Annals of Indian Academy of Neurology
ORIGINAL ARTICLE
Year
: 2018  |  Volume : 21  |  Issue : 4  |  Page : 275--279

Utility of various ultrafast magnetic resonance sequences in the detection of fetal intracranial hemorrhage


Remya Baburaj1, Rajeswaran Rangasami1, Anupama Chandrasekharan1, Indrani Suresh2, Sudarshan Suresh2, Suresh Seshadri2,  
1 Department of Radiology and Imaging Sciences, Sri Ramachandra Medical College and Research Institute, Sri Ramachandra University, Chennai, Tamil Nadu, India
2 Department of Sonography, Mediscan Systems, Chennai, Tamil Nadu, India

Correspondence Address:
Dr. Rajeswaran Rangasami
Department of Radiology and Imaging Sciences, Sri Ramachandra University, Porur, Chennai - 600 116, Tamil Nadu
India

Abstract

Objective: The aim of this study is to compare the images obtained from standard ultrafast magnetic resonance (MR) imaging sequences with gradient (GRE) sequence images in identifying fetal intracranial hemorrhage (ICH). Materials and Methods: MR images of fetal brains with ICH done between October 2012 and September 2015 were reviewed. The images obtained from four ultrafast MR sequences– Turbo Fast Low Angle Shot (Turbo FLASH) T1-weighted images, Half Fourier Acquisition single-shot turbo spin echo (HASTE) T2-weighted images, b0 images of diffusion-weighted imaging (DWI) and b800 images of DWI were compared with images obtained from GRE sequence in depicting fetal ICH. Results: Out of the 212 fetuses during the study period, 15 fetuses had ICH. In the 15 fetuses with ICH as detected on GRE, Grade1 germinal matrix hemorrhage was seen in 5 fetuses, Grade 2 in 4 fetuses, Grade 3 in 3 fetuses, and Grade 4 in two fetuses. Subdural hemorrhage was seen in 1 fetus. In comparison to GRE sequence, b0 of DWI sequence was almost equal in the depiction of ICH. T2 HASTE sequence also delineated hemorrhage, although not as effectively as GRE and b0 images of images DWI. T1 Turbo FLASH and b800 images of DWI were less reliable in the depiction of fetal ICH but were useful in predicting the stage of hemorrhage. Conclusion: As compared to GRE sequence, b0 images of DWI followed by HASTE are the two preferred ultrafast sequences in the diagnosis of fetal ICH.



How to cite this article:
Baburaj R, Rangasami R, Chandrasekharan A, Suresh I, Suresh S, Seshadri S. Utility of various ultrafast magnetic resonance sequences in the detection of fetal intracranial hemorrhage.Ann Indian Acad Neurol 2018;21:275-279


How to cite this URL:
Baburaj R, Rangasami R, Chandrasekharan A, Suresh I, Suresh S, Seshadri S. Utility of various ultrafast magnetic resonance sequences in the detection of fetal intracranial hemorrhage. Ann Indian Acad Neurol [serial online] 2018 [cited 2019 Aug 20 ];21:275-279
Available from: http://www.annalsofian.org/text.asp?2018/21/4/275/244870


Full Text



 Introduction



The incidence of intracranial hemorrhage (ICH) in fetuses in utero is rare; although it is a well-recognized phenomenon in the newborn.[1],[2] Important factors predisposing to in utero ICH include maternal trauma, infection, coagulation disorders, and preeclampsia. Fetal ICHs detected by prenatal neurosonography are widely available in the literature.[2] The sonographic findings are, however, inconsistent, operator dependent, and at times difficult to differentiate from other intracranial lesions/masses. Magnetic resonance imaging (MRI) is a valuable complement to ultrasound in the diagnosis of fetal ICH and provides additional information that is useful in patient management.[3] MRI is much more sensitive in detecting different phases of hemorrhages, making it possible to detect early bleeding stages as well as old blood residue. The standard imaging sequence is gradient (GRE) sequence where blooming is appreciated in the presence of hemorrhage.[4]

The aim of this study is to compare the images obtained from standard ultrafast MR sequences with conventional gradient (GRE) sequence images in depicting the conspicuity of fetal ICH. In our literature search, we could not find the similar study.

 Materials and Methods



MR images of fetal brains with ICH done between October 2012 and September 2015 were reviewed. Institutional Ethical clearance was obtained. Fifteen fetuses with ICH as detected on GRE were included in this study. The gestational age ranged between 19 and 38 weeks.

