Introduction
Intraventricular hemorrhage (IVH) occurs in 10%~45% of patients with ruptured aneurysms.13)25)30) The overall mortality of patients who have ruptured aneurysms associated with IVH has been reported to be 24%-64%.13)25)30) The autopsy findings
of patients with aneurysmal subarachnoid hemorrhages (SAHs) who never reach medical attention include IVH, pulmonary edema, and ruptured posterior circulation aneurysms.33) Intracerebral hemorrhage (ICH) is infrequently
associated with sudden death from aneurysmal SAH.33) IVH in patients with ruptured aneurysms has been associated with poor clinical and radiologic findings, and poor outcome; however, the precise relationship between IVH and these conditions remains unclear. There have been few systematic studies of IVH related to aneurysmal rupture. In addition, there have been few reports focusing on the effect of IVH on outcomes in patients who underwent clipping for ruptured aneurysms.13)24)25)30) The purpose of this study was to determine
the clinical significance of IVH and the effect of IVH on surgical outcome in patients with ruptured aneurysms.
Materials and Methods
Between 1994 and 2007, 1034 patients with ruptured aneurysms were enrolled in our SAH data bank. All patients with ruptured aneurysms were confirmed by conventional transfemoral 4-vessel angiography or 3-dimensional
computed tomographic angiography. Exclusion criteria were as follows: SAH resulting from unknown etiology, patients discharged early prior to definitive treatment, or patients treated with coil embolization. Because the number
of patients treated with coil embolization was too small to include in this study (n = 32 patients) and 2007 was the learning period for coil embolization, we excluded such patients from this study to avoid outcome bias.
Of 1034 patients with ruptured aneurysms, 128 (12.4%) had IVHs. Seventy-two patients (56%) had IVHs involving all 4 ventricles, which was designated as “ll IVH,”whereas 56 (44%) had IVHs involving 1~3 ventricles, which was
designated as “artial IVH.”All patients were divided into two groups (IVH and no-IVH). The clinical and radiologic characteristics of patients with IVHs were compared to patients without IVHs. The clinical parameters for comparison between the two groups were as follows: age, gender, time interval from
onset-to-admission, blood pressure on admission, medical history (e.g., hypertension and alcohol abuse), s-glucose and cholesterol levels on admission, Glasgow Coma Scale (GCS) score and Hunt and Hess grade (HHG) after at least 6 hours from onset, rebleeding or seizures before clipping or during
conservative management, and the incidence of shuntdependant hydrocephalus. The radiologic parameters were as follows: bifrontal index (BFI) on initial computed tomographic (CT) scan, location of ruptured aneurysms, patterns of SAH distribution, and ICH > 10 cc on initial CT scan. Nine hundred thirty-two of 1034 patients (90%) underwent clipping for ruptured aneurysms, of whom 85
(9%) had IVHs on the initial CT scan. One hundred two patients could not be treated by craniotomy because of a moribund state after rebleeding in 32 patients, initial SAH in 68 patients, and fusiform aneurysms in 2 patients.
The 3-month outcomes after surgery were compared between the IVH and no-IVH groups according to the clinical variables.
