- Email: [email protected]
Nucleated red blood cell counts: not associated with brain injury or outcome
Original Articles
Nucleated Red Blood Cell Counts: Not Associated With Brain Injury or Outcome Shannon E. G. Hamrick, MD*, Steven P. Miller, MD†, Nancy R. Newton, MS, RN*, Julian T. Parer, MD, PhD‡, Donna M. Ferriero, MD*†, A. James Barkovich, MD*†§, and J. Colin Partridge, MD* The objective was to determine whether an elevated nucleated red blood cell count at birth after perinatal depression is associated with brain injury as measured by (1) proton magnetic resonance spectroscopy and (2) abnormal neurodevelopmental outcome at 30 months of age. The nucleated red blood cell counts from the first 24 hours of life were statistically analyzed in 33 term infants enrolled in a prospective study of the value of magnetic resonance imaging for the determination of neurodevelopmental outcome after perinatal depression. Nucleated red blood cell counts were elevated in 13/33 (39%). Abnormal outcome (19/33, 54%) was associated with Score for Neonatal Acute PhysiologyPerinatal Extension (P ⴝ 0.04), decreased N-acetylaspartate to choline ratio in the basal ganglia (P ⴝ 0.009), and increased lactate to choline ratio in the basal ganglia (P ⴝ 0.02), but not with cord pH, Apgar score, or nucleated red blood cell value. In a logistic regression model, increasing nucleated red blood cell counts did not increase the odds of an abnormal outcome at 30 months of age (OR 1.02, P ⴝ 0.17). In a population of neonates with perinatal depression, the nucleated red blood cell count at birth does not correlate with magnetic resonance spectroscopy or 30-month neurodevelopmental outcome. The nucleated red blood cell count should not be used as a surrogate marker for subsequent brain injury. © 2003 by Elsevier Inc. All rights reserved. Hamrick SEG, Miller SP, Newton NR, Parer JT, Ferriero DM, Barkovich AJ, Partridge JC. Nucleated red blood cell counts: Not associated with brain injury or outcome. Pediatr Neurol 2003;29:278-283.
From the *Departments of Pediatrics, †Neurology, ‡ Obstetrics/Gynecology, and §Radiology, University of California, San Francisco, San Francisco, California.
278
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
Introduction Perinatal depression is a major cause of neurologic morbidity and mortality in term infants. Approximately 2% of infants are exposed to birth asphyxia with 90% of asphyxial insults occurring in the antepartum or intrapartum periods and the remainder occurring in the postnatal period [1,2]. Approximately 11% of term infants with hypoxic-ischemic encephalopathy do not survive [3]. An estimated 20-40% of infants with moderate hypoxicischemic encephalopathy and nearly 100% with severe hypoxic-ischemic encephalopathy develop long-term neurologic sequelae [4]. If neuroprotective agents are to be used, it is important to identify infants at risk and critical to identify them within a therapeutic window of opportunity. Neurodevelopmental outcome is more accurately predicted for infants with either mild or severe hypoxic-ischemic birth injury, but prognosticating is difficult for those with moderate perinatal depression [2]. Commonly used clinical markers do not accurately predict neurodevelopmental outcome of the term infant after perinatal depression. Low Apgar scores and abnormal acid/base status overestimate sequelae from perinatal depression. Biochemical markers such as cerebrospinal fluid neuron-specific enolase, glial fibrillary acidic protein, brain-specific creatine kinase, lactate dehydrogenase, glutamate, cytokines, superoxide dismutase, glutathione peroxidase, lipid peroxidase, Protein S-100B and S-100A1 have all been investigated as markers of perinatal asphyxia, but they do not necessarily correlate with long-term neurodevelopmental outcome [5-11]. Previous studies have suggested that the nucleated red blood cell (erythrocyte) count is a useful marker of fetal
Communications should be addressed to: Dr. Partridge; Clinical Professor of Pediatrics; San Francisco General Hospital; Mail Stop 6E; 1001 Potrero Avenue; San Francisco, CA 94110. Received November 18, 2002; accepted April 29, 2003.
© 2003 by Elsevier Inc. All rights reserved. doi:10.1016/S0887-8994(03)00266-2 ● 0887-8994/03/$—see front matter
hypoxia [12], can help determine the timing of fetal distress [12,13], and is a useful predictor of outcome after hypoxic-ischemic encephalopathy [12-14]. Two cohorts have addressed the use of the initial nucleated red blood cell count to predict neurodevelopmental outcome. One was limited by the sample size of the patients who were followed long-term [14], and the other was retrospective in design [12,13]. Given this, the value of the nucleated red blood cell count as a predictor of outcome remains uncertain. Our study concurs with the outcomes of other studies in that we found that magnetic resonance spectroscopy can be used to measure cerebral metabolism and neuronal integrity when performed shortly after birth following perinatal asphyxia. Specifically, a decreased ratio of Nacetylaspartate (a marker of neuronal integrity) to choline in the basal ganglia and increased lactate (a marker of cerebral oxidative metabolism) to choline ratio in the basal ganglia were the clinical markers most strongly associated with neurodevelopmental outcome at 30 months of age [15]. Magnetic resonance spectroscopy may also provide the earliest detection of brain injury in infants when compared with other magnetic resonance techniques such as magnetic resonance imaging and diffusion-weighted imaging [16]. As the most reliable in vivo measure of brain injury, we compared the nucleated red blood cell count to magnetic resonance spectroscopy data, as well as the 30-month neurodevelopmental outcome in a group of term neonates with perinatal depression. Our objective was to determine whether an elevated nucleated red blood cell count is associated with brain injury, as measured by [1] magnetic resonance spectroscopy, and [2] abnormal neurodevelopmental outcome at 30 months of age. Methods We studied 35 term infants with perinatal depression as part of a prospective cohort assembled to assess the magnetic resonance predictors of neurodevelopmental outcome. This cohort, enrolled since 1994, has been extensively described elsewhere [17]. Infants were included if they had one of the following indicators of perinatal depression: (1) umbilical artery pH ⬍7.1, (2) umbilical artery base deficit ⬎10, or (3) 5-minute Apgar score ⱕ5. Infants were excluded if (1) gestation was fewer than 36 weeks or (2) there were suspected or confirmed congenital malformations, congenital metabolic disease, or congenital infections. The protocol was approved by the Committee on Human Research at the University of California San Francisco. Infants were studied only after voluntary, informed parental consent was obtained.
