Intraventricular hemorrhage in the preterm neonate: Timing and cerebral blood flow changes

Intraventricular hemorrhage in the preterm neonate: Timing and cerebral blood flow changes Serial cranial ultrasound studies, "Jxenon inhalation cerebral blood flow determinations, and risk factor analyses were performed in 31 preterm neonates. Contrast echocardiographic stffdies were additionally performed in 16 of these 31 infants. Sixty-one percent were found to have germinal matrix or intraventricular hemorrhage. Seventy-four percent of all hemorrhages were detected by the thirtieth postnatal hour. The patients were divided into three groups: earl)' GMH/IVll by the sixth postnatal hour (eight infants) interval GMll/IVll from 6 hours through 5 days (10), and no GMtt/IVH (12). Cerebral blood flow values at 6 postnatal hours were significantly lower for the earl)' GMII/IVtt group than for the no GMti/IVll group (P < 0.01). Progression of GMtt/IVH was observed only in those infants with early hemorrhage, and these infants had a significantly higher incidence of neonatal mortalit)~. Ventriculomegaly as determined by ultrasound studies was noted equally in infants with and without GMtl/IVtl (50%) and was not found to correlate with low cerebral blood flow. The patients with earl), hemorrhage were distinguishable by their need for more vigorous resuscitation at the time of birth and significantly higher ventilator settings during the first 36 postnatal hours, during which time the), also had higher values of PCo,r. At, equal incidence of patent ductus arteriosus was found across all of the groups. We propose that early GMII/IVII may be related to perinatal events and that the significant decrease in cerebral blood flow found in infants with earl)' GMII/IVtl is secondary to the presence of the hemorrhage itself. Progression of earl), GMH/i VII and new interval GMitflVH may be related to later neonatal events known to alter cerebral blood flow. (.I PEDIArR 104:419, 1984)

Laura R. Ment, M.D., Charles C. Duncan, M.D., Richard A. Ehrenkranz, M.D., Robert C. Lunge, Ph.D., Kenneth J. Taylor, M.D., Charles S. Kleinman, M.D., David T. Scott, Ph.D., James Sivo, B.S.R.T., and Patricia Gettner, R.N. New Haven, Conn.

Muctt HAS BEEN WRITTEN about intraventricular hemorrhage in the preterm neonate) -8 but little is understood about the pathophysiology of this process. To approach a better understanding of this problem in its early phases, we undertook a study of preterm infants with real-time From the Departments of Pediatrics. Neurology, Neurosurgery, Obstetrics and Gynecology, and Diagnostic Radiology, Yale University School of Medicine. Supported in part by the Yale University School of Medicine Fluid Fund, by a grant from the National Institutes of llealth (RR 00125). and by the Walter Scott Foundation. Newtown, Colin. Reprint requests: Laura R. Ment. M.D., Department of Pediatrics, Yale University School of Medicine, 333 Cedar St., New llaven. CT 06510.

echoencephalography (ECHO) and xenon cerebral blood flow studies, in addition to risk factor data analysis and neurologie evaluations. These factors were chosen for study in order to assess the timing of IVH, the evolution of CBF CBF~, CBF,,,~, GMII ICP IVH PDA

Cerebral blood flow (CBF ,lib,b,*~pb,r,+ CBE,r.ho~i,~,,,)/2 CBFh~,ho,hom~pho,:/CBF~o,,,b~m~pho,~ General matrix hemorrhage Intracranial pressure Intraventricular hemorrhage Patent ductus arteriosus

the lesion, and the development of ventricular enlargement in children with and without hemorrhage. The CBF studies were performed at the time of the initial cranial ECHO

The Journal of P E D I A T R I C S

419

420

Ment et al.

