Continually Updated Clinical Reference
 
 
  All Sources     eMedicine     Medscape     Drug Reference     MEDLINE
 
eMedicine - Germinal Matrix Hemorrhage : Article by

Quick Find
Authors & Editors
Introduction
Differentials
Radiograph
CT SCAN
MRI
Ultrasound
Nuclear Medicine
Angiography
Intervention
Multimedia
References

Related Articles
Periventricular Leukomalacia




Patient Education
Click here for patient education.



Author: Andrew Leung, MD, Staff Physician, Department of Diagnostic Radiology, Kingston General Hospital

Andrew Leung is a member of the following medical societies: College of Physicians and Surgeons of Ontario

Coauthor(s): Omar Islam, MD, FRCP(C), Assistant Professor of Diagnostic Radiology, Queen's University; Consulting Staff, Department of Diagnostic Radiology, Division of Neuroradiology, Kingston General Hospital

Editors: Charles M Glasier, MD, Professor, Departments of Radiology and Pediatrics, University of Arkansas for Medical Sciences; Chief, Magnetic Resonance Imaging, Vice-Chief, Pediatric Radiology, Arkansas Children's Hospital; Bernard D Coombs, MB, ChB, PhD, Consulting Staff, Department of Specialist Rehabilitation Services, Hutt Valley District Health Board, New Zealand; Kieran McHugh, MBBCh, Honorary Lecturer, The Institute of Child Health; Consultant Pediatric Radiologist, Department of Radiology, Great Ormond Street Hospital for Children, London, UK; Robert M Krasny, MD, Consulting Staff, Department of Radiology, The Angeles Clinic and Research Institute; Eugene C Lin, MD, Consulting Staff, Department of Radiology, Virginia Mason Medical Center

Author and Editor Disclosure

Synonyms and related keywords: periventricular/intraventricular/subependymal hemorrhage, hypoxic-ischemic injury, persistent ventricular enlargement, PVH, IVH, HII, PVE, GM/IVH, GMH, neonatal/fetal cerebral hemorrhage

Background

Germinal matrix hemorrhage (GMH) and intraventricular hemorrhage (IVH) are the most common and most important neurologic injuries in preterm neonates.

The brain of a premature infant lacks the ability to autoregulate cerebral blood pressure. Fluctuations in cerebral blood pressure and flow can rupture the primitive germinal matrix vessels or lead to infarction of the metabolically active germinal matrix. The damage can extend into the periventricular white matter, resulting in significant neurologic sequelae, including cerebral palsy, mental retardation, and seizures. Injury to the germinal matrix has substantial mortality and morbidity rates.

Ultrasonography (US) is the primary imaging modality for the screening and diagnosis of GMH/IVH, and computed tomography (CT) scanning and magnetic resonance imaging (MRI) are used as supplementary tools.

Pathophysiology

A common lesion that characterizes the neuropathology of GMH/IVH is bleeding into the subependymal germinal matrix, with or without subsequent rupture into the lateral ventricle.

The pathogenesis of GMH is multifactorial. For simplification, the influences can be divided into intravascular, vascular, and extravascular factors. The autoregulation of blood flow and blood pressure in the germinal matrix circulation is primitive in premature infants, and the thin microvasculature of the germinal matrix is susceptible to rupture. The primitive mesenchymal and glial supportive tissues influence the extent of GMH.

Large fluctuations in blood flow and blood pressure can result, leading to injury to the germinal matrix vessels and subsequent hemorrhage. Alternatively, hypotension or hypoperfusion can lead to focal or diffuse infarction; the metabolically active germinal matrix is particularly vulnerable. Hemorrhage can occur in infarcted regions after reperfusion, but hemorrhage from any cause can be confined to the subependymal layer, or it can extend into the ventricles or brain parenchyma.

Sequelae of GMH/IVH include germinal matrix destruction, periventricular hemorrhagic infarction with subsequent encephalomalacia, and posthemorrhagic hydrocephalus.

Frequency

United States

Frequency of GMH is inversely related to the neonate's gestational age and birth weight.1 Improvements in obstetric and perinatal care have markedly decreased the incidence and severity of GMH/IVH, which was reported to be as high as 50% in 1977.2  Incidence is now 8-56% in infants; most nurseries report rates of 20-30%. The severity of GMH/IVH has decreased along with the incidence. Sheth et al observed that the rate for severe grade 3 or 4 IVH decreased from 70% in 1986 to 23% in 1995 and that the associated mortality rate decreased by 30% in the same period.2

International

The incidence of GMH varies throughout the world. Rates of GMH are most dependent on the birth rates of premature infants.