Nucleated Red Blood Cell Data Of 35 infants studied at 30 months of age, all had nucleated red blood cell counts reported as part of a standard, initial complete blood count. Two of the 35 infants had nucleated red blood cells reported as “present,” although the absolute value was not recorded. As a result, these two infants were excluded from the analysis. Infants meeting the inclusion criteria for the magnetic resonance study routinely have a complete blood count measured at admission based on clinical practice guidelines. The nucleated red blood cell count was determined by a review of each
patient’s first complete blood count. The initial complete blood count was performed during the first day of life. As is standard practice, the nucleated red blood cell count was reported as a number per 100 white blood cells (leukocytes). A nucleated red blood cell value of ⱖ10/100 leukocytes is considered elevated [18].
Clinical Data We prospectively collected clinical information about each patient to compute a Score for Neonatal Acute Physiology-Perinatal Extension, which is a 37-item score that uses routinely measured physical and laboratory parameters within the first 24 hours of life as a severity of illness index. It has been validated as a predictor of neonatal mortality [19] and, recently, morbidity [20].
Magnetic Resonance Imaging/Magnetic Resonance Spectroscopy Data Proton magnetic resonance imaging (magnetic resonance imaging) and spectroscopy were performed at a median of 6 days of life (range of 2 to 13). The same magnetic resonance imaging scanner and techniques were used for the entire cohort. The infants were imaged as soon as they were stable enough to be transported safely to the magnetic resonance imaging scanner and when imaging time was available. As previously described, the magnetic resonance data were obtained using a standard, research protocol [15-17].
Neurodevelopmental Outcome All patients were examined at 30 months of age. Cognitive development was assessed by administering the Mental Development Index of the Bayley Scales of Infant Development II [21]. A pediatric neurologist, who was not aware of the neonatal course and imaging results of the infant, performed a standardized, neurologic examination. The neurologist scored neuromotor outcome using a validated neuromotor score: (0) normal, (1) abnormal tone or reflexes or primitive reflexes, (2) abnormal tone and reflexes, (3) decreased power and tone or reflex abnormality, (4) cranial nerve involvement and any motor abnormality, and (5) cranial nerve involvement and spastic quadriparesis [22]. Infants were classified as abnormal if the Mental Development Index was less than 85 or the neuromotor score was greater than or equal to 3.
Statistical Analysis Statistical analysis was performed using Stata (Stata Corporation, College Station, Texas). We compared an elevated nucleated red blood cell value (ⱖ10/100 leukocytes) with standard markers of perinatal asphyxia (cord pH, Apgar score, Score for Neonatal Acute PhysiologyPerinatal Extension) and with magnetic resonance spectroscopy findings (ratio of N-acetylaspartate to choline in the basal ganglia and ratio of lactate to choline in the basal ganglia) for predicting the 30-month outcome using the Mann-Whitney U test for nonnormally distributed data and the t test for normally distributed data. Fisher’s Exact Test was used for categoric data. Logistic regression was used to evaluate whether increasing nucleated red blood cell count is associated with an abnormal 30-month outcome.
Results Abnormal outcome on 30-month neurodevelopmental testing was observed in 19 of 33 infants (54%). Of these, 47% were male and 58% were delivered by cesarean section. Of the 14 normal infants, 50% were male and 43% were delivered by cesarean section. These differences
Hamrick et al: NRBCs and Outcome in Brain Injury 279
Table 1.
Relationship between newborn outcome at 30 months of age and clinical and laboratory predictors*
Umbilical artery pH 5-minute Apgar score SNAP-PE NAA/choBG Lac/choBG nRBC/100 WBC Elevated nRBC
Normal (n ⴝ 14)
Abnormal (n ⴝ 19)
P value
7.10 (6.98-7.30) 6 (1-8) 23.5 (11-33) 0.76 (0.59-0.93) 0.06 (0-0.25) 1.5 (0-55) 4/14 (29%)
7.12 (6.82-7.32) 5 (1-9) 27 (15-43) 0.65 (0.38-0.83) 0.13 (0-0.69) 5 (0-109) 9/19 (47%)
0.94† 0.34† 0.04‡ 0.01† 0.02† 0.26† 0.30†
Relationship determined by the Mental Development Index and Neuromotor Score, as well as the clinical or laboratory predictors. Numbers represent median (range or percentage). Abnormal is Mental Development Index ⬍ 85 and Neuromotor Score ⱖ 3. * Fisher’s Exact Test. † Mann-Whitney U test. ‡ t test. Abbreviations: Lac/choBG ⫽ Ratio of lactate to choline in the basal ganglia NAA/choBG ⫽ Ratio of N-acetylaspartate to choline in the basal ganglia
were not statistically significant. The relationship between infant outcome at 30 months of age and clinical or laboratory predictors is shown in Table 1. The normal and abnormal outcome groups did not differ significantly for umbilical artery pH or Apgar score. However, as previously reported, there was a significant association with outcome and Score for Neonatal Acute Physiology-Perinatal Extension, decreased ratio of N-acetylaspartate to choline in the basal ganglia, and increased ratio of lactate to choline in the basal ganglia (Fig 1). Nucleated RBCs were elevated in 13/33 (39%). There were four (29%) in the normal group and nine (47%) in the abnormal group. This was not statistically significant (P ⫽ 0.26). The median nucleated erythrocytes count was not significantly different in the two groups (P ⫽ 0.3). The presence of elevated nucleated red blood cells was not significantly associated with cord pH (P ⫽ 0.6), Apgar score (P ⫽ 0.8), Score for Neonatal Acute PhysiologyPerinatal Extension (P ⫽ 0.1), ratio of N-acetylaspartate to choline in the basal ganglia (P ⫽ 0.053), and ratio of lactate to choline in the basal ganglia (P ⫽ 0.6). In a logistic regression model, increasing nucleated red blood cell counts did not increase the odds of an abnormal outcome at 30 months of age (odds ratio 1.02, P ⫽ 0.17, confidence interval ⫽ 0.99 to 1.05). Using the absolute nucleated red blood cell count (corrected for leukocytes count) did not significantly alter the results. Figure 2 illustrates the nucleated red blood cell value and the time of nucleated red blood cell collection. Discussion In a prospective cohort of term infants with perinatal depression, the nucleated red blood cell count at birth is not associated with brain injury as measured by proton magnetic resonance spectroscopy or neurodevelopmental
