Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print November 8, 2002, 10.1148/radiol.2253011797
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2253011797v1
225/3/859    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zarifi, M. K.
Right arrow Articles by Tzika, A. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zarifi, M. K.
Right arrow Articles by Tzika, A. A.
(Radiology 2002;225:859-870.)
© RSNA, 2002

Pediatric Imaging

Prediction of Adverse Outcome with Cerebral Lactate Level and Apparent Diffusion Coefficient in Infants with Perinatal Asphyxia1

Maria K. Zarifi, MD, Loukas G. Astrakas, PhD, Tina Young Poussaint, MD, Adré du Plessis, MD, David Zurakowski, PhD and A. Aria Tzika, PhD

1 From the Departments of Radiology (M.K.Z., L.G.A., T.Y.P., A.A.T.), Neurology (A.d.P.), and Biostatistics (D.Z.), Children’s Hospital, Harvard Medical School, Boston, Mass. From the 2000 RSNA scientific assembly. Received November 7, 2001; revision requested January 21, 2002; final revision received May 1; accepted May 29. Address correspondence to A.A.T., Department of Surgery, Massachusetts General Hospital, Harvard Medical School, 51 Blossom St, Rm 261, Boston, MA 02114 (e-mail: atzika@partners.org).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To compare the predictive value for adverse outcome of quantitative cerebral lactate level and of apparent diffusion coefficient (ADC) in infants with perinatal asphyxia in the early postnatal period.

MATERIALS AND METHODS: Lactate-choline ratios determined with proton magnetic resonance (MR) spectroscopy and ADC determined with diffusion MR imaging in basal ganglia and thalami in 26 full-term neonates (age range, 1–10 days) were compared with severity of acute hypoxic-ischemic encephalopathy and long-term clinical outcome. Differences in metabolites between outcome groups were evaluated with the nonparametric Kruskal-Wallis test and the Dunn test. Logistic regression was performed to examine the predictive value of each metabolite for differentiating normal from abnormal or fatal clinical outcome. The likelihood ratio test was used to assess the statistical significance of each metabolite.

RESULTS: Logistic regression confirmed that lactate-choline ratio could be used to differentiate normal (n = 5) from abnormal (n = 14) or fatal (n = 6) outcome (P < .001). The probability of an adverse outcome exceeded 95% for a lactate-choline ratio of 1.0. Even when analyses were restricted to the early postnatal period, lactate-choline ratio was still a significant predictor of adverse outcome (P = .001). Although ADC images were useful in clinical examination of these infants, quantitative ADCs were not predictive of outcome (P = .82).

CONCLUSION: Higher lactate-choline ratios in basal ganglia and thalami of infants with perinatal asphyxia were predictive of worse clinical outcomes. Absolute ADC in the same brain regions did not indicate a statistically significant relationship with clinical outcome. Cerebral lactate level is useful in identifying infants who would benefit from early therapeutic intervention.

© RSNA, 2002

Index terms: Asphyxia, 14.591 • Brain, diffusion, 14.12144 • Brain, MR, 14.121411 • Cerebral palsy • Infants, central nervous system, 14.591 • Infants, newborn • Magnetic resonance (MR), spectroscopy, 14.12145


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Early identification of neonates with perinatal asphyxia who are at risk for hypoxic-ischemic encephalopathy (HIE) is crucial because of the cascade of biochemical events (ie, energy failure, acidosis, free radical formation, calcium accumulation, lipid peroxidation, and neurotoxic reactions to glutamate and nitric oxide) that eventually may lead to neuronal necrosis and/or apoptosis (1,2). The introduction of oxygen at the onset of reperfusion may stimulate these processes further, so that ongoing injury ensues. In neonatal animals and, probably, in human infants, these processes may evolve rapidly and may require early intervention before irreversible brain injury occurs.

Therefore, timely intervention within this therapeutic window requires early identification of neonates at risk for HIE (2). Early identification is hindered by a lack of accurate quantifiable predictors of long-term neurodevelopmental outcome (3). A number of biochemical markers in the blood serum and cerebrospinal fluid have been proposed for that purpose, and these include neuron-specific enolase, glial fibrillary acidic protein, and brain-specific creatine kinase (4,5). However, the results are often delayed, and cerebrospinal fluid collection may be difficult in critically ill neonates.

The utility of magnetic resonance (MR) imaging methods in the assessment of HIE has been suggested by some researchers (6–18). Proton MR spectroscopy has a potential for use in accurate prediction of adverse outcome in neonates with perinatal asphyxia (19) through noninvasive evaluation of the in vivo metabolic and biochemical status of the brain, including the brain of the newborn infant (15,20). In some investigations, the value of diffusion-weighted MR imaging in identifying acute or hyperacute changes following HIE has been suggested (16,21–23). Furthermore, quantitative measurements of the apparent diffusion coefficient (ADC) of water have the potential of being used to measure the severity of cellular hypoxic-ischemic injury in the brain, and these ADC measurements may be useful for prediction of clinical outcome (22,24).

The purpose of this study was to compare the predictive value for adverse outcome of the quantitative cerebral lactate level and of the ADC in infants with perinatal asphyxia during the early postnatal period.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Clinical Evaluation
Twenty-six full-term infants (14 girls, 12 boys) aged 1–10 days (gestational age range, 38–42 weeks) who met the clinical criteria for the diagnosis of HIE that was due to perinatal asphyxia (defined later) (18) were included in the study from August 1998 to September 1999. We excluded infants with congenital malformations, inherited metabolic disorders, or infection. Institutional review board approval was obtained for this study. The institutional review board did not require informed consent from the parents, because MR spectroscopy was an approved clinical diagnostic procedure at Children’s Hospital, Boston, Mass.

Perinatal asphyxia was diagnosed by one of the authors (A.d.P.) if three criteria were met: evidence of fetal distress (eg, abnormal fetal heart rate and/or metabolic acidosis with cord blood pH < 7.1), immediate neonatal depression (eg, 5-minute Apgar score <= 5), and neonatal HIE. Neonatal HIE was categorized into three grades by the pediatric neurologist (A.d.P.), with the most severe grade within the first 72 hours of life assigned to each patient. Mild HIE was defined as an altered level of consciousness that included somnolence and/or irritability, with periods of spontaneous eye opening, and jitteriness, abnormal muscle tone, and abnormal reflexes. Moderate HIE was defined as stupor with absence of spontaneous but presence of stimulated eye opening, abnormal muscle tone, seizures, and absence of brainstem dysfunction. Severe HIE was defined as coma with no spontaneous or elicited eye opening, seizures, and brainstem dysfunction with abnormal cranial nerve function and/or respiratory abnormalities.

The electroencephalographic (EEG) recordings were graded by the pediatric neurologist (A.d.P.) into four levels of severity largely on the basis of the background activity. Specifically, an EEG was considered mildly abnormal if the background was less mature than expected for gestational age ("dysmature") (eg, excessive background discontinuity) or showed excessive spikes, sharp discharges, or slowing with normal amplitude. A moderately abnormal EEG had low amplitude (5–15 µV). Severe EEG abnormalities included low voltage (<5 µV), burst suppression pattern, or EEG evidence of status epilepticus. Profound EEG abnormalities included those with an isoelectric (ie, no detectable electrocortical activity) EEG tracing.

