2Ultrasound Measurements of Th

REVIEW ARTICLE

Ultrasound measurements of the lateral ventricles in neonates: why, howand when? A systematic reviewMargaretha J Brouwer1, Linda S de Vries1, Lou Pistorius2, Karin J Rademaker1, Floris Groenendaal1, Manon JNL Benders ([email protected])1

1.Department of Neonatology, Wilhelmina Childrens Hospital, University Medical Centre Utrecht, The Netherlands2.Department of Obstetrics, Wilhelmina Childrens Hospital, University Medical Centre Utrecht, The Netherlands

KeywordsCranial ultrasound, Lateral ventricles, Neonate,Post-haemorrhagic ventricular dilatation, Review

CorrespondenceManon JNL Benders, Wilhelmina Childrens Hospital,University Medical Centre Utrecht,Post box 85090, 3508 AB Utrecht,The Netherlands.Tel: +31-88-7554992 |Fax: +31-88-7555320 |Email: [email protected]

Received18 January 2010; revised 29 March 2010;accepted 30 March 2010.

DOI:10.1111/j.1651-2227.2010.01830.x

ABSTRACTGerminal matrix-intraventricular haemorrhage and subsequent post-haemorrhagic ventri-

cular dilatation (PHVD) are frequently encountered complications in preterm neonates. As

progressive dilatation of the lateral ventricles may be associated with elevated intracranial

pressure, ultrasound measurements of ventricular size play a major role in the evaluation

of neonates at risk of ventricular dilatation as well as in assessing the effect of intervention

for PHVD. A systematic search was carried out in Medline and Embase to identify neonatal

and foetal ultrasound studies on lateral ventricular size. This review presents an overview

of the available data concerning neonatal reference values for lateral ventricular size, the

influence of gender, ventricular asymmetry and the effect of the mode of delivery on the

phenomenon of ventricular reopening following birth.

Conclusion: Serial cranial ultrasound measurements of the lateral ventricles play a key role inthe early recognition and therapeutic evaluation of post-haemorrhagic ventricular dilation and canbe of prognostic value in neonates with ventricular dilatation.

INTRODUCTIONGerminal matrix-intraventricular haemorrhage (GMH-IVH) is a frequently encountered complication in the pre-term neonate. The incidence of GMH-IVH is inverselyrelated to gestational age (GA) (1) and despite a gradualdecline during the last decades, still 2025% among verylow birth weight (VLBW) infants (2,3). For term neonatesincidence rates around 1% have been reported, the haemor-rhage being mainly restricted to the germinal matrix (4).

Especially after a severe haemorrhage [GMH-IVH gradeIIb III according to Levene et al.(5) and de Vries et al.(6)or grade III IV according to Papile et al.(7)] the risk ofdeveloping post-haemorrhagic ventricular dilatation(PHVD) is considerable. PHVD defined as ventricularenlargement above the 97th centile for GA (8) is recog-nized in about one-third of infants with GMH-IVH (9).

Progressive ventricular dilatation can be associated withelevated intracranial pressure (ICP) (10). Raised ICP maylead to alterations in cerebral hemodynamics and tissuedamage (3), and was found to correlate with delayed myeli-nation and neurodevelopment in infantile hydrocephalus(11). Intervention with drainage of cerebrospinal fluid(CSF) improves cerebral perfusion and oxygenation in neo-nates with PHVD (12,13) and may thus prevent furtherdamage to the brain tissue. This underlines the need forsequential cranial ultrasonography following a diagnosis ofGMH-IVH to timely diagnose subsequent PHVD. In thisreview, an outline will be given of the role of sonographicevaluation of cerebral ventricular size in neonates at risk forPHVD.

METHODSA systematic literature search was performed in Medlineand Embase in December 2009 to identify foetal and neona-tal ultrasound studies on lateral ventricular size (detailedsearch strategy available upon request). The papers withabstracts meeting the predefined in- and exclusion criteriawere evaluated for potential inclusion in the review. Thereferences of these studies were screened for additional rele-vant studies. Foetal and neonatal series were included iffrontal and or occipital horn dimensions had been investi-gated. Studies on intra- and interobserver variability, ven-tricular asymmetry, the influence of gender or the effect of

Abbreviations

CSF, cerebrospinal fluid; cUS, cranial ultrasound; ELBW, extre-mely low birth weight; GA, gestational age; GMH-IVH, germinalmatrix-intraventricular haemorrhage; ICP, intracranial pressure;PHVD, post-haemorrhagic ventricular dilatation; VLBW, verylow birth weight; VP, ventriculoperitoneal.Ventricular parameters: AHW, anterior horn width; FHR, frontalhorn ratio; HW, hemispheric width; TOD, thalamo-occipital dis-tance; VA, ventricular axis; VH, ventricular height; VI, ventricularindex.

Acta Pdiatrica ISSN 08035253

1298 2010 The Author(s)/Journal Compilation 2010 Foundation Acta Pdiatrica/Acta Pdiatrica 2010 99, pp. 12981306

the mode of delivery on ventricular size were included aswell. Applied language restrictions were English.

THE IMPORTANCE OF EVALUATION OF VENTRICULAR SIZE INNEONATES AT RISK OF PHVDRole of cUS in diagnosis and evaluation of PHVDTo diagnose and evaluate PHVD, measurements ofventricular size on cranial ultrasound (cUS) are superior toany subjective assessments of elevated ICP, like a tense fon-tanel, sunset phenomena of the eyes or measurements of thehead circumference (8,1417).

