Introduction 

   Clinical deteriorations in patients with hypertensive intracerebral hemorrhage (ICH) are usually attributed to brain edema, mass effect, hydrocephalus, progression of hematoma, or systemic factors.^1)7)28) Of these, the enlargement of hematomas i.e. the continued bleeding in ICH may result in progressive neurological deterioration or even death.^{2)5)6)13)18)19)20)21)22)24)} Therefore, it is important to clarify the factors related with hematoma evolution for clinicians who are searching for appropriate guideline to manage patients with ICH. Several factors such as the level of consciousness level, patient's age, time interval between ictus and admission, irregularity of the hematoma shape, volume of the hematoma, midline shift, blood pressure, the presence of intraventricular hemorrhage (IVH), alcohol consumption, habitual smoking, and liver dysfunction have been suspected to be associated with outcome.^{3)9)11)15)16)} However, the exact mechanism of hematoma enlargement is unclear. Some investigator reported that the growth is not usually attributed to bleeding from the initial vascular source but to bleeding into the necrotic and edematous tissue surrounding the primary hemorrhage.¹²⁾ Meanwhile, there have been some studies for contrast extravasation in ICH using different radiographic modalities and some investigators have reported that in patients with hypertensive ICH, the finding of contrast extravasation from ruptured blood vessels was common when cerebral angiographic studies were performed in the hyperacute stage, within a few hours of onset.^14)23)26)27) Therefore we formulated a hypothesis that the presence of contrast extravasation in and/or around the hematoma on contrast enhanced computerized tomography (CT) scan might be a predictor of ongoing hemorrhage in basal ganglia and thalamus, which are the main locations of hypertensive ICH. The authors, therefore, investigated the clinical significance of extravasation and other risk factors associated with hematoma enlargement. 

Materials and Methods 

Patient population 
   We retrospectively reviewed the records of 384 consecutive patients with ICH whose diagnosis was confirmed by CT scan within 24 hours of the ictus from 2002 to 2005. Recently we usually perform non-enhanced CT scan and computerized tomogrphic angiography (CTA) to rule out secondary ICH due to underlying disease, but at that time only conventional CT scan including enhancement. 
   Patients who met the following criteria were included in this study: 1) presence of ICH in basal ganglia or thalamus on CT scan, 2) initial CT scan with enhancement within 24 hours of ictus, and 3) repeated CT scan within 72 hours of ictus. Patients with ICH in another location, such as lobar (29 patients), posterior fossa (65), and secondary ICH due to underlying diseases, such as aneurysm (10 patients), vascular malformation (8), moyamoya disease (3), and brain tumor (2) were excluded. Besides, of the remaining 267 patients, 196 patients were excluded; 40 did not undergo enhancement on initial CT scan, 41 did not undergo a repeated CT scan because of emergency surgery or hemodynamic instability, and 115 undergo a repeated CT scan more than 72 hours. Finally the remaining 71 patients (43 men and 28 women; mean age, 58.9±12.4 years) served as subjects for the present study. 

Clinical findings 
   Immediately after admission, the patient's neurological status was assessed, and systemic blood pressure was measured. The blood pressure was continuously monitored for at least 72 hours after the onset of ICH. We defined  "maximum systolic blood pressure (SBP)" as the highest value of SBP between first and second CT scan. The intervals from the ictus of ICH to the first and the second CT scans were recorded in each patient. To evaluate hemostatic function and liver function, hemoglobin, hematocrit, platelet count, prothrombin time, activated partial thromboplastin time, glutamic-oxaloacetic transaminase (GOT), and glutamic-pyruvic transaminase (GPT) were assessed. Patient's medical history, such as hypertension and/or diabetes mellitus was also recorded. Hypertension was judged to be present if the patient fulfilled one or both of the following criteria: 1) a history of antihypertensive medication or 2) a systolic blood pressure >160mmHg or diastolic blood pressure >90mmHg on at least two occasions before the onset of ICH. The patients' habitual smoking and/or alcohol consumption were also recorded. 

Computerized tomography findings 
   Hematoma volume was determined using an area calculation program built into the CT scanner. The results were expressed in cubic centimeters. The presence or absence of intraventricular hemorrhage (IVH) was recorded. When IVH was present, accurate measurement of the parenchymal hemorrhage was considered to be impossible because of the spread of blood in the cerebrospinal fluid. Therefore we measured the parenchymal hemorrhage only, and assessed the IVHs influence on the hematoma enlargement. The presence or absence of contrast extravasation was determined by visualization of highdensity contrast within and/or around the hematoma. 
   Hematomas were classified into two types according to the shape; 1) round, with round and smooth margins, and 2) Irregular, with irregular and geographic margins. Insofar as we know, there have no objective radiographic criteria for hematoma enlargement. In this study, therefore, we defined the cut-off point of hematoma enlargement as the volume increase more than 12.5 or ratio more than 1.4 between the hematoma volume on the first CT scan and that on the second CT scan.¹⁸⁾ We divided the 71 patients into enlargement and non-enlargement groups, and the data from the patients with the presence of contrast enhancement and without that were compared. 