Initial MR imaging studies were performed as early in postnatal life as possible, from 7 hours to 10 days old. Eight patients underwent at least one follow-up MR imaging study from 6 days to 12 months old.

After discharge, all infants were reexamined clinically for their neurodevelopmental progress. Clinical outcome was classified as normal, abnormal (ie, on the basis of the absence or presence of an abnormality at the follow-up neurodevelopmental examination), or fatal.

MR Imaging and Proton MR Spectroscopy
All MR imaging and proton MR spectroscopic examinations were performed by using a 1.5-T whole-body MR system (GE Medical Systems, Milwaukee, Wis) with a quadrature head coil. The MR imaging protocol included spin-echo and fast spin-echo MR pulse sequences. Spin-echo images were sagittal T1-weighted MR images (repetition time [TR] msec/echo time [TE] msec, 600/20; two signals acquired). Fast spin-echo images were transverse T2-weighted MR images (3,200/85, one signal acquired, echo train length of eight). The field of view was 20 cm, the section thickness was 4 mm with a gap of 1 mm, and the acquisition matrix was 256 x 192. Where necessary, sedation was induced with midazolam hydrochloride (Versed; Hoffman-LaRoche, Nutley, NJ) administered intravenously at a dose of 0.1 mg per kilogram of body weight or with chloral hydrate (Major Pharmaceuticals, Rosemont, Ill) administered orally at a dose of 50 mg/kg.

Use of proton MR spectroscopy added approximately 4 minutes to the MR imaging examination, and the total examination time did not exceed 40 minutes; proton MR spectroscopy was performed by using either a single-voxel technique and/or a multivoxel long-echo (TE, 144 msec) point-resolved spatially localized spectroscopic method (25). Ideally, we planned to perform both single- and multivoxel MR spectroscopy; however, typically the protocol was reduced to either single- or multivoxel MR spectroscopy, depending on the time requirements. In any case, the voxels were placed by the technologist who was guided by the radiologist.

Other typical parameters for single-voxel MR spectroscopy were as follows: TR, 1,500 msec; 192 signals acquired; 2,500 ± 1,250 SD; and 1,048 complex data points with a voxel size of approximately 2 cm3. For multivoxel MR spectroscopy, other parameters were as follows: TR, 1,000 msec; 16 x 16 x 1 phase-encoding steps; and 160-mm field of view with a nominal voxel size of 1.0–1.2 cm3. Long-echo point-resolved spatially localized spectroscopy with its high signal-to-noise ratio includes no contributions from resonances caused by fat; thus, this technique helps to reveal accumulated lactate. Metabolites of biological importance, such as N-acetylaspartate (NAA), choline-containing compounds, total creatine, and lactate, may be detected by using this technique (26). For metabolite quantification, we used peak area ratios.

For several reasons, we selected the lactate-choline ratio, although others have described the predictive value of the ratio of lactate to NAA (10). Specifically, the temporal pattern of changes in lactate and NAA likely proceed along different, varying, and poorly understood time courses, features that may generate potentially misleading ratios of lactate to NAA at different time points after injury. Although early NAA changes may be modest, caused by means other than neuronal loss (27), the more delayed changes in choline concentration expected from the later postnecrotic myelinolysis are likely to provide a more stable denominator. Others have used total creatine as a proton MR spectroscopic reference metabolite because of its relative stability after hypoxia-ischemia and better reflection of deranged energy metabolism (17,19). However, the total creatine peak is modulated by varying TE and may therefore not reflect lactate accumulation as accurately as does the lactate-choline ratio. Our lactate-choline ratios were in agreement with published values of others (15).

Diffusion-weighted MR imaging was performed by using a line-scan method that resulted in transverse images (field of view, 20 x 15 cm; effective section thickness, 7 mm; gap, 0 mm; acquisition matrix, 128 x 128, interpolated to 256 x 192) obtained with b values of 5 sec/mm2 and of 750 sec/mm2 as the maximum b values applied along the three orthogonal directions (16). The minimum imaging time was 5 minutes 49 seconds. In contrast to methods used in previous studies (21,22,28), we evaluated the quantitative ADC value in a consistent region of interest (ROI) rather than delineate a topographic "map" of abnormal diffusion. The ROI, that is, the area that included the basal ganglia and the thalami, was chosen because of the known vulnerability of these regions to hypoxic-ischemic injury (29). The second author (L.G.A.) selected the ROIs for the ADC measurements. In this task, he was supervised by the first (M.K.Z.), third (T.Y.P.), and last (A.A.T.) authors. The ROIs were square (0.7–1.1 cm3) and matched as closely as possible the MR spectroscopic voxels of interest and thus served the comparative aspects of our analysis. ADC values were compared with our own normative data and with the data of others (23,30).

Biostatistical Analyses
The Pearson product moment correlation coefficient (r) was used to assess the strength of association between continuous variables, which included the lactate-choline ratios and the ADC values. The Spearman rank correlation ({rho}) was used to correlate the ADC and the proton MR spectroscopic measurements, as well as the severity of HIE and EEG results. Differences in lactate-choline ratio, ADC, and neurologic outcome were illustrated by using box plots (31). Differences in metabolites between outcome groups were evaluated by using the nonparametric Kruskal-Wallis test and the Dunn test. Analysis of variance (ANOVA) was also performed by using parametric and nonparametric methods to ensure that differences or lack of differences between groups was consistent. Logistic regression analysis was performed to examine the predictive value of each metabolite for differentiating normal from abnormal or fatal clinical outcome. The likelihood ratio test was used to assess the statistical significance of each metabolite, and probabilities of an adverse outcome were derived according to the maximum likelihood estimation (32). The associations between clinical outcome and grade of HIE and EEG abnormality were ascertained by using the Pearson {chi}2 test for trend. Statistical analysis was conducted by using software packages (SAS, version 6.12, SAS Institute, Cary, NC; SPSS, version 11.0, SPSS, Chicago, Ill). A two-tailed P value of less than .05 was considered to indicate a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Clinical Evaluation
The maximal severity of clinical HIE was mild in one infant, moderate in 10 infants, and severe in 12 infants. Insufficient early clinical data precluded reliable retrospective grading in three infants (patients 19, 25, 26; Table 1). We performed EEG in all but one infant (patient 16; Table 1). In the neonatal period, the EEG results were categorized as mild in four infants, moderate in three infants, severe in 14 infants, and profound in four infants.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Staging of Clinical HIE in 26 Infants with Perinatal Asphyxia

 
Neurodevelopmental outcome was evaluated in all but one infant (patient 3), who was lost to follow-up. Age at the time of follow-up neurodevelopmental assessment and outcome are summarized in Table 2. We assigned the infants to one of three groups according to their neurodevelopmental outcome. Of the 25 infants with available information about neurodevelopmental outcome, five infants (patients 5, 6, 14, 16, 18) had a normal outcome (group 1), 14 infants (patients 1, 4, 7–13, 15, 19–21, 25) had an abnormal outcome (group 2), and six infants (patients 2, 17, 22, 23, 24, 26) had a fatal outcome (group 3).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Neonatal MR Imaging and MR Spectroscopic Findings and Neurodevelopmental Outcome

 
The distribution of outcomes suggests that the neurologic injury in our sample of infants was particularly severe. Among the six infants who had a fatal outcome, four (patients 2, 22, 23, 26) died at 2 days old, one (patient 17) died at 5 days old,and one (patient 24) died at 24 days old. Apart from patients 15 and 16, patients in group 1 were all older than 12 months when they were reexamined. All patients in group 2 were 12 months or older when they were reexamined, with the exception of patient 15, whose last examination was at 41/2 months old, and patients 13, 19, and 20, whose last examinations were at 7, 11, and 9 months old, respectively. A {chi}2 test of trend indicated a significant association between severity of HIE and outcome (P < .001).