In clinical practice, the ventricular index (VI) of thelateral ventricles as measured on cUS (Fig. 1A) is most fre-quently used for monitoring ventricular size followingGMH-IVH. In some studies (1820), measurements of theVI are expressed as a ratio to the hemispheric width (HW).The use of a ratio is however considered to be less useful fordetecting ventricular dilatation according to some articles(8,21). Absolute measurements of the VI are therefore usedin our centre.

Although a rise in ICP may be accompanied by anincrease in VI (10), often the VI only starts to enlarge inmore severe hydrocephalus and consequently may fail toidentify neonates with mild dilatation (22,23). The first signof an increase in ICP is more frequently a change in ventric-ular shape with rounding of the frontal horns and anincrease in anterior horn width (AHW, Fig. 1A) (15,22) aphenomenon referred to as ballooning(24) (Fig. 2). Hence,the AHW has been suggested to be a more sensitive markerfor early or mild ventricular enlargement than the VI(22,23).

The depth of the occipital horn of the lateral ventricle[thalamo-occipital distance (TOD), Fig. 1B] has also beenrecommended as a valuable addition for the evaluation ofventricular size following GMH-IVH. Occipital hornenlargement is often visible before any increase in frontalhorn dimensions (19,2530). The occipital horns are usuallydilated to a greater extent than the frontal horns and mayrepresent the only site of ventricular dilatation (19,2527,30). Isolated occipital horn dilatation is mainly recog-nized in extremely low birth weight (ELBW) infants with,but also without GHM-IVH. Whether it needs intervention,even if the VI and AHW are normal, or whether it is justi-fied to wait, still has to be evaluated.

The occipital horns, as well as the atria, also appear todilate prior to other parts of the lateral ventricles in earlycongenital hydrocephalus (31,32). The underlying mecha-nism of this phenomenon has not been identified so far.Sauerbrei et al.(22) observed that in neonates with GMH-IVH, small intraventricular collections of blood tended toaccumulate in the occipital horns while large collections ofblood were distributed uniformly throughout the ventricles.Whether this finding plays a role, remains speculative.

Visualizing the occipital horn may however be challeng-ing however and more prone to measurement errors(28,30,33). Nevertheless, non-visualization of the occipital

A

B

Figure 1 Ventricular parameters measured in the (A) coronal and (B) sagittalplanes by cranial ultrasonography. AHW = anterior horn width, FHR = frontal

horn ratio, HW = hemispheric width, TOD = thalamo-occipital distance,

VA = ventricular axis, VH = ventricular height, VI = ventricular index.

Figure 2 Ventricular ballooning on cranial ultrasound in a 13-day old pretermneonate (gestational age 27+5 weeks) with bilateral intraventricular haemor-

rhage grade IIb and post-haemorrhagic ventricular dilatation. Both the anterior

horn width (AHW) and ventricular index (VI) were significantly enlarged.

Brouwer et al. Ultrasound measurements of the lateral ventricles

2010 The Author(s)/Journal Compilation 2010 Foundation Acta Pdiatrica/Acta Pdiatrica 2010 99, pp. 12981306 1299

horn in a neonate with an easily accessible fontanel is animportant negative finding (28).

The clinical value of measurements of the third andfourth ventricle is unclear. When combined with the dimen-sions of the lateral ventricles, assessment of the size of thethird and fourth ventricle size may help in differentiatingbetween communicating and non-communicating hydro-cephalus (34). Enlarged third and fourth ventricular sizemay also contribute to distinguishing PHVD from ex-vacuodilatation, which has important therapeutic implications(35). The observation of isolated dilatation of the thirdand or fourth ventricle can be associated with posterior fos-sa haemorrhage and may provide the clue to the properdiagnosis. Dimensions of both ventricles are, however, noteasy to determine on cUS and reference values are scarce(20,34). The third ventricle is usually not visible in the coro-nal plane and must be dilated to perform measurements(36), while the fourth ventricle has a quite complex shaperesulting in a higher interobserver variability (34). Thereforeevaluation of third- and fourth ventricular size is notincluded in this review.

Ventricular size and timing of treatmentAbout 35% of the neonates who develop PHVD requiresome form of intervention (9). The decision whether to initi-ate treatment or not tends to be based on measurements ofventricular size.

Cerebrospinal fluid drainage has a central role in thetreatment of PHVD. It can be achieved by repeat lumbarpunctures, ventricular taps or the placement of a subcuta-neous reservoir, subgaleal shunt or ventriculoperitoneal(VP) shunt. Optimal timing for CSF drainage in infantswith PHVD is still a matter of debate (35). It has beendescribed that when the VI in neonates with progressiveventricular enlargement exceeds 4 mm above the 97th cen-tile for GA (8), PHVD carries a poor prognosis with about4060% of the children being shunt dependent, and over60% of the infants being deceased or either disabled orimpaired at the age of 1 year (3739). Whether early inter-vention decreases the need for shunting and improveslong-term neurodevelopmental outcome is still uncertain(9,3840). In a retrospective study in five Dutch neonatalintensive care units, low threshold intervention initiatedwhen the VI crossed the p97 line was associated with asignificant reduced risk of VP-shunting and a lower num-ber of infants with a moderate or severe handicap com-pared with high threshold intervention started aftercrossing the p97 + 4 mm line (9). A randomized multicen-tre intervention study (Early versus Late Ventricular Inter-vention Study, ELVIS) is carried out at the moment toexplore whether there is a beneficial role for low comparedwith high-threshold intervention in neonates with PHVDwith regard to both the risk of VP-shunting and neuro-developmental outcome.

