EHD-03009; No of Pages 7
ARTICLE IN PRESS Early Human Development xxx (2008) xxx–xxx
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / e a r l h u m d e v
Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors Margit Bauer a,⁎, Christa Fast b, Josef Haas a, Bernhard Resch b, Uwe Lang a, Barbara Pertl a a b
Department of Obstetrics and Gynecology, Medical University Graz, Auenbruggerplatz 14, 8036 Graz, Austria Department of Pediatrics, Medical University Graz, Auenbruggerplatz 30, 8036 Graz, Austria
a r t i c l e
i n f o
Article history: Received 6 June 2008 Received in revised form 22 July 2008 Accepted 22 July 2008 Available online xxxx Keywords: PVL Obstetric risk factors PPROM Preterm infants
a b s t r a c t Objective: To identify obstetric risk factors and to elucidate the effect of prolonged rupture of the membranes on the development of cystic periventricular leukomalacia (PVL) in preterm infants. Methods: A retrospective case–control study of 95 preterm infants with the diagnosis of PVL and 245 healthy controls matched for gestational age. A total of 52 antenatal, intrapartum and neonatal characteristics were studied by univariate methods and logistic regression. Results: Preterm premature rupture of membranes (PPROM) (odds ratio 2.1 [95% CI 1.3–3.4], P = .003), gestational age at PPROM (P = .025), prolonged rupture of membranes (P b.0001), administration of tocolytic agents (1.8 [1.1–3.0], P = .019) and antibiotics (1.9 [1.2–3.1], P = .008) were associated with PVL. The use of tocolytic agents N 24 h (P = .008), prolonged latency between the increase in maternal leukocyte count and birth (P = .034), spontaneous onset of labor (1.8 [1.0–2.9], P = .026), vaginal delivery (1.7 [1.1–2.8], P = .029) and male gender (1.5 [1.0–2.0], P = .04) were found more frequently in PVL cases. Preeclampsia (0.4 [0.1–0.9], P = .034), hypertension at booking (P = .009), sonographic IUGR (P = .020), abnormal blood flow of the umbilical artery (P = .032) and cesarean section without labor (0.5 [0.3–0.8], P = .006) were found less frequently. In logistic regression analysis, prolonged rupture of the membranes (P = .748), preeclampsia (P = .973), the use of antibiotics (P = .617) and beta-sympathomimetic tocolytic agents (P = .563) lost statistical significance, whereas birth weight (P = .036) became significant. Conclusion: PPROM and prolonged rupture of the membranes may provoke adverse effects on the neurodevelopmental outcome of the preterm fetus. These findings may have implications on the obstetric management of PPROM beyond 30 weeks of gestation. Cesarean section without labor was less likely associated with the diagnosis of PVL. © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Periventricular leukomalacia (PVL) is the most common cause of cerebral palsy in preterm infants [1]. The prognosis is poor since 60– 100% of affected infants develop cerebral palsy [2,3]. Studies strongly suggest that intrauterine events are responsible for brain injuries associated with PVL [4–6]. In premature infants two events have been linked to the onset of PVL, namely intrauterine infection and hypoxemia. In studies on sheep, white matter at midgestation was sensitive to injuries following both endotoxemia and systemic asphyxia [7]. It was found that cerebral damage in preterm infants is rather the result from a sequence of events than from one specific [4], which was supported by experimental studies on fetal lambs showing
⁎ Corresponding author. Tel.: +43 316 385 81070; fax: +43 316 385 3061. E-mail address:
[email protected] (M. Bauer).
