Increased plasma concentrations of activin A predict intraventricular hemorrhage in preterm newborns.
We investigated the association of IVH (detected by ultrasonographic findings) and biochemical markers of perinatal hypoxia. Because perinatal hypoxia leads to increased absolute nucleated erythrocyte (NRBC) counts (6) and plasma concentrations of hypoxanthine (Hyp) (5, 7) and xanthine (Xan) (5, 7), [both indicators of oxidative stress (4, 5) ], we chose these as potential markers. We also measured concentrations of plasma activin A. Activin A is a glycoprotein composed of 2 [beta]A subunits that belong to the transforming growth factor-[beta] superfamily of differentiation factors and is expressed in the central nervous system (8). Several models of hypoxic-ischemic brain injury have demonstrated induction of activin A in various brain regions (9 -11). Moreover, intrauterine hypoxia increases activin A in the fetal circulation of preterm newborns (12), and high activin A concentrations have been found in cerebrospinal fluid (13) and urine (unpublished) of asphyxiated full-term newborns who experienced brain damage.
Materials and Methods
We evaluated 53 consecutively born preterm infants born at [less than or equal to] 32 weeks of gestation and admitted to Siena University Hospital at the Department of Pediatrics, Obstetrics and Reproductive Medicine between July 2002 and June 2003. We excluded from the study infants with births complicated by congenital malformations, inborn errors of metabolism, blood group incompatibility, sepsis, maternal diabetes, multiple births, or anemia. Vaginal delivery occurred in 28 women; emergency cesarean section was performed in 25 cases for indications including threatened fetal status as defined by the American College of Obstetricians and Gynecologists (bradycardia, fetal heart rate late decelerations, severe and repetitive fetal heart rate variable decelerations, reduced beat-to-beat variability) (14). We administered corticosteroid therapy (intramuscular [beta]-methasone,12 mg/24 h for 2 days) to 27 women, and 14 of 28 women who delivered vaginally received tocolytic therapy (intravenous ritodrine, 30-50 mg/h). No newborns required delivery by forceps.
Perinatal hypoxia was defined as the presence of at least 2 of the following conditions: intrapartum distress, as indicated by fetal bradycardia (heart rate <100 beats per min, late decelerations, or an absence of heart rate variability), an Apgar score of [less than or equal to] 6 at 5 min, need for resuscitation for >1 min with positive-pressure ventilation and oxygen immediately after birth, and a blood pH [less than or equal to] 7.20 in the umbilical vein (12). The degree of hypoxia was ascertained by measurement of Hyp and Xan concentrations in cord blood.
The study was approved by the Human Ethics Committee of the Medical Faculty, University of Siena, and informed written parental consent was obtained before enrollment of each infant.
To measure erythrocyte count, arterial blood pH, ion concentrations, plasma glucose, and base deficit in the newborns during the first hour after birth, we collected arterial blood samples (1 mL) from the umbilical cord into heparin-containing tubes without the use of catheters.
To check for IVH, we performed cerebral ultrasound scanning on newborns, within 48 h of birth, with a real-time ultrasound machine (Interspec Apogee) with a transducer frequency emission of 7.5 mHz and repeated the scan serially at 5- or 6-day intervals until the age of 4 weeks (15). A single neonatologist experienced in ultrasonography and blind to patient history evaluated the neonatal scans. Clinical management and intensive care procedures were the same for all babies. The occurrence of IVH was diagnosed according to Papile et al. (16).
NRBC COUNT ASSESSMENT We performed a complete blood cell count and a total leukocyte count. NRBC count was expressed as absolute erythroblast count (NRBCs/mm3) and calculated by light microscopic examination of May-Grunwald-Giemsastained blood smears. The blood was immediately centrifuged and subdivided into aliquots for chemical determinations.
