Predictors of Hematoma Growth?
To the Editor:
I read with much interest the article, “Multivariate Analysis of Predictors of Hematoma Enlargement in Spontaneous Intracerebral Hemorrhage,” by Fujii and colleagues.1 I fully agree with the authors on the importance of identifying factors responsible for the increase in volume of intracerebral hemorrhage, since the hematoma “growth” is common1 2 and because this is associated with neurological deterioration.2 While their multivariate analyses revealed 5 independent “predictors” for hematoma enlargement, Fujii and colleagues pointed out in their Discussion section that 3 of the 5 factors (ie, a short time interval from onset, the presence of disturbed consciousness, and irregularly shaped hematoma) were related to the natural time course rather than acting as risk factors.1 The authors further postulated that the stabilization of hematoma formation takes some time, that disturbed consciousness can be a consequence of hematoma enlargement rather than a cause, and that irregularly shaped hematomas may indicate bleeding from multiple arterioles.1 Regarding their findings and interpretations, I would like to make the following comments.
First, the great majority of intracerebral hematomas are caused by bleeding from arteries or arterioles under systemic arterial pressure, and so hematomas will “grow” for some time, until the hematoma enlargement is counteracted by increasing regional intracranial pressure; eventually, bleeding ceases because of hemostasis.3 In a similar study, Brott and colleagues2 performed baseline CT scans in patients with intracerebral hemorrhage within 3 hours of onset and repeated the scans at regular intervals after the first scans. Brott and colleagues reported 26% of substantial hematoma growth between the baseline and 1-hour CT scans and 12% of substantial hematoma growth between the 1- and 20-hour CT scans.2 In the article by Fujii and colleagues, the patients had their first CT scans within 24 hours of onset of symptoms and their second scans within 24 hours of admission.1 Considering the postulation by Fujii and colleagues that active bleeding of hematoma formation largely stabilizes within 6 hours of onset, the study by Brott and colleagues reflects the early stage of hematoma stabilization rather than growth. Similarly, the patients of Fujii and colleagues should be divided into 2 groups, according the time of onset (ie, <6 or >6 hours) before the univariate and multivariate analyses are performed to reveal the respective factors for the initial hematoma stabilization and the subsequent hematoma growth.
Second, the symptomatology of intracerebral hematoma is affected by its location, volume, and rate of formation.4 In the study by Brott and colleagues,2 hematoma growth was defined by an increase in hematoma volume by >33% of the baseline volume (ie, 10% increase in diameter); the definition was simple and clear. In contrast, Fujii and colleagues defined hematoma growth by either (1) hematoma volume increased by >50% of the initial scans (ie, 14.5% increase in diameter) plus an absolute increase of >2 cm3 or (2) an absolute increase of hematoma volume by >20 cm3. It seems that the 2 definitions are rather unequal. For example, a hematoma enlarged from 0.9 cm3 to 3 cm3 (ie, an increase of 2.1 cm3 or 233%) may be well tolerated but an enlargement by >20 cm3 is likely to cause clinical deterioration. In fact, the mortality rate is affected by the hematoma volume4: about 5% of patients with hematomas <30 cm3 will die, about 35% at volumes between 30 and 50 cm3, and about 85% at volumes of >50 cm3. I would therefore suggest that Fujii and colleagues should use either the relative change in hematoma volume or the absolute increase, but not both, in their analyses.
Finally, I wonder whether Fujii and colleagues can include some data on the conscious state and neurological deficits within the first 24 hours of admission. It is clinically important to see if the initial hematoma stabilization in the first 6 hours and subsequent hematoma enlargement are correlated with clinical deterioration.
- Copyright © 1998 by American Heart Association
Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Tanaka R. Multivariate analysis of predictors of hematoma enlargement in spontaneous intracerebral hemorrhage. Stroke. 1998;29:1160–1166.
Brott T, Broderick J, Kothari R, Barsan W, Tomsick T, Sauerbeck L, Spilker J, Duldner J, Khoury J. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke. 1997;28:1–5.
Olson JD. Mechanisms of hemostasis: effect on intracerebral hemorrhage. Stroke. 1993;24[suppl I]:I-109–I-114.
Weir B. The clinical problem of intracerebral hematoma. Stroke. 1993;24[suppl I]:I-93.
We thank Dr Cheung for his comments on our recent report. First, we did not divide our patients into 2 groups according to the time of onset before performing the analyses because of the following reasons: (1) the majority of the 627 patients were admitted within 6 hours of onset (Table 1⇑ ), (2) only 2 of the 88 patients with hematoma growth were admitted more than 6 hours after onset, and (3) we intended to emphasize the fact that the incidence of hematoma growth was extremely rare in the patients admitted >6 hours after onset. Although we performed multiple logistic regression analysis of the 519 patients admitted within 6 hours of onset (Table 2⇑ ), we could not find major differences in the statistical results between the 2 patient groups admitted within 6 hours of onset and between 6 and 24 hours after onset.
Second, we initiated these studiesR1 R2 R3 to find possible ways of preventing patients with spontaneous intracerebral hemorrhage from clinical deterioration due to hematoma growth. Thus, we determined the definition of hematoma growth after looking at a number of cases to assess predictors of definite enlargement of hematoma likely to result in neurological deterioration. As Dr Cheung mentioned, an enlargement of >20 cmR3 is likely to cause clinical deterioration. However, if we use only the relative change (>50%), we cannot regard a hematoma enlarged from 45 cmR3 to 65 cmR3 as hematoma growth. On the other hand, in pontine hemorrhage, a hematoma enlarged from 4 cmR3 to 6 cm3, (ie, an increase by 2 cm3) is likely to cause neurological deterioration. Hence, we should use both parameters in assessing definite hematoma growth likely to result in clinical deterioration in intracerebral hemorrhages at unspecified sites, although we can use either the relative change in hematoma volume or the absolute increase in dealing with a particular site of hematoma (for example, putaminal hemorrhage). Finally, our databases for the previous studiesR1 R2 R3 do not contain parameters regarding the conscious state and neurological deficits within the first 24 hours after admission, although we have the data on the relationship between the level of consciousness disturbance at admission and hematoma growth (Table 1⇑). Thus, we cannot provide any actual data on the conscious state and neurological deficits within the first 24 hours after admission. However, hematoma growth according to our definition seems to be clearly correlated with clinical deterioration because almost all of the 88 patients with hematoma growth had clinical deterioration after admission as a matter of course (owing to the concept of our definition).