- Email: [email protected]
Hypoxic-Ischemic Injury: Utility of Susceptibility-Weighted Imaging
Pediatric Neurology 45 (2011) 220e224
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
Hypoxic-Ischemic Injury: Utility of Susceptibility-Weighted Imaging Gene Kitamura MD a, Daniel Kido MD a, Nathaniel Wycliffe MD a, J. Paul Jacobson MD a, Udochukwu Oyoyo MPH a, Stephen Ashwal MD b, * a b
Department of Radiology, Loma Linda University School of Medicine, Loma Linda, California Division of Child Neurology, Department of Pediatrics, Loma Linda University School of Medicine, Loma Linda, California
article information
abstract
Article history: Received 22 February 2011 Accepted 23 June 2011
Magnetic resonance imaging is increasingly used to assess neonatal hypoxic-ischemic injury, and several scoring systems were developed to predict neurologic outcomes in these patients. We examined the magnetic resonance imaging studies of 33 neonates/infants who manifested acute perinatal hypoxicischemic injuries. Using a seven-point susceptibility-weighted imaging categorical grading scale, each patient received a “prominence of vein” score, which was dichotomized into a “normal” or “abnormal” group. Six-month outcomes were assessed using the Pediatric Cerebral Performance Category Scale. We then determined whether “prominence of vein” scores correlated with neurologic outcomes in patients with hypoxic-ischemic injuries, and compared these results with the Barkovich magnetic resonance imaging scoring system. Patients with “normal” “prominence of vein” scores demonstrated better outcomes (mean Pediatric Cerebral Performance Category Scale value ¼ 2) than patients with “abnormal” “prominence of vein” scores (mean Pediatric Cerebral Performance Category Scale value ¼ 4). The dichotomized “prominence of vein” groups demonstrated correlations with the Barkovich magnetic resonance imaging scores of the proton density-weighted basal ganglia, watershed, and combined basal ganglia/watershed regions. The susceptibility-weighted imaging categorical grading scale may aid in predicting neurologic outcomes after hypoxic-ischemic injuries. Ó 2011 Elsevier Inc. All rights reserved.
Introduction Neonatal hypoxic-ischemic injury is a major cause of morbidity and mortality. Its prognoses correlate with the severity of hypoxicischemic encephalopathy [1]. The evaluation of neonates with hypoxic-ischemic injury has become increasingly reliant on magnetic resonance imaging to assess injury severity and predict outcomes [2-7]. Susceptibility-weighted imaging involves a magnetic resonance imaging sequence that is used to evaluate trauma, tumors, venous diseases, and arteriovenous malformations [8]. This sequence is particularly sensitive at detecting extravascular blood products and intravascular deoxygenated blood [9]. Previous studies indicated an increased level of deoxygenated blood with hypoxic-ischemic injury [10], and susceptibility-weighted imaging will likely detect neonatal cerebral insults with more sensitivity than currently used magnetic resonance imaging sequences. We developed a susceptibility-weighted imaging categorical grading scale derived from the level of deoxyhemoglobin in the deep * Communications should be addressed to: Dr. Ashwal; Department of Pediatrics, Loma Linda University School of Medicine; 11175 Campus Street, Room A1120G; Loma Linda, CA 92350. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pediatrneurol.2011.06.009
cerebral venous system, because the quantity of deoxyhemoglobin reflects the magnitude of oxygen extraction or veno-occlusion. Our study sought to determine whether susceptibility-weighted imaging could predict outcomes in neonates and very young infants with hypoxic-ischemic injuries, and to compare these results with the well-accepted and standardized magnetic resonance imaging scoring system of Barkovich et al. [6]. Patients and Methods Subjects This study was approved by the Institutional Review Board of Loma Linda University Medical Center. A cohort of neonates and infants who had undergone magnetic resonance imaging evaluations at Loma Linda University Children’s Hospital from 2002-2007 was retrospectively reviewed. Inclusion criteria comprised: (1) evidence of neonatal hypoxic-ischemic injury within the first 3 days after birth, defined by at least one of several findings that included neonatal encephalopathy, 5-minute Apgar score of 5, umbilical artery pH 7.1, or umbilical cord base excess >10 [11]; and (2) the acquisition of magnetic resonance imaging and susceptibility-weighted imaging between 0-60 days after birth. Exclusion criteria comprised proven congenital malformations, infection, or metabolic disease. Relevant demographic data from the antenatal, perinatal, and postnatal periods were collected from medical records. Maternal substance use, including cigarettes and drugs, was defined by a positive patient history or urine drug screen. Maternal inflammatory state was defined as a documented maternal infection or presence of
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224 antepartum or peripartum maternal fever. Complicated vaginal delivery was defined as failed vacuum delivery or an arrest of descent. Fetal distress was defined as nonreassuring decelerations, as documented by the treating physician. A placental or cord insult was defined as cord prolapse, vasa previa, nuchal cord, cord rupture, abruptio placentae, or uterine rupture. The resuscitation score was based on the severity of resuscitation at birth (1, no intervention; 2, blow-by oxygen; 3, endotracheal suctioning; 4, bag-mask positive pressure ventilation; 5, endotracheal intubation with positive pressure ventilation; and 6, endotracheal intubation with ventilation and the administration of sodium bicarbonate) [11]. Magnetic resonance imaging and grading Magnetic resonance imaging was performed using a conventional 1.5-T wholebody magnetic resonance imaging system (Magnetom Vision, Siemens Medical Solutions, Iselin, NJ). The susceptibility-weighted imaging sequence consists of a strongly susceptibility-weighted, low-bandwidth (78 Hz/pixel), three-dimensional fast low-angle shot sequence (TR/TE ¼ 57/40 ms, FA ¼ 20 ), with first-order flow compensation in three orthogonal directions. Thirty-two partitions of 2 mm were acquired using a rectangular FOV (5/8 of 256 mm) and a matrix size of 160 512, resulting in 16 slices with an effective slice thickness of 4 mm. The acquisition time was 4.25 minutes. Images underwent additional post-processing, using a phase mask that led to an enhancement of the phase differences between paramagnetic substances and the surrounding tissue [12]. The susceptibility-weighted imaging sequence was used to evaluate the cerebral venous system. Emphasis was placed on examining the prominence of deep medullary veins instead of the entire venous system, because they were more consistently present and easier to evaluate than the subependymal or cortical veins. Each susceptibility-weighted imaging sequence was evaluated with a categorical grading scale, ranging from 1-7 (Table 1 and Fig 1). Two neuroradiologists (D.K. and J.P.J.), blinded to patients’ neurologic status, assigned “prominence of vein” scores to each patient’s susceptibility-weighted imaging sequence. When disagreement in the “prominence of vein” score of a patient occurred between the two neuroradiologists, an average “prominence of vein” score was used for analyses. In addition to receiving a “prominence of vein” score, each patient’s magnetic resonance imaging was evaluated by a neuroradiologist (N.W.) using a previously validated system for the prognostic evaluation of patients with hypoxic-ischemic injuries (i.e., the Barkovich score). The T1-weighted (TR/TE ¼ 9000/110 ms; 5 mm thick), proton density-weighted (TR/TE ¼ 3000/22 ms; 5 mm thick), T2-weighted (TR/TE ¼ 300/120 ms; 5 mm thick), and diffusion-weighted (TR/TE ¼ 4000/11 ms; 5 mm thick) magnetic resonance imaging sequences were evaluated to assess injury to the basal ganglia, watershed, and combined basal ganglia and watershed areas. Signal abnormalities were evaluated in the basal ganglia and combined basal ganglia and watershed regions with a scale ranging from 0-4, and insults to the watershed region were evaluated with a scale ranging from 0-5 [6]. Assessment of outcomes Neurologic outcomes at age 6 months were evaluated with the previously validated Pediatric Cerebral Performance Category Scale: (1) normal, at age-appropriate level; (2) mild disability, i.e., conscious, alert, able to interact at age-appropriate level, with possible mild neurologic deficit; (3) moderate disability, i.e., conscious, with age-appropriate independent activities, and with learning disabilities among schoolage children or enrollment in special education; (4) severe disability, i.e., conscious, and dependent on others for daily support; (5) coma or vegetative state, i.e., any degree of coma without the criteria of brain death; and (6) brain death, i.e., apnea,
221
areflexia, and/or electroencephalographic silence [13]. The Pediatric Cerebral Performance Category Scale was previously implemented to assess the neurologic status of pediatric patients at less than 1 year of age [14]. A neurologist blinded to the “prominence of vein” and Barkovich scores assigned a Pediatric Cerebral Performance Category Scale value to each patient during 6-month follow-up, based on a retrospective chart review. Statistical analysis Before analyses, the “prominence of vein” values were dichotomized into a “normal” group with “prominence of vein” values ranging from 2-4, and an “abnormal” group with “prominence of vein” values of 1 and 5-7. “Prominence of vein” values were dichotomized into the two groups according to the concept that the outlying “prominence of vein” values most likely represented irregular cerebral metabolism. The Pediatric Cerebral Performance Category Scale values were also dichotomized into good (1 and 2) and poor (3-6) outcome groups for analyses. Differences in antenatal, perinatal, and postnatal factors between the “prominence of vein” and Pediatric Cerebral Performance Category Scale groups were analyzed using the Pearson c2 test for nominal values, and the Mann-Whitney U test for ordinal and parametric values. The difference in Pediatric Cerebral Performance Category Scale values between the two “prominence of vein” groups, i.e., “normal” and “abnormal,” was analyzed using the Mann-Whitney U test. Differences in individual Barkovich scores between the dichotomized Pediatric Cerebral Performance Category Scale groups were analyzed using the Mann-Whitney U test. In addition, differences in individual Barkovich scores between the “prominence of vein” groups were analyzed using the Mann-Whitney U test. Interobserver variability for the susceptibility-weighted imaging categorical grading scale was analyzed by calculating a k value. P < 0.05 was considered statistically significant.
