<p>The purpose of this work was to improve the diagnosis of intracranial hemorrhage with MRI, using, among others, T2*-w GE sequences. Various sequences were tested in rabbits at two magnetic field strengths. Then, the most effective technique was applied to stroke patients. </p><p>Experimental studies: The MR detectability of small experimental haematomas in the brain and of blood in the cerebrospinal fluid (CSF) spaces of 30 rabbits was evaluated. MRI examinations were performed at determined intervals. The last MR images were compared to formalin fixed brain sections and, in 16 rabbits, also to the histological findings. T2*-weighted GE sequences revealed all the intraparenchymal haematomas at 1.5 T, appearing strongly hypointense. Their signal patterns remained unchanged during the follow-up. Blood in the CSF spaces was best detected at 1.5T with T2*-weighted GE sequences during the first 2 days. FLAIR and SE sequences were rather insensitive.</p><p>Clinical studies: MR examinations were performed at 1.5T, including T1- and T2-w SE, FLAIR and T2*-w GE sequences. In the first clinical study, 66 intraparenchymal hematomas (IPH) of different sizes and ages were examined. T2*-w GE sequence was the most sensitive. On all the sequences, we found a big variety of signal patterns, without a clear relationship to the age of the hematomas. </p><p>In a second clinical study, MR examinations were performed to 83 patients with acute stroke: 43 presented acute IPH and 40 were used as controls. Old microhemorrhages (OMHs) were found in 60% of the patients with IPH, and in 15% of the controls. </p><p>Conclusion: T2*-weighted GE sequences are capable of revealing very small intraparenchymal hemorrhages at any stage, and blood in CSF spaces during at least the first 2 days. The age of IPHs cannot reliably be estimated with MRI. We have found a correlation between the presence of OMHs and acute intraparenchymal hematomas.</p>

The purpose of this work was to improve the diagnosis of intracranial hemorrhage with MRI, using, among others, T2*-w GE sequences. Various sequences were tested in rabbits at two magnetic field strengths. Then, the most effective technique was applied to stroke patients. Experimental studies: The MR detectability of small experimental haematomas in the brain and of blood in the cerebrospinal fluid (CSF) spaces of 30 rabbits was evaluated. MRI examinations were performed at determined intervals. The last MR images were compared to formalin fixed brain sections and, in 16 rabbits, also to the histological findings. T2*-weighted GE sequences revealed all the intraparenchymal haematomas at 1.5 T, appearing strongly hypointense. Their signal patterns remained unchanged during the follow-up. Blood in the CSF spaces was best detected at 1.5T with T2*-weighted GE sequences during the first 2 days. FLAIR and SE sequences were rather insensitive. Clinical studies: MR examinations were performed at 1.5T, including T1- and T2-w SE, FLAIR and T2*-w GE sequences. In the first clinical study, 66 intraparenchymal hematomas (IPH) of different sizes and ages were examined. T2*-w GE sequence was the most sensitive. On all the sequences, we found a big variety of signal patterns, without a clear relationship to the age of the hematomas. In a second clinical study, MR examinations were performed to 83 patients with acute stroke: 43 presented acute IPH and 40 were used as controls. Old microhemorrhages (OMHs) were found in 60% of the patients with IPH, and in 15% of the controls. Conclusion: T2*-weighted GE sequences are capable of revealing very small intraparenchymal hemorrhages at any stage, and blood in CSF spaces during at least the first 2 days. The age of IPHs cannot reliably be estimated with MRI. We have found a correlation between the presence of OMHs and acute intraparenchymal hematomas.