MEASUREMENT OF HYP AND XAN
Hyp and Xan in plasma were measured in a blinded fashion within 2 h after blood sampling, to avoid storage effects. After centrifugation, the plasma and buffy coat were removed. Plasma Hyp and Xan concentrations were measured with a Vista 5500 HPLC equipped with a variable-wavelength ultraviolet detector (model 4290; Varian). The analytical system also included a ready-touse prepacked Supelcosil LC-18 column by Supelco [250 x 4.6 mm (i.d.); 5 /,tm bead size], with a precolumn [20 x 4.6 mm (i.d.)] filled with the same packing (Supelguard; Supelco). Two solvents, 0.01 mol/L potassiumosphate buffer at pH 5.5 (A) and methanol (B), were used. The mobile phase gradient was 0% B at 0 min, 10% B at 10 min, 20% Bat 20 min, and 0% B at 30 min. The next sample was injected 10 min later. The flow rate was 1 mL/min, and the wavelength was 220 run.
ACTIVIN A ASSAY
Activin A measurements were performed blinded and in duplicate with a specific 2-site enzyme immunoassay from Serotec (12). The analytic detection limit of the activin A assay was <10 ng/L; intra- and interassay CVs were 5.0% and 9.0%, respectively. Cross-reactivities with various inhibin-related proteins were <0.5%.
We used the Kolmogorov-Smirnov test to confirm that the distributions of activin A, Hyp, and Xan concentrations and NRBC counts were not gaussian. We expressed the data as mean (SE) and analyzed for statistically significant differences by unpaired t-test. We used Pearson correlations and analyzed the differences between groups for each categorical variable with X' and Fisher exact tests.
Using ROC curves (17), we estimated the probability of developing IVH after having none or 1 positive test (higher than the cutoff point) and compared it with the pretest probability, defined as the prevalence of brain damage in the whole group of newborns (18).
Characteristics of preterm newborns are shown in Table 1. Of 53 newborns enrolled, 11 developed grade I or grade II IVH, according to the criteria of Papile et al. (16), in the first 10 days of life. None had ultrasonographic signs of IVH at birth or within 24 h of birth. Infants with and without IVH differed significantly in Apgar score at 5 min (P <0.0001), blood pH (P <0.001), prevalence of hypoxia (P = 0.016), and heart rate (P <0.01; Table 1).
FETAL PLASMA ACTIVIN A, HYP, AND XAN CONCENTRATIONS, AND NRBC COUNT
Activin A, Hyp, and Xan concentrations and NRBC counts were significantly higher in preterm newborns who developed IVH than in those who did not (Fig. 1). Fetal plasma activin A correlated significantly with Hyp (r = 0.948; P <0.0001) and Xan (r = 0.897; P <0.0001) concentrations and the NRBC count (r = 0.895; P <0.0001) in newborns without IVH as well as in those who developed IVH at follow-up (Hyp, r = 0.890, P = 0.0002; Xan, r = 0.950, P <0.0001; NRBC count, r = 0.897, P = 0.0002; data not shown).
PREDICTION OF IVH IN PRETERM NEWBORNS
Activin A was >0.8 [micro]g/L in 11 of 11 patients with IVH [sensitivity, 100%; 95% confidence interval (CI), 71%-100%] and <0.8 [micro]g/L in 39 of 42 without IVH (sensitivity, 93%; 95% CI, 81%-98%). The area under the ROC curve was 0.98, and the positive and negative likelihood ratios (LRs) were 14 and 0, respectively (Fig. 2 and Table 2).
In contrast, Hyp was >0.8 mg/L in 8 of 11 patients with IVH (sensitivity, 73%; 95% CI, 39%-94%) and <0.8 mg/L in 33 of 42 without IVH (sensitivity, 79%; 95% CI, 63%-90%), with an area under the ROC curve of 0.78, and the positive and negative LRs of 3.39 and 0.35, respectively (Fig. 2 and Table 2).
Xan was >0.99 mg/L in 9 of 11 patients with IVH (sensitivity, 82%; 95% CI, 48%-97.2%) and <0.99 mg/L in 33 of 42 newborns without IVH (sensitivity, 79%; 95% CI, 63%-90%). The area under the ROC curve was 0.76, and the positive and negative LRs were 3.82 and 0.23, respectively (Fig. 2 and Table 2).
NRBC count at a cutoff of 1529 cells/[mm.sup.3] showed a sensitivity of 73% in 8 of 11 patients (95% CI, 39%-94%) and a specificity of 83% (35 of 42; 95% CI, 69%-93%) as a single marker of IVH, with an ROC area under the curve of 0.78, and positive and negative LRs of 4.36 and 0.33, respectively (Fig. 2 and Table 2).
