J Neurosurg 87:870–875, 1997

The relationship of preoperative magnetic resonance imaging

findings and closed system drainage in the recurrence of
chronic subdural hematoma
KAZUO TSUTSUMI, M.D., KEIICHIROU MAEDA, M.D., AKIRA IIJIMA, M.D.,
MASAAKI USUI, M.D., YOSHIHUMI OKADA, M.D., AND TAKAAKI KIRINO, M.D.
Department of Neurosurgery, Aizu Chuou Hospital, Aizuwakamatsu, Fukushima, Japan; and Department
of Neurosurgery, University of Tokyo, Tokyo, Japan

U Although chronic subdural hematoma (CSDH) is a well-known entity, its recurrence rate has remained uncertain. There
is little knowledge concerning whether the results of radiological imaging can be used to predict CSDH recurrence or
whether surgical methods can influence this rate. The first aim of this study is to evaluate the relationship between the
recurrence rate of CSDHs and their appearance on preoperative magnetic resonance (MR) or computerized tomography
images. The second aim is to evaluate by means of a prospective randomized method the usefulness of closed-system
drainage.
From January 1988 through June 1996, the authors surgically treated 257 consecutive adult patients with CSDHs. Data
obtained in 199 patients who were evaluated preoperatively by MR imaging were analyzed. Thirty-one of these patients
underwent bilateral operations and thus 230 operative sites of CSDH were included in the analyses. The cases of CSDH
were separated into high- and nonhigh-intensity groups on the basis of the appearance on T1-weighted MR images. From
July 1992 to June 1996, the authors conducted a prospective randomized study on the recurrence rate of CSDH in patients
undergoing burr-hole irrigation with or without closed system drainage.
The recurrence rate of 3.4% in the high-intensity group was significantly lower than the 11.6% rate found in the non-
high-intensity group (p , 0.05). The recurrence rates following irrigation with and without closed system drainage were
significantly different (p , 0.025): 3.1% with closed system drainage and 17% following burr-hole irrigation alone. The
surgical procedures were correlated with the MR findings. In the high-intensity group, 1.1% of CSDHs recurred in patients
in whom closed system drainage was used and 11.1% in patients without closed system drainage. In the nonhigh-intensi-
ty group, 8.1% of CSDHs recurred in patients in whom drainage was used and 23.1% in patients without closed system
drainage.
Magnetic resonance T1-weighted imaging was useful in predicting the propensity of CSDHs to recur. Closed system
drainage significantly reduced the recurrence rate of CSDHs regardless of MR findings.

KEY WORDS • chronic subdural hematoma • magnetic resonance imaging •

computerized tomography • recurrence • closed system subdural drainage

LTHOUGH chronic subdural hematoma (CSDH) is Clinical Material and Methods

A well known as a curable disease in the elderly, it
remains uncertain whether the recurrence rate of
CSDH can be predicted by radiological imaging or im-
proved by surgical method.2,8,9,18,20,21 To our knowledge,
there is insufficient information in the literature to deter-
mine these issues.
It was our aim in this study to evaluate the relationship
between the recurrence rate of CSDHs in adult patients
and the appearance of these lesions on preoperative mag-
netic resonance (MR) or computerized tomography (CT)
images. We also conducted a prospective randomized
study on the recurrence rate of CSDH by comparing two
different surgical methods: one burr-hole irrigation of the
hematoma cavity with or without closed system drainage.
Patient Population
From January 1988, when the MR imaging system was
installed in Aizu Chuou Hospital, through June 1996 we
surgically treated 257 consecutive adult patients who had
a symptomatic CSDH. In 199 of these patients, we evalu-
ated preoperative MR images that were obtained on the
day of, or 1 day prior to, surgery. Because 31 patients un-
derwent bilateral operations, we analyzed a total of 230
operative sites of CSDH. The age and sex distribution of
the patients is shown in Table 1. The mean age of the pa-
tients was 67 years (range 19–92 years). One hundred thir-
ty-two patients (66%) had a history of head trauma, most
of which were minor.