Results Although the magnetic resonance imaging and susceptibilityweighted imaging sequences were performed up to 60 days after birth, all patients in this study demonstrated hypoxic-ischemic injury based on the inclusion criteria during the postnatal period, within 3 days after birth. No significant differences were demonstrated in the antenatal, perinatal, and postnatal clinical variables between the “normal” and “abnormal” “prominence of vein” groups (Table 2). The mean age of patients at the time of susceptibilityweighted imaging scan was comparable between the “normal” and “abnormal” “prominence of vein” groups (P ¼ 0.98). Neonates with “normal” “prominence of vein” values (2-4) demonstrated improved neurologic outcomes compared with those scoring “abnormal” “prominence of vein” values (1 and 5-7), based on their Pediatric Cerebral Performance Category Scale values (P ¼ 0.03). The average Pediatric Cerebral Performance Category Scale value for the “normal” “prominence of vein” group was 2 vs the “abnormal” “prominence of vein” group with an average Pediatric Cerebral Performance Category Scale value of 4. Agreement in
Table 1. Grading scale of deep medullary veins demonstrated on susceptibility-weighted imaging sequences Grade
Category
Appearance of Deep Medullary Veins on Susceptibility-Weighted Imaging
1
Absent DMVs
2
Faint DMVs
3
Minimal DMVs
4
Mildly prominent DMVs
5
Moderately prominent DMVs
6
Severely prominent DMVs
7
Extremely prominent DMVs
No visible deep medullary veins, absent or nearly absent signal in the subependymal veins, and a few nonprominent cortical veins. Equivocal low signal in the medullary veins, some low signal in the subependymal veins, and a few nonprominent cortical veins. Several definite, fine, light gray deep medullary veins, limited to the deep white matter, are visible, the subependymal veins are visible but not prominent, and cortical veins are occasionally prominent. Many dark, distinct deep medullary veins are visible, either diffusely or regionally. The deep medullary veins are wider and better demarcated, but do not extend to the most superficial layers of the deep white matter. The subependymal veins are usually prominent, and a few cortical veins may be prominent. The deep medullary veins are dark, numerous, and wider, and extend into the superficial white matter, either diffusely or regionally. The subependymal veins are prominent, and numerous cortical veins may be prominent. Numerous deep medullary veins are very dark and extend through the superficial white matter nearly to the cortex, either diffusely or regionally. Subependymal veins and cortical veins are very prominent. A dark background blush in the white matter may be evident. Numerous, thick, dark deep medullary veins extend to the cortex. They may be irregular. The subependymal and cortical veins are prominent.
Abbreviation: DMVs ¼ Deep medullary veins
222
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224
Figure 1. Susceptibility-weighted axial images illustrate the seven “prominence of vein” categories, with emphasis on the disc-modulating veins system. (A) Category 1, absent (no visible deep medullary are visible but not prominent). (B) Category 2, faint (visible deep medullary veins, nearly absent subependymal veins, and a few nonprominent cortical veins). (C) Category 3, minimal (several definite, fine, light gray deep medullary veins are visible and limited to the deep white matter; the subependymal veins are present but not prominent, and cortical veins are occasionally prominent). (D) Category 4, mildly prominent (many dark, distinct deep medullary veins are visible, either diffusely or regionally; the deep medullary veins are wider and better demarcated, but do not extend to the most superficial layers of the deep white matter; the subependymal veins are usually prominent, and a few cortical veins may be prominent). (E) Category 5, moderately prominent (the deep medullary veins are even darker, more numerous, and wider, extending into the superficial white matter, either diffusely or regionally; the subependymal veins are prominent and more cortical veins may be prominent). (F) Category 6, severely prominent (numerous deep medullary veins are very dark and extend through the superficial white matter nearly to the cortex, either diffusely or regionally; subependymal veins and cortical veins are very prominent; a dark background blush may be evident in the white matter). (G) Category 7, extremely prominent (numerous, thick, dark deep medullary veins extending to the cortex; they may be irregular; the subependymal and cortical veins are prominent). Separating categories 1 and 2 on the basis of deep medullary veins is difficult and usually requires examination of the subependymal veins. The category 6 image contains varying sizes of hypodensity, caused by multiple hemorrhages. Images A-F were taken of patients in this study, aged 12, 5, 51, 8, 2, and 7 days, respectively. Image G was taken of a 3-month old infant who was not part of this study but was included because no category 7 patients were available in the study cohort. This patient had undergone nonaccidental trauma and was being examined for brain death.
“prominence of vein” values between the two neuroradiologists (D.K. and J.P.J.) was acceptable, with a k value of 0.82. Table 3 summarizes the differences in clinical variables between the good vs poor outcome groups. The presence of clinical seizures was the only variable associated with worse neurologic outcomes at 6-month follow-up (P ¼ 0.02). Differences in Barkovich scores between the outcome groups are summarized in Table 4. Patients with good outcomes demonstrated lower proton density-weighted Barkovich scores of the basal ganglia (P ¼ 0.015) and combined basal ganglia and watershed region (P ¼ 0.035), compared with patients manifesting poor outcomes. The comparison of Barkovich scores between “normal” vs “abnormal” “prominence of vein” groups is summarized in Table 5. The “normal” “prominence of vein” group demonstrated lower proton density-weighted Barkovich scores of the basal ganglia (P ¼ 0.004), watershed (P ¼ 0.04), and combined basal ganglia and watershed regions (P ¼ 0.012), compared with the “abnormal” “prominence of vein” group. In contrast, no differences were evident in T1-weighted, T2-weighted, and diffusion-weighted Barkovich scores between the “normal” vs “abnormal” “prominence of vein” groups. Discussion Our findings demonstrated that susceptibility-weighted imaging based “prominence of vein” scores are helpful in predicting intermediate-term neurologic outcomes after neonatal hypoxic-ischemic injuries. Patients with “normal” “prominence of vein” scores demonstrated better outcomes (mean Pediatric Cerebral Performance Category Scale value of 2) compared with patients with “abnormal” “prominence of vein” scores (mean Pediatric Cerebral Performance Category Scale value of 4).