Imaging protocols

MR imaging was performed with a 1.5 T superconducting system (AvantoSiemens, Erlangen, Germany) with an eight element torso array coil. The routine MR sequences obtained in our institution in suspected fetal ICH were[1] T2-weighted Half Fourier acquisition single shot turbo spin echo (HASTE) (TR - 900 ms, TE-90, FOV: 24–28 cm, matrix - 256 × 205, number of excitations-1, slice thickness-4.5 mm, intersection gap-0.2 mm),[2] T1-weighted TURBO FLASH (TR-100 ms, TE-4.7 ms, flip angle - 70°, FOV: 24–28 cm, matrix-256 × 173, number of excitations-1, slice thickness-4.5 mm, intersection gap-0.2 mm)[3] Echo-planar diffusion-weighted imaging (DWI) (TR-5000 ms, TE-101, FOV-24-28 cm, matrix– 192 × 192, slice thickness-4.5 mm, intersection gap-0.2 mm). Gradients were applied in 3 orthogonal directions using a b value of 0 and 800 seconds/mm2[4] Conventional GRE (TR-600 ms, TE-26 ms, flip angle - 20°, FOV: 24–28 cm, matrix-256 × 135, number of excitations-1, slice thickness-4.5 mm, intersection gap-0.2 mm, Acquisition time −1 min 30 s).

Only the TURBO FLASH sequence was obtained with maternal breath-hold, the other sequences were obtained with maternal free breathing. Axial, coronal, and sagittal images relative to the fetal head were obtained using HASTE sequence. Only axial images relative to the fetal head were obtained in the reminder of the sequences. For the study purpose, only the axial images of the various sequences were compared. Exclusion criteria included pure subarachnoid hemorrhage in the sulcal spaces and studies with nondiagnostic quality images. Two neuroradiologists with more than 15 years' experience each reviewed the images in a blinded fashion. When there was discrepancy in reporting, the final decision was achieved based on consensus. The images of various sequences from each fetus were evaluated at separate sessions. For example, first the GRE sequence of the all the patient fetuses were studied. Next, the HASTE sequence of all patients was studied in a random manner. This was repeated with T1 TURBO FLASH, bo, and B800 images of DWI. Subsequently, the images of all the sequences were analyzed side-by-side, in conjunction with GRE images, for relative conspicuity of hemorrhage and accuracy. The associated hemorrhage within sulcal spaces was not used in calculation due to technical difficulty in computation.

Volume calculation

The visualization of fetal ICH on each sequence was first recorded as “Present” or “Absent” as per the conspicuity in each sequence. The ICH volume was calculated according to the formula ABC/2, as proposed by Kothari et al.[5] The MRI section with the largest area of ICH was taken into account with A representing the largest diameter, B representing the diameter perpendicular to A. The C value was calculated by comparing the size of hemorrhage in the adjacent sections with the slice showing the largest hemorrhage. A value >75% area of hemorrhage in the subjacent slice as compared to the slice with the largest hemorrhage was considered as 1 hemorrhage slice, between 25% and 75% as half hemorrhage slice and <25% as nil hemorrhage slice. C value was obtained by the formula: C = corrected number of slices the hemorrhage is visualized x (slice thickness + interslice gap). All measurements were recorded in millimeter scale and the product obtained was divided by 2, which yielded the ICH volume in cubic millimeter. All positive cases were further classified based on the volume of hemorrhage visualized in that sequence (1) volume less than or equal to one-half that seen in GRE sequence. (2) volume more than one-half that seen in GRE sequence.

Fetal intracranial hemorrhage grading

In the 15 fetuses, the grading was obtained using the classification proposed by Papile et al.[2] classification system.

Grade I: Limited to subependymal matrixGrade II: Subependymal germinal matrix hemorrhage with intraventricular extension but without ventricular dilatationGrade III: Subependymal germinal matrix hemorrhage with intraventricular extension and ventricular dilatationGrade IV: Grade III with parenchymal extension.

Stage of fetal intracranial hemorrhage

Stage of fetal intracranial hemorrhage was next recorded based on a combination of sequences.

 Results



Out of the 212 fetuses during the study period, 15 fetuses had ICH. In our series of 15 fetuses, grade1 germinal matrix hemorrhage was seen in 5 fetuses, Grade 2 hemorrhage in 4 fetuses, Grade 3 hemorrhage in 3 fetuses, Grade 4 hemorrhage in 2 fetuses and subdural hemorrhage in 1 fetus. The visibility of hemorrhage was compared on four different MR sequences with standard GRE sequence [Figure 1] and [Table 1]. The stage of hemorrhage could be ascertained confidently in 7 fetuses based on the appearance in different sequences. Early subacute hemorrhage was seen in 5 fetuses, late subacute hemorrhage in 2 fetuses. In 8 fetuses, the stage of the hemorrhage could not be predicted as they were not visible on T1-weighted.{Figure 1}{Table 1}