Mortality/Morbidity

The extent of intracranial hemorrhage and any associated parenchymal injury or hydrocephalus are the most important factors in predicting mortality and morbidity.

  • In 1983, Papile et al found major neurologic deficits in 10% of premature infants without GMH/IVH and in 28% of infants with GMH/IVH.3 Major deficits were present in 9% of infants with grade 1 GMH, in 11% with grade 2 injury, in 36% with grade 3 injury, and in 76% with grade 4 GMH/IVH. For grading, see Clinical Details. Cerebral palsy occurs in as many as 15% of these children.
  • The mortality rate in infants with GMH/IVH has been reported as high as 35%, compared with 13% in premature infants without GMH/IVH.4

Race

No significant racial predilection is known.

Sex

GMH affects the 2 sexes equally.

Age

Because maturation of the germinal matrix vasculature occurs in the perinatal period for premature infants, most cases of GMH/IVH occur in the first week of life for these infants. Approximately 50% of cases of GMH/IVH occur in the first day of life, and about 65% occur in the first week.

  • Neonates born at 32 weeks' gestation or earlier and those born with birth weights less than 1500 g are at particular risk.
  • It is unusual to find GMH/IVH in infants born after 34 weeks' gestation.

Anatomy

The germinal matrix is located in the subependyma of the ventricular walls and initially extends along the ventricles. At 8-28 weeks' gestation, the germinal matrix produces neurons and glial cells, which migrate to populate the cerebral cortex. Neurons are produced earlier in gestation, and glial cells are produced later. Involution of the germinal matrix toward the caudothalamic groove begins late in the second trimester and is nearly complete by 32 weeks' gestation.

The germinal matrix is metabolically active with a rich supply of blood via a thin, fragile capillary network. Arterial supply of the germinal matrix is provided by branches of the anterior cerebral artery: the arterioles from the recurrent artery of Heubner at the level of the foramen of Monro and the terminal branches of the lateral striate arteries, located more superiorly. Venous blood flows through the terminal vein, which drains via the internal cerebral vein into the vein of Galen. 

Clinical Details

The major risk factors for GMH include a young gestational age, low birth weight, acute amnionitis, and exposure to antenatal steroids for less than 48 hours. Other risk factors include the use of general anesthesia for cesarean delivery; Apgar scores that are less than 4 in the first minute or are less than 8 by 5 minutes; respiratory distress; persistent ductus arteriosus; anemia; and arterial catheterization.

The grading system created by Burstein et al in 1979 relies on the detection of blood in the subependymal germinal matrix and the ventricles as follows5:

  • Grade 1: Hemorrhage that is confined to the germinal matrix
  • Grade 2: Extension of the hemorrhage into the lateral ventricles without hydrocephalus
  • Grade 3: Ventricular hemorrhage with the presence of associated hydrocephalus
  • Grade 4: Parenchymal hemorrhage

Preferred Examination

US is the preferred screening and diagnostic tool for GMH. The portability of US allows imaging in the nursery with minimal disturbance of the infant. US also depicts GMHs that are larger than 5 mm, with a sensitivity of nearly 100% and specificity of 91%. Smaller GMHs, however, are more difficult to identify.

Power and pulsed-wave Doppler US can be used to identify preterm neonates who are at risk for GMH/IVH during their first week of life. Using this modality, clinicians can detect autoregulatory fluctuations in the preterm neonate's cerebral blood flow with examination of the lenticulostriate arteries; measurements of the peak velocity, resistive index, and coronal vascular cross-sectional area; and product of the peak velocity and vascular cross-sectional area.

CT scanning and MRI are also used and have better interobserver agreement. Because these modalities more readily depict small GMHs, CT scanning and MRI have a higher sensitivity than that of US. However, these imaging modalities require that the infant be moved from the nursery; there is also the possibility that sedation would be required.

Limitations of Techniques

All imaging modalities have relatively low negative predictive values (NPVs). In a 2000 study, Blankenberg et al found NPVs of 53% and 59% (irrespective of the modality) at 2-month and at 2-year follow-up, respectively.6 However, the absence of neuroimaging abnormalities in the infant does not exclude the possibility of later neurodevelopmental problems.