When treatment for PHVD is initiated, serial cUS mea-surements of the lateral ventricles still play an important role.Monitoring ventricular size enables us to assess the effect ofthe intervention and the need for other therapeutic options.

Relevance of ventricular dilatation for long-term outcomeEvaluation of ventricular size is also of prognostic value. Inaddition to concomitant mortality and shunt dependency,PHVD is often associated with impaired neurodevelopmen-tal outcome in preterm neonates (9,3740). Several studieshave demonstrated that ventricular dilatation is predictiveof cognitive impairment (41,42) as well as cerebral palsy(41,43) in (V)LBW infants. Often though, it is hard to distin-guish the mechanisms underlying the origin of ventriculo-megaly either post-haemorrhagic resulting from increasedICP or atrophic resulting from white matter loss (or a com-bination) what makes it difficult to judge the impact ofmild or moderate PHVD. The extent of parenchymalinvolvement and associated white matter injury may be themain predictor for neurodevelopmental outcome in infantswith PHVD, especially with regard to motor outcome(6,40,44).

Importance of accurate reference rangesWith a better understanding of normal and abnormal size ofthe neonatal lateral ventricles, it may be possible to improveboth the diagnostic and therapeutic evaluation of neonateswith GMH-IVH PHVD. Infants with ventriculomegaly athigh risk of neurodevelopmental impairment may be identi-fied with greater accuracy.

An overview on neonatal reference ranges for the lateralventricles, including several variables influencing those val-ues, will be provided in the following sections.

(AB) NORMAL VENTRICULAR SIZE ON CUS: NEONATAL REFERENCEVALUESSix studies (8,15,20,21,30,34) which met the selection crite-ria, reported cross-sectional reference ranges for lateral ven-tricular size on cUS in neonates (Table 1). Data areavailable for both term and preterm infants, although refer-ence values for ELBW infants are scarce and based on smallnumbers in most studies.

Considerable variation exists between the studies withrespect to sample size, patient characteristics, timing of thefirst cUS and measured parameters. In the study of Levene(8) ultrasound measurements of the VI were performedthrough the temporoparietal bone, which correlated wellwith measurements through the anterior fontanel accordingto the author.

Most commonly measured parametersSeveral dimensions of the lateral ventricles have been inves-tigated. Levene (8) was the first to publish a reference curvefor the VI the distance between the falx and the lateralwall of the anterior horn in the coronal plane at the level ofthe third ventricle (Fig. 1A). The VI has also been expressedas a ratio to the width of the corresponding hemisphere,referred to as the frontal horn ratio (FHR, Fig. 1A) (20).Because ratios are considered to be less useful for monitor-ing ventricular size (8,21), as outlined before, the FHR is leftout of consideration. Another parameter that can be mea-sured in the coronal plane is the maximal diagonal width of

Ultrasound measurements of the lateral ventricles Brouwer et al.


the anterior horn (AHW, Fig. 1A) (34). A growth curve hasalso been published for the longitudinal axis of anteriorhorn [ventricular axis (VA), Fig. 1A] (15). Because thisparameter is not frequently used in clinical practice anddoes not have any apparent advantage over the VI or AHW,it will not be discussed in this review either.

In the parasagittal plane, the ventricular height (VH)adjacent to the foramen of Monro can be determined, aswell as the thalamo-occipital distance the distancebetween the outermost point of the thalamus at its junctionwith the choroid plexus, to the outermost part of the occipi-tal horn (Fig. 1B). While for the VH no reference values areavailable, data on occipital horn size were limited to pre-term neonates (30,34) till Sondhi et al.(20) quite recentlyreported reference ranges for the TOD in both term and pre-term infants.

What is (ab)normal ventricular size?Ventricular Index. In addition to Levene (8), Liaoet al.(15) also published a nomogram for the VI in new-borns (Fig. 3A). Both demonstrated that the upper limitfor the VI depends on GA. The reference values of thesestudies correlate well for term neonates, while those forpreterm newborns show more variation. This may beattributable to the fact that the lower the post-menstrualage the fewer infants were included, making the resultsmore prone to variation. For premature newborns below26 weeks of gestation, who are most at risk of developingGMH-IVH and subsequent PHVD, only the study ofLiao et al.(15) is applicable. Therefore more data areneeded for ELBW infants.

Anterior Horn Width. Whether normal values for the widthof the anterior horn correlate with GA is controversial.While several authors (15,21,34) reported no or just

minimal change in AHW with GA, an evident increase inAHW was demonstrated with ongoing maturity by Sondhiet al.(20) (Fig. 3B). Remarkable is the small standard devia-tion in the last study in comparison with previously pub-lished data. Whether this can be completely explained bythe large sample size and the fact that all neonates wereexamined on the third day post-partum in Sondhis study(20), remains uncertain.