870 J. Neurosurg. / Volume 87 / December, 1997

Magnetic resonance imaging and drainage in CSDH
TABLE 1
Age and sex distribution in 199 patients with CSDH
Age (yrs)
Sex 10–19 20–29 30–39 40–49
male 1 1 1 9
female 0 0 0 2
total no. of cases 1 1 1 11
no. of recurrent cases 0 1 0 0
Chronic subdural hematoma was defined as including:
1) the presence of a typical neomembrane; 2) typical liqui-
fied blood within the hematoma cavity; and 3) if following
acute SDH, at least 3 weeks had passed. All CSDHs were
confirmed during surgery. Hygromas, infantile CSDHs,
calcified or ossified CSDHs, and asymptomatic CSDHs
were excluded from this study.
Chronic subdural hematoma was considered to have re-
curred when neurological symptoms reappeared within
6 months after surgery and the hematoma cavity volume
on the operative side increased. All patients who fulfilled
both criteria underwent repeated operation. Reaccumu-
lation of blood within the hematoma cavity without ac-
companying symptoms and contralateral CSDH following
the first operation were not recognized as recurrence in
this study.
The CSDHs were classified into five groups according
to their density and appearance on CT scans: high, iso-,
low, mixed, and layered (niveau). Isodensity was defined
as equal or near-equal density to that of the gray matter.
Three cases were not included in this study because two
patients had been examined 3 days or more before the op-
eration and one patient’s record was lost.
All CSDHs were evaluated preoperatively by MR im-
aging (0.5 tesla) on the day of, or 1 day before, surgery.
Images were obtained using a spin-echo sequence with
repetition/echo times of 500/30 msec for T1-weighted,
2000/30 or 2500/40 msec for proton density–, and 2000/
120 or 2500/80 msec for T2-weighted images. On proton
density– and T2-weighted images, most CSDHs were
clearly demonstrated as high-intensity areas and rarely as
a mixture of iso-/low-intensity areas. Approximately one-
half of the CSDHs also were demonstrated as high-in-
tensity areas on T1-weighted images. The others, howev-
er, exhibited variable intensity and appearance. Two
hundred thirty operative sites of CSDH were classified
into five groups according to their MR intensity and ap-
pearance on T1-weighted images: high, mixed high/iso-,
iso-, mixed iso-/low, and low intensity (Fig. 1). Isoin-
tensity was defined as the intensity of the lesion on MR
imaging appearing equal to the intensity of the brain
(ranged from gray matter to white matter).
Surgical Methods
From January 1988 to June 1992, we used one burr-hole
irrigation of the hematoma cavity accompanied by closed
system drainage in our patients. From July 1992 to June
1996, we conducted a prospective randomized study on
the recurrence rate of CSDH based on two different meth-
ods: one burr-hole irrigation of the hematoma cavity with
J. Neurosurg. / Volume 87 / December, 1997
50–59 60–69 70–79 80–89 $90 Totals
33 42 54 14 1 156 (78.4%)
9 11 13 7 1 43 (21.6%)
42 53 67 21 2 199
4 4 6 1 0 16 (8.0%)
and without closed system drainage. After obtaining the
consent of the patients, the randomized selection of surgi-
cal methods (that is, whether to use drainage or not) was
decided by the toss of a coin. In cases that required bilat-
eral surgery, we decided to use the same surgical method
for each side. Exceptions to this selection process includ-
ed three patients who did not suffer recurrence and were
treated without randomization: two patients who showed
a tendency to bleed (hemodialysis and ticlopidine intake)
were treated with drainage and one patient in whom MR
imaging revealed a contralateral small CSDH was treated
without drainage. In addition, one of the 31 patients with
bilateral CSDHs was treated on the right side with drain-
age and on the left without drainage. Fifty-three patients
were treated with drainage and 37 were treated without
drainage. The drain (SL-C ventricular catheter; Sonne-Ika
Co., Tokyo, Japan) placed within the cavity was connect-
ed to a ventricular drainage bag with an antireflux valve
(Hanaco Medical Co., Saitama, Japan), which was placed
on the bed for approximately 1 day, in the longest instance
for 3 days.
Follow Up and Statistical Analysis
Every patient received follow-up care for at least 6
months. Statistical analysis was performed using the chi-
square test, Student’s t-test, and analysis of variance. Sta-
tistical significance was assumed if probability was mea-
sured at less than 0.05.
Results
Chronic subdural hematoma recurred in 16 patients at
17 operative sites (Table 2). The age distribution in these
patients was similar to that of all patients (Table 1). Four
(12.9%) of 31 patients with bilateral CSDH experienced
recurrence and underwent reoperation. One of these
patients displayed bilateral recurrence. The difference be-
tween recurrences in patients with bilateral (five [8.1%] of
62) and unilateral (12 [7.1%] of 168) sites of CSDH was
not significant. Recurrence was demonstrated in three
(18.8%) of 16 patients who had a suspected bleeding ten-
dency (that is, patients who had restarted a medication
regimen of warfarin sodium or ticlopidine, required hemo-
dialysis, or had liver cirrhosis). Chronic subdural hema-
toma recurred in three (13%) of 23 patients with suspect-
ed bleeding tendency and 14 (6.8%) of 207 of those
without such a tendency. These differences did not reach
statistical significance in these small groups.
Two hundred twenty-seven CSDHs were classified into
five groups according to their density on CT scan: high
871