within the first 6 postnatal hours and on the fifth postnatal day, to assess the possible relationship between CBF and IVH. In addition, a subpopulation of the patients underwent contrast echoeardiography at the time of the CBF determinations in order to assess the status of the ductus arteriosus. Multiple risk factor analysis was carried out in order to relate the timing of the hemorrhage and changes in CBF to prenatal and postnatal events. PATIENTS Between October 1, 1982, and March 31, 1983, 31 infants with birth weight 600 to 1250 gm and no major congenital anomalies were admitted to the Yale-New Haven Hospital (YNHH) Newborn Special Care Unit (NBSCU). The study protocol was approved by the Yale University Human Investigations Committee, and informed parental consent for the study procedures were obtained by one of the principal investigators. All 31 infants were included in the evaluations. The mean birth weight of the study infants was 981 gm (range 610 to 1250 gm). Gestational age was assessed by the method of Ballard et al., 9 and the mean was 28.3 weeks (range 25 to 33 weeks). All infants were appropriate in size for gestational age, and none had an occipitofrontal head circumference less than the 10th percentile for gestational age. ~~ Seventeen patients were boys, and five were the products of multiple gestations. The patients received standard care according to the practices of the Y N H H NBSCU, and therapy regimens were not altered for the purpose of this study. METItODS Details of the gestation, labor, and delivery of all patients were obtained by chart review and interview of the attending physicians and nurses. Resuscitation scores were assigned to each infant depending on the nature of the delivery room resuscitation required by each (0, no resuscitation; 1, blow by 02; 2, bag and mask ventilation; 3, tracheal intubation; 4, intubation and ventilation). The standard neonatal neurologic examinations of the preterm infant described by Amiel-Tison" and by Saint-Anne Dargassies ~2 were performed in all study patients within the first 5 postnatal hours. Data collected in the immediate neonatal period included blood pressure determinations performed either through indwelling arterial catheter or by the Doppler method, any blood chemistry values, maximal and minimal values of Po2, Pco2, and pH. Transfusions and volume infusions (including time, rate, and composition of the fluid given) were also noted. All medications were recorded, and all significant events, including seizures, pneumothoraces, resuscitations, and surgical procedures, were recorded.

The Journal of Pediatrics March 1984

Echoencephaiography. Serial cranial ultrasound studies were performed at the bedside utilizing a portable realtime sector ultrasound machine (Advanced Technology Laboratories, Belleyiew, WA). Studies were performed within the first 6 postnatal hours to coincide with the first CBF study, and every 12 hours thereafter for the first 3 postnatal days. Subsequent studies were performed at least on the fourth, fifth, seventh, fourteenth, and twentieth postnatal days. Examinations were performed through the anterior fontanelle in coronal and sagittal projections and through the temporal squamosal bone in an axial projection using a 5 MHz transducer. All ultrasound studies were reviewed by observers unaware of the infant's clinical course and other study findings. The grading system for the hemorrhages detected was adapted from the CT scan grading system of Papile et al. ~3 as follows: grade 0, no germinal matrix hemorrhage or IVH; grade I, blood in the periventricular germinal matrix region, or GMH; grade II, blood within the lateral ventricles without ventricular dilation; grade III, blood within and distending the lateral ventricles; and grade IV, blood within the ventricles in addition to parenchymal extension. We divided the patient population into four groups for assessment: (1) early G M H / I V H , present at the time of the first ultrasound study; (2) interval G M H / I V H , detected after 6 hours of age through 5 postnatal days; (3) late G M H / I V H , detected after the fifth postnatal day, as defined by Perlman and Volpe~; and (4) no G M H / I V H . Ventricular dilation and the progression of ventricular size were noted from each study to the next. Excessive ventricular enlargement was defined as a ventricular to hemispheric ratio as determined in the axial plane of >35%. Hemorrhages were defined as stable when there had been no change on ultrasound studies for 3 consecutive days or longer. Infants with ECHO evidence for stable G M H / I V H and in whom excessive ventricular enlargement was noted on two consecutive studies underwent lumbar puncture for opening pressure as well as routine cell counts and sugar and protein analysis. Opening pressures of > 130 mm H20 were considered to be elevated ~s.~ and consistent, in patients with evidence for hemorrhage, with posthemorrhagic hydrocephalus. Infants without documented G M H / I V H who were similarly found to have ventriculomegaly on two successive ECHO studies also underwent lumbar puncture. In those infants with blood in the CSF and no ECHO evidence for G M H ] I V H , CT scans were performed for investigation of possible subarachnoid hemorrhage, as this could not be readily determined by ultrasound. Hemorrhages were considered to have progressed if (1) a GMH developed an intraventricular or parenchymal component, (2) an IVH developed a parenchymal compo-