The infant (patient 16) with mild HIE had a normal outcome. Of the 10 infants with moderate HIE, four (patients 5, 6, 14, 18) had normal outcomes and six (patients 1, 4, 8, 12, 13, 21) had abnormal outcomes. Of the 12 infants with severe HIE, five (patients 2, 17, 22–24) had fatal outcomes and six (patients 7, 9–11, 15, 20) had abnormal outcomes.

Of the three infants with a moderate EEG abnormality, the outcome was normal in one (patient 5) and abnormal in two (patients 10, 21). Of the 14 infants with severely abnormal EEG results, four died (patients 22–24, 26), nine had an abnormal outcome (patients 1, 4, 7–9, 12, 13, 19, 20), and one (patient 6) had a normal outcome at follow-up. Of the four infants with profound EEG abnormalities, two had fatal outcomes (patients 2, 17), and one had an abnormal outcome (patient 11); in one patient (patient 3), the neurodevelopmental outcome was not known. A {chi}2 test of trend indicated that a significant association was observed between severity of EEG results and outcome (P = .008).

Findings at MR Imaging and Proton MR Spectroscopy
In 20 of 26 patients in this study, T2 abnormalities were observed at initial or follow-up conventional MR imaging (Table 2). Qualitative evaluation of conventional T2-weighted and diffusion-weighted MR imaging findings at initial and follow-up examinations are presented in Table 2 together with the long-term neurodevelopmental outcome. Table 2 also shows the lactate-choline ratios and ADC measurements in the basal ganglia and thalami in 26 patients. In some patients, more than one measurement was performed. Correlations were determined on the basis of the total measurements across all patients. Figures 1–3 show MR images and proton MR spectra in two infants with fatal (Figs 1, 2) and abnormal (Fig 3) neurodevelopmental outcomes. Figure 2 shows multivoxel (at day 1 of life) and single-voxel (at day 2 of life) proton MR spectroscopic images, respectively. Multivoxel proton MR spectroscopic measurements were determined; however, for our analyses, we included only measurements from the basal ganglia and thalami.



View larger version (57K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Patient 2. Transverse T2-weighted MR image, ADC map, and proton MR spectrum obtained in a 1-day-old full-term male infant. This infant had severe neonatal HIE, persistent seizures, perinatal depression (Apgar scores 2 and 1 at 1 and 5 minutes, respectively), and metabolic acidosis that was due to placental abruption. The infant died at 1 day old. T2-weighted fast spin-echo MR image (3,000/140) shows subtle thalamic hyperintensity (arrow). MR spectrum from an ROI ({square}) within the deep gray matter shows a prominent lactate (Lac) peak, with a lactate-choline (Cho) ratio of 1.51. ADC map (calculated by using a line-scan diffusion sequence with 2,850/106 and b = 1,000 sec/mm2) at the level of the basal ganglia shows a reduced ADC value of 0.5 µm2/msec. In this infant, both clinical data and MR imaging findings combined with proton MR spectroscopic variables correlated well with the fatal outcome. tCr = total creatine.

 


View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Patient 26. Transverse T2-weighted fast spin-echo (3,000/140) and diffusion-weighted MR images and proton MR spectra at 1 and 2 days old in a full-term male infant. This infant had perinatal depression (Apgar scores 8 and 9 at 1 and 5 minutes, respectively) that was due to placental abruption and developed persistent neonatal seizures. T2-weighted MR images show minimal hyperintensity (arrow) in the thalami bilaterally. At multivoxel proton MR spectroscopy (MRS) at 1 day old, lactate (Lac) accumulation in the basal ganglia was revealed; ROI ({square}) includes the basal ganglia. At 2 days old, single-voxel MR spectrum from within deep gray matter showed high levels of lactate (lactate-choline [Cho] ratio = 1.56), and the patient died on that day. Favorable clinical variables could not be used to predict the fatal outcome in this infant, but MR imaging findings combined with proton MR spectroscopic variables were used to do so. tCr = total creatine.

 


View larger version (52K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Patient 10. Transverse T2-weighted fast spin-echo (3,000/140) and diffusion-weighted MR images and proton MR spectrum in a 3-day-old full-term male infant. This infant had severe HIE, perinatal depression (Apgar scores 3 and 7 at 1 and 5 minutes, respectively) after placental abruption, persistent seizures, and metabolic acidosis. T2-weighted images show subtle hyperintensity (arrow) in the thalami. This infant had an abnormal neurodevelopmental outcome at 13 months. At proton MR spectroscopy (ROI is indicated [{square}]), an increased lactate (Lac) level in the thalamus was revealed. ADC map (calculated by using a line-scan diffusion sequence with 2,850/106 and b = 1,000 sec/mm2) showed restricted diffusion in the thalamus (ADC = 0.45 µm2/msec). Cho = choline, tCr = total creatine.

 
MR Imaging Findings and Clinical Evaluation
In this study, there was a positive correlation between time at initial MR imaging and Apgar score (Spearman {rho} = 0.42, P = .05), and this correlation suggested that patients who underwent imaging at an earlier age had lower scores. In addition, we tested compared differences in time at initial MR imaging among the three clinical outcomes, and the Kruskal-Wallis test indicated that the patients with a fatal outcome underwent imaging at an earlier age (P = .04). The median age at initial MR imaging was 1 day old for the patients with a fatal outcome, 3 days old for those with an abnormal outcome, and 4 days old for those who had a normal clinical outcome.

The lactate-choline ratio correlated significantly with the Apgar score at 5 minutes (Pearson r = -0.50, P = .02) but not at 1 minute (Pearson r = -0.22, P = .33). The lactate-choline ratio correlated positively with EEG results (Spearman {rho} = 0.45, P = .02) and grading of HIE (Spearman {rho} = 0.72, P < .001). However, the lactate-choline ratio did not correlate with the ADC (Pearson r = -0.10; P = .59).

Results of multiple comparisons testing, performed by using the Kruskal-Wallis test and the Dunn test, indicated that the lactate-choline ratio could be used to differentiate neonates with a normal outcome from those with an abnormal or a fatal outcome. Figure 4 presents the median values and interquartile ranges of lactate-choline ratios for the three outcome groups. No significant differences for ADC (P = .82) were detected among the three groups (Fig 4). The logistic regression model confirmed a significant positive relationship between the lactate-choline ratio and the likelihood of an adverse outcome (likelihood ratio test = 13.5, P < .001; Fig 5). For instance, a lactate-choline ratio of 0 was associated with a 25% probability of an adverse outcome; a ratio of 0.25, with a 75% probability of an adverse outcome; a ratio of 0.75, with a 90% probability of an adverse outcome; and a ratio of 1.00, with a 95% probability of an adverse outcome (Fig 5). Logistic regression demonstrated no significant relationship between ADC and outcome (P = .91).