FIG. 1. Magnetic resonance T1-weighted images demonstrating examples of five types of CSDH: high (A), iso- (B),
low (C), mixed high/iso- (D), and mixed iso-/low (E) intensity. Isointensity means equal or near-equal intensity to that
of the brain.
(55 cases), iso- (79 cases), low (26 cases), mixed (56
cases), and layered (11 cases). The incidences of recur-
rence were three, six, three, three, and two in each group,
respectively (Table 3). The recurrence rate in the layered
group was higher than that found in other groups, but the
difference was not statistically significant in this small
group.
Two hundred thirty CSDHs were classified into five
groups according to the intensity of their appearance on
T1-weighted MR images: high (118 cases), mixed high/
iso- (46 cases), iso- (32 cases), mixed iso-/low (16 cases),
and low (18 cases). The incidences of recurrence were
four, four, three, three, and three in each group, respec-
tively (Table 3). The recurrence rate in the high-intensity
group (four [3.4%] of 118) was much lower than that in
other groups. When the other groups were considered as
one group (nonhigh-intensity group), there were no statis-
tically significant differences between the two groups with
regard to age, surgical methods, bilateral operations, or
suspected bleeding tendency (Table 4). In the non-
high-intensity group, the recurrence rate was 11.6%.
The difference in recurrence rates observed between the
two groups was statistically significant (chi-square test
p , 0.05).
In the prospective study that was undertaken from July
1992 to June 1996, the recurrence rates of CSDH follow-
ing the surgeon’s use of two different surgical methods
were clearly different. The rate was 3.1% (two of 65
cases) following one burr-hole irrigation with closed sys-
872
K. Tsutsumi, et al.
tem drainage, but increased to 17% (nine of 53 cases) fol-
lowing one burr-hole irrigation alone (chi-square test, p ,
0.025). There were no statistically significant differences
between the two groups, with and without closed system
drainage, with regard to patient age, bilateral operations,
suspected bleeding tendency, or intensity on T1-weighted
MR images (Table 5).
From January 1988 to June 1992, six (5.4%) of 112
CSDH sites subjected to a single operative method (one
burr-hole irrigation with closed system drainage) exhibit-
ed recurrence. The operative methods were correlated
with MR findings. In the high-intensity group, 1.1% of
CSDHs recurred in patients in whom closed system drain-
age was used and 11.1% in patients without drainage. In
the nonhigh-intensity group, 8.1% of CSDHs recurred in
patients in whom closed system drainage was used and
23.1% in patients without drainage. Regardless of the
findings on MR imaging, the use of closed system drain-
age significantly reduced the risk of recurrence (Table 6).
Discussion
Recurrences of CSDH
As risk factors of recurrence for CSDHs, bleeding ten-
dency and intracranial hypotension (for example, shunt
operation and cerebrospinal fluid [CSF] leakage) are well
known.1,2 However, most of our patients who suffered
recurrence had no such risk. Older age of the patient and
bilateral sites of CSDH have also been considered to be
J. Neurosurg. / Volume 87 / December, 1997
Magnetic resonance imaging and drainage in CSDH

TABLE 2
Clinical characteristics of patients with recurrent CSDH*
Age Recurrence
Case (yrs), CSDH T1-Weighted Density on
No. Sex Site Risk Factor MR Imaging CT Scanning Drainage Symptom(s) Day of Reop

