burr-hole craniostomy (5 to 30 mm), and craniotomy (more than 30 mm).22 Burr-hole craniostomy is the most popular surgical technique worldwide.22
Burr-Hole Craniostomy
Burr-hole craniostomy is a widely practiced technique with an overall morbidity and recurrence rates of 0% to 9% and 5% to 30%, respectively.1,22 Both the systematic review by Weigel and colleagues22 and a decision analysis model based on the data reported in the literature by Lega and colleagues23 have identified burr-hole craniostomy as the most efficient choice to treat a primary and uncomplicated CSDH because it balances a low recurrence rate against a better morbidity and mortality profile compared to craniotomy and twist-drill craniostomy.
Technique
Two 14-mm burr holes approximately 7 cm apart are drilled over the maximum width of the hematoma (usually one frontal and one parietal, with the exact location determined from the CT scan). The dura mater is opened with a cruciate incision and coagulated with bipolar diathermy. The subdural collection is washed out with warmed Ringer’s solution using a 50-ml syringe until the subdural fluid runs clear. A soft catheter may be used to irrigate distant areas of the hematoma cavity, although we do not consider it always necessary. The subdural membrane loculations accessible via the burr holes are disrupted. A soft silicon drain with three smooth side holes and a blunt tip (external diameter 4.7 mm and length 90 cm) is inserted into the subdural space through the burr hole overlying the large part of the subdural cavity, and the drain is tunneled for a minimum of 5 cm away from the scalp incision.
First, the parietal burr hole is closed in layers to achieve watertight closure. The burr hole may be plugged with a thick wafer of an absorbable sponge (e.g., gelatin sponge) to aid hemostasis and improve seal. The head is manipulated so that the frontal burr hole is at the highest point, the subdural cavity is filled with Ringer’s solution, and the wound is closed, trying to minimize air trapping. The drain is connected to a soft collection bag (subdural-external, no-suction, closed drainage system) that is kept in a dependent position for 48 hours. The drain is then carefully removed and a single stitch is applied at the drain insertion site.2
Discussion
Since the 1990s, evidence has been emerging that the use of drains with burr-hole craniostomy is associated with lower recurrence rates.3,24–27 In the review by Weigel and colleagues, the use of drains was endorsed with type B recommendation.22 The results of a Monte Carlo simulation in the paper by Lega and colleagues suggests a trend toward better outcomes with insertion of an indwelling drain.23
In 2009, we published the results of a randomized, controlled trial of the use of drains versus no drains after burr-hole craniostomy.2 Hematoma recurred in 10 of 108 (9.3%) patients with a drain and in 26 of 107 (24%) without one (p = 0.003). At 6 months, mortality was 9 of 105 (8.6%) and 19 of 105 (18.1%), respectively (p = 0.042). Medical and surgical complications did not differ between the two arms. In accordance with the aforementioned studies, the results of this trial provide strong evidence for the use of drains with burr-hole craniostomy.
Several groups have attempted to address the role of irrigation in the treatment of CSDH and have demonstrated a trend toward reduced recurrence with the use of intraoperative irrigation.28–31 Only Kuroki and colleagues32 found more than six times higher recurrence (p = 0.49) in cases with irrigation (5/45, 11.1%) than without irrigation (1/55, 1.8%). Conversely, Aoki33 in a study of twist-drill craniostomies found a lower recurrence rate in the cases where intraoperative irrigation was used (1/15 vs 7/24, p = 0.096). None of the studies cited here were sufficiently powered to demonstrate significant differences.
Most surgeons use two burr holes, mainly because this allows better wash out of the subdural cavity. Taussky and colleagues34 observed a higher recurrence rate if one rather than two burr holes was used. By contrast, in the series by Han and colleagues35 the recurrence rate was 1/51 (2%) with one burr hole and 9/129 (7%) with two (p