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Box plots show relationship between (A) lactate-choline ratio (Lac/Cho) obtained by using proton MR spectroscopy and (B) ADC obtained by using diffusion-weighted MR imaging in the first 10 days of life and neurologic outcome. Outcome was normal in five infants, abnormal in 14, and fatal in six. In a box plot, the box is divided at the median (50th percentile). The box length is the interquartile range, and the top and bottom of the box represent the upper and lower quartiles. The error bars extend to the largest and smallest observed data points within 1.5 interquartile ranges from the top and bottom of the box, respectively. A shows the significant difference (P < .001) in the lactate-choline ratio of the three groups (ie, fatal, abnormal, and normal). The n represents the number of measurements. The lactate-choline ratio was higher in the fatal group, compared with that in both normal and abnormal groups (P < .001). The lactate-choline ratio was higher in the abnormal group, compared with that in the normal group (P = .004). B shows similar ADCs in the three groups (P = .82).

 


View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Graph shows the significant inverse relationship between lactate-choline ratio (Lac/Cho) and the probability of an adverse outcome (likelihood ratio test = 13.5, 1 df, P < .001). Note that at a lactate-choline ratio of 1.00, the probability of an adverse outcome exceeds 95%.

 
Lactate-choline levels were compared between groups by using ANOVA and post hoc Bonferroni tests, and results were similar to those achieved by using the nonparametric approach (Table 3). ANOVA revealed highly significant differences (P < .001). Patients with a fatal outcome had significantly higher lactate-choline ratios compared with the ratios of those who had normal and abnormal outcomes (both P < .001). Patients who had an abnormal outcome had significantly higher mean lactate-choline values compared with the values in those who had a normal outcome (P = .004). The F test from the ANOVA revealed no significant differences in ADC among the three groups (P = .87) (Table 3).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Lactate-Choline Ratios and ADC Measurements from Basal Ganglia and Thalami according to Clinical Outcome

 
When we restricted our analyses to the MR imaging measurements determined during the early postnatal period (mean, within 50 hours of birth) in 24 infants, the differences in the lactate-choline ratios among the three outcome groups remained statistically significant. Specifically, the nonparametric comparisons determined by using the Kruskal-Wallis test and the Dunn test revealed higher lactate-choline median ratios for patients with a fatal outcome compared with those for patients with a normal or abnormal outcome (P < .01), as well as a higher lactate-choline ratio for patients with an abnormal compared with a normal outcome (P = .02). A comparison of mean values by using ANOVA produced significant and comparable results. In addition, logistic regression analysis still indicated that the lactate-choline ratio was a highly significant predictor that could be used to differentiate normal from adverse outcomes (likelihood ratio test = 11.21, P = .001).

With respect to the ADC, the Kruskal-Wallis test and the parametric ANOVA revealed no significant differences among the three groups (P = .90 and P = .85, respectively). Logistic regression analysis indicated that the ADC was not a significant predictor that could be used to differentiate normal from adverse outcomes (P = .82).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study, we observed a strong association between the lactate-choline ratio in the basal ganglia and the thalami and both the severity of acute neonatal HIE and long-term neurodevelopmental outcome. These relationships remained statistically significant even when analyses were confined to MR imaging measurements within the first 5 days of life, a period of critical importance for clinical decision making. Furthermore, the majority of the infants had reduced quantitative ADC values in the basal ganglia and thalami. However, these absolute ADC values were not predictors of neurologic outcome. These findings suggest that diffusion-weighted MR imaging measurements are sensitive to hypoxic-ischemic injuries but lack specificity for use in predicting outcome in the population of asphyxiated newborn infants in this study.

To discuss our findings in the context of the cerebral cellular events and their temporal evolution, a brief review of the relevant pathophysiology of HIE is warranted. Cellular hypoxia triggers a mitochondrial response (33) that involves anaerobic glycolysis and lactate accumulation (34). Cellular energy failure inactivates critical membrane ionic pumps (eg, sodium, potassium, and adenosine triphosphatase), which leads to osmoregulatory failure and causes cytotoxic edema (35). Also, disruption of outer mitochondrial membrane permeability, which is regulated with physiologic conditions, has important implications for cell bioenergetics and cell survival (36).

With the onset of oxygenated reperfusion, recovery of cerebral energy status ensues while the lactate level begins to decrease. However, even with normal postreperfusion hemodynamics and oxygenation, a delayed form of energy failure with a secondary increase in lactate level and cytotoxic edema (37) may develop, with onset between 8 and 24 hours of reperfusion (38–43). The debate regarding the precise mechanisms of delayed energy failure is discussed elsewhere (2). After irreversible cellular injury, brain cells die by necrotic lysis or apoptosis. Axonal degeneration is more delayed and followed by myelinolysis, which releases substances such as choline. After 3 days, activated phagocytic cells infiltrate the injured area and remove cellular debris. These cells, which may persist in the areas for months, generate high lactate concentrations (19,39).

Finally, during the period of active gliosis, increased production by astrocytes could increase cerebral lactate levels (44,45). The tissue water characteristics following reperfusion evolve over time and are determined by factors such as cell lysis, release of osmotically active debris, postischemic vasogenic edema, and compromised blood-brain barrier permeability (46), all of which result in an increase in extracellular water. MR imaging methods that provide insights into the pathophysiologic characteristics of HIE include diffusion MR imaging, which depicts cytotoxic edema; T2-weighted MR imaging, which maps vasogenic edema; and MR spectroscopy, which is sensitive to tissue bioenergetics (47).

In most patients in this study, initial or follow-up MR imaging demonstrated abnormalities, and this finding is in agreement with data in previous studies (48,49) but does not support findings in other reports on the use of MR imaging in neonates (21) and adults (50). According to the study by Ning et al (48), early T2 abnormalities in the infants in our study may be attributed to vasogenic and/or cytotoxic edema or to an alteration in brain protein. Although the diagnosis of neonatal HIE presents a challenge to the neuroradiologist because of the undermyelinated brain (51), absence of T2 abnormalities in four of the infants in this study may imply asynchronous performance of MR imaging and development of vasogenic edema or other alterations.

In studies with animal models of HIE injury (40,41,45,52–55), the close temporal relationship between changes in cerebral lactate level and ADC during both the primary and secondary periods of energy failure were described. In a number of studies in animals and humans, spectroscopy, diffusion imaging, and other MR imaging techniques have been used to examine the temporal evolution of cerebral lactate levels and water diffusion after focal and global hypoxic-ischemic injuries (56–59). For the sake of clarity, we will discuss separately the postischemic changes in the cerebral lactate level and findings of diffusion-weighted MR imaging in our study and in those of others.