1 66, M lt none mixed H&I iso- 1 rt hemiparesis 24

2 51, F rt none mixed H&I mixed 1 headache & nausea 18
3 76, M bilat ticlopidine mixed I&L (bilat) iso- (bilat) 1 confusion 46 (lt)
4 59, M rt none iso- layered 1 lt hemiparesis 21
5 77, M rt none mixed H&I mixed 1 diplopia & headache 17
6 24, M bilat none iso- (lt), low (rt) high (bilat) 1 headache 68 (lt)
7 72, F lt none mixed H&I mixed 2 gait disturbance 48
8 75, M lt none high low 1 epilepsy 76
9 57, M bilat ticlopidine mixed I&L (bilat) iso- (bilat) 6 rt hemiparesis 30 (lt)
10 68, F rt liver cirrhosis low low 2 headache 13
11 81, M rt none mixed I&L layered 1 lt hemiparesis 33
12 60, F bilat none low (bilat) iso- (bilat) 2 headache 18 (bilat)
13 58, M rt none high high 2 lt hemiparesis 70
14 76, M lt none iso- high 2 rt hemiplegia 18
15 67, M lt none high low 2 rt hemiparesis 19
16 77, M lt none high iso- 2 gait disturbance 80
* H&I = high and iso-; I&L = iso- and low; 1 = present; 2 = absent; 6 = each side of CSDH was treated differently (with [rt] and
without [lt] drainage).

risks for recurrence by some authors.14,16,21 In this study,
older age was not a risk factor for recurrence. It seems that
the older the patient, the longer it takes for the brain to be
restored. In studies found in the literature, it is possible
that prolonged reaccumulations of blood within the hema-
toma cavity might have been misunderstood and unneces-
sarily subjected to reoperation. In the patients with bilat-
eral CSDH, the rate of recurrence (12.9%) was higher than
that of patients with unilateral CSDH. With respect to the
recurrence rate of operative sites, however, the difference
between bilateral (five [8.1%] of 62) and unilateral (12
[7.1%] of 168) CSDH was not significant.
In a discussion of the reasons for and quantification of
CSDH recurrences, the most important points to be de-
fined involve the nature of the hematoma and the method-
ology used to judge the recurrence. In this study, hy-
gromas, infantile CSDH, calcified or ossified CSDH, and
asymptomatic CSDH were excluded because they were
considered to be different clinical entities. After surgery
for CSDH, reaccumulation of blood in the hematoma cav-
ity often spontaneously disappears regardless of its vol-
ume. Reoperation, therefore, is not indicated for recur-
rences in patients not showing positive symptoms.11 Our
postoperative management strategy for asymptomatic pa-
tients in whom CT scanning revealed persistent CSDH
was to evaluate them at regular intervals over several
months to monitor the appearance of any clinical symp-
toms. In this study, each site of bilateral CSDH was
considered a single CSDH, although bilateral CSDH was
distinguished from unilateral CSDH. Because of this clas-
sification mechanism, some bilateral CSDHs showed ra-
diological differences from one site to the other and in
some there were no consistent findings between the sites.
Role of CT Scanning
After the introduction of CT scanning, outcome of
CSDHs became much improved.15 In our series of 257 pa-
tients, the postoperative mortality rate (within 1 month
after surgery) was 0%. Computerized tomography scan-
ning remains the most important diagnostic test for
CSDH, although MR imaging has an advantage over CT
scanning with respect to the quality of the images.

TABLE 3 TABLE 4
Relationship between recurrence rates and findings on Comparison of potential risk factors and recurrence rates of
preoperative MR T1-weighted images and CT scans CSDH in 230 patients according to findings on
in 230 patients with CSDH* preoperative MR T1- weighted imaging

Intensity Density on CT Scans (no. of recurrences per op site) Op Method

on MR Intensity No. Mean Bleeding CSDH Re-
Images High Iso- Low Mixed Layered NA Total on MR of Age W/ W/O Bilat Ten- currence
Images Patients (yrs) Drain Drain Ops dency Rate
high 1 of 25 1 of 52 2 of 18 0 of 17 0 of 3 0 of 3 4 of 118
mixed H/I 0 of 7 1 of 3 0 of 3 3 of 33 0 of 0 — 4 of 46 high 118 66.9 91 27 24 11 4 of 118
iso 2 of 17 0 of 9 0 of 0 0 of 4 1 of 2 — 3 of 32 (3.4%)
mixed I/L 0 of 5 2 of 4 0 of 0 0 of 1 1 of 6 — 3 of 16 nonhigh 112 67.1 86 26 38 12 13 of 112
low 0 of 1 2 of 11 1 of 5 0 of 1 0 of 0 — 3 of 18 (11.6%)
total 3 of 55 6 of 79 3 of 26 3 of 56 2 of 11 0 of 3 17 of 230 p value* NS NS NS NS NS ,0.05
* H&I = high and iso-; I&L = iso- and low; NA = not available; — = not * Comparison between high-intensity group and nonhigh-intensity
applicable. group. Abbreviation: NS = not significant.