In the majority of neonates, the AHW is less than 3 mm(15,20,21,34). The clinical significance of an AHW above3 mm in the absence of GMH-IVH is not clear. An AHWbetween 3 and 5 mm was not associated with neurodevel-opmental impairment at a single follow up visit in a smallgroup of 13 infants within the first year of life (21). Valuesexceeding 6 mm, however, are associated with ventricularballooning and suggest the need for treatment according toGovaert et al.(24)

Thalamo-Occipital Distance. No explicit conclusions canbe drawn from data on occipital horn size. Controversyalso exists whether normal values of occipital horn sizedepend on GA. Furthermore, the variation in reported ref-erence curves is considerable (Fig. 3C). Reeder et al.(30)compared occipital horn measurements in normal prema-ture infants and in premature neonates with intracranialhaemorrhage or neural tube defects and concluded that,regardless of the degree of prematurity, an occipital hornlength exceeding 16 mm suggests intracranial pathology inpreterms. Davies et al.(34) reported much higher referencevalues for the TOD in preterm neonates, with an upperlimit of 24.7 mm. Neonates with a GMH-IVH grade I IIwere, however, not excluded in the latter study, which mayhave resulted in higher values. A smaller occipital hornsize was measured in premature newborns by Sondhiet al.(20) Again the small range in this study is remarkable

Table 1 Cross-sectional studies on lateral ventricular size in preterm and term neonates performed within 1 week post-partum

StudySamplesize

GA(weeks) Exclusion criteria

First cUSpost-partum

Followup cUS

FrequencycUS

Method ofscanning

Ventricularparameters

Davies et al.(34)

(2000)

120 23

and does not agree with the other studies (30,34) and ourown experience.

Reliability of ventricular measurementsTo develop reliable reference ranges for clinical practice,ventricular measurements need to be reproducible for

subsequent measurements with good interobserver agree-ment. The few studies (8,20,34) which reported data onintra- and interobserver variability for several ventricularparameters are reassuring. For all above mentioned ventric-ular dimensions (VI, AHW and TOD) good intra- (8,34)and interobserver (8,20,34) correlation could be achieved.

7

8

9

10

11

12

13

14

15

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Post-menstrual age (weeks)

Ven

tric

ula

r in

dex

(m

m)

Levene (1981) Liao et al. (1986)

1

0

1

2

3

4

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


Ant

erio

r hor

n w

idth

(mm)

Perry et al. (1985) Liao et al. (1986) Davies et al. (2000) Sondhi et al. (2008)

0

5

10

15

20

25

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42


Thal

amo-

occi

pita

l dis

tanc

e (m

m)

Reeder et al. (1983) Davies et al. (2000) Sondhi et al. (2008)

A

B

C

Figure 3 Overview of the reference curves for the (A) ventricular index, (B) anterior horn width and (C) thalamo-occipital distance according to Davies et al. (34)(regression line with 95% confidence interval), Levene (8) (p3, p50, p97), Liao et al. (15) (mean 2 SD), Perry et al. (21) (p97), Reeder et al. (30) (mean 2

SD) and Sondhi et al. (20) (p5, p50, p95). Adapted from Davies et al. (34), Levene (8), Liao et al. (15) and Sondhi et al. (20).



With respect to assessment of occipital horn size, it has beenour experience that reproducible measurements of the TODare harder to obtain in some neonates because of difficultiesin visualizing the occipital horn and obliquity of the trans-ducer.

Subjective assessment of ventricular size is lessreproducible, especially in case of ventricular enlargement(45,46).

The use of cross-sectional data for follow upCompared with term-born infants, preterm infants tend tohave increased volumes of lateral ventricles and CSF onMRI at term-equivalent age (4749). This phenomenon hasbeen demonstrated during infancy (50,51) and adolescence(52) as well and may be due to cerebral atrophy. Hence it isquestionable whether cross-sectional reference ranges forventricular size can be used for the follow up of prematureneonates, especially in neonates with PHVD. Longitudinaldata on neonatal ventricular size are nevertheless scarceand only available for the VI and AHW. While an increasein VI was demonstrated in both preterm and term infantspostnatally (8,15), the AHW remained stable in preterminfants during the first 6 weeks of life (15).

VENTRICULAR SIZE: DOES GENDER MATTER?In foetuses ventricular size seems to be correlated to gender.Normal male foetuses appear to have slightly larger atriathan female foetuses throughout the second and third tri-mester (5356). Although statistically significant, reportedmean differences are small, ranging from 0.1 to 0.6 mm.

Less is known about the relationship between gender andventricular size in neonates. A good correlation has beenobserved between male and female newborns for the FHR(20), AHW (20,34) and TOD (20).

Volumetric MRI-measurements demonstrated largermean ventricular volumes in boys during the first 6 years oflife (57). The difference disappeared however when cor-rected for intracranial volume. No difference in ventricularvolume between both sexes could be demonstrated with fur-ther growth (51,57,58).

According to these data gender seems to play only aminor role in determining neonatal ventricular size. Maleneonates may have slightly larger ventricles than femaleneonates, but if present, the difference does not seem to beclinically significant.

ASYMMETRY OF THE LATERAL VENTRICLESThe phenomenon of lateral ventricular asymmetry on cUSwas first described in the 1980s (59). Since then it wasdemonstrated by others as well (30,33,34,60; 6164 in Sup-porting Information). Reported prevalence rates range from12 to 77% and this wide range was partly explained by thedefinition which was used.

When asymmetry is present, the left ventricle is usuallylarger than the right ventricle.(33,34,60; 6163 in Support-ing Information). The same trend has been noted in the

foetus (65 in Supporting Information). Volumetric cUS-and MRI measurements in neonates (6670 in SupportingInformation) as well as children and adolescents (58) alsoshowed that left ventricular volumes tend to be larger thanright ventricular volumes. Asymmetry tends to be more pro-nounced in the occipital horns than in the anterior part ofthe lateral ventricle (33; 62,64 in Supporting Information)and side to side differences exceeding 5 mm have beendescribed (34). The frontal horns may show some asymme-try, but reported differences in left- and right frontal horndimensions are small and unlikely to be of clinical relevance(20,34).