280
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
nRBC ⫽ Nucleated red blood cell count SNAP-PE ⫽ Score for Neonatal Acute Physiology-Perinatal Extension
outcome at 30 months of age. Although there was a trend toward significance for the presence of nucleated red blood cells and decreased ratio of N-acetylaspartate to choline in the basal ganglia, nucleated red blood cell count was not associated with other magnetic resonance predictors or neurodevelopmental outcome. Despite the finding that there is a compensatory increase in erythropoiesis after hypoxia that can be estimated by the circulating nucleated red blood cell count, no correlation was observed between the nucleated red blood cell count and brain injury or outcome. A nucleated red blood cell count of less than 10 per 100 leukocytes is common at birth, but there should be no nucleated red blood cells after the first few days of life [17,23]. Tissue hypoxia stimulates erythropoietin, a 34-kD glycoprotein well characterized for its production by the kidney and regulation of hematopoiesis. Erythropoietin does not cross the placenta, therefore, elevated levels in a fetus should reflect fetal hypoxia. Animal studies suggest that plasma erythropoietin levels are increased as early as 4 hours after an episode of hypoxia [24]. In a human study, plasma erythropoietin increased 1.5 hours after exposure to acute hypobaric hypoxia [25]; however, it remains unclear how quickly the nucleated red blood cell count rises in response to erythropoietin. The results from previous studies investigating the value of the nucleated red blood cell count in predicting neurodevelopmental outcome after fetal asphyxia are not consistent with our results. In one cohort [12,13], the neurodevelopmental outcome data were obtained by retrospective chart review of term infants from the National Registry for Brain Injured Infants. The length of neurodevelopmental follow-up of these patients is unclear. The other study [14] was limited by the sample size of the patients followed for long-term outcome and did not accurately account for multiple comparisons. Our cohort
Figure 1. (A) T2-weighted magnetic resonance imaging illustrating voxel placement for spectra analysis. (B) Representative spectra from a patient with a normal outcome (Cho ⫽ choline, C ⫽ creatine, NAA ⫽ N-acetylaspartate). (C) Representative spectra from a patient with an abnormal outcome (Lac ⫽ lactate).
was selected for perinatal depression and not for abnormal outcomes; therefore, the severity of perinatal depression may differ from these previous studies. Though this study is limited by size, within the same cohort other variables, including the Score for Neonatal Acute Physiology-Perinatal Extension, diminished N-acetylaspartate in the basal ganglia, and elevated lactate in the basal ganglia, are each associated with brain injury and predictive of outcome at 30 months of age. The Score for Neonatal Acute Physiology-Perinatal Extension, a measure of illness severity, scores the worst physiologic abnormalities in each organ system during the first 24 hours after admission, with increasing scores reflecting more severe derangement [26]. Such systemic abnormalities often accompany the neurologic syndrome and can have a negative effect on neurodevelopmental prognosis [2]. N-acetylaspartate is a marker of neuronal integrity, and lactate is a marker for impaired cerebral oxidative metabolism; abnormalities of both of these variables directly affect brain injury. Lactate metabolism has also been investigated in other organ systems in the setting of
perinatal asphyxia. The ratio of urinary lactate to creatinine has been shown to identify infants at risk for hypoxic-ischemic encephalopathy and to correlate with outcomes at 1 year of age [27]. However, no direct causal pathway links nucleated red blood cells to brain injury or neurodevelopmental outcome. This study is also limited by the number of nucleated red blood cell assessments made. Because it is not known when the nucleated erythrocytes count rises in response to hypoxia and erythropoietin, it is conceivable that the lack of association with outcome results from acquiring the blood samples at an inappropriate time in relation to the asphyxial injury. On the other hand, this timing of collection reflects the clinical reality of when specimens are obtained and their prognostic importance. Given the data presented in Figure 2, it remains unlikely that the timing of the nucleated red blood cell counts obscured a meaningful relationship with the outcome. In a study investigating normoblast and lymphocyte counts in 16 neonates, for whom the timing of antenatal hypoxia-ischemia was known and who subsequently developed cerebral palsy, it
Hamrick et al: NRBCs and Outcome in Brain Injury 281
Figure 2. Nucleated red blood cell count in relation to age at assay, by outcome group.
was noted that the normoblast count rose within 2 hours after the event and returned to normal within 24 to 36 hours [28]. They also recognized that if hypoxia persisted after birth, the normoblast count remained elevated and was not useful in predicting timing of hypoxic-ischemic insult. Therefore, in a population of depressed term neonates, the nucleated erythrocytes count at birth does not correlate with proton magnetic resonance spectroscopy or with 30-month neurodevelopmental outcome. A better understanding of the timing of nucleated red blood cell elevation in response to hypoxia and erythropoietin is required. Although an elevated nucleated red blood cell count is often used as a marker that perinatal hypoxia has occurred, it should not be used as a surrogate predictor for subsequent brain injury. The authors thank Jeffrey L. Ecker, MD, for his earlier contribution to this project. The authors also thank the National Institutes of Health (NIH) NS35902 and NIH Pediatric Clinical Research Center Grant M01-RR01271 (J.T.P., D.M.F., A.J.B., J.C.P.) for the support of this research. S.E.G.H. is supported by National Institute of Child Health and Human Development Training Grant HD-07162. S.P.M. is supported by the Canadian Institutes of Health Research Clinician Scientist Program (Phase 1) and the Larry L. Hillblom Foundation.