Our analyses demonstrated that seizures were more frequent in patients with poor outcomes, which was expected, because seizures are associated with poor neurologic outcomes after hypoxicischemic brain injuries [15]. Higher Barkovich scores, particularly of the proton density-weighted basal ganglia and combined basal ganglia and watershed regions, were demonstrated in the poor outcome group, which was expected, because higher Barkovich scores, particularly of the proton density-weighted combined basal ganglia and watershed region, are associated with poor neurologic outcomes after hypoxic-ischemic brain injury [6]. Patients with abnormal “prominence of vein” values (values of 1 and 5-7) A “prominence of vein” value of 1, representing decreased levels of deoxyhemoglobin in the deep medullary veins, may be secondary to cerebral swelling compressing the vessels after hypoxic-ischemic injury or because of cerebral hemodynamic changes after hypoxicischemic injury. Previous studies using perfusion-weighted imaging demonstrated marked cerebral hyperperfusion in newborns with hypoxic-ischemic encephalopathy [16,17]. Early scans demonstrated diffuse involvement of the brain, but later scans indicated increased relative cerebral blood flow and relative cerebral blood volume only in the more severely abnormal regions [16]. Early hyperperfusion represents a reperfusion phase, and is likely attributable to impaired cerebral autoregulation secondary to vasoparalysis [18]. Late hyperperfusion is thought to persist because the more severely infarcted areas may take longer to regain cerebral blood flow autoregulation [16]. Previous studies also investigated the level of cerebral fractional oxygen extraction in newborns with hypoxic-ischemic injury. One
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224 Table 2. Comparison of clinical characteristics and laboratory values between normal and abnormal POV groups
Number Male sex Six-month PCPCS S.D. Antenatal factors Gestational diabetes Preeclampsia Maternal substance use Maternal inflammatory state Perinatal factors Fetal distress Complicated vaginal delivery Cesarean section delivery Emergency cesarean section delivery Placental or cord insult Postnatal factors Gestational age (weeks) S.D. Birth weight (g) S.D. Head circumference (cm) S.D. Length (cm) S.D. Resuscitation score (1-6) S.D. One-minute Apgar score S.D. Five-minute Apgar score S.D. Meconium staining Clinical seizures Age at time of magnetic resonance imaging (days) S.D. Hemoglobin S.D. Hematocrit S.D. Mean arterial pressure S.D. Blood glucose S.D.
“Abnormal” POV Group*
“Normal” POV Groupy
7 4 (57%) 42
26 19 (73%) 21
0.42 0.03
0 2 1 3
(0%) (29%) (14%) (43%)
1 5 3 4
0.60 0.59 0.84 0.12
5 1 5 5
(71%) (14%) (71%) (71%)
20 4 20 13
(4%) (19%) (11%) (15%) (77%) (15%) (77%) (50%)
0.76 0.94 0.76 0.31
7 (27%)
0.93
37 2 2861 624 35 3 49 3 61 11 31 2 (29%) 5 (71%) 11 5
37 3 2936 1056 33 4 47 6 51 22 42 7 (27%) 10 (38%) 17 16
0.98 0.82 0.31 0.64 0.33 0.22 0.56 0.93 0.12 0.98
3 7 15 11
14 43 53 77
3 10 7 23
Table 3. Comparison of clinical characteristics and laboratory values between good and poor outcome groups
P Value
2 (29%)
13 37 54 75
223
0.10 0.09 0.76 0.59
Abbreviations: PCPCS ¼ Pediatric Cerebral Performance Category Scale POV ¼ “Prominence of vein” * “Abnormal” POV group represents a POV range of 1 and 5-7. y “Normal” POV group represents a POV range of 2-4.
study calculated cross-brain oxygen extraction by comparing the jugular venous bulb oxygen content with the systemic arterial oxygen content, and demonstrated that cerebral fractional oxygen extraction was much lower in neonates who eventually died or sustained severe brain damage [19]. Another study, using nearinfrared spectroscopy and amplitude-integrated electroencephalography, demonstrated decreased cerebral fractional oxygen extraction in infants who eventually manifested poor neurologic outcomes [20]. The decreased level of cerebral fractional oxygen extraction is thought to result from secondary energy failure [20]. When increased cerebral blood flow and decreased cerebral fractional oxygen extraction are taken together, it may explain why a “prominence of vein” value of 1 represents pathology that leads to poor neurologic outcomes in patients with hypoxic-ischemic injury. “Prominence of vein” values of 5-7, representing increased levels of deoxyhemoglobin in the cerebral circulation, were also considered abnormal. Studies of cerebral hemodynamic changes after brain injury suggest possible underlying mechanisms. After cerebral injury, oxygen metabolism is maintained despite a reduced cerebral blood flow secondary to increased cerebral fractional oxygen extraction [21,22]. This hypoperfusion appears to result from increased microvascular resistance, possibly because of occlusion by microthrombi and circulating blood cells [21,23,24]. Moreover, regional hypoxia was demonstrated in injured brain tissue [22,23]. Ischemia, resulting in cerebral hypoxia and increased cerebral fractional oxygen extraction, likely results in increased levels of deoxyhemoglobin in the cerebral circulation, and presents as a high “prominence of vein” value (5-7) on susceptibilityweighted imaging scans, representing abnormal cerebral metabolism and likely leading to poor neurologic outcomes.
Number Male sex Antenatal factors Gestational diabetes Preeclampsia Maternal substance use Maternal inflammatory state Perinatal factors Fetal distress Complicated vaginal delivery Cesarean section delivery Emergency cesarean section delivery Placental or cord insult Postnatal factors Gestational age (weeks) S.D. Birth weight (g) S.D. Head circumference (cm) S.D. Length (cm) S.D. Resuscitation score (1-6) S.D. One-minute Apgar score S.D. Five-minute Apgar score S.D. Meconium staining Clinical seizures Hemoglobin S.D. Hematocrit S.D. Mean arterial pressure S.D. Blood glucose S.D.
Good Outcomes*
Poor Outcomesy
16 12 (75%)
17 11 (65%)
P Value
0.52
0 2 2 4
(0%) (13%) (13%) (25%)
1 5 2 3
(6%) (29%) (12%) (18%)
0.33 0.24 0.95 0.61
10 3 11 6 3
(63%) (19%) (69%) (38%) (19%)
15 2 14 12 6
(88%) (12%) (83%) (71%) (35%)
0.09 0.58 0.36 0.06 0.29
38 2 3078 773 34 2 49 3 51 22 42 3 (18%) 11 (65%) 14 3 42 11 54 11 76 15
0.07 0.55 0.75 0.27 0.14 0.20 0.64 0.20 0.02 0.55 0.75 0.44 0.46
36 4 2761 1127 32 5 46 6 51 32 42 6 (38%) 4 (25%) 14 2 41 7 52 7 81 19
* Good outcomes represent a Pediatric Cerebral Performance Category Scale range of 1-2. y Poor outcomes represent a Pediatric Cerebral Performance Category Scale range of 3-6.
Comparison of “prominence of vein” values to Barkovich scores The susceptibility-weighted imaging categorical grading scale was compared with the Barkovich magnetic resonance imaging scoring system, which was previously validated to predict intermediate-term neurologic outcomes in neonates with hypoxicischemic injuries [6,11]. The “abnormal” “prominence of vein” group was expected to exhibit higher Barkovich scores because higher Barkovich scores had been correlated with worse neuromotor outcomes [6,11]. One subset of the Barkovich score, the proton density-weighted combined basal ganglia and watershed region score, was observed to best discriminate between patients who developed good vs poor intermediate-term neurologic Table 4. Barkovich scores for good and poor outcome groups Good Outcomes* (value S.D.) Basal ganglia scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Basal ganglia/watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted
Poor Outcomesy (value S.D.)
P Values
1.1 0.7 0.7 0.6
1.1 1.1 1.0 1.0
2.2 2.0 1.7 1.5
0.6 1.2 1.4 1.5
0.082 0.015 0.050 0.204
0.5 0.7 0.7 0.8
1.0 1.3 1.3 1.3
1.1 1.5 1.4 1.4
1.8 2.0 2.0 2.0
0.623 0.355 0.449 0.552
0.9 0.9 0.8 0.9
1.1 1.3 1.3 1.4
1.7 1.8 1.6 1.4
1.1 1.4 1.4 1.5
0.076 0.034 0.050 0.284
* Good outcomes represent a Pediatric Cerebral Performance Category Scale range of 1-2. y Poor outcomes represent a Pediatric Cerebral Performance Category Scale range of 3-6.