Patient age was divided into two groups (> or < 70 years). Hypertension history was classified into two categories (treated or untreated). Alcohol abuse was defined as consumption of ethyl-alcohol > 5 times per week. GCS and
HHG were divided into two categories each (GCS, >8 or 8; and HHG, 4-5 or 1~3). Shunt-dependent hydrocephalus was defined as required shunt surgery within 3 months after the initial SAH. The location of ruptured aneurysms was
divided into the following four groups: internal carotid artery (ICA), anterior cerebral artery (ACA), middle cerebral artery (MCA), and vertebrobasilar artery (VBA). The ICA aneurysms include all of the aneurysms arising from the
paraclinoid portion of the ICA, posterior communicating artery, anterior choroidal artery, and ICA bifurcation. The patterns of SAH on initial CT scans were divided into the following four types: thick diffuse, thick focal, thin diffuse,
and thin focal. Thick or thin SAHs were defined as > or < 1 mm in actual thickness of the SAH, respectively. Diffuse or
focal SAHs were defined as a bilateral distribution of SAH or SAH in the interhemispheric fissure and/or unilateral distribution of SAH, respectively. All patients who did not have ICH on the initial CT scan were included in the group of ICHs with < 10 cc volume. The 3-month outcome after
surgery was divided into 2 groups (good and poor). Patients with good outcomes were defined as those with excellent or good Glasgow Outcome Scale (GOS) scores (5 or 4), whereas poor outcome was defined as those with fair, poor,
or dead GOS scores (3, 2, or 1). Data were analyzed with SPSS software for Windows (version 11.0.1; SPSS, Inc., Chicago, IL, USA). Differences
between the IVH and no-IVH groups were assessed by an independent t-test for continuous variables or chi-square test for categorical variables. A p value <0.05 was considered significant.
Results
Table 1 shows the mean difference between the IVH and no-IVH groups in the patients with ruptured aneurysms. The mean age was 2 year older in the IVH group than the no-
IVH group. The mean time interval of time from onset of symptoms to admission was shorter in the IVH group than the no-IVH group. In the IVH group, the systolic and diastolic blood pressures and the s-glucose level at the time of admission were significantly higher than the no-IVH group. The GCS scores and HHGs reflected poor clinical states in the IVH group compared with the no-IVH group. The BFI on the initial CT scan was also greater in the IVH group than no-IVH group. All of the parameters, except the s-cholesterol level, suggested poor clinical status in patients in the IVH group compared with the patients in the no-IVH
group (Table 1). The clinical and radiologic characteristics of the 1034
patients with ruptured aneurysms are shown in Tables 2, 3, and 4. The risk of IVH in patients with ruptured aneurysms was significantly higher in patients > 70 years of age (p=0.030), males (p=0.000), alcohol abusers (p=0.039), patients with a GCS 8 (p=0.000), and patients with HHGs
of 4-5 (p=0.000) based on the chi-square test. The risk of IVH was approximately 8 times higher in patients with GCSs 8 and 7 times higher in patients with HHGs of 4~5
than patients with GCSs > 8 and HHGs of 1~3, respectively. The incidence of rebleeding (p=0.010) and shunt-dependant hydrocephalus (p=0.000) was higher in the IVH group
compared with the no-IVH group. A history of hypertension and seizures was more common in the IVH group than the no-IVH group, but the differences were not statistically
significant (Table 2). The overall mortality was 2 times higher in the partial IVH group (p=0.004) and 6.6 times higher in the all IVH group (p=000) than the no-IVH group
(Table 3). The incidence of IVH based on the locations of ruptured
aneurysms was most common in the VBA (p=0.000), followed by the ACA (p=0.036). In VBA and ACA aneurysms, the risk of IVH was 4.5 and 1.5 times higher than the non-VBA and non-ACA aneurysms, respectively. However, in patients with MCA or ICA aneurysms, the risk
of occurrence of IVH was significantly lower than patients with other aneurysms. The risk of occurrence of IVH based on the SAH distribution was 1.9-fold higher in the thick,
diffuse pattern (p=0.004) compared with the other SAH patterns, and 2.3-fold higher in the thick, focal pattern (p=0.003) compared with the other patterns. In patients with an ICH volume > 10 cc (p=0.010), the risk of occurrence of IVH was significantly higher than the patients with an ICH volume < 10 cc (Table 3). In the patients who underwent clipping, IVHs influenced
the poor outcome in patients < 70 years and 70 years of age. An IVH significantly influenced the poor outcome in patients with a GCS > 8 without pre-operative rebleeding,
or seizures, and in patients with an ICH volume < 10 cc on the initial CT scan. However, an IVH did not affect the poor outcome in patients with a GCS 8, pre-operative
rebleeding, seizures, and in patients with an ICH volume 10 cc (Table 4). These findings suggest that an IVH does not affect the outcome in patients with a poor clinical status preoperatively.