J. Neurosurg. / Volume 87 / December, 1997 873


TABLE 5
Comparison of potential risk factors and recurrence rates of
CSDH in 118 patients treated with one burr-hole
irrigation with or without drainage
Intensity on MR
No. Mean (T1) Images Bleeding CSDH Re-
of Age Bilat Ten-
Op Group Patients (yrs) High Nonhigh Ops dency
w/ drainage 65 68.0 36 29 13 9 2 of 65
w/o drainage 53 69.7 27 26 17 9 9 of 53
p value* NS NS NS NS NS
* Comparison between two groups with or without drainage. Abbrevia-
tion: NS = not significant.
Chronic subdural hematoma is often classified accord-
ing to CT findings; however, these classification systems
can be subjective. Nomura, et al.,13 biochemically ana-
lyzed hematoma samples and discussed the relationship
between CT classifications and the coagulative, fibrin-
olytic activities of each type of CSDH. They concluded
that mixed- and layered-density hematomas had a greater
tendency to rebleed and that the low-density type had less
tendency to rebleed. Unfortunately, these authors have
never reported the clinical outcomes based on their hy-
potheses. In our study, the recurrence rate in the layered-
density group was higher than that found in other groups;
however, the difference was not statistically significant in
this small group. It seems that prediction of recurrence
based on CT scanning has little relation to actual rates of
CSDH recurrence. We could not find a positive correla-
tion between CT classifications and T1-weighted MR im-
ages: homogeneous areas of high intensity on T1-weight-
ed MR images were present in two-thirds of the iso- and
low-density types of CSDH, whereas they were present in
approximately 40% of mixed- and layered-density types
(Table 3).
Role of MR Imaging
The advantages of MR imaging over CT scanning in
the evaluation of CSDHs have already been shown, espe-
cially in cases of isodense and bilateral CSDHs.3,17 One
disadvantage, however, is that MR imaging is a time-
consuming, costly examination. In this series, 58 patients
with CSDHs (including four recurrences and 67 opera-
tive sites) were excluded from MR examination because
the study was contraindicated (cardiac pacemakers, metal
implants, emergency operations) or MR imaging was not
available.
In general, CSDHs displayed more hyperintensity than
normal brain tissue on both T1- and T2-weighted MR
images.3,17,21 Also in this study, most CSDHs were clearly
demonstrated as high-intensity areas on proton density–
and T2-weighted images and rarely as a mixture of iso-/
low-intensity areas. Because CSF is demonstrated as rela-
tively low in intensity on proton density–weighted
images, it was very helpful to distinguish CSDH from
hygroma, after subdural fluid collection was shown to be
similar to CSF on CT and on T1- and T2-weighted MR
874
K. Tsutsumi, et al.
TABLE 6
Relationship between recurrence rates and combinations of
preoperative findings on MR T1-weighted imaging and
operative methods in 230 patients with CSDH*
MR Finding/ January 1988– July 1992– January 1988–
Op Method June 1992 June 1996 June 1996
currence
Rate high intensity
w/ drainage 0 of 55 1 of 36 1 of 91 (1.1%)
w/o drainage — 3 of 27 3 of 27 (11.1%)
(3.1%) nonhigh intensity
w/ drainage 6 of 57 1 of 29 7 of 86 (8.1%)
(17%) w/o drainage — 6 of 26 6 of 26 (23.1%)
,0.025
* — = not applicable.
images. We had surgically treated some patients in whom
lesions appeared as low-intensity areas on proton densi-
ty–weighted images; all of these patients were confirmed
to have subdural collections of CSF or xanthochromia
without a neomembrane. On T1-weighted images, approx-
imately one-half of CSDHs also appeared as homoge-
neous high-intensity areas. The others, however, were
shown to have variable intensity and appearance. Hosoda
and colleagues3 showed that 30% (six of 20 cases) of
CSDHs were iso- or hypointense on T1-weighted images;
this is similar to our findings shown in Table 3. Their six
cases, however, included three infants. The authors sug-
gested that these iso- or hypointense areas on T1-weighted
images indicated fresh rebleeding into the hematoma cav-