Periventricular Leukomalacia


Findings

Plain radiographs are not used in the evaluation of GMH.



Findings

IVH evolves in a predictable pattern. Acutely, it appears to hyperattenuate. After 7-10 days, the hemorrhage becomes isoattenuating relative to the brain parenchyma. Later, with clot retraction, a subependymal hematoma may develop into a fluid-filled cyst. The affected brain parenchyma may undergo atrophy and gliosis (see Image 1).

Degree of Confidence

Blankenberg et al found that CT scanning had nearly twice the sensitivity of US in the detection of GMH/IVH, and interobserver agreement with this modality was also improved relative to US.6

False Positives/Negatives

A normal CT scan finding for GMH/IVH does not exclude abnormal neurodevelopment; the NPV is 50-60% at age 2 years.



Findings

In the first 3 days after IVH, subependymal hematomas are isointense to slightly hypointense on T1-weighted MRIs (T1WI) and markedly hypointense on T2-weighted MRIs (T2WI). In the early subacute stage during days 4-7, the signal intensity increases on T1WIs. In the late subacute stage during days 7-14, the signal intensity increases on T2WIs. Over the next several months, the hemorrhage becomes hypointense on images obtained with both sequences, and ferromagnetic effects secondary to hemosiderin and ferritin predominate (see Images 3-8).

Degree of Confidence

As with CT scanning, Blankenberg et al found that MRI had nearly twice the sensitivity of US in the detection of GMH/IVH, and interobserver agreement with this modality was also improved relative to US.6

False Positives/Negatives

A normal image finding for GMH/IVH does not exclude abnormal neurodevelopment. The NPV is 50-60% at age 2 years.



Findings

Neurosonography is the primary modality for both screening and follow-up of GMH/IVH in neonates. US is portable, allowing imaging in the comfortable environment of the neonatal intensive care unit (NICU). US has NPVs similar to those of CT scanning and MRI. Current screening protocols recommend performing US studies on days 7-14 of life and between the fourth and sixth weeks of life. Many centers offer more frequent screening.

On US, acute subependymal hemorrhage appears as a homogeneous echogenic mass, often in the caudothalamic groove (see Image 2). The hematoma becomes less echogenic over time, beginning with the central portion. Subsequent to eventual clot retraction, a subependymal cyst may develop, or a linear echo may result (see Image 9).

Acutely, IVH also appears echogenic. Cerebrospinal fluid (CSF)–blood fluid levels may be observed. When large, the clot forms a cast of the ventricle (see Image 10) and may break up in the ventricle, resulting in low-level echoes that float in the CSF. The clot may also move when the patient's head position changes. With clot evolution, the hematoma becomes echolucent, starting centrally (see Images 11-12). Scanning through the posterior fontanelle may optimize visualization of occipital horn clots.

Intraparenchymal hemorrhage is usually located in the frontal and parietal lobes and appears acutely as an echogenic homogeneous mass. As the hemorrhage evolves, an echogenic rim with a sonolucent center forms. After 2-3 months, a porencephalic cyst (if the lesion communicates with a ventricle) or encephalomalacia may develop (see Image 13).

Power and pulsed-wave Doppler US may be useful in identifying preterm neonates who are at risk of GMH/IVH during their first week of life. The sonograms may depict autoregulatory fluctuations in cerebral blood flow.

Degree of Confidence

Neurosonography depicts GMHs that are larger than 5 mm with a sensitivity of 100% and a specificity of 91%.

False Positives/Negatives

IVH may blend imperceptibly with the choroid plexus, which has a similar echo texture; thus, asymmetric thickness of the choroid plexus should be viewed with suspicion. The lack of abnormality with US does not exclude the possibility of later neurodevelopmental problems.



Findings

Scintigraphy is not used in the evaluation of GMH.



Findings

Cerebral angiography is not used in the evaluation of GMH.



Medical/Legal Pitfalls

  • False-negative findings

    • US, CT scanning, and MRI all have low NPVs of approximately 60%.
    • Normal imaging findings must be viewed with caution.

Special Concerns

  • Prenatal GMH
    • Not all cases of hemorrhage are related to perinatal or postnatal stresses alone.
    • With fetal US and fetal MRI, GMH/IVH can be identified in utero, remote in time from delivery.
    • Though the incidence is unknown, prenatal factors may affect the development of hemorrhage.



