Detection and appearance of intraparenchymal haematomas of the brain at 1.5 T with spin-echo, FLAIR and GE sequences: poor relationship to the age of the haematoma

The specific appearance of blood related to time at T1- and T2-weighted spin-echo (SE) sequences is generally accepted; thus, these sequences are classically used for estimating the age of haematomas. Magnetic resonance imaging at 1.5 T, including T1- and T2-weighted SE fluid-attenuated inversion recovery (FLAIR) and T2*-weighted gradient-echo (GE) sequences, was performed on 82 intraparenchymal haematomas (IPHs) and 15 haemorrhagic infarcts (HIs) in order to analyse the appearance at different stages and with different sequences, and to investigate how reliably the age of hematomas can be estimated. The IPHs had been previously detected by CT, were spontaneous (n = 72) or traumatic (n = 10) in origin and were of different sizes (2 mm to 7 cm) and ages (from 7.5 h to 4 years after acute haemorrhagic event). The age of the lesion was calculated from the moment when clinical symptoms started or the traumatic event occurred. The 15 patients with HIs were patients with ischaemic stroke in whom there was either a suspicion of haemorrhagic transformation on CT, or haemorrhage was detected as an additional finding on MR performed for other indications. Patients with conditions that could affect the SI of blood, such as anticoagulant therapy or severe anaemia, were excluded. The signal intensity pattern of the lesions was analysed and related to their ages without prior knowledge of the clinical data. All lesions were detected with T2*-weighted GE. T1-weighted SE missed 13 haematomas and T2-weighted SE and FLAIR sequences missed five. Haemorrhagic transformation was missed in three infarcts by T1-, T2-weighted SE and FLAIR. The signal pattern on FLAIR was identical to that on T2-weighted SE. For all sequences, a wide variety of signal patterns, without a clear relationship to the age of the haematomas, was observed. There was a poor relationship between the real MR appearance of IPHs and the theoretical appearance on SE sequences. T2*-weighted GE was effective for detecting small bleedings but was not useful for estimating the age of a lesion. The FLAIR does not provide any more information than T2-weighted SE.