Volume 104 Number 3

nent, or (3) a second site of GMH was noted in the opposite hemisphere from that with an existing G M H / IVH. Cerebral blood flow studies. The CBF determinations were performed as soon after the first cranial E C H O as possible, at a mean age of 6 postnatal hours (range 4 to 11 hours), and on the fifth postnatal day. During all studies, heart rate, respiratory rate, and temperature were recorded. At the time of the first CBF determinations, in all infants who had indwelling arterial catheters for medical care, samples were drawn for determination of arterial blood gases. At the time of the second CBF determination, blood for gas analysis was obtained either from indwelling arterial catheters, if these were medically indicated, or by capillary techniques. The experimental protocol for CBF determinations has been previously described. ~7The studies were performed at the infant's bedside. A portable gamma camera and computer for data acquisition and analysis (Ohio Nuclear, model Sigma 420) were used. The technique for these studies, as well as the mathematics, methodology, and limitations, have been previously reported. 182~ 133Xenon was administered by inhalation from a to-and-fro system containing Soda lime to remove expired CO2. Patients with endotracheal tubes received manual ventilation via a closed anesthesia bag system with an air pressure gauge, with the same oxygen concentrations, inspiratory and peak endexpiratory pressures, and rate as delivered by the ventilators. After 2 minutes, mechanical ventilation was resumed. Patients who were spontaneously breathing inhaled mXenon by mask connected to a flexible vinyl bag with the same oxygen concentration as was present in the infant's head box/incubator. Five mCi mXenon were administered to all patients, and the radiation dose to the lungs received during each CBF study was 240 mR. All studies were carried out in a large, well-ventilated room approved by the Yale Radiation Safety Committee. We report CBF .... and CBFio,~,h~m~phcr~cr~,~o values, as previously defined." Cardiac studies. Echocardiography was performed in 16 consecutive patients at age 6 postnatal hours to coincide with the first cranial ECHO and CBF determinations, and on the fifth postnatal day. M-mode echocardiographie evaluation of left ventricular and left atrial size, left atrial/aortic root ratio, and left ventricular ejection fraction were performed (Picker ECHO View 80 CI Cardiac Imager). Two-dimensional studies for visualizaton of the patent ductus arteriosus with umbilical artery contrast eehocardiography to determine left-to-right aortopulmonary shunting also performed (Advanced Technology Laboratory Mark III system). Ductai patency was inferred from the presence of L A / A o ratio >__1.3 or an increase of

I V H in the preterm neonate

4 21

20% since the first study, or ECHO contrast passing from the descending aorta to the pulmonary arterial circulation on contrast echocardiography. Statistics. As described above, infants were grouped for purposes of data analysis according to the presence and timing of G M H / I V H , as detected by echoencephalography. Dichotomous variables, such as the presence of mechanical ventilation at the time of the first CBF determination, were then analyzed with reference to the chi-square distribution. Continuous variables, such as birth weight, were analyzed using omnibus analyses of variance. Significant f tests were then further investigate d in a posteriori comparisons of the group means using the studentized range statistic in the Duncan Multiple Range Test. RESULTS Incidence of IVH: Timing and evolution of hemorrhages. Nineteen, or 61% of all study infants, were found to have G M H / I V H . When classified for the highest grade of G M H / I V H observed by ECHO, six infants were found to have grade I, six were thought tO have grade II, three had grade III, and four had grade IV hemorrhage. Seven of the 19 with G M H / I V H died, compared to two of the 12 patients with no known G M H / I V H . Eight infants were found to have evidence for G M H / IVH at the time of the first ultrasound: two infants with grade I, five with grade II, and one with grade III hemorrhage. Two of these infants died, and subsequent studies were not performed. Serial ECHOs in the remaining patients demonstrafed a progressive process in five, with three of those patients eventually developing grad e IV hemorrhage. Only one infant with early grade I hemorrhage experienced no change. In addition, six of the eight with early G M H / I V H died during the first 30 postnatal days. 9 Eleven infants were found to have evidence of new G M H / I V H after the time of the first ultrasound (Table I). None of these infants was noted to have progression of hemorrhage. The patient with GMH on day 14, or late G M H / I V H , was excluded from further data analysis. Two of the 10 with interval G M H / I V H died within the 9neonatal period. Seven of the 12 infants with G M H / I V H who survived longer than 5 postnatal days were found to have ventriculomegaly 3 to 14 days (mean 8 days) after stabilization of the G M H / I V H . Two of these patients were found to have increased ICP (140 and 160 mm H20, respectively, at the time of lumbar puncture. Four of the 10 patients without hemorrhage who survived were also found to have late ventriculomegaly (mean age I0 days). Lumbar puncture demonstrated no evidence for increasing ICP or bleeding in

422

Ment et al.