Cerebral Lactate Levels after Hypoxia-Ischemia
Our findings support results mentioned in previous studies (13,15,19,39,44,45,60,61). Increased cerebral lactate levels have been found at proton MR spectroscopy in infants who were a few hours to several months old (19,39,44,45,60,62). As outlined elsewhere, there are at least three mechanisms, separated in time, that may produce an increase in cerebral lactate levels after a global hypoxic-ischemic injury; some of these mechanisms include anaerobic glycolysis during the injury, secondary energy failure (45) associated with delayed mitochondrial dysfunction, and lactate generated by infiltrating macrophages during the repair phase (63). In the infants in our study, the intrauterine and/or intrapartum injury and resuscitation at birth help to exclude a primary form of energy failure as a mechanism for cerebral lactate level increase.

Any subsequent episodes of hypotension in the infants in our study were treated aggressively and promptly, and because of this treatment, ongoing or recurrent HIE was unlikely. In our study, since initial lactate-choline ratios were determined within the first 72 hours of life in most cases, a substantial contribution from lactate-generating macrophages was unlikely (63). Although the time course of the cerebral lactate level increase is attributable to delayed mitochondrial failure and has been defined clearly as a biphasic response in experimental models (45), a systematic follow-up study in infants is lacking. In this study, it is likely that the infants with increased cerebral lactate levels were in this phase of secondary energy failure at initial proton MR spectroscopy.

Furthermore, our data highlight another important point, namely, that it is not merely the presence but rather the level of cerebral lactate, as assessed by using noninvasive in vivo proton MR spectroscopy, that is the critical determinant of outcome. This is clearly depicted in the graph in Figure 5, which suggests that a threshold level of lactate-choline ratio, determined early in postnatal life to identify neonates at risk for irreversible injury, can be achieved. As shown in Figure 5, a lactate-choline ratio of approximately 1.00 indicates a greater than 95% probability of adverse neurodevelopmental outcome.

Even for lactate-choline ratios less than 1.00, the probability for adverse outcome is still high. Therefore, our study findings suggest, and it is well accepted, that absence of a cerebral lactate level at proton MR spectroscopy generally can be used to predict a normal outcome (9). Nevertheless, some infants with normal developmental outcome (three infants in our study) had low levels of cerebral lactate, a finding that is likely secondary to the milder perinatal asphyxia in these patients. Similar findings have been reported in previous studies (39,64).

ADC at Diffusion-weighted MR Imaging
In our study, the majority of infants had cerebral areas of restricted diffusion at the time of initial diffusion-weighted MR imaging, findings that are consistent with those in previous reports of focal and diffuse hypoxic-ischemic injury in infants and adults (21,50). Findings of previous studies in newborn infants (21,22) have suggested a relationship between diffusion changes on diffusion-weighted MR images and adverse outcome. Other researchers have indicated that regions of restricted diffusion can be used to predict the topography of brain lesions on later conventional MR images (21,65) and correlate well with short-term clinical outcome (22,65).

Parsons et al (28) combined diffusion-weighted MR imaging and proton MR spectroscopy in adults with stroke and achieved results similar to ours, namely, that the lactate-choline ratio could be used to predict outcome, but findings at diffusion-weighted MR imaging could not be used to do so. However, a fundamental difference exists between the application of diffusion-weighted MR imaging in our study and that used in previous studies (21,22,28). In previous studies, diffusion-weighted MR imaging was used to detect the presence or absence of abnormal diffusion or to measure its extent.

Our approach was based on data from animal models (40,41) that indicated a close temporal association between changes in the cerebral energy status and changes in the cerebral ADC. As such, we used the absolute ADC as a surrogate indicator of cellular energy status in a specific cerebral region (ie, the basal ganglia and thalami) known to have particularly high energy demands and vulnerability to energy failure in the full-term newborn. We found that although ADCs in the basal ganglia and thalami were reduced in the majority of the infants in this study, these values were not predictive of neurodevelopmental outcome. The reasons for this finding are not clear but may relate to a threshold effect of energy failure on osmoregulation, cytotoxic edema, and water diffusion.

Currently, there is no known threshold ADC below which irreversible cell damage occurs. In fact, the basic question of whether areas of restricted diffusion on diffusion-weighted MR images remain potentially viable or are irreversibly committed to infarction remains unanswered (66,67). Nevertheless, false-negative results in human neonates in the early postnatal period may be explained by experimental MR imaging studies in animals. Findings in these studies suggest that after the acute hypoxic-ischemic injury, which is accompanied by primary cytotoxic edema, cells recover transiently because of reversible alterations of physiologic water compartmentation, and both T2- and diffusion-weighted MR images may be normal (48,53–55,68). Another explanation is that apoptosis that is caused early in HIE is not detected by using diffusion-weighted or perfusion MR imaging (69).

On the other hand, our true-positive results may be explained by a secondary cytotoxic swelling, which follows the primary cytotoxic edema and is characterized by a combination of swollen glial cells and lysed neurons that are already necrotic. This swelling allows both cytotoxic and vasogenic edema to coexist and results in a low ADC and an increased T2 (53,55). Therefore, in agreement with findings in our study, it appears that single measurements of the ADC have limited prognostic value following perinatal asphyxia. Our results confirm findings in preliminary reports (70,71) as well as observations of others that suggest that diffusion-weighted MR imaging may help to identify (23), but may cause underestimation (62,72) of, the extent of injury that results from perinatal asphyxia.

Current Limitations and Future Directions
Our study, in general, has limitations inherent to studies of perinatal asphyxia in human newborns. Nevertheless, we were focused on identifying early (ie, in less than 72 hours of life) cerebral MR imaging predictors of neurodevelopmental outcome rather than on tracking changes in these measurements over time. Since we lacked an a priori protocol for follow-up MR imaging studies, we are unable to comment meaningfully on the evolution of MR changes over time in these infants. In addition, our study is limited by the absence of age-matched control subjects, although such control subjects were not required by our study design. However, we compared our values with those in control subjects in the literature (23,30,73). Unlike data in animal studies but common to data in other clinical studies of perinatal asphyxia, the exact timing, duration, and intensity of injury were not known in the patients in this study. Also, since lactate may not be visualized with the currently used techniques for detection, the importance of using lactate editing techniques (74), which help to avoid contributions from protons other than those from lactate, must be emphasized.

Since the lactate level and ADC follow a biphasic change after hypoxic-ischemic injury, although, in longitudinal experimental animal studies, the lactate level did not normalize (45) in a manner similar to that of the ADC (53,54), the appropriate timing of the MR imaging examination should be decided (62,75). In addition, a threshold value of the lactate level, as suggested by the results of the present study, that could reliably be used to identify irreversible brain injury must be determined. Meanwhile, quantitative diffusion anisotropy, provided by diffusion tensor MR imaging, which promises to be more sensitive to tissue microstructure (30), may be a pertinent quantitative predictor, in addition to the ADC, in newborns with perinatal asphyxia. Finally, imaging with radioactively labeled annexin V and proton lipid MR spectroscopy may have a role in the detection of apoptosis (69,76) that develops after perinatal asphyxia and, thus, may be useful in identification of neonates who might benefit from neuroprotective strategies.