224
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224
Table 5. Barkovich scores for normal and abnormal POV groups “Normal” POV Group* (value S.D.) Basal ganglia scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Basal ganglia/watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted
“Abnormal” POV Groupy (value S.D.)
P Values
1.5 1.0 0.9 0.8
1.0 1.0 1.0 1.0
2.7 3.0 2.3 2.0
0.9 1.0 1.7 1.9
0.216 0.004 0.074 0.167
0.6 0.7 0.7 0.7
1.2 1.3 1.3 1.3
1.7 3.0 2.6 2.6
2.4 2.2 2.3 2.3
0.689 0.038 0.074 0.085
1.2 1.0 1.0 0.9
1.1 1.3 1.2 1.3
2.0 2.7 2.1 2.0
1.4 1.3 1.6 1.8
0.358 0.012 0.109 0.182
Abbreviation: POV ¼ “Prominence of vein” * “Normal” POV group represents a POV range of 2-4. y “Abnormal” POV group represents a POV range of 1 and 5-7.
outcomes [6]. When “prominence of vein” groups were compared, a much higher Barkovich score was demonstrated in the “abnormal” “prominence of vein” group with proton densityweighted images of the basal ganglia, watershed, and combined basal ganglia and watershed regions. This comparison demonstrated that the susceptibility-weighted, imaging based “prominence of vein” grading scale correlated well with the Barkovich magnetic resonance imaging scoring system. Study limitations Several limitations restrict the clinical utility of our study. The sample size, though adequate to achieve statistically significant findings, was small (n ¼ 33). In addition, the distribution of patients was skewed, with only seven patients in the “abnormal” “prominence of vein” group. The two “prominence of vein” groups were shown to be comparable in terms of demographic data analyses. However, a more robust study would have been possible with a similar group size for the “normal” vs “abnormal” “prominence of vein” groups. The lack of long-term follow-up with patients comprised another limiting factor. Our sample of 33 patients all received at least a 6-month follow-up, but many did not receive 12month or 24-month follow-up evaluations. Finally, an inherent weakness of the susceptibility-weighted imaging categorical grading scale involves its use of ordinal rather than parametric values, which restricts the application of certain statistical tools in analyses. Conclusion Susceptibility-weighted imaging may eventually become a useful tool in predicting intermediate-term to long-term neurologic outcomes in patients with hypoxic-ischemic injuries. Patients with “normal” “prominence of vein” values (2-4) are much likelier to experience normal neurologic outcomes than patients with “abnormal” “prominence of vein” values (1 and 5-7). Because susceptibility-weighted imaging can be obtained together with conventional magnetic resonance imaging sequences, the
susceptibility-weighted imaging categorical grading scale may eventually be used in conjunction with other grading systems to aid in predicting intermediate-term to long-term neurologic outcomes in patients with hypoxic-ischemic injuries. References [1] Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress: A clinical and electroencephalographic study. Arch Neurol 1976;33:696e705. [2] Jyoti R, O’Neil R, Hurrion E. Predicting outcome in term neonates with hypoxic-ischaemic encephalopathy using simplified MR criteria. Pediatr Radiol 2006;36:38e42. [3] Rutherford MA, Pennock JM, Counsell SJ, et al. Abnormal magnetic resonance signal in the internal capsule predicts poor neurodevelopmental outcome in infants with hypoxic-ischemic encephalopathy. Pediatrics 1998;102:323e8. [4] Liauw L, van der Grond J, van den Berg-Huysmans AA, Palm-Meinders IH, van Buchem MA, van Wezel-Meijler G. Hypoxic-ischemic encephalopathy: Diagnostic value of conventional MR imaging pulse sequences in term-born neonates. Radiology 2008;247:204e12. [5] Chau V, Poskitt KJ, Sargent MA, et al. Comparison of computer tomography and magnetic resonance imaging scans on the third day of life in term newborns with neonatal encephalopathy. Pediatrics 2009;123:319e26. [6] Barkovich AJ, Hajnal BL, Vigneron D, et al. Prediction of neuromotor outcome in perinatal asphyxia: Evaluation of MR scoring systems. AJNR 1998;19:143e9. [7] Liauw L, van der Grond J, van den Berg-Huysmans AA, Laan LA, van Buchem MA, van Wezel-Meijler G. Is there a way to predict outcome in (near) term neonates with hypoxic-ischemic encephalopathy based on MR imaging? AJNR 2008;29:1789e94. [8] Tong KA, Ashwal S, Obenaus A, Nickerson JP, Kido D, Haacke EM. Susceptibility-weighted MR imaging: A review of clinical applications in children. AJNR 2008;29:9e17. [9] Haacke EM, Xu Y, Cheng YC, Reichenbach JR. Susceptibility weighted imaging (SWI). Magn Reson Med 2004;52:612e8. [10] Calvert JW, Zhang JH. Pathophysiology of an hypoxic-ischemic insult during the perinatal period. Neurol Res 2005;27:246e60. [11] Miller SP, Ramaswamy V, Michelson D, et al. Patterns of brain injury in term neonatal encephalopathy. J Pediatr 2005;146:453e60. [12] Reichenbach JR, Venkatesan R, Schillinger DJ, Kido DK, Haacke EM. Small vessels in the human brain: MR venography with deoxyhemoglobin as an intrinsic contrast agent. Radiology 1997;204:272e7. [13] Fiser DH. Assessing the outcome of pediatric intensive care. J Pediatr 1992;121:68e74. [14] Angeles DM, Ashwal S, Wycliffe ND, et al. Relationship between opioid therapy, tissue-damaging procedures, and brain metabolites as measured by proton MRS in asphyxiated term neonates. Pediatr Res 2007;61:614e21. [15] Glass HC, Glidden D, Jeremy RJ, Barkovich AJ, Ferriero DM, Miller SP. Clinical neonatal seizures are independently associated with outcome in infants at risk for hypoxic-ischemic brain injury. J Pediatr 2009;155:318e23. [16] Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Temporal evolution of MR perfusion in neonatal hypoxic-ischemic encephalopathy. J Magn Reson Imag 2008;27:1229e34. [17] Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Perfusion-weighted magnetic resonance imaging patterns of hypoxic-ischemic encephalopathy in term neonates. J Magn Reson Imag 2008;28:1019e25. [18] Rutherford M, Ward P, Allsop J, Malamatentiou C, Counsell S. Magnetic resonance imaging in neonatal encephalopathy. Early Hum Dev 2005;81: 13e25. [19] Frewen TC, Kissoon N, Kronick J, et al. Cerebral blood flow, cross-brain oxygen extraction, and fontanelle pressure after hypoxic-ischemic injury in newborn infants. J Pediatr 1991;118:265e71. [20] Toet MC, Lemmers PM, van Schelven LJ, van Bel F. Cerebral oxygenation and electrical activity after birth asphyxia: Their relation to outcome. Pediatrics 2006;117:333e9. [21] Abate MG, Trivedi M, Fryer TD, et al. Early derangements in oxygen and glucose metabolism following head injury: The ischemic penumbra and pathophysiological heterogeneity. Neurocrit Care 2008;9:319e25. [22] Narotam PK, Morrison JF, Nathoo N. Brain tissue oxygen monitoring in traumatic brain injury and major trauma: Outcome analysis of a brain tissue oxygen-directed therapy. J Neurosurg 2009;111:672e82. [23] Longhi L, Pagan F, Valeriani V, et al. Monitoring brain tissue oxygen tension in brain-injured patients reveals hypoxic episodes in normal-appearing and in peri-focal tissue. Intensive Care Med 2007;33:2136e42. [24] Martin NA, Patwardhan RV, Alexander MJ, et al. Characterization of cerebral hemodynamic phases following severe head trauma: Hypoperfusion, hyperemia, and vasospasm. J Neurosurg 1997;87:9e19.