The areas under the ROC curve for activin A were significantly higher than those for Hyp (P = 0.001; difference between areas, 0.196; 95% CI, 0.1-0.3), Xan (P = 0.015; difference between areas, 0.221; 95% CI, 0.04-0.4), and NRBC count (P = 0.030; difference between areas, 0.201; 95% CI, 0.02-0.4), as evaluated by pairwise comparison of ROC curves.
Of 53 newborns, 11 had IVH, giving an overall prevalence of 21% (95% CI, 10%-32%), which was the probability of IVH in preterm newborns before measurement of activin A, Hyp, Xan, and NRBC count (pretest probability). If plasma activin A concentrations were high (i.e., above the thresholds defined by the ROC curve analysis), the probability of IVH (positive predictive value) was as high as 79% (95% CI, 61%-100%; Fig. 3), but if activin A concentrations were within the reference interval, that probability was as low as 0% (95% CI, 0%-2%). The probability of predicting IVH by a high Hyp concentration was 67% (95% CI, 36%-98%; Fig. 3), whereas the probability was 9% (95% CI, 1%-17%) for concentrations below the thresholds identified through ROC curve analysis. With respect to Xan concentration and NRBC count, the probabilities of early detection of IVH in the presence of values above the thresholds defined by the ROC curve analysis were 50% (95% CI, 27%-73%) and 53% (95% CI, 28%-78%), respectively (Fig. 3), and 6% (95% CI, 0%-14%) and 8% (95% Cl, 0%-16%), respectively, if concentrations were within the reference intervals.
[FIGURE 1 OMITTED]
The detection of both [beta]A mRNA and activin protein almost exclusively in neurons adjacent to lesion sites suggests that local activin A production increases in response to neuronal damage (9-11, 19). Furthermore, recent in vivo data have suggested that, in the presence of hypoxia and/or asphyxia, activin A concentrations increase in the central nervous system, as in full-term infants who have suffered asphyxia, later developed brain damage, and were found to have increased activin A concentrations in cerebrospinal fluid (13) and urine (P. Florio et al. High urinary concentrations of activin A in asphyxiated full-term newborns with moderate or severe hypoxic ischemic encephalopathy; submitted for publication). On the other hand, increased activin-A concentrations have been observed in the amniotic fluid of patients who subsequently died of intrauterine fetal hypoxia (20) and in the plasma of hypoxic preterm newborns (12).
These reports and our findings of increased plasma concentrations of activin A in newborns who developed IVH suggest that increased activin A concentrations are a direct indication of increased production in the central nervous system as a response to acute hypoxic neuronal damage. This hypothesis is further reinforced by the observation that similar activin A induction occurs in vitro after hypoxic-ischemic injury and in response to cytotoxic lesions (9-11) and by our finding that activin A, Hyp, and Xan concentrations and NRBC counts, all biochemical markers of hypoxia, were increased in newborns who developed IVH.
Recent studies have revealed the role of activin A as a critical modulator of growth and survival in cytoprotection and tissue repair (21). The neurotrophic function of activin A after hypoxic-ischemic brain injury (11, 22-25) enhances the survival of midbrain and hippocampal neurons (24, 25), decreases ischemic brain injury in infant rats (11), and shields striatal and midbrain neurons against neurotoxic damage (22, 23, 25). These findings and the evidence that activin A prevents apoptosis (26) and inhibits caspase (27), 2 important pathways involved in neuronal death (28), support the hypothesis that oversecretion of activin A in the presence of hypoxia/asphyxia may serve a neuroprotective function, reducing neuronal loss from brain injury. Moreover, the evidence that activin A directly potentiates the proliferation and differentiation of erythroid progenitors (29-31) and the present findings of a positive correlation between NRBC count and activin A concentrations suggest a role for activin A in activating erythropoiesis to increase the oxygen supply to the brain after hypoxia (32).