The Journal of Pediatrics March 1984

Table I. Incidence and timing of intraventricular hemorrhage

i

Time Hr 6 Hr 18

Hr 30 tlr 42 Day 3 Day 4 Day 5 Late IVtl Total No GMH/IVH

% Total population

Cumulative % total population

Cumulative % all GMH/IVH

8 2

26 6

26 32

42 53

4 0 2 1 I 1 19 12

13

45 45 51 54 58 61 6! 39

74 74 84 89 95 I00 100

n

I

6 3 3 3 39

Table II. Risk factor data

Group

n

Gastational age (wk)

Early Interval No GMIi/IvH

8 10 12

26.9 __. 1.7 27.6 + 2.0 30.1 + 1.8

Resuscitation score 3.2 2.6 1.7

Peak inspiratory I pressure (ram Hg) 29.4 __. 5.0 20.0 + 4.4 20.4 __- 5.4

Fio:=,.=*

Peak Bcoz* (trim Hg) -

Incidence of seizures* (%)

86.1 +-. 13.3 59.6 +-- 24.8 43.2 +-- 21.6

68.6 + 7.1 56.0 +- 7.8 40.9 + 10.5

75 20 8

Values expressedas mean +- SD. *First 36 postnatalhours.

any of these patients (CSF opening pressure was <70 mm HzO in all). Ventriculomegaly persisted in all of these patients during the remainder of the studies. Risk factor analysis. The infants were evaluated for the presence and absence of a number of risk factors according to G/VIH/IVH grouping. The infants with no G M H / I V H were found to have significantly higher gestational ages, compared with infants in the early and interval groups (P < 0 . 0 5 ) a n d significantly lower resuscitation scores (P < 0.05) (Table II). The infants with early G M H / I V H required significantly higher peak inspiratory pressures and maximal Fio2 values during the first 36 postnatal hours (P < 0.05 for both). Similarly, the early group had higher: peak Pcoz values (P < 0.05) and a greater incidence o f seizures (P < 0.05) during this same interval (Table II). There were no significant differences in birth weights, gender, Apgar scores, blood pressure, and incidence of neonatal transfer, caesarian section, breech presentation, exchange transfusions, or pneumothoraces among the three groups. Cerebral blood flow studies. Initial CBF studies were available for 28 infants; one patient with inteiwal G M H / IVH and two with no G M H / I V H did not undergo CBF studies. In all surviving infants with initial CBF studies, a second CBF determination was obtained on postnatal day

5; no infant without an initial CBF underwent a later study. At the time of the first CBF determination, there were no differences in the blood pressures or Po2 or Pco2 values for any Of the groups of infants (Table III). The pH in the infants with early G M H / I V H was significantly less than that in the other groups (P < 0.05). The CBF~ .... in the early G M H / I V H .group was significantly lower than ihat of the infants with no G M H / I V H (P < 0.01, Table IV). Seven infants had CBF~ values to one or both hemispheres of 10 ml/100 gm/min, or less~ Five were in the early G M H / I V H group, and two were in the interval group. Five of these seven died in the neonatal period. At the time for the second CBF determination, five of the patients with early G M H / I V H had died, comps/red to one with interval G M H / I V H and two with no G M H / IVH. There were no differences in the arterial blood gases or blood pressure values across any of the groups, and the CBFzmd, values for all of tfiese groups had increased from the first CBF determination, but the ~?elationship noted at the time of CBF~ remained unchanged (Table IV). The CBF~,~o in the patients with early G M H / I V H was significantly higher than in the interval and no G M H / I V H groups (P < 0.05). There was no significant difference in the CBF2,,,~o across any of the groups. In addition, there was no correlation between the development of ventricu-

Volume 104 Number 3

I V H in t h e preterm neonate

423

T a b l e IlL Cerebral blood flow study 1: Physiologic measurements

Early Interval No GMH/IVH

8 10 10

65.4 -+ 10.0 102.6 • 6.1 88.6 • 9.3

35.1 _+ 9.7 29.3 _+ 6.6 29.1 • 4.8

7.32 • 0.08* 7.36 • 0.06 7.42 • 0.06

BPmto,, (ram Hg)

BP#i,tots, (ram Hg)

41.0 • 6.0 42.0 • 7.I 48.0 __. 6.3

26.0 -4- 7.9 28.0 • 6.0 25.0 • 4.0

Values expressedas mean - SD. *Early less than no GMH/IVtt (P < 0.05).