References [1] Low JA. Intrapartum fetal asphyxia: Definition, diagnosis, and classification. Am J Obstet Gynecol 1997;176:957-9. [2] Volpe JJ. Hypoxic-Ischemic encephalopathy. In: Volpe JJ, ed. Neurology of the newborn, 4th ed. Philadelphia: WB Saunders, 2001: 217-394. [3] Vannucci R, Perlman J. Interventions for perinatal hypoxicischemic encephalopathy. Pediatrics 1997;100:1004-14. [4] Hill A, Volpe JJ. Perinatal asphyxia: Clinical aspects. Clin Perinatol 1989;16:435-57. [5] Nagdyman N, Komen W, Ko H, Muller C, Obladen M. Early biochemical indicators of hypoxic-ischemic encephalopathy after birth asphyxia. Pediatr Res 2001;49(4):502-6.
282
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
[6] Blennow M, Savman K, Ilves P, Thoresen M, Rosengren L. Brain-specific proteins in the cerebrospinal fluid of severely asphyxiated newborn infants. Acta Paediatr 2001;90(10):1171-5. [7] Thornberg E, Thiringer K, Hagberg H, Kjellmer I. Neuron specific enolase in asphyxiated newborns: Association with encephalopathy and cerebral function monitor trace. Arch Dis Child Fetal Neonatal Ed 1995;72:F39-42. [8] Hollander DI, Wright L, Nagey DA, Wright JN, Pupkin MJ, Koch T. Indicators of perinatal asphyxia. Am J Obstet Gynecol 1987; 157:839-43. [9] Hagberg H, Thornberg E, Blennow M. Excitatory amino acids in the cerebrospinal fluid of asphyxiated infants: Relationship to hypoxicischemic encephalopathy. Acta Paediatr 1993;82:925-9. [10] Savman K, Blennow M, Gustafson K, Tarkowski E, Hagberg H. Cytokine response in cerebrospinal fluid after birth asphyxia. Pediatr Res 1998;43:746-51. [11] Nangia S, Saili A, Dutta AK, Batra S, Ray GN. Free oxygen radicals—predictors of neonatal outcome following perinatal asphyxia. Indian J Pediatr 1998;65(3):419-27. [12] Korst LM, Phelan JP, Ahn MO, Martin GI. Nucleated red blood cells: An update on the marker for fetal asphyxia. Am J Obstet Gynecol 1996;175:843-846. [13] Phelan JP, Ahn MY, Korst LM, Martin GI. Nucleated red blood cells: A marker for fetal asphyxia? Am J Obstet Gynecol 1995;173: 1380-4. [14] Buonocore G, Perrone S, Gioia D, et al. Nucleated red blood cell count at birth as an index of perinatal brain damage. Am J Obstet Gynecol 1999;181:1500-5. [15] Miller SP, Newton N, Ferriero DM, et al. Predictors of 30-month outcome following perinatal depression: Role of proton MRS and socio-economic factors. Pediatr Res 2002;52(1):71-7. [16] 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. Am J Neuroradiol 2001;22(9):1786-94. [17] Barkovich AJ, Baranski K, Vigneron D, et al. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. Am J Neuroradiol 1999;20:1399-405. [18] Hermansen MC. Nucleated red blood cells in the fetus and newborn. Arch Dis Child Fetal Neonatal Ed 2001;84:F211-5. [19] Richardson DK, Phibbs CS, Gray JE, McCormick MC, Workman-Daniels K, Goldmann DA. Birth weight and illness severity: Independent predictors of neonatal mortality. Pediatrics 1993;91:969-75. [20] Newton NR, Miller SP, Ferriero DM, Kaufman SA, Barkovich
AJ, Partridge JC. SNAP-PE as a predictor of neurodevelopmental outcome at 1 and 2.5 years of age following perinatal depression. Pediatr Res 2001;49(4):310A. [21] Bayley N. The Bayley scales of infant development II. New York: New York Psychological Corporation, 1993. [22] Hajnal BL, Sahebkar-Moghaddam F, Barnwell AJ, Barkovich AJ, Ferriero DM. Early prediction of neurologic outcome after perinatal depression. Pediatr Neurol 1999;21:788-93. [23] Green DW, Mimouni F. Nucleated erythrocytes in healthy infants and in infants of diabetic mothers. J Pediatr 1990;116:129-31. [24] Widness JA, Teramo KA, Clemons GK, et al. Temporal response of immunoreactive erythropoietin to acute hypoxemia in fetal sheep. Pediatr Res 1986;20(1):15-9.
[25] Eckardt K, Boutellier U, Kurtz A, Schopin M, Koller EA, Bauer C. Rate of erythropoietin formation in humans in response to acute hypobaric hypoxia. J Appl Physiol 1989;66:1785-8. [26] Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA. Score for neonatal acute physiology: A physiology severity index for neonatal intensive care. Pediatrics 1993;91:617-23. [27] Huang C, Wang S, Chang Y, Lin K, Wu P. Measurement of urinary lactate: Creatinine ratio for the early identification of newborn infants at risk for hypoxic-ischemic encephalopathy. N Engl J Med 1999;341:328-35. [28] Naeye RL, Localio AR. Determining the time before birth when ischemia and hypoxemia initiated cerebral palsy. Obstet Gynecol 1995; 86:713-9.