[FIGURE 2 OMITTED]
Our study results also suggest that indications of hypoxic-ischemic brain insult and high risk for IVH might be detectable at an earlier stage with activin A measurement than with ultrasound and other diagnostic procedures. We found that activin A showed 100% sensitivity and 88% specificity as a single marker for the prediction of hypoxic-ischemic brain insult in preterm newborns. Therefore, activin A concentrations above the threshold defined by the ROC curve analysis indicated a probability of IVH as high as 79%, but concentrations with positive and negative predictive values that differed from the pretest probability of the disease in our population (21%) indicated a probability as low as 0%. The high accuracy of activin A might be explained by the strong up-regulation of activin A in the brain within 6 h of injury (11,19,33), whereas a longer period is needed for the appearance of nucleated erythrocytes, Hyp, and Ha in blood when fetal Poz decreases and oxidative stress is activated (32, 34). In any case, the availability of indicators of subclinical lesions at a time when other monitoring indicators are unable to detect bleeding might allow early recognition of the hemorrhaged lesion, permitting the prevention and/or treatment of clinical neurologic damage. Because of the delay of hours or days between hypoxemic events and detection of the hemorrhagic lesion by conventional sonograms, any increase in activin A might be useful for identifying infants at high risk for hemorrhage before the damage becomes visible, so that interventions to prevent or lessen the severity of hemorrhage might be initiated as soon as possible.
[FIGURE 3 OMITTED]
In conclusion, activin A concentrations are increased in preterm newborns who develop IVH. Thus, activin A measurement might be useful for identification of hypoxicischemic brain insult and high risk for IVH in infants before the appearance of related biophysical signs.
Received December 25, 2005; accepted May 12, 2006.
Previously published online at DOI: 10.1373/clinchem.2005.065979
(1.) Philip AG, Allan WC, Tito AM, Wheeler LR. Intraventricular hemorrhage in preterm infants: declining incidence in the 1980s. Pediatrics 1989;84:797-801.
(2.) Thorp JA, Jones PG, Clark RH, Knox E, Peabody JL. Perinatal factors associated with severe intracranial hemorrhage. Am J Obstet Gynecol 2001;185:859-62.
(3.) Lou HC. Hypoxic-hemodynamic pathogenesis of brain lesions in the newborn. Brain Dev 1994;16:423-31.
(4.) Lewen A, Matz P, Chan PH. Free radical pathways in CNS injury. J Neurotrauma 2000;17:871-90.
(5.) Buonocore G, Perrone S, Longini M, Vezzosi P, Bracci R. Total hydroperoxide and advanced oxidation protein products in preterm hypoxic babies. Pediatr Res 2000;47:221-4.
(6.) Buonocore G, Perrone S, Gioia D, Gatti MG, Massafra C, Agosta R, et al. Nucleated red blood cell count at birth as an index of perinatal brain damage. Am J Obstet Gynecol 1999;181:1500-5.
(7.) Buonocore G, Perrone S, Longini M, Vezzosi P, Marzocchi B, Paffetti P, et al. Oxidative stress in preterm neonates at birth and on the seventh day of life. Pediatr Res 2002;52:46-9.
(8.) Luisi S, Florio P, Reis FM, Petraglia F. Expression and secretion of activin A: possible physiological and clinical implications. Eur J Endocrinol 2001;145:225-36.
(9.) Lai M, Gluckman P, Dragunow M, Hughes PE. Focal brain injury increases activin RA mRNA expression in hippocampal neurons. Neuroreport 1997;8:2691-4.
(10.) Lai M, Sirimanne E, Williams CE, Gluckman PD. Sequential patterns of inhibin subunit gene expression following hypoxicischemic injury in the rat brain. Neuroscience 1996;70:1013-24.
(11.) Wu DD, Lai M, Hughes PE, Sirimanne E, Gluckman PD, Williams CE. Expression of the activin axis and neuronal rescue effects of recombinant activin A following hypoxic-ischemic brain injury in the infant rat. Brain Res 1999;835:369-78.
(12.) Florio P, Perrone S, Luisi S, Longini M, Tanganelli D, Petraglia F, et al. Activin A plasma levels at birth: an index of fetal hypoxia in preterm newborn. Pediatr Res 2003;54:696-700.