Table IV. Cerebral blood flow data GMH-IVH

n

CBF, ~,~,o (ml/lO0 gm/rnin)

(ml/lO0 gm/min)

CB Fj ,~ao (ml/lO0 gm/min)

CB F2,~ao (ml/lO0 gm/min)

Early Interval No GMH/IVH

8 10 10

21.4 --- 12.6 27.8 • 9.9 41.9 • 14.3

29.2 • 7.3 (n = 3) 39.8 _+ 9.9 (n = 7) 48.4 ___14.9 (n = 8)

4.2 • 1.6 1.7 • 0.9 1.7 _+ 0.4

t.3 - 0.2 1.4 • 0.2 1.3 • 0.2

CBF2~,~

Data expressedas mean _+SD.

lomegaly and IVH or ventriculomegaly and CBF~ . . . . CBF~ ..... or either of the CBFra,~ovalues. Cardiac studies. Eehocardiographic evidence for PDA was present in four of five infants with early bleeding, five of seven with interval hemorrhage, and three of four with no G M H / I V H at the time of the first ultrasound and C S F study. In this small sample of patients, there were no correlations between the development of G M H / I V H and P D A or CBF and PDA. DISCUSSION Hemorrhage into the primordial tissues of the developing brain is a major problem in preterm neonates? 4.~-28 Although Bejar et al. 2~ noted a very high incidence of G M H / I V H on the first postnatal day and an overall incidence of almost 90% in infants of less than 34 Weeks' gestational age, most other investigators have reported an incidence of G M H / I V H o f 40 to 50% in similar populations. 5.'~3. 2528 Perlman and Volpe '4 noted that approximately 50% of all infants with hemorrhage were found to have this lesion on the first postnatal day; only 3% of their patients experienced hemorrhage after the fourth day. Donn and Strick z9 first reported progression of G M H / IVH, and both Perlman and Vo!pe~4and Shankaran et alY noted that in l0% to 20% of infants with G M H / I V H , progression of their hemorrhages will be demonstrated. Nutnerous studiesS.,~3.57-2~.3e-3~ have correlated, events known to alter CBF, such as hypercarbia, volume expansion, and pneumothoraces, with IVH.oCooke 39 a n d Milligan 4~ have reported increases in the CBF index in infants with hypercarbia and volume expansion who subsequently developed IVH~ Similarly, others ~'34 have noted increased

CBF velocity in neonates with seizures prior to the development of IVH. We reported marked alterations in CBF in neonates with IVH on the first postnatal day, although we were unable to correlate the timing of the hemorrhage With changes in flow. ~7 Other investigators have examined CBF after G M H / IVH. Bada et al. 4 and Bashiru et al. ~ have reported alterations in pulsatility index and CBF velocity, respectively, following G M H / I V H in preterm neonates, andthus postulated decreased CBF following hemorrhage. Similar~ ly, by means of positron emission tomography, Volpe et al. 42 demonstrated marked areas of ischemia in preterm infants following grade IV hemorrhage. These areas of ischemia were found in the hemispheres ipsilatera! and contralateral to the hemorrhage. Finally, using the nerOborn beagle puppy model, IVH neuropathologically similar to that in human preterm neonates can be produced by the acute onset of hypertension or hypercarbia or by hemorrhagic hypotension followed by rapid volume expansion. ~3-46 In this last model, CBF was increased to the germinal matrix but significantly decreased to cortical gray a n d white matter tissues 1 hour after the insult. 46 We have serially studied 3 1 p r e t e r m neonates; the incidence of G M H / I V H was 61%, and by the thirtieth postnatal hour, 74% of all hemorrhages had been detected. Only those patients with G M H / I V H at the time of the first ultrasound study at 6 postnatal hours were found to experience progression of hemorrhage. These patients had lower gestational ages than their peers without G M H / IVH. In addition, they required more vigorous delivery room resuscitation and h a d greater ventilatory Cequire" ments than those without G H M / I V H , and'they had higher Pco2 values in the first 36 postnatal hours. Six of eight

4 24

Ment et al.

infants with early G M H / I V H died within the first 30 postnatal days, compared to two of 11 infants with GMH/1VH detected at 18 postnatal hours or later (P < 0.05). Although, like Perlman and Folpe, ~4we report no statistically significant relation between time for the first G M H / I V H and the severity of the initial lesion, grade IV lesions developed only in those infants with early