Discussion
The results of the current study showed that the incidence of IVH in patients with ruptured aneurysms increased in patients > 70 years of age, males, alcohol abusers, patients with a GCS 8, and patients with HHGs of 4~5. The time interval from initial hemorrhage to admission was shorter and the blood pressure and s-glucose level were higher in the IVH group than the no-IVH group. The rate of pre-operative rebleeding was also significantly higher in the IVH group than the no-IVH group. The incidence of SAHs with a thick, diffuse pattern and an ICH volume 10cc was higher in the IVH group than the no-IVH group. The overall mortality was highest in patients with an IVH involving all ventricles. In patients who underwent clipping, IVH influenced the poor
outcome in patients with a GCS > 8 and no pre-operative rebleeding and seizures.
In patients with ruptured aneurysms, the incidence of IVH is 10%~45%9)13)24)30) and the overall mortality in patients with an IVH is 24%~64%.24)25)30) The mortality is 75%~100% in
patients with an IVH involving all of the ventricles, and 46%~50% in patients with an IVH involving 1~3 ventricles.13)29) Based on an autopsy study, the typical clinical
profile of sudden death in patients with SAH includes IVH, pulmonary edema, and a ruptured posterior circulation aneurysm.33) In the present study, the incidence of IVH in
patients with ruptured aneurysms was 12.3% and the overall mortality was 48.4% in the IVH group, 59.7% in patients with an IVH involving all of the ventricles, and 33.9% in
patients with an IVH involving 1~3 ventricles. These clinical and autopsy findings, including the findings in the present study, demonstrate that IVH in patients with ruptured
aneurysms reflects a severe clinical entity. Age is a prognostic factor, including surgical mortality, in patients with rupture of an intracranial aneurysm.5)31) The
incidence of IVH according to age has been reported to be 32.5%, 30%, and 25.6% in 50-, 60-, and 40-year-old patients, respectively; mortality increases with higher age.13)
Another study reported that the mean age of patients with a ruptured aneurysm and IVH was 54 years, and 4 years older than the no-IVH group, which was a statistically significant
difference.30) A study have reported that the incidence of IVH in patients with ruptured aneurysms was the highest in the age group between 40 and 60 years of age.25) An aging brain
with a ruptured aneurysm has a less optimal response to initial bleeding. Symptomatic vasospasm is more frequently reported in an older age group, resulting in a poor
prognosis.22) In the present study, the mean age was 58 years in the IVH group and 56 years in the no-IVH group. An IVH commonly accompanies ruptured aneurysms in patients >
70 years of age compared with patients < 70 years of age. Regardless of age, IVH has an affects on poor outcome after surgery. The time interval between initial hemorrhage and
admission has an important influence on outcome.18) Based on an epidemiologic study of SAH, 12% of patients with ruptured aneurysms died without reaching medical attention
and they never reached the hospital while alive; > 90 % of the patients had IVHs.33) Thus, the interval of time to arrival at the hospital is important, particularly in patients with an
IVH. In the present study, the interval of time to admission in patients with IVHs was significantly shorter than the no- IVH group because the people who brought the patients to
the hospital were more likely around patients in poor health, who in turn were more likely tohave IVHs. Hypertension as a risk factor of the aneurysmal SAH is still under debate. Blood pressure, influenced by the Cushing reflex, can abruptly increase in patients without hypertension
as a consequence of SAH, which makes it difficult to determine whether hypertension is caused by SAH or precedes SAH. Another problem appears in many studies involving risk factors for hemorrhagic events when no difference is made between SAH and hypertensive ICH.