Out of the 4 fetuses with Grade 1 ICH, bo images of DWI and HASTE depicted hemorrhage in 3 cases, while b800 images of DWI and Turbo FLASH depicted in only one case each [Figure 1] and [Figure 2]. Out of the 5 fetuses with Grade 2 ICH, bo images of DWI and HASTE depicted hemorrhage in 5 and 4 cases, respectively, while b800 images of DWI and Turbo FLASH depicted in 3 and 2 cases, respectively [Figure 1] and [Figure 3]. Grade 3 ICH seen in 2 fetuses was depicted in all the sequences [Figure 1] and [Figure 4]. Grade 4 ICH seen in 2 fetuses was depicted in all the sequences except in Turbo FLASH where only one was depicted [Figure 1] and [Figure 5]. The large subdural hemorrhage encountered, was seen in all the sequences.{Figure 2}{Figure 3}{Figure 4}{Figure 5}

In 4 fetuses with Grade 1 hemorrhage, more than 50% the volume of hemorrhage was seen in two and three fetuses by HASTE and b0 sequences, respectively [Table 1]. Out of the 5 fetuses with Grade 2 hemorrhage, more than 50% the volume of hemorrhage was seen in three, two, and five fetuses by HASTE, b800, and b0 diffusion sequences, respectively. Out of the 2 fetuses with Grade 3 hemorrhage, more than 50% the volume of hemorrhage was seen by bo diffusion sequence in both the cases. More than 50% the volume of hemorrhage was seen in one fetus each by T1, HASTE, and b800 diffusion sequences. Out of the 2 fetuses with grade 4 hemorrhage, more than 50% the volume of hemorrhage was seen by HASTE and bo diffusion in both the cases. More than 50% the volume of hemorrhage was seen in one fetus each by T1 and b800 diffusion sequences. In a fetus with large subdural hemorrhage, all the sequences depicted more than 50% the volume of hemorrhage. Thus, b0 diffusion sequence gave a good prediction of volume of hemorrhage in comparison with GRE. T1 and b800 diffusion images showed ICH in few fetuses and mostly depicted reduced volume of hemorrhage. The depiction of volume of hemorrhage by T1, HASTE, and b800 diffusion sequences improved with increasing grades of hemorrhage.

 Discussion



Ultrasound remains the modality of choice in the examination of pregnant patients.[6],[7] However, the sonographic findings of fetal ICH are at times subtle or mimic a mass. Various studies have underlined the complementary role of MRI in the diagnosis of fetal ICH.[8],[9] The parenchymal/ventricular extension, associated cysts, and the presence of parenchymal ischemia may not be well depicted on sonography. Kirkinen et al. in a study of fetal ICH in 4 fetuses have concluded that MRI is more useful than sonography in predicting the duration of hemorrhage and the anatomical structures involved.[8] They had used T2-weighted fast spin echo, and T1-weighted fast low angle shot imaging sequences. Elchalal et al. studied 33 fetuses with ICH and concluded that MRI is useful to make an accurate diagnosis of fetal ICH especially in grades 2 and 3 where the management is very important.[9]

MRI is much more sensitive in detecting different stages of hemorrhages– making it possible to detect acute as well as chronic hemorrhage and its sequelae. The appearance of ICH primarily depends on the timing of the examination, the imaging sequence, and parameters in use.[4] The other factors deciding the visibility of ICH on MRI are the volume of the hemorrhage and contrast difference between the hemorrhage and the adjacent parenchyma. During the aging of blood from acute to chronic, hemoglobin proceeds through a cascade of natural degradation. In acute stages of hemorrhage, hemoglobin loses its oxygen content forming deoxyhemoglobin which appears hypointense on T1-weighted and hyperintense on T2-weighted. In subacute stages of hemorrhage, intracellular or extracellular methemoglobin is formed with the disappearance of erythrocytes. They contribute to the hyperintense signal on T1-weighted. In its chronic sequelae, hemorrhage is broken down to products of ferritin and hemosiderin producing low signals on both T1-weighted and T2-weighted.[4]

B0 images of diffusion-weighted imaging

The visibility of hemorrhage and volume depiction was very good and comparable to GRE sequence [Figure 1], [Figure 2]d, 3d, 4d, 5d, and [Table 1]. Only one case of grade 1 ICH was not depicted by this sequence. The hemorrhages mostly appeared hypointense on this sequence due to the magnetic susceptibility of Echo Planar DWI.[10] Doris et al. compared images obtained from b0 of EPI DWI and conventional GRE sequences in adult patients and concluded that b0 of DWI was good in depicting hemorrhage but could miss small streaks of hemorrhages.[11] The study results correlate with their study as b0 of DWI depicted the fetal ICH in a good number of cases and did not show hemorrhage in one case of Grade1 ICH.