CMIa/IVH. Furthermore, we have found very low CBF values at 6 postnatal hours in these same infants with early G M H / IVH. The CBF in the patients with interval G M H / I V H was also somewhat lower than in the group with no G M H / I V H , indicating possible disturbances in flow in these infants as well. As we have previously reported, the CBF ratios for the infants with hemorrhage are very different than those for the infants with no G M H / I V H . By the fifth postnatal day, the values for CBF have increased for all groups of infants, although they are still lower in the early G M H / I V H than in the other two groups. In addition, although various authors have commented on the relationship of the patent ductus arteriosus to IVH, 4~.4s we detected a high overall incidence of PDA in infants with or without G M H / I V H and no relationship to CBF, although all infants were not assessed for this factor. Profoundly low values of CBF in the immediate perinatal period have been reported to correlate with both diminished survival and poor neurodevelopmental outcome? 9-5~Older infants and children with retardation and hydrocephalus are also known to have low values for CBF? ~ As mentioned, six of our eight patients with early G M H / I V H and low CBF values died within the first 30 postnatal days. Finally, several investigators have serially studied infants with IVH at risk for posthemorrhagie hydrocephalus, and the incidence of ventriculomegaly or ultrasounddiagnosed enlargement of the ventricular system appears to be approximately 50%.52 "54 O f the infants with ventriculomegaly, one half to three fourths have beeri defined as having hydrocephalus with both increased ventricular size and elevated intracranial pressure, s2"53 We detected no difference in the incidence of ventriculomegaly in the patients with G M H / I V H compared with those without. However, none of the latter infants was found to have evidence for increased ICP, compared with two of the infants with IVH, in whom this problem was transient and was treated with repeated lumbar punctures. Although ventriculomegaly is believed by some to be associated with developmental and neurologic problems,5S and thus to suggest ischemia to the developing brain, Fitzhardinge et al. 56 demonstrated the widespread prevalence of ventriculomegaly in infants with IVH and found no correlation

The Journal o f Pediatrics 2llarch 1984

between this finding and neurodevelopmental outcome. Similarly, we were unable to document any association between low neonatal CBF and ventriculomegaly in infants with and without G M H / I V H . We thank Constance Hanrahan, R.N., Michael Planckey, R.T., and Diane Erica, R.T., for assistance. REI~ERENCES 1. Volpe J: Neonatal periventricular hemorrhage: Past, present and future. J PEDtA'rR92:693, 1978. 2. Hambleton G, Wigglesworth JS: Origin of intraventricular hemorrhage in the preterm infant. Arch Dis Child 51:651, 1976. 3. Lou 11C: Perinatal hypoxic-ischemiebrain damage and intraventricular hemorrhage. Arch Neurol 37:585, 1980. 4. Bada ttS, Hajiar MS, Chan C, Sumner DS: Noninvasive diagnosis of neonatal asphyxia and intraventricular hemorrhage by Doppler ultrasound. J PEDIATR95:775, 1979. 5. Dykes FD, Lazzara A, Ahmann P, et al: lntraventricular hemorrhage: A prospective evaluation of etiopathogenesis. Pediatrics 66:42, 1980. 6. Volpe J J: Neurology of the newborn. Philadelphia, 1981, WB Saunders. 7. Donat JF, Okazaki H, Kleinberg F, et al: Intraveatricular hemorrhages in full-term and premature infants. Mayo Clin Proc 53:437, 1978. 8. Wigglesworth JS, Pape KE: An integrated model for haemorrhagic and ischaemic lesions in the newborn brain. Early tlum Dev 2:179, 1978. 9. Ballard JL, Kazmaier K, Driver M: A simplified assessment of gestational age. Pediatr Res 11:374, 1977. 10. Lubchenco LO, ltansman C, Boyd E: Intrauterine growth: Weight, length and head circumferenceas estimated from live births at gestational ages from 26 to 42 weeks. Pediatrics 37:403, 1966. 11. Amiel-Tison C: Neurologie evaluation of the maturity of newborn infants. Arch Dis Child 43:89, 1968. 12. Saint-Anne Dargassies S: Neurodevelopmental symptoms during the first year of life. I. Essential landmarks for each key-age. 11. Practical examination and the application of this assessment method to the abnormal infant. Dev Med Child Neurol 14:235, 1972. 13. Papile LS, Burstein J, Burstein R, et al: Incidence and evolution of the subependymal intraventricular hemorrhage: A study of infants with weights less than 1500 grams. J pEmAXR92:529, 1978. 14. Perlman JM, Volpe J J: Cerebral blood flow velocity in relation to intraventricular hemorrhage in the premature newborn infant. J Pzt~lAa'R100:956, 1982. 15. Vidyasagar D, Raju TN, Chiang J: Clinical significance of monitoring anterior fontanel pressure in sich neonates and infants. Pediatrics 62:996, 1972. 16. Salmon JH, ttajjar W, Bada HS: The fonogram: A noninvasive intracranial pressure monitor. Pediatrics 60:721, 1977. 17. Ment LR, Ehrenkranz RA, Lange RC, Rothstein PT, Duncan CC: Alterations in cerebral blood flowin preterm infants with intraventricular hemorrhage. Pediatrics 68:763, 1981. 18. Wyper DJ, Lennox GA, Rowan JO: Two minute slope inhalation technique for cerebral blood flow measurements in man. I. Method. J Neurol Neurosurg Psychiatry 39:141, 1976.