Indeed, hypertension can falsely show an association with SAH if one considers hypertension to have a more important role as a risk factor for ICH than SAH.15) Even though
persistent hypertension is known to affect hematoma enlargement in spontaneous ICH1), little is known regarding the influence of hypertension on IVH in patients with
ruptured aneurysms. Rosen30) reported that blood pressure at the time of admission was significantly higher in the no-IVH group and IVH was highly associated with patients who had
a history of hypertension compared with patients without IVH. The current study findings were also similar to the results of Rosen et al.30), but a history of hypertension did not
affect the occurrence of IVH in patients with ruptured aneurysms. Thus, hypertension at the time of admission is
clinically meaningful and can affect prognosis. Further study may be necessary to determine the relationship between hypertension and IVH in patients with ruptured aneurysms.
The role of alcohol as a risk factor in patients with ruptured aneurysms has not been well-established. Heavy alcohol consumption might contribute to the formation of an
aneurysm through hypertension, or some unknown mechanism2)17). Heavy alcohol consumption impairs outcome mainly through severe rebleeding and delayed ischemia, and
to a lesser extent through poor health status and the presence of an ICH. Thus, alcohol consumption is a known risk factor for poor outcome after aneurysmal SAH16). In the present
study, the occurrence of IVH was significantly higher in alcohol abusers. Indeed, IVH in alcohol abusers may reflect clinical severity.
Twenty to forty percent of the patients admitted to the hospital after a ruptured aneurysm have poor clinical status (HHGs of 4~5). However, if the patient has an IVH with a ruptured aneurysm, the incidence of HHGs of 4~5 is increased to 33%~53%.13)25)30) In the current study, a GCS 8, HHGs of 4~5, and pre-operative rebleeding or seizure were related to the high occurrence of IVH at the time of the attack. During aneurysmal rupture, the loss of consciousness is caused by global ischemia, resulting from a lack of
perfusion pressure.8) Loss of consciousness is a cause of poor grade on admission, and IVH induces hydrocephalus, a rise in intracerebral pressure rise, aggravation of global ischemia,
and delayed perfusion. Poor clinical grade, posterior circulation aneurysm, excessive ventricular drainage, and abnormal hemostatic parameters appear to be associated with an increased risk of rebleeding.7)23)27) The effect of IVH on rebleeding has not been clearly delineated. When
considering the effect of IVH on poor clinical state and acute hydrocephalus, IVH may be a possible cause of an increased risk of rebleeding. With respect to the effect of IVH on
surgical outcome, the results of this study showed that the presence of IVH in patients with a pre-operative GCS>8 and no rebleeding or seizures led to a poor outcome. However,
IVH did not affect poor outcome in patients with a preoperative
GCS 8, rebleeding, or seizures because high mortality and poor outcome in those patients may have been the reasons for the poor outcome, irrespective of the IVH.
Chronic hydrocephalus is observed in 6%~37% of patients after SAH.21)25)26) IVH is known as a cause of shunt-dependent hydrocephalus; other factors include patient age, poor
clinical grade on admission, aneurysm size, aneurysm location in the posterior circulation, acute hydrocephalus, and radiologic hydrocephalus on admission.21)26) The great majority of patients with IVH have hydrocephalus, even at the time of admission. The higher the CSF pressure at the time of aneurysmal rupture, the more likely the ventricular system will be distended acutely.25) In the present study, the mean BFI on the initial CT scan was increased in the IVH
group and shunt-dependant hydrocephalus occurred in 36.6% in the IVH group compared with 11.3 % in the no- IVH group. IVHs are highly associated with ruptured aneurysms from
the posterior circulation and ACA.13)30) Penetration to the floor of the third ventricle and rupture into the lateral recess of the fourth ventricle by hematomas of ruptured aneurysms
in the posterior circulation produce IVHs.34)35) In ACA aneurysms, IVHs result from extension of frontal lobe hematomas into the ipsilateral frontal horn, rather than downward extension into the third ventricle. In the present study, IVH was most common in VBA, followed by ACA.