(13.) Florio P, Luisi S, Bruschettini M, Grutzfeld D, Dobrzanska A, Bruschettini P, et al. Cerebrospinal fluid activin a measurement in asphyxiated full-term newborns predicts hypoxic ischemic encephalopathy. Clin Chem 2004;50:2386-9.
(14.) American College of Obstetrics and Gynecology Committee Opinion. Inappropriate use of the terms fetal distress and birth asphyxia. Int J Gynecol Obstet 1998;61:309-10.
(15.) Perlman JM, Rollins N. Surveillance protocol for the detection of intracranial abnormalities in premature neonates. Arch Pediatr Adolesc Med 2000;154:822-6.
(16.) Papile LA, Burnstein J, Burnstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weight less than 1.500 gm. J Pediatr 1978;92:529-34.
(17.) Stephan C, Wesseling S, Schink T, Jung K. Comparison of eight computer programs for receiver-operating characteristic analysis. Clin Chem 2003;49:433-9.
(18.) Richardson WS, Wilson MC, Guyatt GH, Cook DJ, Nishiyama J. Users' guides to the medical literature: XV. How to use an article about disease probability for differential diagnosis: evidence-based medicine working group. JAMA 1999;281:1214-9.
(19.) Tretter YP, Munz B, Hubner G, ten Bruggencate G, Werner S, Alzheimer C. Strong induction of activin expression after hippocampal lesion. Neuroreport 1996;7:1819-23.
(20.) Petraglia F, Gomez R, Luisi S, Florio P, Tolosa JE, Stomati M, et al. Increased midtrimester amniotic fluid activin A: a risk factor for subsequent fetal death. Am J Obstet Gynecol 1999;180:194-7.
(21.) Wankell M, Werner S, Alzheimer C, Werner S. The roles of activin in cytoprotection and tissue repair. Ann N Y Acad Sci 2003;995: 48-58.
(22.) Tretter YP, Hertel M, Munz B, ten Bruggencate G, Werner S, Alzheimer C. Induction of activin A is essential for the neuroprotective action of basic fibroblast growth factor in vivo. Nat Med 2000;6:812-5.
(23.) Schubert D, Kimura H, LaCorbiere M, Vaughan J, Karr D, Fischer WH. Activin is a nerve cell survival molecule. Nature 1990;344: 868-70.
(24.) Iwahori Y, Saito H, Torii K, Nishiyama N. Activin exerts a neurotrophic effect on cultured hippocampal neurons. Brain Res 1997; 760:52-8.
(25.) Krieglstein K, Suter-Crazzolara C, Fischer WH, Unsicker K. TGF-R superfamily members promote survival of midbrain dopaminergic neurons and protect them against MPP+ toxicity. EMBO J 1995; 14:736-42.
(26.) Alzheimer C, Werner S. Fibroblast growth factors and neuroprotection. Adv Exp Med Biol 2002;513:335-51.
(27.) Tessier C, Prigent-Tessier A, Bao L, Telleria CM, FergusonGottschall S, Gibori GB, Gu Y, et al. Decidual activin: its role in the apoptotic process and its regulation by prolactin. Biol Reprod 2003;68:1687-94.
(28.) Troy CM, Salvesen GS. Caspases on the brain. J Neurosci Res 2002;69:145-50.
(29.) Eramaa M, Hurme M, Stenman UH, Ritvos 0. Activin A/erythroid differentiation factor is induced during human monocyte activation. J Exp Med 1992;176:1449-52.
(30.) Broxmeyer HE, Lu L, Cooper S, Schwall RH, Mason AJ, Nikolics K. Selective and indirect modulation of human multipotenttial and erythroid hematopoietic progenitor cell proliferation by recombinant human activin and inhibin. Proc Natl Acad Sci U S A 1988; 85:9052-6.
(31.) Yu J, Shao L, Vaughan J, Vale W, Yu AL. Characterization of the potentiation effect of activin on human erythroid colony formation in vitro. Blood 1989;73:952-60.
(32.) Clemons GK, Fitzsimons SL, DeManincor D. Immunoreactive erythropoietin concentrations in fetal and neonatal rats and the effects of hypoxia. Blood 1986;68:892-9.