Hamrick et al: NRBCs and Outcome in Brain Injury 283
Nucleated Red Blood Cell Counts: Not Associated With Brain Injury or Outcome Shannon E. G. Hamrick, MD*, Steven P. Miller, MD†, Nancy R. Newton, MS, RN*, Julian T. Parer, MD, PhD‡, Donna M. Ferriero, MD*†, A. James Barkovich, MD*†§, and J. Colin Partridge, MD* The objective was to determine whether an elevated nucleated red blood cell count at birth after perinatal depression is associated with brain injury as measured by (1) proton magnetic resonance spectroscopy and (2) abnormal neurodevelopmental outcome at 30 months of age. The nucleated red blood cell counts from the first 24 hours of life were statistically analyzed in 33 term infants enrolled in a prospective study of the value of magnetic resonance imaging for the determination of neurodevelopmental outcome after perinatal depression. Nucleated red blood cell counts were elevated in 13/33 (39%). Abnormal outcome (19/33, 54%) was associated with Score for Neonatal Acute PhysiologyPerinatal Extension (P ⴝ 0.04), decreased N-acetylaspartate to choline ratio in the basal ganglia (P ⴝ 0.009), and increased lactate to choline ratio in the basal ganglia (P ⴝ 0.02), but not with cord pH, Apgar score, or nucleated red blood cell value. In a logistic regression model, increasing nucleated red blood cell counts did not increase the odds of an abnormal outcome at 30 months of age (OR 1.02, P ⴝ 0.17). In a population of neonates with perinatal depression, the nucleated red blood cell count at birth does not correlate with magnetic resonance spectroscopy or 30-month neurodevelopmental outcome. The nucleated red blood cell count should not be used as a surrogate marker for subsequent brain injury. © 2003 by Elsevier Inc. All rights reserved. Hamrick SEG, Miller SP, Newton NR, Parer JT, Ferriero DM, Barkovich AJ, Partridge JC. Nucleated red blood cell counts: Not associated with brain injury or outcome. Pediatr Neurol 2003;29:278-283.
From the *Departments of Pediatrics, †Neurology, ‡ Obstetrics/Gynecology, and §Radiology, University of California, San Francisco, San Francisco, California.
278
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
Introduction Perinatal depression is a major cause of neurologic morbidity and mortality in term infants. Approximately 2% of infants are exposed to birth asphyxia with 90% of asphyxial insults occurring in the antepartum or intrapartum periods and the remainder occurring in the postnatal period [1,2]. Approximately 11% of term infants with hypoxic-ischemic encephalopathy do not survive [3]. An estimated 20-40% of infants with moderate hypoxicischemic encephalopathy and nearly 100% with severe hypoxic-ischemic encephalopathy develop long-term neurologic sequelae [4]. If neuroprotective agents are to be used, it is important to identify infants at risk and critical to identify them within a therapeutic window of opportunity. Neurodevelopmental outcome is more accurately predicted for infants with either mild or severe hypoxic-ischemic birth injury, but prognosticating is difficult for those with moderate perinatal depression [2]. Commonly used clinical markers do not accurately predict neurodevelopmental outcome of the term infant after perinatal depression. Low Apgar scores and abnormal acid/base status overestimate sequelae from perinatal depression. Biochemical markers such as cerebrospinal fluid neuron-specific enolase, glial fibrillary acidic protein, brain-specific creatine kinase, lactate dehydrogenase, glutamate, cytokines, superoxide dismutase, glutathione peroxidase, lipid peroxidase, Protein S-100B and S-100A1 have all been investigated as markers of perinatal asphyxia, but they do not necessarily correlate with long-term neurodevelopmental outcome [5-11]. Previous studies have suggested that the nucleated red blood cell (erythrocyte) count is a useful marker of fetal
Communications should be addressed to: Dr. Partridge; Clinical Professor of Pediatrics; San Francisco General Hospital; Mail Stop 6E; 1001 Potrero Avenue; San Francisco, CA 94110. Received November 18, 2002; accepted April 29, 2003.
© 2003 by Elsevier Inc. All rights reserved. doi:10.1016/S0887-8994(03)00266-2 ● 0887-8994/03/$—see front matter
hypoxia [12], can help determine the timing of fetal distress [12,13], and is a useful predictor of outcome after hypoxic-ischemic encephalopathy [12-14]. Two cohorts have addressed the use of the initial nucleated red blood cell count to predict neurodevelopmental outcome. One was limited by the sample size of the patients who were followed long-term [14], and the other was retrospective in design [12,13]. Given this, the value of the nucleated red blood cell count as a predictor of outcome remains uncertain. Our study concurs with the outcomes of other studies in that we found that magnetic resonance spectroscopy can be used to measure cerebral metabolism and neuronal integrity when performed shortly after birth following perinatal asphyxia. Specifically, a decreased ratio of Nacetylaspartate (a marker of neuronal integrity) to choline in the basal ganglia and increased lactate (a marker of cerebral oxidative metabolism) to choline ratio in the basal ganglia were the clinical markers most strongly associated with neurodevelopmental outcome at 30 months of age [15]. Magnetic resonance spectroscopy may also provide the earliest detection of brain injury in infants when compared with other magnetic resonance techniques such as magnetic resonance imaging and diffusion-weighted imaging [16]. As the most reliable in vivo measure of brain injury, we compared the nucleated red blood cell count to magnetic resonance spectroscopy data, as well as the 30-month neurodevelopmental outcome in a group of term neonates with perinatal depression. Our objective was to determine whether an elevated nucleated red blood cell count is associated with brain injury, as measured by [1] magnetic resonance spectroscopy, and [2] abnormal neurodevelopmental outcome at 30 months of age. Methods We studied 35 term infants with perinatal depression as part of a prospective cohort assembled to assess the magnetic resonance predictors of neurodevelopmental outcome. This cohort, enrolled since 1994, has been extensively described elsewhere [17]. Infants were included if they had one of the following indicators of perinatal depression: (1) umbilical artery pH ⬍7.1, (2) umbilical artery base deficit ⬎10, or (3) 5-minute Apgar score ⱕ5. Infants were excluded if (1) gestation was fewer than 36 weeks or (2) there were suspected or confirmed congenital malformations, congenital metabolic disease, or congenital infections. The protocol was approved by the Committee on Human Research at the University of California San Francisco. Infants were studied only after voluntary, informed parental consent was obtained.