The clinical significance of ventricular asymmetry is notclear. Observations in some studies (30,33; 61 in Support-ing Information) suggest that there might be a relationshipwith intracranial pathology. A higher percentage ofGMH-IVH was found in neonates with asymmetrical ven-tricles raising the possibility that intracranial haemorrhage(in utero) may lead to ventricular asymmetry in someinfants (33; 61 in Supporting Information). However, theGMH-IVH was more commonly seen on the side of thesmaller ventricle (61 in Supporting Information) and themajority of neonates with asymmetry did not have anyevidence of a haemorrhage (33; 61 in Supporting Infor-mation). Other investigators also demonstrated that asym-metry is not necessarily a pathological entity, because ofGMH-IVH or impaired liquor drainage, but exists in thenormal population and may represent normal physiologi-cal variation (63,64 in Supporting Information). This isconsistent with observations in utero where asymmetry ofthe lateral ventricles also was observed, although less fre-quently, and was considered to be a normal variant afterexcluding underlying pathology (65 in Supporting Infor-mation).

Ventricular asymmetry may be influenced by genetic fac-tors or environmental events that occur during brain growth(71 in Supporting Information). It has also been reportedthat in case of asymmetry, the enlarged ventricle is oftenassociated with a larger ipsilateral choroid plexus (62 inSupporting Information). No relationship was foundbetween ventricular asymmetry and gender, mode of deliv-ery or the infants age at the time of the ultrasound (61,64 inSupporting Information).

There appears to be an effect of head position at the timeof scanning on ventricular asymmetry. In preterm neonatesas well as in term newborns with brain damage, the depth ofthe anterior horn clearly changed with head position within3 h after turning the head. While the upper frontal hornincreased in size, the depth of lower frontal horn decreased.(71 in Supporting Information) Area (63 in SupportingInformation) and volume measurements (70 in SupportingInformation) in preterm and term infants also showed thathead position plays a role. Its influence seems to depend onthe time interval between changing the position of the head,for turning the head did not significantly alter ventricularasymmetry within 1 h (61 in Supporting Information). Theinfluence of head position on ventricular size appears todiminish with increasing GA, which contributes to the idea



that maturation of the brain plays a role. It has been sug-gested that the structure of the brain of preterm infantsmight be less firm when compared with term infants, mak-ing it more sensitive to the effect of gravity.(71 in Support-ing Information)

DIFFERENCES IN ANTENATAL AND POSTNATAL VENTRICULAR SIZEFoetal reference curvesBoth abdominal and transvaginal ultrasonography are usedfor monitoring foetal ventricular size during intra-uterinedevelopment to detect congenital hydrocephalus. Like inneonates, the foetal atria and occipital horns appear to bemore sensitive for early ventricular enlargement than thefrontal horns and midbody of the lateral ventricles (31,32).Because reference values for the occipital horn are harderto obtain (72,73 in Supporting Information) most studies(56; 7480 in Supporting Information) have focussed onthe size of the atria. Foetal ventriculomegaly has beendefined as an atrial size exceeding 10 mm, which represents2.54 standard deviations above the mean. Less data areavailable with respect to foetal dimensions of the frontal(31; 72,73,8183 in Supporting Information) and occipitalhorn (83,84 in Supporting Information).

Foetal ventricular size has mainly been determined in theaxial plane. Monteagudo et al.(83 in Supporting Informa-tion), however, performed frontal horn measurements inthe coronal plane and occipital horn measurements in theparasagittal plane, thus enabling comparison with neonataldata. Differences between intra-uterine and extra-uterinereference ranges are remarkable, especially with regard tothe AHW and TOD. Upper limits for the foetal AHW andTOD are 7.5 mm and 2530 mm (depending on GA)respectively (83 in Supporting Information), which is higherthan neonatal measurements for those parameters(15,20,21,30,34) (Fig. 3B-C). This raises the question whathappens with ventricular size during and following birth.

Changes in ventricular size following birthWhile the lateral ventricles are open in most foetuses (83 inSupporting Information), nearly or completely closed ven-tricles are quite a common phenomenon in term and pre-term neonates in the first week post-partum and notnecessarily related to cerebral ischaemia (33,36; 8587 inSupporting Information).

A rapid increase in ventricular area (36; 63 in SupportingInformation) and volume (69 in Supporting Information) atthe end of the first week and in the second week of life hasbeen reported. Nelson et al. (85 in Supporting Information)investigated the postnatal change in ventricular size in apopulation of healthy term newborns following a vaginaldelivery. By combining cross-sectional and longitudinalclassifications of the lateral ventricles, it was assessed thatwithin 12 h of birth the ventricles were completely closedin 80%, partially opened in 18%, and completely open inonly 1% of the newborns. Gradual reopening was seen overthe following days. The estimated median time to partiallyopen ventricles following a vaginal delivery was 2.53 days.

The effect of the mode of deliveryThe difference in ante- and postnatal ventricular size andthe process of reopening seen postnatally may be partlyexplained by the effect of a vaginal delivery (33; 85 in Sup-porting Information). It has been suggested that the com-pression on the skull of the baby during its passage throughthe birth canal possibly forces the CSF out of the ventricles.(85 in Supporting Information) For term infants, a statisti-cally significant correlation between the mode of deliveryand the visibility of CSF in the lateral ventricles wasreported by Winchester et al (33). While open ventricleswere found in two out of 16 neonates born by caesareansection within the first 6 days of life, it was seen in only twoout of 28 infants born vaginally. For preterm infants, theassociation between compressed ventricles and mode ofdelivery is less clear.