In conclusion, in this study, findings of advanced combined in vivo MR imaging techniques were used to predict adverse outcome in infants with perinatal asphyxia. Our data suggest the added value of using the cerebral lactate level. Determined by using proton MR spectroscopic findings as noninvasive predictors of clinical outcome, the cerebral lactate level can be used in the identification of infants who would benefit from therapeutic intervention. Therapeutic intervention would include use of neuroprotective strategies during the early postnatal period. Absolute ADC measurements were not predictive of adverse outcome. Both conventional and diffusion-weighted MR imaging demonstrated abnormal findings as early as 1 day old.


   ACKNOWLEDGMENTS
 
The authors thank Joseph J. Volpe, MD, for his invaluable review of the manuscript and Pek L. Khong, MBBS, for assistance with MR imaging studies.


   FOOTNOTES
 
Abbreviations: ADC = apparent diffusion coefficient, ANOVA = analysis of variance, EEG = electroencephalogram, HIE = hypoxic-ischemic encephalopathy, NAA = N-acetylaspartate, ROI = region of interest, TE = echo time, TR = repetition time

Author contributions: Guarantors of integrity of entire study, all authors; study concepts, A.A.T., M.K.Z.; study design, all authors; literature research, M.K.Z., A.A.T., L.G.A.; clinical studies, T.Y.P., A.A.T.; data acquisition, M.K.Z., T.Y.P., A.A.T.; data analysis/interpretation, A.A.T., L.G.A., D.Z.; statistical analysis, D.Z.; manuscript preparation, M.K.Z., A.A.T.; manuscript definition of intellectual content, all authors; manuscript editing, A.A.T., A.d.P.; manuscript revision/review and final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Johnston MV. Cellular alterations associated with perinatal asphyxia. Clin Invest Med 1993; 16:122-132.[Medline]
  2. du Plessis AJ, Johnston MV. Hypoxic-ischemic brain injury in the newborn: cellular mechanisms and potential strategies for neuroprotection. Clin Perinatol 1997; 24:627-654.[Medline]
  3. Roth SC, Edwards AD, Cady EB, et al. Relation between cerebral oxidative metabolism following birth asphyxia and neurodevelopmental outcome and head growth at one year. Dev Med Child Neurol 1992; 34:285-295.[Medline]
  4. Volpe JJ. Hypoxic-ischemic encephalopathy: biochemical and physiological aspects In: Neurology of the newborn. 4th ed. Philadelphia, Pa: Saunders, 2001; 217-276.
  5. Huang CC, Wang ST, Chang YC, Lin KP, Wu PL. Measurement of the urinary lactate: creatinine ratio for the early identification of newborn infants at risk for hypoxic-ischemic encephalopathy. N Engl J Med 1999; 341:328-335.[Abstract/Free Full Text]
  6. McArdle CB, Richardson CJ, Hayden CK, Nicholas DA, Amparo EG. Abnormalities of the neonatal brain: MR imaging. Part II. Hypoxic-ischemic brain injury. Radiology 1987; 163:395-403.
  7. Rutherford MA, Pennock JM, Dubowitz LM. Cranial ultrasound and magnetic resonance imaging in hypoxic-ischaemic encephalopathy: a comparison with outcome. Dev Med Child Neurol 1994; 36:813-825.[Medline]
  8. Menkes JH, Curran J. Clinical and MR correlates in children with extrapyramidal cerebral palsy. AJNR Am J Neuroradiol 1994; 15:451-457.[Abstract]
  9. Groenendaal F, Veenhoven RH, van der Grond J, Jansen GH, Witkamp TD, de Vries LS. Cerebral lactate and N-acetyl-aspartate/choline ratios in asphyxiated full-term neonates demonstrated in vivo using proton magnetic resonance spectroscopy. Pediatr Res 1994; 35:148-151.[Medline]
  10. Penrice J, Cady EB, Lorek A, et al. Proton magnetic resonance spectroscopy of the brain in normal preterm and term infants, and early changes after perinatal hypoxia-ischemia. Pediatr Res 1996; 40:6-14.[Medline]
  11. Martin E, Buchli R, Ritter S, et al. Diagnostic and prognostic value of cerebral 31P magnetic resonance spectroscopy in neonates with perinatal asphyxia. Pediatr Res 1996; 40:749-758.[Medline]
  12. Cady EB. Metabolite concentrations and relaxation in perinatal cerebral hypoxic-ischemic injury. Neurochem Res 1996; 21:1043-1052.[Medline]
  13. Ashwal S, Holshouser BA, Tomasi LG, et al. 1H-magnetic resonance spectroscopy-determined cerebral lactate and poor neurological outcomes in children with central nervous system disease. Ann Neurol 1997; 41:470-481.[CrossRef][Medline]
  14. Barkovich AJ, Hajnal BL, Vigneron D, et al. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. AJNR Am J Neuroradiol 1998; 19:143-149.[Abstract]
  15. Barkovich AJ, Baranski K, Vigneron D, et al. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. AJNR Am J Neuroradiol 1999; 20:1399-1405.[Abstract/Free Full Text]
  16. Robertson RL, Ben-Sira L, Barnes PD, et al. MR line-scan diffusion-weighted imaging of term neonates with perinatal brain ischemia. AJNR Am J Neuroradiol 1999; 20:1658-1670.[Abstract/Free Full Text]
  17. Hanrahan JD, Cox IJ, Azzopardi D, et al. Relation between proton magnetic resonance spectroscopy within 18 hours of birth asphyxia and neurodevelopment at 1 year of age. Dev Med Child Neurol 1999; 41:76-82.[CrossRef][Medline]
  18. Volpe JJ. Hypoxic-ischemic encephalopathy: clinical aspects In: Neurology of the newborn. Philadelphia, Pa: Saunders, 2001; 331-394.
  19. Hanrahan JD, Cox IJ, Edwards AD, et al. Persistent increases in cerebral lactate concentration after birth asphyxia. Pediatr Res 1998; 44:304-311.[Medline]
  20. Volpe JJ. Specialized studies in the neurological evaluation In: Neurology of the newborn. Philadelphia, Pa: Saunders, 2001; 134-177.
  21. Cowan FM, Pennock JM, Hanrahan JD, Manji KP, Edwards AD. Early detection of cerebral infarction and hypoxic ischemic encephalopathy in neonates using diffusion-weighted magnetic resonance imaging. Neuropediatrics 1994; 25:172-175.[Medline]
  22. Johnson AJ, Lee BC, Lin W. Echoplanar diffusion-weighted imaging in neonates and infants with suspected hypoxic-ischemic injury: correlation with patient outcome. AJR Am J Roentgenol 1999; 172:219-226.[Abstract/Free Full Text]
  23. Wolf RL, Zimmerman RA, Clancy R, Haselgrove JH. Quantitative apparent diffusion coefficient measurements in term neonates for early detection of hypoxic-ischemic brain injury: initial experience. Radiology 2001; 218:825-833.[Abstract/Free Full Text]