Nucleated Red Blood Cell Data Of 35 infants studied at 30 months of age, all had nucleated red blood cell counts reported as part of a standard, initial complete blood count. Two of the 35 infants had nucleated red blood cells reported as “present,” although the absolute value was not recorded. As a result, these two infants were excluded from the analysis. Infants meeting the inclusion criteria for the magnetic resonance study routinely have a complete blood count measured at admission based on clinical practice guidelines. The nucleated red blood cell count was determined by a review of each
patient’s first complete blood count. The initial complete blood count was performed during the first day of life. As is standard practice, the nucleated red blood cell count was reported as a number per 100 white blood cells (leukocytes). A nucleated red blood cell value of ⱖ10/100 leukocytes is considered elevated [18].
Clinical Data We prospectively collected clinical information about each patient to compute a Score for Neonatal Acute Physiology-Perinatal Extension, which is a 37-item score that uses routinely measured physical and laboratory parameters within the first 24 hours of life as a severity of illness index. It has been validated as a predictor of neonatal mortality [19] and, recently, morbidity [20].
Magnetic Resonance Imaging/Magnetic Resonance Spectroscopy Data Proton magnetic resonance imaging (magnetic resonance imaging) and spectroscopy were performed at a median of 6 days of life (range of 2 to 13). The same magnetic resonance imaging scanner and techniques were used for the entire cohort. The infants were imaged as soon as they were stable enough to be transported safely to the magnetic resonance imaging scanner and when imaging time was available. As previously described, the magnetic resonance data were obtained using a standard, research protocol [15-17].
Neurodevelopmental Outcome All patients were examined at 30 months of age. Cognitive development was assessed by administering the Mental Development Index of the Bayley Scales of Infant Development II [21]. A pediatric neurologist, who was not aware of the neonatal course and imaging results of the infant, performed a standardized, neurologic examination. The neurologist scored neuromotor outcome using a validated neuromotor score: (0) normal, (1) abnormal tone or reflexes or primitive reflexes, (2) abnormal tone and reflexes, (3) decreased power and tone or reflex abnormality, (4) cranial nerve involvement and any motor abnormality, and (5) cranial nerve involvement and spastic quadriparesis [22]. Infants were classified as abnormal if the Mental Development Index was less than 85 or the neuromotor score was greater than or equal to 3.
Statistical Analysis Statistical analysis was performed using Stata (Stata Corporation, College Station, Texas). We compared an elevated nucleated red blood cell value (ⱖ10/100 leukocytes) with standard markers of perinatal asphyxia (cord pH, Apgar score, Score for Neonatal Acute PhysiologyPerinatal Extension) and with magnetic resonance spectroscopy findings (ratio of N-acetylaspartate to choline in the basal ganglia and ratio of lactate to choline in the basal ganglia) for predicting the 30-month outcome using the Mann-Whitney U test for nonnormally distributed data and the t test for normally distributed data. Fisher’s Exact Test was used for categoric data. Logistic regression was used to evaluate whether increasing nucleated red blood cell count is associated with an abnormal 30-month outcome.
Results Abnormal outcome on 30-month neurodevelopmental testing was observed in 19 of 33 infants (54%). Of these, 47% were male and 58% were delivered by cesarean section. Of the 14 normal infants, 50% were male and 43% were delivered by cesarean section. These differences
Hamrick et al: NRBCs and Outcome in Brain Injury 279
Table 1.
Relationship between newborn outcome at 30 months of age and clinical and laboratory predictors*
Umbilical artery pH 5-minute Apgar score SNAP-PE NAA/choBG Lac/choBG nRBC/100 WBC Elevated nRBC
Normal (n ⴝ 14)
Abnormal (n ⴝ 19)
P value
7.10 (6.98-7.30) 6 (1-8) 23.5 (11-33) 0.76 (0.59-0.93) 0.06 (0-0.25) 1.5 (0-55) 4/14 (29%)
7.12 (6.82-7.32) 5 (1-9) 27 (15-43) 0.65 (0.38-0.83) 0.13 (0-0.69) 5 (0-109) 9/19 (47%)
0.94† 0.34† 0.04‡ 0.01† 0.02† 0.26† 0.30†
Relationship determined by the Mental Development Index and Neuromotor Score, as well as the clinical or laboratory predictors. Numbers represent median (range or percentage). Abnormal is Mental Development Index ⬍ 85 and Neuromotor Score ⱖ 3. * Fisher’s Exact Test. † Mann-Whitney U test. ‡ t test. Abbreviations: Lac/choBG ⫽ Ratio of lactate to choline in the basal ganglia NAA/choBG ⫽ Ratio of N-acetylaspartate to choline in the basal ganglia
were not statistically significant. The relationship between infant outcome at 30 months of age and clinical or laboratory predictors is shown in Table 1. The normal and abnormal outcome groups did not differ significantly for umbilical artery pH or Apgar score. However, as previously reported, there was a significant association with outcome and Score for Neonatal Acute Physiology-Perinatal Extension, decreased ratio of N-acetylaspartate to choline in the basal ganglia, and increased ratio of lactate to choline in the basal ganglia (Fig 1). Nucleated RBCs were elevated in 13/33 (39%). There were four (29%) in the normal group and nine (47%) in the abnormal group. This was not statistically significant (P ⫽ 0.26). The median nucleated erythrocytes count was not significantly different in the two groups (P ⫽ 0.3). The presence of elevated nucleated red blood cells was not significantly associated with cord pH (P ⫽ 0.6), Apgar score (P ⫽ 0.8), Score for Neonatal Acute PhysiologyPerinatal Extension (P ⫽ 0.1), ratio of N-acetylaspartate to choline in the basal ganglia (P ⫽ 0.053), and ratio of lactate to choline in the basal ganglia (P ⫽ 0.6). In a logistic regression model, increasing nucleated red blood cell counts did not increase the odds of an abnormal outcome at 30 months of age (odds ratio 1.02, P ⫽ 0.17, confidence interval ⫽ 0.99 to 1.05). Using the absolute nucleated red blood cell count (corrected for leukocytes count) did not significantly alter the results. Figure 2 illustrates the nucleated red blood cell value and the time of nucleated red blood cell collection. Discussion In a prospective cohort of term infants with perinatal depression, the nucleated red blood cell count at birth is not associated with brain injury as measured by proton magnetic resonance spectroscopy or neurodevelopmental