Statistical Analysis 
   Univariate and multivariate analyses were performed to assess the relationship between hematoma enlargement and time from ictus, consciousness level, CT finding, blood pressure (systolic and diastolic) and hematologic parameters. A chi-square analysis or Fishers exact test was used to the association between hematoma enlargement and hematoma site, hematoma shape, IVH and medical history. The student t-test was used to assess the differences in age and other continuous variables between the patient groups with and without hematoma enlargement. Values are expressed as mean±standard deviation. Analyses resulting in p values of less than 0.05 were considered statistically significant. 

Results 

   The mean volume of hematoma on the initial CT scan was 14.6±17.9cc, and that on the follow-up CT scan was 19.7±20.6cc. Seventeen (23.9%) of the 71 patients demonstrated enlargements of the hematoma after the initial CT scan. The mean increment of hematoma volume between the initial CT scan and the follow-up CT scan of the group with the enlargement of hematoma was 24.6±24.0cc, and that of the group without the enlargement of hematoma was 0.7±1.9cc (p=0.001). The data comparison between these two groups is shown in Table 1. The enlargement of hematoma was predominant in male patients, and patients with alcohol consumption and habitual smoking. There were lower GCS score, larger volume of hematoma on initial CT scan, and much more frequent contrast extravasations in the group of hematoma enlargement than the other (Table 1). Time interval from initial CT scan to follow-up CT scan was much shorter in the former group than the other. Because only 3 female patients showed hematoma enlargements (only one female patient had habitual alcohol consumption and no female patient had smoked), the statistical analysis is inappropriate in the female patients. Therefore, the authors reassessed the results only in the male patients about the alcohol consumption, habitual smoking, etc. There were more contrast extravasations and shorter time interval from ictus to follow-up CT scan in the group of hematoma enlargement than the other (p=0.000, p=0.010). All other factors, such as the mean Glasgow coma scale (GCS), hematoma volume on initial CT scan, alcohol consumption, and habitual smoking were not significantly different between these two groups (Table 2). Even though the statistical analysis is inappropriate, of 9 female patients with contrast enhancement 2 patients (22.2%) showed hematoma enlargement, on the contrary, of 19 patients without enhancement only 1 patient (5.3%) showed enlargement. 

Discussion 

   In most reports, the incidence of hematoma enlargement was approximately 12% to 20%, though Bae et al.¹⁾ have estimated the incidence of hematoma enlargement to be 3%, on the contrary, Brott et al.⁵⁾ have reported that growth of the hematoma volume occurred in 38% of patients with ICH.^12)13)17) In our study, hematoma enlargement occurred in 23.9% of the 71 patients presenting ICH. However, the real incidence of hematoma expansion suspected to be higher than this result, because the patients who died of rapid expansion of the hematoma before the second CT scan, and the patients who underwent operation immediate after initial CT scan due to huge volume of hematoma were excluded from this study. We think, therefore, the interpretation of this result should be limited to the patients with spontaneous ICH whose clinical and radiological status is not in need of immediate intervention. 

Risk factors 

   In our study, male sex itself seemed to be a significant risk factor for the enlargement of hematoma. A possible explanation for this observation may be that the majority of patients with alcohol and/or smoking habits were male, and these two factors were associated with hematoma enlargement though not reach statistical significance (Table 2). As a matter of fact, they showed strong association with hematoma enlargement as a whole (Table 1), but 20 of 23 male patients (46.5%) were habitual alcohol consumers and 16 of 27 male patients were habitual smokers, on the contrary, only one patient of 28 female patients was the former and no patients were habitual smoker. Therefore, when the patients were confined to male patients, these risk factors did not reach statistical significance (Table 2). Although there was no report statistically confirming alcohol consumption as an independent risk factor for hematoma enlargement, some investigators reported that heavy alcohol intake was a risk factor for the occurrence of ICH.⁸⁾²⁵⁾ However, there was only a trend for the hematoma enlargement in our study. In summary, it is suspected that because most patients with above habits were male patients in Korea and accordingly the enlargement of hematoma was predominant in male patients, but the male sex itself might not affect the results as an independent contributing factor for hematoma enlargement. 
   Bases on some small series, both Chen, et al.⁶⁾ and Kelly, et al.¹⁹⁾ reported that the thalamus is the most frequent site of hematoma growth because of its close proximity to the third ventricle, which is more distensible than brain tissue. However, there was no significant difference in the frequency of hematoma enlargement between thalamus and basal ganglia in this study (p=0.572). The authors think that it cannot be ascertained because of the many different circumstances, such as inclusion criteria above mentioned, hematoma volume, and so on described subsequently. 
   As in some previous studies, hematoma enlargement had relatively strong association with lower GCS (p=0.044, Table 1), however, its association lessened in the analysis in the male patients group (p=0.217, Table 2).^10)16)28) Even though only 3 of 28 patients showed hematoma enlargement in the female patients, because the difference of mean GCS was much larger than in the male (8.7±3.5, 13.0±2.7, p=0.018), its statistical significance seemed to lessen in male patients group (p=0.217). 
   Time interval from initial CT scan to follow-up CT scan was much shorter in the group of hematoma enlargement than the other group. We assumed that follow-up CT scans were performed emergently because of the change of the neurological status especially in the group of hematoma enlargement, but when the clinical status of the patient was stable after admission, the follow-up CT scan was performed the next day mostly, and therefore, the time interval was much shorter in the patients of group of hematoma enlargement (p=0.000 as a whole, p=0.010 in the male patients). 
   Finally, the extravasations of radiographic contrast into the hematoma on the first CT scans were observed much frequently in the group of hematoma enlargement than in the other group (p=0.000, Table 1), and it continued strong association in the male patient group (p=0.000, Table 2). There have been some studies for contrast extravasation in ICH using different radiographic modalities and some investigators have reported that in patients with hypertensive ICH, the finding of contrast extravasation from ruptured blood vessels was common when cerebral angiographic studies were performed in the hyperacute stage, within a few hours of onset.^23)26)27) Yamaguchi et al.²⁹⁾ found that 42% of patients with primary ICH had contrast extravasation during intra-arterial angiography when performed within 5 hours of symptom onset. Similar rates of extravasation into cerebral hematomas have been observed during CT angiogram (CTA) and MR scanning.²⁾²⁶⁾ In contrast, Huckman, et al.¹⁴⁾ found that 12.5% showed extravasation during angiographic study. The exact mechanisms of contrast extravasation were unclear. However, some authors explained that because intravenously administered contrast medium cannot cross the normal blood-brain barrier (BBB), contrast medium leakage seen on CT scans has been proven to originate from the ruptured vessels or the vessels with ischemic changes in their walls caused by compression by the hematoma, resulting in increased vascular permeability.⁴⁾²⁷⁾ 