With respect to the relationship between IVHs and the distribution of SAHs, it is probable that the greater the volume of blood released at the time of rupture the more
likely the ventricular system will be involved. Thus, the occurrence of IVH can produce more frequent thick, diffuse pattern SAHs. The SAH grades on CT scans were originally
developed to estimate the effect on vasospasm. However, the Fisher CT grading scale identified thick SAHs, but does not separately account for IVH or ICH (both grade 4). Frontera
et al.6) modified the Fisher grade and classified the CT grading scale of SAH into four grades by SAH pattern (thin or thick) and the presence of IVH (with or without). In this system, thin or thick is not specified; it accounts for thick cisternal and ventricular blood, and predicts symptomatic vasospasm after SAH more accurately than the original Fisher scale. Claassen et al.3) also suggested a new scale that was considered for IVHs. This system was classified into
four grades by SAH pattern (thin or thick) and the presence or absence of IVHs in both lateral ventricles. In this system, the definition of thick is complete filling of 1 cistern or
fissure. SAHs completely filling any cistern or fissure and IVH in the lateral ventricles are both risk factors for delayed infarction.3) With the modified Fisher scale, each unit
increase in score produced a higher odds ratio for all of the outcomes. This was also true for the Claassen scale in relation to the outcomes of vasospasm and delayed
infarction, but a score of 2 (thin SAH with IVH in both lateral ventricles) was more strongly predictive of poor outcome than a score of 3. In the modified Fisher grade and
Claassen scale, a thick SAH and IVH were closely related to surgical outcome.20) Hijdra10) suggested that the scale score correlated with the amount of IVH, which also reflect
delayed ischemia and outcome. The amount of subarachnoid and intraventricular blood seen on CT scans was scored semi-quantitatively as the sum of scores ranging from 0~3
for each of 10 cisterns.10)12) IVH could contribute to the risk that delayed cerebral ischemia could develop by causing ventricular dilation, which may further impair blood flow in
patients with marginal vascular reserve as a result of vasospasm.19) The present study showed a significant increase in SAH amount in patients who had IVH. However, the relationship between infarction due to vasospasm and IVH was not evaluated because hemorrhagic infarction,
surgical contusion, pre-operative rebleeding, or early deaths create some biases and provide unreliable data. The occurrence of IVH was most commonly associated with the
thick, diffuse pattern of SAH. IVH probably results from an intracerebral and
subarachnoid clot rupturing into the lateral ventricles rather than retrograde passage of blood through the subarachnoid space. ICH occurs in up to 39%~62% of patients with
ruptured intracranial aneurysms;14)32) the overall mortality rate ranges from 36%~58%.28) However, it is well-known that the presence of an ICH after a ruptured aneurysm negatively
influences the clinical course and outcome.4,11) Based on autopsy, ICH is uncommonly associated with sudden death from aneurysmal SAH.33) With respect to IVH in cases of
ICH, Yoshioka36) reported that the total mortality rate of patients analyzed was 32.7%, and if the intraventricular clot occupied over two-thirds of the ventricles with ruptured
intracranial aneurysms, the prognosis was poor with a higher mortality. The combination of ICH and IVH in ruptured aneurysms probably has a worse prognosis than either alone.25) In the present study, the occurrence of IVH was higher in patients with an ICH volume 10 cc compared to those with an ICH volume < 10 cc. IVH did not influence
surgical outcome in patients with an ICH volume 10 cc, whereas IVH influenced poor outcome in patients with an ICH volume < 10 cc. In patients with an ICH volume
10 cc, the pre-operative poor clinical state, including radiologic severity, may further contribute to the surgical mortality and poor outcome regardless of IVH. However, in patients with a small ICH or without ICH, clinical and radiologic severity increase with the occurrence of IVH.
Conclusions
An IVH in patients with ruptured aneurysms reflects the clinical severity. Even though surgical outcomes can be expected to be good in patients with a pre-operative GCS > 8, an ICH less thanvolume < 10 cc, and without no
rebleeding or seizures, the presence of IVH in those patients affect onleads to a poor surgical outcome. The severity of IVH is an independent predictor of mortality. The outcome after aneurysm rupture is at least in part determined by the
severity of IVH. Knowledge of the effect of IVH has an important predictive value in the care of patients with aneurysmal bleeding.