  24. Warach S, Boska M, Welch KM. Pitfalls and potential of clinical diffusion-weighted MR imaging in acute stroke (editorial). Stroke 1997; 28:481-482.
  25. Bottomley PA. Spatial localization in NMR spectroscopy in vivo. Ann N Y Acad Sci 1987; 508:333-348.[Abstract]
  26. Tzika AA, Vigneron DB, Ball WS, Jr, Dunn RS, Kirks DR. Localized proton MR spectroscopy of the brain in children. J Magn Reson Imaging 1993; 3:719-729.[Medline]
  27. Sager TN, Fink-Jensen A, Hansen AJ. Transient elevation of interstitial N-acetylaspartate in reversible global brain ischemia. J Neurochem 1997; 68:675-682.[Medline]
  28. Parsons MW, Li T, Barber PA, et al. Combined (1)H MR spectroscopy and diffusion-weighted MRI improves the prediction of stroke outcome. Neurology 2000; 55:498-505.[Abstract/Free Full Text]
  29. Johnston M. Neurotransmitters and vulnerability of the developing brain. Brain Dev 1995; 17:301-306.[CrossRef][Medline]
  30. Neil JJ, Shiran SI, McKinstry RC, et al. Normal brain in human newborns: apparent diffusion coefficient and diffusion anisotropy measured by using diffusion tensor MR imaging. Radiology 1998; 209:57-66.[Abstract/Free Full Text]
  31. McGill R, Tukey JW, Larsen WA. Variations of box plots. Am Statistician 1978; 3:12-16.
  32. Breslow NE, Day NE. Statistical methods in cancer research Vol 1: The analysis of case-control studies. Lyon, France: International Agency for Research on Cancer, 1980; 200-232.
  33. Aw TY, Andersson BS, Jones DP. Mitochondrial transmembrane ion distribution during anoxia. Am J Physiol 1987; 252:C356-C361.[Abstract/Free Full Text]
  34. Zaidan E, Sims NR. Reduced activity of the pyruvate dehydrogenase complex but not cytochrome c oxidase is associated with neuronal loss in the striatum following short-term forebrain ischemia. Brain Res 1997; 772:23-28.[CrossRef][Medline]
  35. Klatzo I. Evolution of brain edema concepts. Acta Neurochir Suppl 1994; 60:3-6.[Medline]
  36. Vander Heiden MG, Chandel NS, Li XX, Schumacker PT, Colombini M, Thompson CB. Outer mitochondrial membrane permeability can regulate coupled respiration and cell survival. Proc Natl Acad Sci U S A 2000; 97:4666-4671.[Abstract/Free Full Text]
  37. Staub F, Baethmann A, Peters J, Weigt H, Kempski O. Effects of lactacidosis on glial cell volume and viability. J Cereb Blood Flow Metab 1990; 10:866-876.[Medline]
  38. Lorek A, Takei Y, Cady E, et al. Delayed ("secondary") cerebral energy failure after acute hypoxia-ischemia in the newborn piglet: continuous 48-hour studies by phosphorus magnetic resonance spectroscopy. Pediatr Res 1994; 36:699-706.[Medline]
  39. Hanrahan JD, Sargentoni J, Azzopardi D, et al. Cerebral metabolism within 18 hours of birth asphyxia: a proton magnetic resonance spectroscopy study. Pediatr Res 1996; 39:584-590.[Medline]
  40. Thornton JS, Ordidge RJ, Penrice J, et al. Anisotropic water diffusion in white and gray matter of the neonatal piglet brain before and after transient hypoxia-ischaemia. Magn Reson Imaging 1997; 15:433-440.[CrossRef][Medline]
  41. Thornton JS, Ordidge RJ, Penrice J, et al. Temporal and anatomical variations of brain water apparent diffusion coefficient in perinatal cerebral hypoxic-ischemic injury: relationships to cerebral energy metabolism. Magn Reson Med 1998; 39:920-927.[Medline]
  42. Blumberg R, Cady E, Wigglesworth J, McKenzie J, Edwards A. Relation between delayed impairment of cerebral energy metabolism and infarction following transient focal ischemia in the developing brain. Exp Brain Res 1997; 113:130-137.[CrossRef][Medline]
  43. Buchli R, Rumpel H, Maeder U, Nedelcu H, Martin E. Hypoxic ischemic brain damage in neonatal rats monitored by 31P magnetic resonance spectroscopy (MRS) (abstr). Pediatr Res 1995; 38:428A.
  44. Leth H, Toft P, Peitersen B, Lou H, Henriksen O. Use of brain lactate levels to predict outcome after perinatal asphyxia. Acta Paediatr 1996; 85:859-864.[Medline]
  45. Penrice J, Lorek A, Cady EB, et al. Proton magnetic resonance spectroscopy of the brain during acute hypoxia-ischemia and delayed cerebral energy failure in the newborn piglet. Pediatr Res 1997; 41:795-802.[Medline]
  46. Betz AL, Keep RF, Beer ME, Ren XD. Blood-brain barrier permeability and brain concentration of sodium, potassium, and chloride during focal ischemia. J Cereb Blood Flow Metab 1994; 14:29-37.[Medline]
  47. Moseley ME, Cohen Y, Mintorovitch J, et al. Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magn Reson Med 1990; 14:330-346.[Medline]
  48. Ning G, Malisza KL, Del Bigio MR, Bascaramurty S, Kozlowski P, Tuor UI. Magnetic resonance imaging during cerebral hypoxia-ischemia: T2 increases in 2-week- old but not 4-week-old rats. Pediatr Res 1999; 45:173-179.[Medline]
  49. Barkovich AJ, Westmark K, Partridge C, Sola A, Ferriero DM. Perinatal asphyxia: MR findings in the first 10 days. AJNR Am J Neuroradiol 1995; 16:427-438.[Abstract]
  50. Beauchamp NJ, Jr, Barker PB, Wang PY, vanZijl PC. Imaging of acute cerebral ischemia. Radiology 1999; 212:307-324.[Abstract/Free Full Text]
  51. Huppi PS, Barnes PD. Magnetic resonance techniques in the evaluation of the newborn brain. Clin Perinatol 1997; 24:693-723.[Medline]
  52. Lorek A, Takei Y, Cady EB, et al. Delayed ("secondary") cerebral energy failure after acute hypoxia-ischemia in the newborn piglet: continuous 48-hour studies by phosphorus magnetic resonance spectroscopy. Pediatr Res 1994; 36:699-706.
  53. Rumpel H, Nedelcu J, Aguzzi A, Martin E. Late glial swelling after acute cerebral hypoxia-ischemia in the neonatal rat: a combined magnetic resonance and histochemical study. Pediatr Res 1997; 42:54-59.[Medline]
  54. Tuor UI, Kozlowski P, Del Bigio MR, et al. Diffusion- and T2-weighted increases in magnetic resonance images of immature brain during hypoxia-ischemia: transient reversal posthypoxia. Exp Neurol 1998; 150:321-328.[CrossRef][Medline]