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
Hypoxic-Ischemic Injury: Utility of Susceptibility-Weighted Imaging Gene Kitamura MD a, Daniel Kido MD a, Nathaniel Wycliffe MD a, J. Paul Jacobson MD a, Udochukwu Oyoyo MPH a, Stephen Ashwal MD b, * a b
Department of Radiology, Loma Linda University School of Medicine, Loma Linda, California Division of Child Neurology, Department of Pediatrics, Loma Linda University School of Medicine, Loma Linda, California
article information
abstract
Article history: Received 22 February 2011 Accepted 23 June 2011
Magnetic resonance imaging is increasingly used to assess neonatal hypoxic-ischemic injury, and several scoring systems were developed to predict neurologic outcomes in these patients. We examined the magnetic resonance imaging studies of 33 neonates/infants who manifested acute perinatal hypoxicischemic injuries. Using a seven-point susceptibility-weighted imaging categorical grading scale, each patient received a “prominence of vein” score, which was dichotomized into a “normal” or “abnormal” group. Six-month outcomes were assessed using the Pediatric Cerebral Performance Category Scale. We then determined whether “prominence of vein” scores correlated with neurologic outcomes in patients with hypoxic-ischemic injuries, and compared these results with the Barkovich magnetic resonance imaging scoring system. Patients with “normal” “prominence of vein” scores demonstrated better outcomes (mean Pediatric Cerebral Performance Category Scale value ¼ 2) than patients with “abnormal” “prominence of vein” scores (mean Pediatric Cerebral Performance Category Scale value ¼ 4). The dichotomized “prominence of vein” groups demonstrated correlations with the Barkovich magnetic resonance imaging scores of the proton density-weighted basal ganglia, watershed, and combined basal ganglia/watershed regions. The susceptibility-weighted imaging categorical grading scale may aid in predicting neurologic outcomes after hypoxic-ischemic injuries. Ó 2011 Elsevier Inc. All rights reserved.
Introduction Neonatal hypoxic-ischemic injury is a major cause of morbidity and mortality. Its prognoses correlate with the severity of hypoxicischemic encephalopathy [1]. The evaluation of neonates with hypoxic-ischemic injury has become increasingly reliant on magnetic resonance imaging to assess injury severity and predict outcomes [2-7]. Susceptibility-weighted imaging involves a magnetic resonance imaging sequence that is used to evaluate trauma, tumors, venous diseases, and arteriovenous malformations [8]. This sequence is particularly sensitive at detecting extravascular blood products and intravascular deoxygenated blood [9]. Previous studies indicated an increased level of deoxygenated blood with hypoxic-ischemic injury [10], and susceptibility-weighted imaging will likely detect neonatal cerebral insults with more sensitivity than currently used magnetic resonance imaging sequences. We developed a susceptibility-weighted imaging categorical grading scale derived from the level of deoxyhemoglobin in the deep * Communications should be addressed to: Dr. Ashwal; Department of Pediatrics, Loma Linda University School of Medicine; 11175 Campus Street, Room A1120G; Loma Linda, CA 92350. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pediatrneurol.2011.06.009
cerebral venous system, because the quantity of deoxyhemoglobin reflects the magnitude of oxygen extraction or veno-occlusion. Our study sought to determine whether susceptibility-weighted imaging could predict outcomes in neonates and very young infants with hypoxic-ischemic injuries, and to compare these results with the well-accepted and standardized magnetic resonance imaging scoring system of Barkovich et al. [6]. Patients and Methods Subjects This study was approved by the Institutional Review Board of Loma Linda University Medical Center. A cohort of neonates and infants who had undergone magnetic resonance imaging evaluations at Loma Linda University Children’s Hospital from 2002-2007 was retrospectively reviewed. Inclusion criteria comprised: (1) evidence of neonatal hypoxic-ischemic injury within the first 3 days after birth, defined by at least one of several findings that included neonatal encephalopathy, 5-minute Apgar score of 5, umbilical artery pH 7.1, or umbilical cord base excess >10 [11]; and (2) the acquisition of magnetic resonance imaging and susceptibility-weighted imaging between 0-60 days after birth. Exclusion criteria comprised proven congenital malformations, infection, or metabolic disease. Relevant demographic data from the antenatal, perinatal, and postnatal periods were collected from medical records. Maternal substance use, including cigarettes and drugs, was defined by a positive patient history or urine drug screen. Maternal inflammatory state was defined as a documented maternal infection or presence of
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224 antepartum or peripartum maternal fever. Complicated vaginal delivery was defined as failed vacuum delivery or an arrest of descent. Fetal distress was defined as nonreassuring decelerations, as documented by the treating physician. A placental or cord insult was defined as cord prolapse, vasa previa, nuchal cord, cord rupture, abruptio placentae, or uterine rupture. The resuscitation score was based on the severity of resuscitation at birth (1, no intervention; 2, blow-by oxygen; 3, endotracheal suctioning; 4, bag-mask positive pressure ventilation; 5, endotracheal intubation with positive pressure ventilation; and 6, endotracheal intubation with ventilation and the administration of sodium bicarbonate) [11]. Magnetic resonance imaging and grading Magnetic resonance imaging was performed using a conventional 1.5-T wholebody magnetic resonance imaging system (Magnetom Vision, Siemens Medical Solutions, Iselin, NJ). The susceptibility-weighted imaging sequence consists of a strongly susceptibility-weighted, low-bandwidth (78 Hz/pixel), three-dimensional fast low-angle shot sequence (TR/TE ¼ 57/40 ms, FA ¼ 20 ), with first-order flow compensation in three orthogonal directions. Thirty-two partitions of 2 mm were acquired using a rectangular FOV (5/8 of 256 mm) and a matrix size of 160 512, resulting in 16 slices with an effective slice thickness of 4 mm. The acquisition time was 4.25 minutes. Images underwent additional post-processing, using a phase mask that led to an enhancement of the phase differences between paramagnetic substances and the surrounding tissue [12]. The susceptibility-weighted imaging sequence was used to evaluate the cerebral venous system. Emphasis was placed on examining the prominence of deep medullary veins instead of the entire venous system, because they were more consistently present and easier to evaluate than the subependymal or cortical veins. Each susceptibility-weighted imaging sequence was evaluated with a categorical grading scale, ranging from 1-7 (Table 1 and Fig 1). Two neuroradiologists (D.K. and J.P.J.), blinded to patients’ neurologic status, assigned “prominence of vein” scores to each patient’s susceptibility-weighted imaging sequence. When disagreement in the “prominence of vein” score of a patient occurred between the two neuroradiologists, an average “prominence of vein” score was used for analyses. In addition to receiving a “prominence of vein” score, each patient’s magnetic resonance imaging was evaluated by a neuroradiologist (N.W.) using a previously validated system for the prognostic evaluation of patients with hypoxic-ischemic injuries (i.e., the Barkovich score). The T1-weighted (TR/TE ¼ 9000/110 ms; 5 mm thick), proton density-weighted (TR/TE ¼ 3000/22 ms; 5 mm thick), T2-weighted (TR/TE ¼ 300/120 ms; 5 mm thick), and diffusion-weighted (TR/TE ¼ 4000/11 ms; 5 mm thick) magnetic resonance imaging sequences were evaluated to assess injury to the basal ganglia, watershed, and combined basal ganglia and watershed areas. Signal abnormalities were evaluated in the basal ganglia and combined basal ganglia and watershed regions with a scale ranging from 0-4, and insults to the watershed region were evaluated with a scale ranging from 0-5 [6]. Assessment of outcomes Neurologic outcomes at age 6 months were evaluated with the previously validated Pediatric Cerebral Performance Category Scale: (1) normal, at age-appropriate level; (2) mild disability, i.e., conscious, alert, able to interact at age-appropriate level, with possible mild neurologic deficit; (3) moderate disability, i.e., conscious, with age-appropriate independent activities, and with learning disabilities among schoolage children or enrollment in special education; (4) severe disability, i.e., conscious, and dependent on others for daily support; (5) coma or vegetative state, i.e., any degree of coma without the criteria of brain death; and (6) brain death, i.e., apnea,
221
areflexia, and/or electroencephalographic silence [13]. The Pediatric Cerebral Performance Category Scale was previously implemented to assess the neurologic status of pediatric patients at less than 1 year of age [14]. A neurologist blinded to the “prominence of vein” and Barkovich scores assigned a Pediatric Cerebral Performance Category Scale value to each patient during 6-month follow-up, based on a retrospective chart review. Statistical analysis Before analyses, the “prominence of vein” values were dichotomized into a “normal” group with “prominence of vein” values ranging from 2-4, and an “abnormal” group with “prominence of vein” values of 1 and 5-7. “Prominence of vein” values were dichotomized into the two groups according to the concept that the outlying “prominence of vein” values most likely represented irregular cerebral metabolism. The Pediatric Cerebral Performance Category Scale values were also dichotomized into good (1 and 2) and poor (3-6) outcome groups for analyses. Differences in antenatal, perinatal, and postnatal factors between the “prominence of vein” and Pediatric Cerebral Performance Category Scale groups were analyzed using the Pearson c2 test for nominal values, and the Mann-Whitney U test for ordinal and parametric values. The difference in Pediatric Cerebral Performance Category Scale values between the two “prominence of vein” groups, i.e., “normal” and “abnormal,” was analyzed using the Mann-Whitney U test. Differences in individual Barkovich scores between the dichotomized Pediatric Cerebral Performance Category Scale groups were analyzed using the Mann-Whitney U test. In addition, differences in individual Barkovich scores between the “prominence of vein” groups were analyzed using the Mann-Whitney U test. Interobserver variability for the susceptibility-weighted imaging categorical grading scale was analyzed by calculating a k value. P < 0.05 was considered statistically significant.