The optimal study design for demonstrating a possibleeffect of the mode of delivery on ventricular size would be alongitudinal study with antenatal and postnatal ventricularmeasurements in neonates born after a vaginal delivery orprimary caesarean section. In addition to Nelsons researchin term infants (85 in Supporting Information), it would beinteresting to investigate whether the phenomenon ofreopening can also be observed in preterm infants.

Other hypothesesPostnatal changes in ventricular size can not exclusively beexplained by the mode of delivery as compressed ventriclesmay persist after the first week after birth (85 in SupportingInformation) and can also be seen following a caesarean sec-tion (33; 86 in Supporting Information). Another factor thatat least partly may be responsible for the difference in ante-and postnatal ventricular size, is the mild degree of dehydra-tion of the newborn during the first days of life, which is alsoconsidered to explain the smaller renal pyela in neonatescompared with foetal pyelum size (88,89 in SupportingInformation). Dehydration may lead to a reduced CSF pro-duction and hence to a smaller ventricular size.

The transition from the foetal low pressure to the neona-tal high pressure circulatory state, probably accompanied bychanges in the secretion or reabsorption of CSF, has alsobeen proposed as an explanation for the changes in ventric-ular size following birth (36).

CONCLUSIONEvaluation of the size of the lateral ventricles is importantfor the early detection of PHVD in neonates, which mayhave important therapeutic and prognostic implications.During the last three decades, cUS played a key role in thediagnosis and evaluation of ventricular dilatation.Reference values were established for several ventriculardimensions. Furthermore, several variables influencingventricular size have been investigated.

The main conclusions are that, while reference values forthe VI increase with ongoing maturity, the correlation withGA remains controversial for measurements of the AHWand TOD. Asymmetry between the lateral ventricles is a



common phenomenon and appears to represent a normalphysiological variation rather than a sign of cerebral pathol-ogy in most neonates. The difference in foetal and neonatalreference curves is remarkable, the ventricular size appear-ing larger in foetuses. This may partly be explained by theeffect of a vaginal delivery.

Several questions remain unanswered. There is still aneed for more cross-sectional as well as longitudinal data onnormal ventricular size, especially for ELBW infants. Somestudies have been carried out with only a small study popu-lation or were performed quite a long time ago with lessadvanced techniques. No data are available with respect toreopening of the ventricles in premature infants or neonatesborn by primary caesarean section, so the effect of the modeof delivery on postnatal ventricular size is still not clear.

The main question, however, which needs to be answeredis which ventricular parameter gives the most accuratereflection of intracranial pressure in neonates with PHVD.To our knowledge, this has only been investigated for theVI.(10) As important therapeutic decisions are based onventricular size measurements, there still remains a need forprospective studies.

ACKNOWLEDGEMENTSThis study was funded by a grant from the Alexandre Suer-man programme for MD PhD students of the UniversityMedical Centre Utrecht.

References

1. Kliegman RM, Behrman RE, Jenson HB. Nelson textbook ofpediatrics. Philadelphia: Saunders Elsevier, 2007:715.

2. Murphy BP, Inder TE, Rooks V, Taylor GA, Anderson NJ, Mog-ridge N, et al. Posthaemorrhagic ventricular dilatation in thepremature infant: natural history and predictors of outcome.Arch Dis Child Fetal Neonatal Ed 2002; 87: F3741.

3. Volpe JJ. Intracranial hemorrhage: germinal matrix-intraven-tricular hemorrhage of the premature infant. Neurology of thenewborn. 5th ed. Philadelphia: Saunders Elsevier, 2008.

4. Jaeger M, Grussner SE, Omwandho CO, Klein K, TinnebergHR, Klingmuller V. Cranial sonography for newborn screening:a 10-year retrospective study in 11,887 newborns. Rofo 2004;176: 8528.

5. Levene MI, Fawer CL, Lamont RF. Risk factors in the develop-ment of intraventricular haemorrhage in the preterm neonate.Arch Dis Child 1982; 57: 4107.

6. de Vries LS, Dubowitz LM, Dubowitz V, Kaiser A, Lary S, Silv-erman M, et al. Predictive value of cranial ultrasound in thenewborn baby: a reappraisal. Lancet 1985; 2: 13740.

7. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evo-lution of subependymal and intraventricular hemorrhage: astudy of infants with birth weights less than 1,500 gm. J Pediatr1978; 92: 52934.

8. Levene MI. Measurement of the growth of the lateral ventriclesin preterm infants with real-time ultrasound. Arch Dis Child1981; 56: 9004.

9. de Vries LS, Liem KD, van Dijk K, Smit BJ, Sie L, RademakerKJ, et al. Early versus late treatment of posthaemorrhagic ven-tricular dilatation: results of a retrospective study from five neo-natal intensive care units in The Netherlands. Acta Paediatr2002; 91: 2127.

10. Kaiser AM, Whitelaw AG. Cerebrospinal fluid pressure duringpost haemorrhagic ventricular dilatation in newborn infants.Arch Dis Child 1985; 60: 9204.

11. Hanlo PW, Gooskens RJ, van Schooneveld M, Tulleken CA,van der Knaap MS, Faber JA, et al. The effect of intracranialpressure on myelination and the relationship with neurodevel-opment in infantile hydrocephalus. Dev Med Child Neurol1997; 39: 28691.