  55. Li F, Silva MD, Liu KF, et al. Secondary decline in apparent diffusion coefficient and neurological outcomes after a short period of focal brain ischemia in rats. Ann Neurol 2000; 48:236-244.[CrossRef][Medline]
  56. Gadian DG, Frackowiak RS, Crockard HA, et al. Acute cerebral ischaemia: concurrent changes in cerebral blood flow, energy metabolites, pH, and lactate measured with hydrogen clearance and 31P and 1H nuclear magnetic resonance spectroscopy. I. Methodology. J Cereb Blood Flow Metab 1987; 7:199-206.
  57. Crockard HA, Gadian DG, Frackowiak RS, et al. Acute cerebral ischaemia: concurrent changes in cerebral blood flow, energy metabolites, pH, and lactate measured with hydrogen clearance and 31P and 1H nuclear magnetic resonance spectroscopy. II. Changes during ischaemia. J Cereb Blood Flow Metab 1987; 7:394-402.
  58. Hossmann KA. Viability thresholds and the penumbra of focal ischemia. Ann Neurol 1994; 36:557-565.[CrossRef][Medline]
  59. Hesselbarth D, Franke C, Hata R, Brinker G, Hoehn-Berlage M. High resolution MRI and MRS: a feasibility study for the investigation of focal cerebral ischemia in mice. NMR Biomed 1998; 11:423-429.[CrossRef][Medline]
  60. Azzopardi D, Wyatt J, Cady E, et al. Prognosis of newborn infants with hypoxic-ischemic brain injury assessed by phosphorus magnetic resonance spectroscopy. Pediatr Res 1989; 25:445-451.[Medline]
  61. Amess PN, Penrice J, Wylezinska M, et al. Early brain proton magnetic resonance spectroscopy and neonatal neurology related to neurodevelopmental outcome at 1 year in term infants after presumed hypoxic-ischaemic brain injury. Dev Med Child Neurol 1999; 41:436-445.[CrossRef][Medline]
  62. Barkovich AJ, Westmark KD, Bedi HS, Partridge JC, Ferriero DM, Vigneron DB. Proton spectroscopy and diffusion imaging on the first day of life after perinatal asphyxia: preliminary report. AJNR Am J Neuroradiol 2001; 22:1786-1794.[Abstract/Free Full Text]
  63. Duijn JH, Matson GB, Maudsley AA, Hugg JW, Weiner MW. Human brain infarction: proton MR spectroscopy. Radiology 1992; 183:711-718.[Abstract/Free Full Text]
  64. Shu SK, Ashwal S, Holshouser BA, Nystrom G, Hinshaw DB, Jr. Prognostic value of 1H-MRS in perinatal CNS insults. Pediatr Neurol 1997; 17:309-318.[CrossRef][Medline]
  65. Tong DC, Yenari MA, Albers GW, O’Brien M, Marks MP, Moseley ME. Correlation of perfusion- and diffusion-weighted MRI with NIHSS score in acute (<6.5 hour) ischemic stroke. Neurology 1998; 50:864-870.[Abstract/Free Full Text]
  66. Marks MP, de Crespigny A, Lentz D, Enzmann DR, Albers GW, Moseley ME. Acute and chronic stroke: navigated spin-echo diffusion-weighted MR imaging. Radiology 1996; 199:403-408.[Abstract/Free Full Text]
  67. Connelly A, Chong WK, Johnson CL, Ganesan V, Gadian DG, Kirkham FJ. Diffusion weighted magnetic resonance imaging of compromised tissue in stroke. Arch Dis Child 1997; 77:38-41.[Abstract/Free Full Text]
  68. Rumpel H, Buchli R, Gehrmann J, Aguzzi A, Illi O, Martin E. Magnetic resonance imaging of brain edema in the neonatal rat: a comparison of short and long term hypoxia-ischemia. Pediatr Res 1995; 38:113-118.[Medline]
  69. D’Arceuil H, Rhine W, de Crespigny A, et al. 99mTc annexin V imaging of neonatal hypoxic brain injury. Stroke 2000; 31:2692-2700.[Abstract/Free Full Text]
  70. Tzika A, Zarifi M, Poussaint TY, Astrakas L, Zurakowski D, Duplessis A. MRI and MRS in perinatal asphyxia: predictive for long-term outcome (abstr). Radiology 2000; 217(P):429.
  71. Zarifi M, Astrakas L, Zurakowski D, Poussaint T, du Plessis A, Tzika AA. Predicting clinical outcome in perinatal asphyxia by MRI and MRS (abstr) In: Proceedings of the Ninth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 2001; 580.
  72. Soul J, Robertson R, Tzika A, du Plessis A, Volpe J. Time course of changes in the apparent diffusion coefficient in a case of neonatal encephalopathy with defined onset and duration of hypoxic-ischemic insult. Pediatrics 2001; 108:1211-1214.[Abstract/Free Full Text]
  73. Barkovich AJ, Baranski K, Vigneron D, et al. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. AJNR Am J Neuroradiol 1999; 20:1399-1405.
  74. Terpstra M, High WB, Luo Y, de Graaf RA, Merkle H, Garwood M. Relationships among lactate concentration, blood flow and histopathologic profiles in rat C6 glioma. NMR Biomed 1996; 9:185-194.[CrossRef][Medline]
  75. Barkovich AJ, Ferriero DM. Line-scan diffusion imaging of term neonates with perinatal brain ischemia. AJNR Am J Neuroradiol 1999; 20:1577-1579.[Free Full Text]
  76. Hakumaki JM, Poptani H, Sandmair AM, Yla-Herttuala S, Kauppinen RA. 1H MRS detects polyunsaturated fatty acid accumulation during gene therapy of glioma: implications for the in vivo detection of apoptosis. Nat Med 1999; 5:1323-1327.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Am. J. Neuroradiol.Home page
Z.J. Wang, D.B. Vigneron, S.P. Miller, P. Mukherjee, N.N. Charlton, Y. Lu, and A.J. Barkovich
Brain Metabolite Levels Assessed by Lactate-Edited MR Spectroscopy in Premature Neonates with and without Pentobarbital Sedation
AJNR Am. J. Neuroradiol., April 1, 2008; 29(4): 798 - 801.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
L. Liauw, J. van der Grond, A. A. van den Berg-Huysmans, I. H. Palm-Meinders, M. A. van Buchem, and G. van Wezel-Meijler
Hypoxic-Ischemic Encephalopathy: Diagnostic Value of Conventional MR Imaging Pulse Sequences in Term-born Neonates
Radiology, April 1, 2008; 247(1): 204 - 212.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
S. Shanmugalingam, J. S. Thornton, O. Iwata, A. Bainbridge, F. E. O'Brien, A. N. Priest, R. J. Ordidge, E. B. Cady, J. S. Wyatt, and N. J. Robertson
Comparative Prognostic Utilities of Early Quantitative Magnetic Resonance Imaging Spin-Spin Relaxometry and Proton Magnetic Resonance Spectroscopy in Neonatal Encephalopathy
Pediatrics, October 1, 2006; 118(4): 1467 - 1477.
[Abstract] [Full Text] [PDF]

Home page
 RadioGraphicsHome page
C. P. Chao, C. G. Zaleski, and A. C. Patton
Neonatal Hypoxic-Ischemic Encephalopathy: Multimodality Imaging Findings
RadioGraphics, October 1, 2006; 26(suppl_1): S159 - S172.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
J.L.Y. Cheong, E.B.