Results Although the magnetic resonance imaging and susceptibilityweighted imaging sequences were performed up to 60 days after birth, all patients in this study demonstrated hypoxic-ischemic injury based on the inclusion criteria during the postnatal period, within 3 days after birth. No significant differences were demonstrated in the antenatal, perinatal, and postnatal clinical variables between the “normal” and “abnormal” “prominence of vein” groups (Table 2). The mean age of patients at the time of susceptibilityweighted imaging scan was comparable between the “normal” and “abnormal” “prominence of vein” groups (P ¼ 0.98). Neonates with “normal” “prominence of vein” values (2-4) demonstrated improved neurologic outcomes compared with those scoring “abnormal” “prominence of vein” values (1 and 5-7), based on their Pediatric Cerebral Performance Category Scale values (P ¼ 0.03). The average Pediatric Cerebral Performance Category Scale value for the “normal” “prominence of vein” group was 2 vs the “abnormal” “prominence of vein” group with an average Pediatric Cerebral Performance Category Scale value of 4. Agreement in
Table 1. Grading scale of deep medullary veins demonstrated on susceptibility-weighted imaging sequences Grade
Category
Appearance of Deep Medullary Veins on Susceptibility-Weighted Imaging
1
Absent DMVs
2
Faint DMVs
3
Minimal DMVs
4
Mildly prominent DMVs
5
Moderately prominent DMVs
6
Severely prominent DMVs
7
Extremely prominent DMVs
No visible deep medullary veins, absent or nearly absent signal in the subependymal veins, and a few nonprominent cortical veins. Equivocal low signal in the medullary veins, some low signal in the subependymal veins, and a few nonprominent cortical veins. Several definite, fine, light gray deep medullary veins, limited to the deep white matter, are visible, the subependymal veins are visible but not prominent, and cortical veins are occasionally prominent. Many dark, distinct deep medullary veins are visible, either diffusely or regionally. The deep medullary veins are wider and better demarcated, but do not extend to the most superficial layers of the deep white matter. The subependymal veins are usually prominent, and a few cortical veins may be prominent. The deep medullary veins are dark, numerous, and wider, and extend into the superficial white matter, either diffusely or regionally. The subependymal veins are prominent, and numerous cortical veins may be prominent. Numerous deep medullary veins are very dark and extend through the superficial white matter nearly to the cortex, either diffusely or regionally. Subependymal veins and cortical veins are very prominent. A dark background blush in the white matter may be evident. Numerous, thick, dark deep medullary veins extend to the cortex. They may be irregular. The subependymal and cortical veins are prominent.
Abbreviation: DMVs ¼ Deep medullary veins
222
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224
Figure 1. Susceptibility-weighted axial images illustrate the seven “prominence of vein” categories, with emphasis on the disc-modulating veins system. (A) Category 1, absent (no visible deep medullary are visible but not prominent). (B) Category 2, faint (visible deep medullary veins, nearly absent subependymal veins, and a few nonprominent cortical veins). (C) Category 3, minimal (several definite, fine, light gray deep medullary veins are visible and limited to the deep white matter; the subependymal veins are present but not prominent, and cortical veins are occasionally prominent). (D) Category 4, mildly prominent (many dark, distinct deep medullary veins are visible, either diffusely or regionally; the deep medullary veins are wider and better demarcated, but do not extend to the most superficial layers of the deep white matter; the subependymal veins are usually prominent, and a few cortical veins may be prominent). (E) Category 5, moderately prominent (the deep medullary veins are even darker, more numerous, and wider, extending into the superficial white matter, either diffusely or regionally; the subependymal veins are prominent and more cortical veins may be prominent). (F) Category 6, severely prominent (numerous deep medullary veins are very dark and extend through the superficial white matter nearly to the cortex, either diffusely or regionally; subependymal veins and cortical veins are very prominent; a dark background blush may be evident in the white matter). (G) Category 7, extremely prominent (numerous, thick, dark deep medullary veins extending to the cortex; they may be irregular; the subependymal and cortical veins are prominent). Separating categories 1 and 2 on the basis of deep medullary veins is difficult and usually requires examination of the subependymal veins. The category 6 image contains varying sizes of hypodensity, caused by multiple hemorrhages. Images A-F were taken of patients in this study, aged 12, 5, 51, 8, 2, and 7 days, respectively. Image G was taken of a 3-month old infant who was not part of this study but was included because no category 7 patients were available in the study cohort. This patient had undergone nonaccidental trauma and was being examined for brain death.
“prominence of vein” values between the two neuroradiologists (D.K. and J.P.J.) was acceptable, with a k value of 0.82. Table 3 summarizes the differences in clinical variables between the good vs poor outcome groups. The presence of clinical seizures was the only variable associated with worse neurologic outcomes at 6-month follow-up (P ¼ 0.02). Differences in Barkovich scores between the outcome groups are summarized in Table 4. Patients with good outcomes demonstrated lower proton density-weighted Barkovich scores of the basal ganglia (P ¼ 0.015) and combined basal ganglia and watershed region (P ¼ 0.035), compared with patients manifesting poor outcomes. The comparison of Barkovich scores between “normal” vs “abnormal” “prominence of vein” groups is summarized in Table 5. The “normal” “prominence of vein” group demonstrated lower proton density-weighted Barkovich scores of the basal ganglia (P ¼ 0.004), watershed (P ¼ 0.04), and combined basal ganglia and watershed regions (P ¼ 0.012), compared with the “abnormal” “prominence of vein” group. In contrast, no differences were evident in T1-weighted, T2-weighted, and diffusion-weighted Barkovich scores between the “normal” vs “abnormal” “prominence of vein” groups. Discussion Our findings demonstrated that susceptibility-weighted imaging based “prominence of vein” scores are helpful in predicting intermediate-term neurologic outcomes after neonatal hypoxic-ischemic injuries. Patients with “normal” “prominence of vein” scores demonstrated better outcomes (mean Pediatric Cerebral Performance Category Scale value of 2) compared with patients with “abnormal” “prominence of vein” scores (mean Pediatric Cerebral Performance Category Scale value of 4).