(33.) Inokuchi K, Kato A, Hiraia K, Hishinuma F, Inoue M, Ozawa F. Increase in activin R A mRNA in rat hippocampus during long-term potentiation. FEBS 1996;382:48-52.
(34.) Buonocore G, Perrone S. Biomarkers of hypoxic brain injury in the neonate. Clin Perinatol 2004;31:107-16.
(1) Nonstandard abbreviations: IVH, intraventricular hemorrhage; NRBC, nucleated erythrocyte; Hyp, hypoxanthine; Xan, xanthine; CI, confidence interval; and LR, likelihood ratio.
PASQUALE FLORIO, SERAFINA PERRONE, STEFANO LUISI, PIERO VEZZOSI, MARIANGELA LONGINI, BARBARA MARZOCCHI, FELICE PETRAGLIA, and GIUSEPPE BUONOCORE *
Department of Pediatrics, Obstetrics and Reproductive Medicine, University of Siena, Siena, Italy.
* Address correspondence to this author at: Department of Pediatrics, Obstetrics and Reproductive Medicine, University of Siena, Policlinioo "Le Scotte", We Bracci 36, 53100 Siena, Italy. Fax 39-0577-586182; e-mail email@example.com.
Table 1. Preterm newborn patient data recorded at admission to the newborn intensive care unit. (a) IVH (n 11) Controls (n = 42) P Gestational age, weeks 28.12 (1.5) 28.39 (0.05) NS(b) Birth weight, g 1120.44 (259.16) 1163.31 (109.42) NS 5-min Apgar score 6.09 (1.44) 8.89 (1.76) 0.0001 Emergency cesarean 9 16 NS section, n Prenatal corticosteroid 9 18 NS therapy, n Preeclampsia, n 4 8 NS Perinatal hypoxia, n 8 6 0.019 Blood pH 7.10 (0.18) 7.29 (0.10) 0.0004 Plasma glucose, g/L 470 (161.3) 462.1 (114.2) NS Base excess, mmol/L -11.67 (9.3) 5.86 (5.27) NS [Na.sup.+] mmol/L 134.27 (2.19) 134.89 (2.62) NS [K.sup.+] mmol/L 4.30 (0.45) 4.51 (0.41) NS [Ca.sup.2+] mmol/L 1.25 (0.12) 1.31 (0.09) NS Heart rate, beats/min 148.4 (10.99) 138.5 (8.16) 0.01 Systolic arterial 49.55 (6.57) 52.78 (7.08) NS pressure, mmHg Diastolic arterial 25.77 (4.52) 29.92 (5.32) NS pressure, mmHg (a) Data are the mean (SE). (b) NS, not significant (P 0.05). Table 2. Sensitivity, specificity, and predictive values of activin A, Hyp, and Xan concentrations and NRBC count at birth for early detection of IVH in preterm infants. Cutoff value Sensitivity, % Specificity, % Activin A 0.8 [micro] g/L 100 93 95% Cl 71-100 81-98 Hyp 0.8 mg/L 73 79 95% Cl 39-94 63-90 Xan 0.99 mg/L 82 79 95% Cl 48-97 63-90 NRBC count 1529 cells/m[m.sup.3] 73 83 95% Cl 39-94 69-93 Positive LR Negative LR AUC (a) Activin A 14.0 0 0.98(b) 95% Cl 0.9-1.0 Hyp 3.39 0.35 0.78 95% Cl 0.6-0.9 Xan 3.82 0.23 0.76 95% Cl 0.6-0.9 NRBC count 4.36 0.33 0.78 95% Cl 0.6-0.9 (a) AUC, area under the receiver operating characteristic curve. (b) P< 0.05 vs other areas under the curves.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Proteomics and Protein Markers|
|Author:||Florio, Pasquale; Perrone, Serafina; Luisi, Stefano; Vezzosi, Piero; Longini, Mariaangela; Marzocchi|
|Date:||Aug 1, 2006|
|Previous Article:||Predicting tissue HER2 status using serum HER2 levels in patients with metastatic breast cancer.|
|Next Article:||Increased plasma concentration of matrix metalloproteinase-7 in patients with coronary artery disease.|