Media file 1:  Axial nonenhanced computed tomography scan. The image shows bilateral grade 2 subependymal and intraventricular germinal matrix hemorrhage without hydrocephalus.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  CT

Media file 2:  Coronal sonogram of a 1-day-old neonate. The image demonstrates a bilateral grade 2 subependymal hemorrhage. Note the echogenic clot in the left temporal horn (arrow, right side).
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound

Media file 3:  Sagittal spoiled gradient-recalled echo magnetic resonance image (MRI) of a fetus at 34 weeks' gestation in a 29-year-old woman. The MRI was performed to investigate fetal tachycardia; the image demonstrates intraventricular hemorrhage and parenchymal hematoma.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 4:  Sagittal spoiled gradient-recalled echo (SPGR) magnetic resonance image (MRI) of a fetus at 34 weeks' gestation in a 29-year-old woman (same patient as in Image 3). A sonogram was initially performed to investigate fetal tachycardia and revealed hydrocephalus. Subsequent SPGR MRI demonstrated a grade 4 germinal matrix hemorrhage in utero.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 5:  Sagittal spoiled gradient-recalled echo magnetic resonance image of a fetus. The image demonstrates a grade 4 germinal matrix hemorrhage in utero.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 6:  Axial single-shot, fast spin-echo, T2-weighted magnetic resonance image of a fetus. The image demonstrates a grade 4 germinal matrix hemorrhage in utero.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 7:  Coronal single-shot, fast spin-echo, T2-weighted magnetic resonance image of a fetus. The image demonstrates a grade 4 germinal matrix hemorrhage in utero.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 8:  Axial diffusion-weighted magnetic resonance image (MRI) of a fetus. The MRI demonstrates no evidence of an acute infarct. Abnormal parenchymal signal intensity seen on MRIs obtained with other sequences is most compatible with hemorrhage.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  MRI

Media file 9:  Sagittal sonogram. The image demonstrates a grade 3 germinal matrix hemorrhage and shows a subependymal focus of hyperechogenicity (arrow) that represents hemorrhage, as well as intraventricular hemorrhage and hydrocephalus.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound

Media file 10:  Sagittal sonogram. The image shows intraventricular hemorrhage; the clot forms a cast of the ventricle.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound

Media file 11:  Sagittal neurosonogram. The image demonstrates a grade 3 germinal matrix hemorrhage with an intraventricular clot and hydrocephalus.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound

Media file 12:  Coronal neurosonogram. The image demonstrates (right) grade 4 germinal matrix and intraventricular hemorrhage and (left) grade 1 germinal matrix hemorrhage.
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound

Media file 13:  Sagittal sonogram. The images shows cystic encephalomalacia as a sequela of previous germinal matrix hemorrhage. In a region of an earlier parenchymal hemorrhage, a cerebrospinal fluid-filled cyst has formed (arrow).
Click to see larger pictureClick to see detailView Full Size Image
Media type:  Ultrasound