12. Soul JS, Eichenwald E, Walter G, Volpe JJ, du Plessis AJ. CSFremoval in infantile posthemorrhagic hydrocephalus results insignificant improvement in cerebral hemodynamics. PediatrRes 2004; 55: 8726.

13. van Alfen-van der Velden AA, Hopman JC, Klaessens JH, Feu-th T, Sengers RC, Liem KD. Cerebral hemodynamics and oxy-genation after serial CSF drainage in infants with PHVD. BrainDev 2007; 29: 6239.

14. Korobkin R. The relationship between head circumference andthe development of communicating hydrocephalus in infantsfollowing intraventricular hemorrhage. Pediatrics 1975; 56:747.

15. Liao MF, Chaou WT, Tsao LY, Nishida H, Sakanoue M. Ultra-sound measurement of the ventricular size in newborn infants.Brain Dev 1986; 8: 2628.

16. Lorber J. Systematic ventriculographic studies in infants bornwith meningomyelocele and encephalocele. The incidence anddevelopment of hydrocephalus. Arch Dis Child 1961; 36: 3819.

17. Muller WD, Urlesberger B. Correlation of ventricular sizeand head circumference after severe intra-periventricularhaemorrhage in preterm infants. Childs Nerv Syst 1992; 8:335.

18. Graziani LJ, Pasto M, Stanley C, Steben J, Desai H, Desai S,et al. Cranial ultrasound and clinical studies in preterm infants.J Pediatr 1985; 106: 26976.

19. Johnson ML, Mack LA, Rumack CM, Frost M, Rashbaum C. B-mode echoencephalography in the normal and high risk infant.AJR Am J Roentgenol 1979; 133: 37581.

20. Sondhi V, Gupta G, Gupta PK, Patnaik SK, Tshering K.Establishment of nomograms and reference ranges for intra-cranial ventricular dimensions and ventriculo-hemisphericratio in newborns by ultrasonography. Acta Paediatr 2008;97: 73844.

21. Perry RN, Bowman ED, Murton LJ, Roy RN, de Crespigny LC.Ventricular size in newborn infants. J UltrasoundMed 1985; 4:4757.

22. Sauerbrei EE, Digney M, Harrison PB, Cooperberg PL. Ultra-sonic evaluation of neonatal intracranial hemorrhage and itscomplications. Radiology 1981; 139: 67785.

23. Grasby DC, Esterman A, Marshall P. Ultrasound grading ofcerebral ventricular dilatation in preterm neonates. J PaediatrChild Health 2003; 39: 18690.

24. Govaert P, de Vries LS. An atlas of neonatal brain sonography.London: Mac Keith Press, 1997:10.

25. Brann BS, Qualls C, Wells L, Papile L. Asymmetric growth ofthe lateral cerebral ventricle in infants with posthemorrhagicventricular dilation. J Pediatr 1991; 118: 10812.

26. Grant EG, Borts FT, Schellinger D, McCullough DC, Sivasubr-amanian KN, Smith Y. Real-time ultrasonography of neonatalintraventricular hemorrhage and comparison with computedtomography. Radiology 1981; 139: 68791.

27. London DA, Carroll BA, Enzmann DR. Sonography of ventric-ular size and germinal matrix hemorrhage in premature infants.AJR Am J Roentgenol 1980; 135: 55964.

28. Poland RL, Slovis TL, Shankaran S. Normal values for ventric-ular size as determined by real time sonographic techniques.Pediatr Radiol 1985; 15: 124.



29. Silverboard G, Horder MH, Ahmann PA, Lazzara A, SchwartzJF. Reliability of ultrasound in diagnosis of intracerebral hem-orrhage and posthemorrhagic hydrocephalus: comparison withcomputed tomography. Pediatrics 1980; 66: 50714.

30. Reeder JD, Kaude JV, Setzer ES. The occipital horn of the lat-eral ventricles in premature infants. An ultrasonographic study.Eur J Radiol 1983; 3: 14850.

31. Lustig-Gillman I, Snyder JR, Silverman F, Young BK. Sono-graphic anatomy of the fetal cerebral ventricles, with referenceto the early diagnosis of hydrocephaly. J Perinat Med 1984; 12:18591.

32. Kurjak A. Ultrasonic grey-scale technique in early and latediagnosis of foetuses with hydrocephalus. Ultraschall Med1980; 1: 2707.

33. Winchester P, Brill PW, Cooper R, Krauss AN, Peterson HD.Prevalence of compressed and asymmetric lateral ventriclesin healthy full-term neonates: sonographic study. AJR Am JRoentgenol 1986; 146: 4715.

34. Davies MW, Swaminathan M, Chuang SL, Betheras FR. Refer-ence ranges for the linear dimensions of the intracranial ventri-cles in preterm neonates. Arch Dis Child Fetal Neonatal Ed2000; 82: F21823.

35. du Plessis AJ. Posthemorrhagic hydrocephalus and brain injuryin the preterm infant: dilemmas in diagnosis and management.Semin Pediatr Neurol 1998; 5: 16179.

36. Saliba E, Bertrand P, Gold F, Vaillant MC, Laugier J. Area oflateral ventricles measured on cranial ultrasonography in pre-term infants: reference range. Arch Dis Child 1990; 65: 102932.

37. International PHVD Drug Trial Group. International rando-mised controlled trial of acetazolamide and furosemide in post-haemorrhagic ventricular dilatation in infancy. Lancet 1998;352: 43340.