B800 images of diffusion-weighted imaging

This sequence was inadequate in visualizing ICH, especially with Grades 1 and 2 hemorrhage [Figure 1], [Figure 2]e and [Figure 3]e. This sequence also underestimated the volume of hemorrhage [Table 1]. However, this sequence was useful in diagnosing late subacute stage hemorrhage based on its hyperintense appearance.[10] Kang et al. studied the appearance of ICH with DWI with b800 in 38 adult patients.[10] They concluded that ICH appeared hyperintense on late subacute and hyperacute stages of hemorrhage and hypointense in a reminder of the stages due to ADC value and the magnetic susceptibility effect.

Half Fourier Acquisition single-shot turbo spin echo

Although not as good as GRE and bo DWI in detecting fetal ICH, HASTE was a better sequence than T1 TURBO FLASH and b0 DWI in identifying the presence and extent of hemorrhage [Figure 1], [Figure 2]c, [Figure 3]c, [Figure 4]b,[Figure 4]c, [Figure 5]c an[Figure 5]d [Table 1]. It depicted higher grades of hemorrhage better due to the increased volume of haemorrhage.

T1 TURBO FLASH

The visibility of hemorrhage was very poor, especially with Grades 1 and 2 ICH, where the hemorrhage volume is less [Figure 1]. This sequence also underestimated the volume of hemorrhage [Table 1]. However, this sequence was helpful in diagnosing early and late subacute hemorrhage [Figure 3]f and [Figure 4]f based on T1 hyperintensity due to the presence of methemoglobin.[4]

The limitation of this study is that the number of fetuses with ICH is low. Another limitation is that we could not compare other ultrafast sequences such as steady state sequence and GRE EPI with conventional GRE sequence as they are not done routinely done in our institution. Yet, another limitation is that we used conventional gradient sequence as the gold standard, though there are minimal chances of missing hemorrhage on this sequence. We used conventional gradient sequence because it is one of the best sequences to detect hemorrhage as it is highly sensitive to magnetic susceptibility.[4]

 Conclusion



Fetal MRI helps in grading fetal ICH and may also predict the approximate time of hemorrhage. In b0 of DWI sequence the depiction of ICH was good as compared to GRE sequence. HASTE sequence is the next preferred sequence in the detection fetal ICH. T1 TURBO FLASH and b800 images of DWI are less reliable in depiction of fetal ICH; although, they may be useful in diagnosing the stage of hemorrhage.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Reiss I, Gortner L, Moller J, Gehl HB, Baschat AA, Gembruch U, et al. Fetal intracerebral hemorrhage in the second trimester: Diagnosis by sonography and magnetic resonance imaging. Ultrasound Obstet Gynecol 1996;7:49-51.
2Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529-34.
3Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: Beyond T2-weighted imaging. Eur J Radiol 2006;57:172-81.
4Gomori JM, Grossman RI. Mechanisms responsible for the MR appearance and evolution of intracranial hemorrhage. Radiographics 1988;8:427-40.
5Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 1996;27:1304-5.
6Glenn OA, Barkovich AJ. Magnetic resonance imaging of the fetal brain and spine: An increasingly important tool in prenatal diagnosis, part 1. AJNR Am J Neuroradiol 2006;27:1604-11.
7Rajeswaran R, Chandrasekharan A, Joseph S, Venkata Sai PM, Dev B, Reddy S, et al. Ultrasound versus MRI in the diagnosis of fetal head and trunk anomalies. J Matern Fetal Neonatal Med 2009;22:115-23.
8Kirkinen P, Partanen K, Ryynänen M, Ordén MR. Fetal intracranial hemorrhage. Imaging by ultrasound and magnetic resonance imaging. J Reprod Med 1997;42:467-72.
9Elchalal U, Yagel S, Gomori JM, Porat S, Beni-Adani L, Yanai N, et al. Fetal intracranial hemorrhage (fetal stroke): Does grade matter? Ultrasound Obstet Gynecol 2005;26:233-43.
10Kang BK, Na DG, Ryoo JW, Byun HS, Roh HG, Pyeun YS, et al. Diffusion-weighted MR imaging of intracerebral hemorrhage. Korean J Radiol 2001;2:183-91.
11Lin DD, Filippi CG, Steever AB, Zimmerman RD. Detection of intracranial hemorrhage: Comparison between gradient-echo images and b(0) images obtained from diffusion-weighted echo-planar sequences. AJNR Am J Neuroradiol 2001;22:1275-81.