  1. Kadri H, Mawla AA, Kazah J. The incidence, timing, and predisposing factors of germinal matrix and intraventricular hemorrhage (GMH/IVH) in preterm neonates. Childs Nerv Syst. Sep 2006;22(9):1086-90. [Medline].
  2. Sheth RD. Trends in incidence and severity of intraventricular hemorrhage. J Child Neurol. Jun 1998;13(6):261-4. [Medline].
  3. Papile LA, Munsick-Bruno G, Schaefer A. Relationship of cerebral intraventricular hemorrhage and early childhood neurologic handicaps. J Pediatr. Aug 1983;103(2):273-7. [Medline].
  4. Hawgood S, Spong J, Yu VY. Intraventricular hemorrhage. Incidence and outcome in a population of very-low-birth-weight infants. Am J Dis Child. Feb 1984;138(2):136-9. [Medline].
  5. Burstein J, Papile LA, Burstein R. Intraventricular hemorrhage and hydrocephalus in premature newborns: a prospective study with CT. AJR Am J Roentgenol. Apr 1979;132(4):631-5. [Medline].
  6. Blankenberg FG, Loh NN, Bracci P, et al. Sonography, CT, and MR imaging: a prospective comparison of neonates with suspected intracranial ischemia and hemorrhage. AJNR Am J Neuroradiol. Jan 2000;21(1):213-8. [Medline][Full Text].
  7. Antoniuk S, da Silva RV. [Periventricular and intraventricular hemorrhage in the premature infants] [Spanish]. Rev Neurol. Aug 1-15 2000;31(3):238-43. [Medline].
  8. Balasubramaniam J, Del Bigio MR. Animal models of germinal matrix hemorrhage. J Child Neurol. May 2006;21(5):365-71. [Medline].
  9. Batton DG, Holtrop P, DeWitte D, Pryce C, Roberts C. Current gestational age-related incidence of major intraventricular hemorrhage. J Pediatr. Oct 1994;125(4):623-5. [Medline].
  10. Blankenberg FG, Norbash AM, Lane B, et al. Neonatal intracranial ischemia and hemorrhage: diagnosis with US, CT, and MR imaging. Radiology. Apr 1996;199(1):253-9. [Medline].
  11. Boal DK, Watterberg KL, Miles S, Gifford KL. Optimal cost-effective timing of cranial ultrasound screening in low-birth-weight infants. Pediatr Radiol. 1995;25(6):425-8. [Medline].
  12. Carson SC, Hertzberg BS, Bowie JD, Burger PC. Value of sonography in the diagnosis of intracranial hemorrhage and periventricular leukomalacia: a postmortem study of 35 cases. AJNR Am J Neuroradiol. Jul-Aug 1990;11(4):677-83. [Medline].
  13. Duncan CC, Ment LR. Intraventricular hemorrhage and prematurity. Neurosurg Clin N Am. Oct 1993;4(4):727-34. [Medline].
  14. Ghazi-Birry HS, Brown WR, Moody DM, et al. Human germinal matrix: venous origin of hemorrhage and vascular characteristics. AJNR Am J Neuroradiol. Feb 1997;18(2):219-29. [Medline][Full Text].
  15. Levy ML, Masri LS, McComb JG. Outcome for preterm infants with germinal matrix hemorrhage and progressive hydrocephalus. Neurosurgery. Nov 1997;41(5):1111-7; discussion 1117-8. [Medline].
  16. Ment LR, Oh W, Philip AG, et al. Risk factors for early intraventricular hemorrhage in low birth weight infants. J Pediatr. Nov 1992;121(5 Pt 1):776-83. [Medline].
  17. Morioka T, Hashiguchi K, Nagata S, et al. Fetal germinal matrix and intraventricular hemorrhage. Pediatr Neurosurg. 2006;42(6):354-61. [Medline][Full Text].
  18. Paneth N, Pinto-Martin J, Gardiner J, et al. Incidence and timing of germinal matrix/intraventricular hemorrhage in low birth weight infants. Am J Epidemiol. Jun 1 1993;137(11):1167-76. [Medline].
  19. Pinto-Martin JA, Riolo S, Cnaan A, et al. Cranial ultrasound prediction of disabling and nondisabling cerebral palsy at age two in a low birth weight population. Pediatrics. Feb 1995;95(2):249-54. [Medline].
  20. Salafia CM, Minior VK, Rosenkrantz TS, et al. Maternal, placental, and neonatal associations with early germinal matrix/intraventricular hemorrhage in infants born before 32 weeks' gestation. Am J Perinatol. Nov 1995;12(6):429-36. [Medline].
  21. Tarby TJ, Volpe JJ. Intraventricular hemorrhage in the premature infant. Pediatr Clin North Am. Oct 1982;29(5):1077-104. [Medline].
  22. Taylor GA. New concepts in the pathogenesis of germinal matrix intraparenchymal hemorrhage in premature infants. AJNR Am J Neuroradiol. Feb 1997;18(2):231-2. [Medline][Full Text].
  23. Townsend SF, Rumack CM, Thilo EH, Merenstein GB, Rosenberg AA. Late neurosonographic screening is important to the diagnosis of periventricular leukomalacia and ventricular enlargement in preterm infants. Pediatr Radiol. May 1999;29(5):347-52. [Medline].
  24. Volpe JJ. Intraventricular hemorrhage in the premature infant--current concepts. Part I. Ann Neurol. Jan 1989;25(1):3-11. [Medline].

Germinal Matrix Hemorrhage excerpt

Article Last Updated: Jun 5, 2007