Conclusion 

   Due to a high risk for hematoma enlargement, patients with spontaneous ICH in basal ganglia and thalamus, especially those with evidence of contrast extravasation on CT scan should be closely observed and short term followup radiological studies are needed for the verification of hematoma enlargement. 

REFERENCES

1. Bae HG, Lee KS, Yun IG, Bae WK, Choi SK, Byun BJ, et al. Rapid expansion of hypertensive intracerebral hemorrhage. Neurosurgery 31:35-41, 1992 

2. Becker KJ, Baxter AB, Bybee HM, Tirschwell DL, Abouelsaad T, Cohen WA. Extravasation of radiographic contrast is an independent predictor of death in primary intracerebral hemorrhage. Stroke 30:2025-32, 1999 

3. Broderick JP, Brott TG, Duldner JE, Tomisick T, Huster G. Volume of intracerebral hemorrhage: a powerful and easy-touse 30-day mortality. Stroke 24:987-93, 1993 

4. Broderick JP, Brott TG, Tomsick T, Barsan W, Spilker J. Ultra-early evaluation of intracerebral hemorrhage. J Neurosurg 72:195-9, 1990 

5. Brott T, Broderick J, Kothari R, Barsan W, Tomsick T, Sauerbeck L, et al. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke 28:1-5, 1997 

6. Chen ST, Chen SD, Hsu CY, Hogan EL. Progression of hypertensive intracerebral hemorrhage. Neurology 39:1509-14, 1989 

7. Daverat P, Castel JP, Dartigues JF, Orgogozo JM. Death and functional outcome after spontaneous intracerebral hemorrhage: a prospective study of 166 cases using multivariate analysis. Stroke 22:1-6, 1991 

8. Donahue RP, Abbott RD, Reed DM, Yano K. Alcohol and hemorrhagic stroke. JAMA 255:2311-4, 1986 

9. Fieshci C, Carolei A, Fiorelli M, Argentino C, Bozzao L, Fazio C, et al. Changing prognosis of primary intracerebral hemorrhage: results of a clinical and computed tomographic follow-up study of 104 patients. Stroke 19:192-5, 1988 

10. Fogelholm R, Acikainen S, Murros K. Prognostic value and determinants of first-day mean arterial pressure in spontaneous supratentorial intracerebral hemorrhage. Stroke 28:1396-400, 1997 

11. Franke CL, van Swieten JC, Algra A, van Gijn J. Prognostic factors in patients with intracerebral hematoma. J Neurol Neurosurg Psychiatry 55:653-7, 1992 

12. Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Tanaka R. Multivariate analysis of predictors of hematoma enlargement in spontaneous intracerebral hemorrhage. Stroke 29:1160-6, 1998 

13. Fujii Y, Tanaka R, Takeuchi S, Koike T, Minakawa T, Sasaki O. Hematoma enlargement in spontaneous intracerebral hemorrhage. J Neurosurg 80:51-7, 1994 

14. Huckman MS, Weinberg PE, Kim KS, Davis DO. Angiographic and clinico-pathologic correlates in basal ganglionic hemorrhage. Radiology 95:79-92, 1970 

15. Hwang SN, Ahn YH, Mok JH, Park K, Kim YB, Min BK, et al. An analysis of factors related to the shape of hypertensive intracerebral hematoma and its clinical signific