Our analyses demonstrated that seizures were more frequent in patients with poor outcomes, which was expected, because seizures are associated with poor neurologic outcomes after hypoxicischemic brain injuries [15]. Higher Barkovich scores, particularly of the proton density-weighted basal ganglia and combined basal ganglia and watershed regions, were demonstrated in the poor outcome group, which was expected, because higher Barkovich scores, particularly of the proton density-weighted combined basal ganglia and watershed region, are associated with poor neurologic outcomes after hypoxic-ischemic brain injury [6]. Patients with abnormal “prominence of vein” values (values of 1 and 5-7) A “prominence of vein” value of 1, representing decreased levels of deoxyhemoglobin in the deep medullary veins, may be secondary to cerebral swelling compressing the vessels after hypoxic-ischemic injury or because of cerebral hemodynamic changes after hypoxicischemic injury. Previous studies using perfusion-weighted imaging demonstrated marked cerebral hyperperfusion in newborns with hypoxic-ischemic encephalopathy [16,17]. Early scans demonstrated diffuse involvement of the brain, but later scans indicated increased relative cerebral blood flow and relative cerebral blood volume only in the more severely abnormal regions [16]. Early hyperperfusion represents a reperfusion phase, and is likely attributable to impaired cerebral autoregulation secondary to vasoparalysis [18]. Late hyperperfusion is thought to persist because the more severely infarcted areas may take longer to regain cerebral blood flow autoregulation [16]. Previous studies also investigated the level of cerebral fractional oxygen extraction in newborns with hypoxic-ischemic injury. One
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224 Table 2. Comparison of clinical characteristics and laboratory values between normal and abnormal POV groups
Number Male sex Six-month PCPCS S.D. Antenatal factors Gestational diabetes Preeclampsia Maternal substance use Maternal inflammatory state Perinatal factors Fetal distress Complicated vaginal delivery Cesarean section delivery Emergency cesarean section delivery Placental or cord insult Postnatal factors Gestational age (weeks) S.D. Birth weight (g) S.D. Head circumference (cm) S.D. Length (cm) S.D. Resuscitation score (1-6) S.D. One-minute Apgar score S.D. Five-minute Apgar score S.D. Meconium staining Clinical seizures Age at time of magnetic resonance imaging (days) S.D. Hemoglobin S.D. Hematocrit S.D. Mean arterial pressure S.D. Blood glucose S.D.
“Abnormal” POV Group*
“Normal” POV Groupy
7 4 (57%) 42
26 19 (73%) 21
0.42 0.03
0 2 1 3
(0%) (29%) (14%) (43%)
1 5 3 4
0.60 0.59 0.84 0.12
5 1 5 5
(71%) (14%) (71%) (71%)
20 4 20 13
(4%) (19%) (11%) (15%) (77%) (15%) (77%) (50%)
0.76 0.94 0.76 0.31
7 (27%)
0.93
37 2 2861 624 35 3 49 3 61 11 31 2 (29%) 5 (71%) 11 5
37 3 2936 1056 33 4 47 6 51 22 42 7 (27%) 10 (38%) 17 16
0.98 0.82 0.31 0.64 0.33 0.22 0.56 0.93 0.12 0.98
3 7 15 11
14 43 53 77
3 10 7 23
Table 3. Comparison of clinical characteristics and laboratory values between good and poor outcome groups
P Value
2 (29%)
13 37 54 75
223
0.10 0.09 0.76 0.59
Abbreviations: PCPCS ¼ Pediatric Cerebral Performance Category Scale POV ¼ “Prominence of vein” * “Abnormal” POV group represents a POV range of 1 and 5-7. y “Normal” POV group represents a POV range of 2-4.
study calculated cross-brain oxygen extraction by comparing the jugular venous bulb oxygen content with the systemic arterial oxygen content, and demonstrated that cerebral fractional oxygen extraction was much lower in neonates who eventually died or sustained severe brain damage [19]. Another study, using nearinfrared spectroscopy and amplitude-integrated electroencephalography, demonstrated decreased cerebral fractional oxygen extraction in infants who eventually manifested poor neurologic outcomes [20]. The decreased level of cerebral fractional oxygen extraction is thought to result from secondary energy failure [20]. When increased cerebral blood flow and decreased cerebral fractional oxygen extraction are taken together, it may explain why a “prominence of vein” value of 1 represents pathology that leads to poor neurologic outcomes in patients with hypoxic-ischemic injury. “Prominence of vein” values of 5-7, representing increased levels of deoxyhemoglobin in the cerebral circulation, were also considered abnormal. Studies of cerebral hemodynamic changes after brain injury suggest possible underlying mechanisms. After cerebral injury, oxygen metabolism is maintained despite a reduced cerebral blood flow secondary to increased cerebral fractional oxygen extraction [21,22]. This hypoperfusion appears to result from increased microvascular resistance, possibly because of occlusion by microthrombi and circulating blood cells [21,23,24]. Moreover, regional hypoxia was demonstrated in injured brain tissue [22,23]. Ischemia, resulting in cerebral hypoxia and increased cerebral fractional oxygen extraction, likely results in increased levels of deoxyhemoglobin in the cerebral circulation, and presents as a high “prominence of vein” value (5-7) on susceptibilityweighted imaging scans, representing abnormal cerebral metabolism and likely leading to poor neurologic outcomes.
Number Male sex Antenatal factors Gestational diabetes Preeclampsia Maternal substance use Maternal inflammatory state Perinatal factors Fetal distress Complicated vaginal delivery Cesarean section delivery Emergency cesarean section delivery Placental or cord insult Postnatal factors Gestational age (weeks) S.D. Birth weight (g) S.D. Head circumference (cm) S.D. Length (cm) S.D. Resuscitation score (1-6) S.D. One-minute Apgar score S.D. Five-minute Apgar score S.D. Meconium staining Clinical seizures Hemoglobin S.D. Hematocrit S.D. Mean arterial pressure S.D. Blood glucose S.D.
Good Outcomes*
Poor Outcomesy
16 12 (75%)
17 11 (65%)
P Value
0.52
0 2 2 4
(0%) (13%) (13%) (25%)
1 5 2 3
(6%) (29%) (12%) (18%)
0.33 0.24 0.95 0.61
10 3 11 6 3
(63%) (19%) (69%) (38%) (19%)
15 2 14 12 6
(88%) (12%) (83%) (71%) (35%)
0.09 0.58 0.36 0.06 0.29
38 2 3078 773 34 2 49 3 51 22 42 3 (18%) 11 (65%) 14 3 42 11 54 11 76 15
0.07 0.55 0.75 0.27 0.14 0.20 0.64 0.20 0.02 0.55 0.75 0.44 0.46
36 4 2761 1127 32 5 46 6 51 32 42 6 (38%) 4 (25%) 14 2 41 7 52 7 81 19
* Good outcomes represent a Pediatric Cerebral Performance Category Scale range of 1-2. y Poor outcomes represent a Pediatric Cerebral Performance Category Scale range of 3-6.
Comparison of “prominence of vein” values to Barkovich scores The susceptibility-weighted imaging categorical grading scale was compared with the Barkovich magnetic resonance imaging scoring system, which was previously validated to predict intermediate-term neurologic outcomes in neonates with hypoxicischemic injuries [6,11]. The “abnormal” “prominence of vein” group was expected to exhibit higher Barkovich scores because higher Barkovich scores had been correlated with worse neuromotor outcomes [6,11]. One subset of the Barkovich score, the proton density-weighted combined basal ganglia and watershed region score, was observed to best discriminate between patients who developed good vs poor intermediate-term neurologic Table 4. Barkovich scores for good and poor outcome groups Good Outcomes* (value S.D.) Basal ganglia scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Basal ganglia/watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted
Poor Outcomesy (value S.D.)
P Values
1.1 0.7 0.7 0.6
1.1 1.1 1.0 1.0
2.2 2.0 1.7 1.5
0.6 1.2 1.4 1.5
0.082 0.015 0.050 0.204
0.5 0.7 0.7 0.8
1.0 1.3 1.3 1.3
1.1 1.5 1.4 1.4
1.8 2.0 2.0 2.0
0.623 0.355 0.449 0.552
0.9 0.9 0.8 0.9
1.1 1.3 1.3 1.4
1.7 1.8 1.6 1.4
1.1 1.4 1.4 1.5
0.076 0.034 0.050 0.284
* Good outcomes represent a Pediatric Cerebral Performance Category Scale range of 1-2. y Poor outcomes represent a Pediatric Cerebral Performance Category Scale range of 3-6.