Volunw 104 Number 3

19. Wyper DJ, Lennox GA, Rowan JO: Two minute slope inhalation technique for cerebral blood flow measurement in man. 11. Clinical appraisal. J Neurol Neurosurg Pychiatry 39:147, 1976. 20. Wyper D J, Rowan JO: The construction and use of homograms for cerebral blood flow calculations using a "3Xe inhalation technique. Phys Med Biol 21:406, 1976. 21. Blauenstein UW, Halsy JH Jr, Wilson M, et al: mXenon inhalation method. Analysis of reproducibility: Some of its physiological implications. Stroke 8:92, 1977. 22. Clark CE, Clyman RI, Roth RS, Sniderman SH, Lane B, Ballard RA: Risk factor analysis of intraventricular hemorrhage in low-birth-weight infants. J PEDIATR 99:625, 1981. 23. Shankaran S, Slovis TL, Bedard MP, Poland RL: Sonographic classification of intracranial hemorrhage: A prognostic indicator of mortality, morbidity, and short-term ncurologic outcome. J PEDIATR 100:469, 1982. 24. Bejar R, Curbelo V, Coen R, Leopold G, James |I, Gluck L: Diagnosis and follow-up of intraventricular and intracerebral hemorrhages by ultrasound studies of infant's brains through the fontanelles and sutures. Pediatrics 66:661, 1980. 25. Ahmann PA, Lazzara A, Dykes FD, Brann AW Jr, Schwartz JF: Intraventricular hemorrhage in the high-risk preterm infant: Incidence and outcome. Ann Neurol 7:118, 1980. 26. Thornburn R J, Lipscomb AP, Stewart AL, Reynolds EOR, Hope PL, Pape KE: Prcdiction of death anad major handicap in very preterm infants by brain ultrasound. Lancet l:l 119, 1981. 27. Hope PL, Rhorhurn R J, Stewart AL, Reynolds EOR: Timing and antecedents of periventricular hemorrhage in very preterm infants. Second Special Ross Laboratories Conference on Perinatal Intracranial Hemorrhage. Columbus, Ohio, 1982, Ross Laboratories, pp 78-101. 28. McDonald MM, Koops BL, Johnson ML, Guggenheim MA, Rumack CM, Mitchell SA, ttathaway WE: Timing and etiology of intraeranial hemorrhage in the newborn. Second Special Ross Laboratories Conference on Perinatal Intracranial Hemorrhage. Columbus, Ohio, 1982, Ross Laboratories, pp 211-232. 29. Donn SM, Strick K J: Neonatal germinal matrix hemorrhage: Evidence of a progressive lesion. J PEDIATR 99:459, 1981. 30. Volpe J J: Neonatal intracranial hemorrhage: Pathophysiology, neuropathology and clinical features. Clin Perinatol 4:77, 1977. 31. Lipscomb A, Thornburn R J, Reynolds EOR, et al: Pneumothorax and cerebral hemorrhage in preterm infants. Lancet 1:414, 1981. 32. Levene MI, Fawer CL, Lamont RF: Risk factors in the development of intraventricular haemorrhage in the preterm neonate. Arch Dis Child 57:410, 1982. 33. Goldberg RN, Chung D, Goldman SL, Bancalari E: The association of rapid volume expansion and intraventricular hemorrhage in the preterm infant. J PEDIA1"R 96:1060, 1979. 34. tlill A, Volpe J J: Seizures, hypoxic-ischemie brain injury and intraventricular hemorrhage in the newborn. Ann Neurol 10:109, 1981. 35. llill A, Perlman MB, Volpe J J: Relationship of pneumothorax to occurrence of intraventricular hemorrhage in the premature newborn. Pediatrics 69:144, 1982. 36. Kenny JD, Garcia-Prats JA, Hilliard JL, et al: llypercarbia