280
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
nRBC ⫽ Nucleated red blood cell count SNAP-PE ⫽ Score for Neonatal Acute Physiology-Perinatal Extension
outcome at 30 months of age. Although there was a trend toward significance for the presence of nucleated red blood cells and decreased ratio of N-acetylaspartate to choline in the basal ganglia, nucleated red blood cell count was not associated with other magnetic resonance predictors or neurodevelopmental outcome. Despite the finding that there is a compensatory increase in erythropoiesis after hypoxia that can be estimated by the circulating nucleated red blood cell count, no correlation was observed between the nucleated red blood cell count and brain injury or outcome. A nucleated red blood cell count of less than 10 per 100 leukocytes is common at birth, but there should be no nucleated red blood cells after the first few days of life [17,23]. Tissue hypoxia stimulates erythropoietin, a 34-kD glycoprotein well characterized for its production by the kidney and regulation of hematopoiesis. Erythropoietin does not cross the placenta, therefore, elevated levels in a fetus should reflect fetal hypoxia. Animal studies suggest that plasma erythropoietin levels are increased as early as 4 hours after an episode of hypoxia [24]. In a human study, plasma erythropoietin increased 1.5 hours after exposure to acute hypobaric hypoxia [25]; however, it remains unclear how quickly the nucleated red blood cell count rises in response to erythropoietin. The results from previous studies investigating the value of the nucleated red blood cell count in predicting neurodevelopmental outcome after fetal asphyxia are not consistent with our results. In one cohort [12,13], the neurodevelopmental outcome data were obtained by retrospective chart review of term infants from the National Registry for Brain Injured Infants. The length of neurodevelopmental follow-up of these patients is unclear. The other study [14] was limited by the sample size of the patients followed for long-term outcome and did not accurately account for multiple comparisons. Our cohort
Figure 1. (A) T2-weighted magnetic resonance imaging illustrating voxel placement for spectra analysis. (B) Representative spectra from a patient with a normal outcome (Cho ⫽ choline, C ⫽ creatine, NAA ⫽ N-acetylaspartate). (C) Representative spectra from a patient with an abnormal outcome (Lac ⫽ lactate).
was selected for perinatal depression and not for abnormal outcomes; therefore, the severity of perinatal depression may differ from these previous studies. Though this study is limited by size, within the same cohort other variables, including the Score for Neonatal Acute Physiology-Perinatal Extension, diminished N-acetylaspartate in the basal ganglia, and elevated lactate in the basal ganglia, are each associated with brain injury and predictive of outcome at 30 months of age. The Score for Neonatal Acute Physiology-Perinatal Extension, a measure of illness severity, scores the worst physiologic abnormalities in each organ system during the first 24 hours after admission, with increasing scores reflecting more severe derangement [26]. Such systemic abnormalities often accompany the neurologic syndrome and can have a negative effect on neurodevelopmental prognosis [2]. N-acetylaspartate is a marker of neuronal integrity, and lactate is a marker for impaired cerebral oxidative metabolism; abnormalities of both of these variables directly affect brain injury. Lactate metabolism has also been investigated in other organ systems in the setting of
perinatal asphyxia. The ratio of urinary lactate to creatinine has been shown to identify infants at risk for hypoxic-ischemic encephalopathy and to correlate with outcomes at 1 year of age [27]. However, no direct causal pathway links nucleated red blood cells to brain injury or neurodevelopmental outcome. This study is also limited by the number of nucleated red blood cell assessments made. Because it is not known when the nucleated erythrocytes count rises in response to hypoxia and erythropoietin, it is conceivable that the lack of association with outcome results from acquiring the blood samples at an inappropriate time in relation to the asphyxial injury. On the other hand, this timing of collection reflects the clinical reality of when specimens are obtained and their prognostic importance. Given the data presented in Figure 2, it remains unlikely that the timing of the nucleated red blood cell counts obscured a meaningful relationship with the outcome. In a study investigating normoblast and lymphocyte counts in 16 neonates, for whom the timing of antenatal hypoxia-ischemia was known and who subsequently developed cerebral palsy, it
Hamrick et al: NRBCs and Outcome in Brain Injury 281
Figure 2. Nucleated red blood cell count in relation to age at assay, by outcome group.
was noted that the normoblast count rose within 2 hours after the event and returned to normal within 24 to 36 hours [28]. They also recognized that if hypoxia persisted after birth, the normoblast count remained elevated and was not useful in predicting timing of hypoxic-ischemic insult. Therefore, in a population of depressed term neonates, the nucleated erythrocytes count at birth does not correlate with proton magnetic resonance spectroscopy or with 30-month neurodevelopmental outcome. A better understanding of the timing of nucleated red blood cell elevation in response to hypoxia and erythropoietin is required. Although an elevated nucleated red blood cell count is often used as a marker that perinatal hypoxia has occurred, it should not be used as a surrogate predictor for subsequent brain injury. The authors thank Jeffrey L. Ecker, MD, for his earlier contribution to this project. The authors also thank the National Institutes of Health (NIH) NS35902 and NIH Pediatric Clinical Research Center Grant M01-RR01271 (J.T.P., D.M.F., A.J.B., J.C.P.) for the support of this research. S.E.G.H. is supported by National Institute of Child Health and Human Development Training Grant HD-07162. S.P.M. is supported by the Canadian Institutes of Health Research Clinician Scientist Program (Phase 1) and the Larry L. Hillblom Foundation.