ity. The presence of high-intensity areas on these images
suggested that several days had passed since the last bleed
and that most of the hemoglobin had changed to free
methemoglobin. It is believed that a late increase in the
size of the CSDH occurs because of repeated microhem-
orrhages from the neocapillary network in the outer mem-
brane.9 If that were the only cause, most CSDHs would
appear as the mixed type including low-intensity or isoin-
tense areas on T1-weighted images. The finding that
approximately one-half of CSDHs appeared as homoge-
neous high-intensity areas on T1-weighted images sug-
gested that another mechanism, not involving hemorrhage
(red blood cells), might be the cause of the enlargement of
CSDH; for example, transportation of plasma and/or CSF
by chemical irritation of the hematoma and/or its product.
In this study, the recurrence rate of CSDHs that exhib-
ited homogeneous high intensity on T1-weighted images
was much lower (four of 118 [3.4%]) than that in the
nonhigh-intensity groups (13 of 112 [11.6%]) (p , 0.05).
Moreover, the recurrences seemed to be more delayed in
the high-intensity group (median 61.3 days after surgery)
than in the nonhigh-intensity groups (median 28.6 days
after surgery). Chronic subdural hematomas are presumed
to have a life history extending from a proliferative to a
degenerative stage on the basis of histological analysis of
the hematoma membrane and periodic follow-up CT
scans showing spontaneously resolving CSDHs.12 The
main cause in the early proliferative stage seems to
be repeated microhemorrhages from microvessels of the
neomembrane. If fresh rebleeding develops in patients,
the fresh component is demonstrated as low or isointensi-
ty on T1-weighted MR imaging. In this stage, microvessels
of neomembrane may be more vulnerable, easily rebleed,
J. Neurosurg. / Volume 87 / December, 1997
Magnetic resonance imaging and drainage in CSDH
and be prone to recur regardless of operative methods. On
the contrary, CSDHs with homogeneous high intensity on
T1-weighted images might enlarge by another mechanism
as mentioned earlier. It seems that this condition is the late
proliferative stage and that the hematoma membrane is
more mature and less likely to rebleed. Therefore, CSDHs
in the high-intensity group may be less prone to recur than
those in nonhigh-intensity groups. Magnetic resonance T1-
weighted imaging could provide us with some predictions
about recurrences of CSDH.
Operative Methods
Recently burr-hole surgery has become established as
the treatment of CSDH rather than craniotomy; however,
controversy still exists whether drainage should be used
after irrigation of the CSDH, although many authors have
recommended drainage.2,5,7,10,19,20 Thus far, there has been
only one attempt to clarify the actual benefit of drainage
by a prospective comparative study. Wakai, et al.,20 con-
ducted a prospective comparative study on this issue and
concluded that closed system drainage after burr-hole irri-
gation reduced the recurrence rate of CSDH. In their
study, however, the recurrence rate in patients treated
without drainage (six [33%] of 18 cases) is much higher
than that given in other reports including ours (eight
[22%] of 37 cases), although they excluded patients with
coagulopathy. In addition, all recurrences occurred within
1 month after surgery. Unfortunately, the conclusion of
their study is limited because the number of the patients
was small (38 cases).
In our prospective study, it was confirmed that closed
system drainage significantly reduced the recurrence rate
of CSDH. Most of the drains were kept in place for 1 day,
with the longest in place for 3 days. As to the duration of
drainage to be prescribed, we agree with Wakai, et al.,20
that a long period of drainage may increase the risk of
infection. Some authors reported cases of tension pneu-
mocephalus as a disadvantage of placing closed system
drainage.4,6 This rare complication, however, can be pre-
vented by caution not to tear the inner membrane of the
CSDH and arachnoid membrane. None of our patients
