Extraparenchymal Bleeding Predicts an Unfavorable Outcome in Patients With Hemorrhagic Transformation
To the Editor:
We read with interest the recent article by Fiorelli et al.1 The authors confirmed the reliability of hemorrhagic transformation (HT) as a diagnosis which, as we have found, could be made by either a neuroradiologist or a trained neurologist.2 However, of greater clinical relevance was the fact that they found that only severe HT (parenchymal hematoma 2 [PH2] in ECASS I1 ) was associated with an unfavorable outcome. As shown in the Table⇓, the same result emerged from the Multicentre Acute Stroke Trial–Italy (MAST-I) analysis.3
We have found that severe HT is very often associated with intraventricular or subarachnoid bleeding. This finding, together with but independent of cerebral edema, made the prognosis unfavorable in our study. The severe HT (PH2) definition of ECASS I points to a “significant space occupying effect” and includes the presence of a “clot remote from the infarct area.” But the ECASS I investigators did not specifically assess the effect of the intraventricular or subarachnoid bleeding on the prognosis. This finding and not HT per se is the real adverse effect of thrombolytic treatment, and we wonder whether any results from ECASS and NIH studies have been published on this point. We need confirmation of our funding on the prognostic significance of intraventricular or subarachnoid bleeding before implementing new strategies to prevent it or to manage it as soon as it develops.
- Copyright © 2000 by American Heart Association
Fiorelli M, Bastianello S, von Kummer R, del Zoppo GJ, Larrue V, Leseffre E, Ringleb AP, Lorenzano S, Manelfe C, Bozzao L, for the ECASS I Study Group. Hemorrhagic transformation within 36 hours of a cerebral infarct: relationships with early clinical deterioration and 3-month outcome in the European Cooperative Acute Stroke Study I (ECASS I) cohort. Stroke.. 1999;30:2280–2284.
Motto C, Aritzu E, Boccardi E, De Grandi C, Piana A, Candelise L. Reliability of hemorrhagic transformation diagnosis in acute ischemic stroke. Stroke.. 1997;28:302–306.
Motto C, Ciccone A, Aritzu E, Boccardi E, De Grandi C, Piana A, Candelise L, and the MAST-I Collaborative Group. Hemorrhage after an acute ischemic stroke. Stroke.. 1999;30:761–764.
We are grateful for the interest shown by Candelise et al in our article concerning the prognosis of hemorrhagic transformation of a cerebral infarct. In the ECASS I cohort, 34 of 71 patients (48%) with parenchymal hematoma (PH) at day 1 had an associated extraparenchymal hemorrhage (EH+). At 3 months all these 34 patients had a poor outcome, defined as a Rankin score (RS) of ≥1 (100%, 95% CI 90% to 100%), as opposed to 90% (79% to 97%) of patients with PH but no EH (EH−). The analysis of risk expressed in terms of odds ratio did not reveal a significant additional risk of poor outcome associated with EH (OR 1.09, 95% CI 0.54 or 2.19). Poor outcome defined as an RS of ≥2 was observed in 91% (77% to 97%) of EH+ patients as opposed to 70% (54% to 83%) of their EH− counterparts (OR 1.3, 0.65 to 2.61).
Our data are therefore not in favor of the hypothesis that the risk for disability and death is increased in association with EH. The numbers are too small, however, to rule out type 2 errors. In any case, comparisons between our data and those collected in the framework of MAST-I should be made with caution. At variance with ECASS I, the MAST-I investigators defined unfavorable outcome as an RS of ≥3 at 6 months after stroke. Our opinion is that rather than EH, the mass effect caused by PH and the resulting compression of functioning brain tissue is the primary determinant of poor outcome. However, topics of crucial importance like this certainly deserve further studies. Between-cohorts analyses can help, especially if made using individual data and after having verified the extent to which the information drawn from different sources is comparable.