References [1] Low JA. Intrapartum fetal asphyxia: Definition, diagnosis, and classification. Am J Obstet Gynecol 1997;176:957-9. [2] Volpe JJ. Hypoxic-Ischemic encephalopathy. In: Volpe JJ, ed. Neurology of the newborn, 4th ed. Philadelphia: WB Saunders, 2001: 217-394. [3] Vannucci R, Perlman J. Interventions for perinatal hypoxicischemic encephalopathy. Pediatrics 1997;100:1004-14. [4] Hill A, Volpe JJ. Perinatal asphyxia: Clinical aspects. Clin Perinatol 1989;16:435-57. [5] Nagdyman N, Komen W, Ko H, Muller C, Obladen M. Early biochemical indicators of hypoxic-ischemic encephalopathy after birth asphyxia. Pediatr Res 2001;49(4):502-6.
282
PEDIATRIC NEUROLOGY
Vol. 29 No. 4
[6] Blennow M, Savman K, Ilves P, Thoresen M, Rosengren L. Brain-specific proteins in the cerebrospinal fluid of severely asphyxiated newborn infants. Acta Paediatr 2001;90(10):1171-5. [7] Thornberg E, Thiringer K, Hagberg H, Kjellmer I. Neuron specific enolase in asphyxiated newborns: Association with encephalopathy and cerebral function monitor trace. Arch Dis Child Fetal Neonatal Ed 1995;72:F39-42. [8] Hollander DI, Wright L, Nagey DA, Wright JN, Pupkin MJ, Koch T. Indicators of perinatal asphyxia. Am J Obstet Gynecol 1987; 157:839-43. [9] Hagberg H, Thornberg E, Blennow M. Excitatory amino acids in the cerebrospinal fluid of asphyxiated infants: Relationship to hypoxicischemic encephalopathy. Acta Paediatr 1993;82:925-9. [10] Savman K, Blennow M, Gustafson K, Tarkowski E, Hagberg H. Cytokine response in cerebrospinal fluid after birth asphyxia. Pediatr Res 1998;43:746-51. [11] Nangia S, Saili A, Dutta AK, Batra S, Ray GN. Free oxygen radicals—predictors of neonatal outcome following perinatal asphyxia. Indian J Pediatr 1998;65(3):419-27. [12] Korst LM, Phelan JP, Ahn MO, Martin GI. Nucleated red blood cells: An update on the marker for fetal asphyxia. Am J Obstet Gynecol 1996;175:843-846. [13] Phelan JP, Ahn MY, Korst LM, Martin GI. Nucleated red blood cells: A marker for fetal asphyxia? Am J Obstet Gynecol 1995;173: 1380-4. [14] Buonocore G, Perrone S, Gioia D, et al. Nucleated red blood cell count at birth as an index of perinatal brain damage. Am J Obstet Gynecol 1999;181:1500-5. [15] Miller SP, Newton N, Ferriero DM, et al. Predictors of 30-month outcome following perinatal depression: Role of proton MRS and socio-economic factors. Pediatr Res 2002;52(1):71-7. [16] 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. Am J Neuroradiol 2001;22(9):1786-94. [17] Barkovich AJ, Baranski K, Vigneron D, et al. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. Am J Neuroradiol 1999;20:1399-405. [18] Hermansen MC. Nucleated red blood cells in the fetus and newborn. Arch Dis Child Fetal Neonatal Ed 2001;84:F211-5. [19] Richardson DK, Phibbs CS, Gray JE, McCormick MC, Workman-Daniels K, Goldmann DA. Birth weight and illness severity: Independent predictors of neonatal mortality. Pediatrics 1993;91:969-75. [20] Newton NR, Miller SP, Ferriero DM, Kaufman SA, Barkovich
AJ, Partridge JC. SNAP-PE as a predictor of neurodevelopmental outcome at 1 and 2.5 years of age following perinatal depression. Pediatr Res 2001;49(4):310A. [21] Bayley N. The Bayley scales of infant development II. New York: New York Psychological Corporation, 1993. [22] Hajnal BL, Sahebkar-Moghaddam F, Barnwell AJ, Barkovich AJ, Ferriero DM. Early prediction of neurologic outcome after perinatal depression. Pediatr Neurol 1999;21:788-93. [23] Green DW, Mimouni F. Nucleated erythrocytes in healthy infants and in infants of diabetic mothers. J Pediatr 1990;116:129-31. [24] Widness JA, Teramo KA, Clemons GK, et al. Temporal response of immunoreactive erythropoietin to acute hypoxemia in fetal sheep. Pediatr Res 1986;20(1):15-9.
[25] Eckardt K, Boutellier U, Kurtz A, Schopin M, Koller EA, Bauer C. Rate of erythropoietin formation in humans in response to acute hypobaric hypoxia. J Appl Physiol 1989;66:1785-8. [26] Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA. Score for neonatal acute physiology: A physiology severity index for neonatal intensive care. Pediatrics 1993;91:617-23. [27] Huang C, Wang S, Chang Y, Lin K, Wu P. Measurement of urinary lactate: Creatinine ratio for the early identification of newborn infants at risk for hypoxic-ischemic encephalopathy. N Engl J Med 1999;341:328-35. [28] Naeye RL, Localio AR. Determining the time before birth when ischemia and hypoxemia initiated cerebral palsy. Obstet Gynecol 1995; 86:713-9.
Hamrick et al: NRBCs and Outcome in Brain Injury 283