224
G. Kitamura et al. / Pediatric Neurology 45 (2011) 220e224
Table 5. Barkovich scores for normal and abnormal POV groups “Normal” POV Group* (value S.D.) Basal ganglia scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted Basal ganglia/watershed scores T1-weighted Proton density-weighted T2-weighted Diffusion-weighted
“Abnormal” POV Groupy (value S.D.)
P Values
1.5 1.0 0.9 0.8
1.0 1.0 1.0 1.0
2.7 3.0 2.3 2.0
0.9 1.0 1.7 1.9
0.216 0.004 0.074 0.167
0.6 0.7 0.7 0.7
1.2 1.3 1.3 1.3
1.7 3.0 2.6 2.6
2.4 2.2 2.3 2.3
0.689 0.038 0.074 0.085
1.2 1.0 1.0 0.9
1.1 1.3 1.2 1.3
2.0 2.7 2.1 2.0
1.4 1.3 1.6 1.8
0.358 0.012 0.109 0.182
Abbreviation: POV ¼ “Prominence of vein” * “Normal” POV group represents a POV range of 2-4. y “Abnormal” POV group represents a POV range of 1 and 5-7.
outcomes [6]. When “prominence of vein” groups were compared, a much higher Barkovich score was demonstrated in the “abnormal” “prominence of vein” group with proton densityweighted images of the basal ganglia, watershed, and combined basal ganglia and watershed regions. This comparison demonstrated that the susceptibility-weighted, imaging based “prominence of vein” grading scale correlated well with the Barkovich magnetic resonance imaging scoring system. Study limitations Several limitations restrict the clinical utility of our study. The sample size, though adequate to achieve statistically significant findings, was small (n ¼ 33). In addition, the distribution of patients was skewed, with only seven patients in the “abnormal” “prominence of vein” group. The two “prominence of vein” groups were shown to be comparable in terms of demographic data analyses. However, a more robust study would have been possible with a similar group size for the “normal” vs “abnormal” “prominence of vein” groups. The lack of long-term follow-up with patients comprised another limiting factor. Our sample of 33 patients all received at least a 6-month follow-up, but many did not receive 12month or 24-month follow-up evaluations. Finally, an inherent weakness of the susceptibility-weighted imaging categorical grading scale involves its use of ordinal rather than parametric values, which restricts the application of certain statistical tools in analyses. Conclusion Susceptibility-weighted imaging may eventually become a useful tool in predicting intermediate-term to long-term neurologic outcomes in patients with hypoxic-ischemic injuries. Patients with “normal” “prominence of vein” values (2-4) are much likelier to experience normal neurologic outcomes than patients with “abnormal” “prominence of vein” values (1 and 5-7). Because susceptibility-weighted imaging can be obtained together with conventional magnetic resonance imaging sequences, the
susceptibility-weighted imaging categorical grading scale may eventually be used in conjunction with other grading systems to aid in predicting intermediate-term to long-term neurologic outcomes in patients with hypoxic-ischemic injuries. References [1] Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress: A clinical and electroencephalographic study. Arch Neurol 1976;33:696e705. [2] Jyoti R, O’Neil R, Hurrion E. Predicting outcome in term neonates with hypoxic-ischaemic encephalopathy using simplified MR criteria. Pediatr Radiol 2006;36:38e42. [3] Rutherford MA, Pennock JM, Counsell SJ, et al. Abnormal magnetic resonance signal in the internal capsule predicts poor neurodevelopmental outcome in infants with hypoxic-ischemic encephalopathy. Pediatrics 1998;102:323e8. [4] Liauw L, van der Grond J, van den Berg-Huysmans AA, Palm-Meinders IH, van Buchem MA, van Wezel-Meijler G. Hypoxic-ischemic encephalopathy: Diagnostic value of conventional MR imaging pulse sequences in term-born neonates. Radiology 2008;247:204e12. [5] Chau V, Poskitt KJ, Sargent MA, et al. Comparison of computer tomography and magnetic resonance imaging scans on the third day of life in term newborns with neonatal encephalopathy. Pediatrics 2009;123:319e26. [6] Barkovich AJ, Hajnal BL, Vigneron D, et al. Prediction of neuromotor outcome in perinatal asphyxia: Evaluation of MR scoring systems. AJNR 1998;19:143e9. [7] Liauw L, van der Grond J, van den Berg-Huysmans AA, Laan LA, van Buchem MA, van Wezel-Meijler G. Is there a way to predict outcome in (near) term neonates with hypoxic-ischemic encephalopathy based on MR imaging? AJNR 2008;29:1789e94. [8] Tong KA, Ashwal S, Obenaus A, Nickerson JP, Kido D, Haacke EM. Susceptibility-weighted MR imaging: A review of clinical applications in children. AJNR 2008;29:9e17. [9] Haacke EM, Xu Y, Cheng YC, Reichenbach JR. Susceptibility weighted imaging (SWI). Magn Reson Med 2004;52:612e8. [10] Calvert JW, Zhang JH. Pathophysiology of an hypoxic-ischemic insult during the perinatal period. Neurol Res 2005;27:246e60. [11] Miller SP, Ramaswamy V, Michelson D, et al. Patterns of brain injury in term neonatal encephalopathy. J Pediatr 2005;146:453e60. [12] Reichenbach JR, Venkatesan R, Schillinger DJ, Kido DK, Haacke EM. Small vessels in the human brain: MR venography with deoxyhemoglobin as an intrinsic contrast agent. Radiology 1997;204:272e7. [13] Fiser DH. Assessing the outcome of pediatric intensive care. J Pediatr 1992;121:68e74. [14] Angeles DM, Ashwal S, Wycliffe ND, et al. Relationship between opioid therapy, tissue-damaging procedures, and brain metabolites as measured by proton MRS in asphyxiated term neonates. Pediatr Res 2007;61:614e21. [15] Glass HC, Glidden D, Jeremy RJ, Barkovich AJ, Ferriero DM, Miller SP. Clinical neonatal seizures are independently associated with outcome in infants at risk for hypoxic-ischemic brain injury. J Pediatr 2009;155:318e23. [16] Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Temporal evolution of MR perfusion in neonatal hypoxic-ischemic encephalopathy. J Magn Reson Imag 2008;27:1229e34. [17] Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Perfusion-weighted magnetic resonance imaging patterns of hypoxic-ischemic encephalopathy in term neonates. J Magn Reson Imag 2008;28:1019e25. [18] Rutherford M, Ward P, Allsop J, Malamatentiou C, Counsell S. Magnetic resonance imaging in neonatal encephalopathy. Early Hum Dev 2005;81: 13e25. [19] Frewen TC, Kissoon N, Kronick J, et al. Cerebral blood flow, cross-brain oxygen extraction, and fontanelle pressure after hypoxic-ischemic injury in newborn infants. J Pediatr 1991;118:265e71. [20] Toet MC, Lemmers PM, van Schelven LJ, van Bel F. Cerebral oxygenation and electrical activity after birth asphyxia: Their relation to outcome. Pediatrics 2006;117:333e9. [21] Abate MG, Trivedi M, Fryer TD, et al. Early derangements in oxygen and glucose metabolism following head injury: The ischemic penumbra and pathophysiological heterogeneity. Neurocrit Care 2008;9:319e25. [22] Narotam PK, Morrison JF, Nathoo N. Brain tissue oxygen monitoring in traumatic brain injury and major trauma: Outcome analysis of a brain tissue oxygen-directed therapy. J Neurosurg 2009;111:672e82. [23] Longhi L, Pagan F, Valeriani V, et al. Monitoring brain tissue oxygen tension in brain-injured patients reveals hypoxic episodes in normal-appearing and in peri-focal tissue. Intensive Care Med 2007;33:2136e42. [24] Martin NA, Patwardhan RV, Alexander MJ, et al. Characterization of cerebral hemodynamic phases following severe head trauma: Hypoperfusion, hyperemia, and vasospasm. J Neurosurg 1997;87:9e19.