I V H in the preterm neonate

37.

38. 39. 40. 41.

42.

43.

44.

45. 46.

47.

48.

49. 50.

51.

52.

53. 54.

55.

56.

425

at birth: A possible role in the pathogenesis of intraventricular hemorrhage. Pediatrics 62:465, 1978. Fujimura M, Salisbury DM, Robinson RO, et al: Clinical events relating to intraventricular hemorrhage in the newborn. Arch Dis Child 54:409, 1979. Guggenheim MA, Young J, Murphy J: Risk factors in germinal matrix hemorrhage. Ann Neurol 8:225, 1980. Cooke RWI: Autoregulation and intraventricular hemorrhage, letter. Lancet 1:1197, 1980. Milligan DWA: Failure of autoregulation and intraventricular hemorrhage in preterm infants. Lancet !:896, 1980. Bashiru M, Russell J, Bada HS, Menke JA, Miles R, Sumner D: Noninvasive detection of neonatal cerebrovascular disease. Bruit 4:42, 1982. Volpe J J, Perlman Jb,l, llerscovitch P, Raichle ME: Position emission tomography in the assessment of RCBF in the newborn. Ann Neurol 12:225, 1982. Goddard J, Lewis RM, Armstrong DL, Zeller RS: Moderate, rapidly induced hypertension as a cause of intraventricular hemorrhage in the newborn beagle model. J PEDIATR96:1057, 1980. Goddard-Finegold J, Armstrong D, Zeller RS: Intraventrieular hemorrhage following volume expansion after hypovolemic hypotension in the newborn beagle. J PEDIATR 100:796, 1982. Goddard J, Lewis RM, Alcala 1t, Zeller RS: Intraventricular hemorrhage: An animal model. Biol Neonate 37:39, 1980. Merit LR, Stewart WB, Duncan CC, et al: Beagle puppy model of intraventricular hemorrhage. J Neurosurg 57:219, 1982. Ellison PH, Denny J: Cerebral blood flow (CBF) by Doppler velocimeter in preterm infants with and without PDA. Ann Neurol 12:223, 1982. Gersony WM, Peckham G J, Ellison RC, Miettinen OS, Nadas AS: Effects of Indomethacin in premature infants with patent duetus arteriosus.: Results of a national collaborative study. J PEDIATR 102:895, 1983. Lou IIC, Skov H, Pedersen H: Low cerebral blood flow: A risk factor in the neonate. J PEDIATR 95:606, 1979. Merit LR, Scott DT, Lange RC, Ehrenkranz RA, Duncan CC, Warshaw JB: Postpartum perfusion of the preterm brain: Relationship to neurodevelopmental outcome. Child's Brain 10:266, 1983. Baird IIW III, Garfunkel JM: A method for the measurement of cerebral blood flow in infants and children. J PEDIATR 42:570, 1953. Allan WC, Holt P J, Sawyer LR, Tito AM, Meade S: Ventricular dilation after neonatal periventricular-intraventricular hemorrhage. Am J Dis Child 136:589, 1982. 11ill A, Volpe J J: Normal pressure hydrocephalus in the newborn. Pediatrics 68:623, 1981. Levene ML, Starte DR: A Ionitudinal study of posthaemorrhagic ventricular dilatation in the newborn. Arch Dis Child 56:905, 198 I. Palmer P, Dubowitz LMS, Levene MI, Dubowitz V: Developmental and neurologic progress of preterm infants with intraventricular hemorrhage and ventricular dilatation. Arch Dis Child 57:748, 1982. Fitzhardinge PM, Flodmark O, Fitz CR, Ashby S: The prognostic value of computed tomography of the brain in asphyxiated premature infants. J PEDIA'rR 100:476, 1982.