38. Ventriculomegaly Trial Group. Randomised trial of early tap-ping in neonatal posthaemorrhagic ventricular dilatation. ArchDis Child 1990; 65: 310.

39. Ventriculomegaly Trial Group. Randomised trial of early tap-ping in neonatal posthaemorrhagic ventricular dilatation:results at 30 months. Arch Dis Child Fetal Neonatal Ed 1994;70: F12936.

40. Brouwer A, Groenendaal F, van Haastert I, Rademaker K,Hanlo P, de Vries LS. Neurodevelopmental outcome of pre-term infants with severe intraventricular hemorrhage and ther-apy for post-hemorrhagic ventricular dilatation. J Pediatr 2008;152: 64854.

41. Ment LR, Vohr B, Allan W, Westerveld M, Katz KH, SchneiderKC, et al. The etiology and outcome of cerebral ventriculomeg-aly at term in very low birth weight preterm infants. Pediatrics1999; 104: 2438.

42. Whitaker AH, Feldman JF, van Rossem R, Schonfeld IS, Pinto-Martin JA, Torre C, et al. Neonatal cranial ultrasound abnor-malities in low birth weight infants: relation to cognitive out-comes at six years of age. Pediatrics 1996; 98: 71929.

43. Kitchen WH, Ford GW, Rickards AL, Doyle LW, Kelly E, Mur-ton LJ. Five-year outcome of infants of birthweight 500 to 1500grams: relationship with neonatal ultrasound data. Am J Per-inatol 1990; 7: 605.

44. Cooke RW. Determinants of major handicap in post-haemor-rhagic hydrocephalus. Arch Dis Child 1987; 62: 5046.

45. Corbett SS, Rosenfeld CR, Laptook AR, Risser R, MaravillaAM, Dowling S, et al. Intraobserver and interobserver reliabil-ity in assessment of neonatal cranial ultrasounds. Early HumDev 1991; 27: 917.

46. Hintz SR, Slovis T, Bulas D, Van Meurs KP, Perritt R, Steven-son DK, et al. Interobserver reliability and accuracy of cranialultrasound scanning interpretation in premature infants. J Pedi-atr 2007; 150: 5926.

47. Inder TE, Warfield SK, Wang H, Huppi PS, Volpe JJ. Abnormalcerebral structure is present at term in premature infants. Pedi-atrics 2005; 115: 28694.

48. Nguyen The Tich S, Anderson PJ, Shimony JS, Hunt RW,Doyle LW, Inder TE. A novel quantitative simple brain metricusing MR imaging for preterm infants. AJNR Am J Neuroradiol2009; 30: 12531.

49. Boardman JP, Counsell SJ, Rueckert D, Hajnal JV, BhatiaKK, Srinivasan L, et al. Early growth in brain volume is pre-served in the majority of preterm infants. Ann Neurol 2007;62: 18592.

50. Peterson BS, Vohr B, Staib LH, Cannistraci CJ, Dolberg A,Schneider KC, et al. Regional brain volume abnormalities andlong-term cognitive outcome in preterm infants. JAMA 2000;284: 193947.

51. Kesler SR, Ment LR, Vohr B, Pajot SK, Schneider KC,Katz KH, et al. Volumetric analysis of regional cerebraldevelopment in preterm children. Pediatr Neurol 2004; 31:31825.

52. Nosarti C, Al-Asady MH, Frangou S, Stewart AL, Rifkin L,Murray RM. Adolescents who were born very preterm havedecreased brain volumes. Brain 2002; 125: 161623.

53. Kramer RL, Yaron Y, Johnson MP, Evans MI, Treadwell MC,Wolfe HM. Differences in measurements of the atria of the lat-eral ventricle: does gender matter? Fetal Diagn Ther 1997; 12:3045.

54. Nadel AS, Benacerraf BR. Lateral ventricular atrium: larger inmale than female fetuses. Int J Gynaecol Obstet 1995; 51: 1236.

55. Patel MD, Goldstein RB, Tung S, Filly RA. Fetal cerebral ven-tricular atrium: difference in size according to sex. Radiology1995; 194: 7135.

56. Salomon LJ, Bernard JP, Ville Y. Reference ranges for fetal ven-tricular width: a non-normal approach. Ultrasound ObstetGynecol 2007; 30: 616.

57. Xenos C, Sgouros S, Natarajan K. Ventricular volume changein childhood. J Neurosurg 2002; 97: 58490.

58. Giedd JN, Snell JW, Lange N, Rajapakse JC, Casey BJ, Ko-zuch PL, et al. Quantitative magnetic resonance imaging ofhuman brain development: ages 418. Cereb Cortex 1996; 6:55160.

59. Horbar JD, Leahy KA, Lucey JF. Ultrasound identification oflateral ventricular asymmetry in the human neonate. J ClinUltrasound 1983; 11: 679.

60. Batton DG, Swails TL. Isolated lateral ventricular asymmetryin very low-birth-weight infants: a left-sided lesion? Am J Per-inatol 1998; 15: 1836.

SUPPORTING INFORMATIONAdditional Supporting Information may be found in theonline version of this article:

Supplementary references

Please note: Wiley-Blackwell is not responsible for thecontent or functionality of any supporting materials sup-plied by the authors. Any queries (other than missing mate-rial) should be directed to the corresponding author for thearticle.



2Ultrasound Measurements of Th

Documents

Transcript of 2Ultrasound Measurements of Th