with closed system drainage suffered tension pneumo-
cephalus or infection related to surgery.
In a comparison of operative methods, it seemed that
CSDHs that showed high intensity on T1-weighted im-
ages rarely recurred if they were drained (1.1%) and
that CSDHs that exhibited other appearances and intensi-
ty on the images often recurred if they were not drained
(23.1%). Regardless of the MR findings, closed system
drainage was significantly effective in reducing the risk
of recurrences. Nonetheless, one burr-hole irrigation with
closed system drainage might be insufficient for the treat-
ment of CSDHs that do not demonstrate high intensity on
T1-weighted images.
References
1. Arbit E, Patterson RH Jr, Fraser RAR: An implantable subdu-
ral drain for treatment of chronic subdural hematoma. Surg
Neurol 15:175–177, 1981
J. Neurosurg. / Volume 87 / December, 1997
2. Asano Y, Hasuo M, Takahashi I, et al: [Recurrent cases of
chronic subdural hematoma—its clinical review and serial CT
findings.] No Shinkei Geka 44:827–831, 1992 (Jpn)
3. Hosoda K, Tamaki N, Masumura M, et al: Magnetic resonance
images of chronic subdural hematomas. J Neurosurg 67:
677–683, 1987
4. Ishiwata Y, Fujino H, Kubokura T, et al: [Subdural tension
pneumocephalus following surgery for chronic subdural hema-
toma.] No Shinkei Geka 15:419–424, 1987 (Jpn)
5. Iwadate Y, Ishige N, Hosoi Y: [Single burr hole irrigation with-
out drainage for chronic subdural hematoma.] Neurol Med
Chir 29:117–121, 1989 (Jpn)
6. Kawakami Y, Tamiya T, Shimamura Y, et al: [Tension pneu-
mocephalus following surgical evacuation of chronic subdural
hematoma.] No Shinkei Geka 13:833–837, 1985 (Jpn)
7. Kotwica Z, Brzeziński J: Chronic subdural haematoma treated
by burr holes and closed system drainage: personal experience
in 131 patients. Br J Neurosurg 5:461–465, 1991
8. Laumer R, Schramm J, Leykauf K: Implantation of a reservoir
for recurrent subdural hematoma drainage. Neurosurgery 25:
991–996, 1989
9. Markwalder TM: Chronic subdural hematomas: a review. J
Neurosurg 54:637–645, 1981
10. Markwalder TM, Seiler RW: Chronic subdural hematomas: to
drain or not to drain? Neurosurgery 16:185–188, 1985
11. Markwalder TM, Steinsiepe KF, Rohner M, et al: The course of
chronic subdural hematomas after burr-hole craniostomy and
closed-system drainage. J Neurosurg 55:390–396, 1981
12. Naganuma H, Fukamachi A, Kawakami M, et al: Spontaneous
resolution of chronic subdural hematoma. Neurosurgery 19:
794–798, 1986
13. Nomura S, Kashiwagi S, Fujisawa H, et al: Characterization of
local hyperfibrinolysis in chronic subdural hematomas by SDS-
PAGE and immunoblot. J Neurosurg 81:910–913, 1994
14. Probst C: Peritoneal drainage of chronic subdural hematomas in
older patients. J Neurosurg 68:908–911, 1988
15. Richter HP, Klein HJ, Schäfer M: Chronic subdural haema-
tomas treated by enlarged burr-hole craniotomy and closed sys-
tem drainage. Retrospective study of 120 patients. Acta Neu-
rochir 71:179–188, 1984
16. Robinson RG: Chronic subdural hematoma: surgical manage-
ment in 133 patients. J Neurosurg 61:263–268, 1984
17. Sipponen JT, Sepponen RE, Sivula A: Chronic subdural hema-
toma: demonstration by magnetic resonance. Radiology 150:
79–85, 1984
18. Tabaddor K, Shulman K: Definitive treatment of chronic sub-
dural hematoma by twist-drill craniostomy and closed-system
drainage. J Neurosurg 46:220–226, 1977
19. Tyson G, Strachan WE, Newman P, et al: The role of craniec-
tomy in the treatment of chronic subdural hematomas. J Neu-
rosurg 52:776–781, 1980
20. Wakai S, Hashimoto K, Watanabe N, et al: Efficacy of closed-
system drainage in treating chronic subdural hematoma: a
prospective comparative study. Neurosurgery 26:771–773,
1990
21. Yamagami T, Higashi K, Handa H, et al: [Magnetic resonance
imaging of chronic subdural hematoma, especially in recurrent
cases.] Progr CT 15:27–33, 1993 (Jpn)
Manuscript received April 1, 1997.
Accepted in final form August 1, 1997.
Address reprint requests to: Kazuo Tsutsumi, M.D., Department
of Neurosurgery, Aizu Chuou Hospital, 1-1, Tsuruga, Aizuwaka-
matsu, Fukushima 965, Japan.
875