that frequent intrauterine episodes of ischemia produced more neuronal loss than one single [8]. Intrauterine infection induces an inflammatory response syndrome activating a cascade of cytokines and chemokines [9,10] which, in turn, may trigger preterm labor, cervical ripening, and in the fetal compartment rupture of the membranes and injury to the lung and central nervous system [10]. Hypoxic–ischemic events may also induce similar cytokine related pathways. On the molecular level tumor necrosis factor α (TNF α) signaling pathways were found to play a central role in apoptotic neural cell death [11], which affects oligodendrocyte progenitors especially during the period of greatest vulnerability between 24–34 postconceptional weeks [9,12]. Recent studies on potential obstetric and neonatal risk factors have partly reached different results, and the number of affected infants studied has often been low. We conducted a retrospective case– control study of 95 infants with the diagnosis of PVL to (i) identify antenatal, intrapartum and early neonatal parameters, which might be either involved or protective, and to (ii) elucidate the role of
0378-3782/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2008.07.007
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS 2
M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx
prolonged rupture of the membranes in PVL with the aim to improve the obstetric management of PPROM. 2. Materials and methods This study was performed as a retrospective, case–control study on preterm infants born at 26 to 35 completed weeks of gestational age at the Medical University Graz, Austria, during the period of 1988 to 2001. The Department of Obstetrics is a tertiary-level regional referral center with 2500 deliveries per year. Since early onset PVL may best reflect intrauterine problems 95 infants with cystic PVL diagnosed within the first two weeks of life were identified from the neonatal database. Periventricular echodensities were diagnosed by cranial ultrasonography in all cases within the first four days of life; cystic lesions were visualized between 7 and 14 days of life. Each case of these 95 pregnancies was matched with two to three eligible pregnancies resulting in a surviving infant without brain injuries according to gestational age, single versus twin pregnancy and year of birth. The time interval of delivery between cases and controls varied between a few days and 7 weeks. Exclusion criteria were fetal malformations, cystic PVL diagnosed within the first 48 h after delivery, prolonged flares and PVL grade 1 as well as brain injuries others than PVL. The infants had neither late onset sepsis nor surgery. A total of 52 parameters were compared including maternal demographic characteristics, medical history, previous obstetric history and history of the relevant pregnancy. Special interest was focused on preterm premature rupture of membranes (PPROM), gestational age at PPROM, prolonged rupture of membranes, evidence of chorioamnionitis or hypoxic changes on placental histology, preeclampsia, intrauterine growth restriction (IUGR), medication (antibiotics, tocolytic agents, antenatal corticosteroids), clinical and laboratory signs of chorioamnionitis. Variables of antenatal monitoring comprised ultrasound scanning for fetal growth and amniotic fluid volume estimation, Doppler flow examinations of the umbilical artery and electronic fetal heart rate monitoring. Oligo-anhydramnios was defined by measuring a single largest amniotic fluid pocket of zero to 2 cm in vertical axis. Intrapartum parameters studied were fetal presentation, electronic fetal heart rate monitoring, mode of delivery and time interval from onset of labor to birth. Non-reassuring fetal heart rates included recurrent late decelerations or absence of variability for more than 40 min.
Neonatal variables of interest were gestational age at birth, birth weight, Apgar score at 5 min of age, pH of umbilical artery and fetal gender. Gestational age was estimated by means of menstrual history and ultrasound examinations performed before 14 weeks of gestation. PPROM was diagnosed when membrane rupture occurred before the onset of spontaneous labor by visualization of amniotic fluid loss by sterile specula examination and confirmed by Actim Prom® (Medix Biochemica Ab, Finland), a test to detect insulin-like growth-factor binding protein 1. Residual amniotic fluid volume was estimated by ultrasound examination. The time from rupture of membranes to delivery was categorized into 4 time periods: b11.9 h, 12 to 23.9 h, 24 to 47.9 h and ≥48 h. Prolonged rupture of membranes was defined as rupture of membranes at least 24 h prior to delivery. Preeclampsia was defined as diastolic blood pressure ≥90 mm Hg at least twice and proteinuria N300 mg/24 h. In the selected study period it was not the policy to administer magnesium sulfate to cases of severe preeclampsia, though a differentiation between mild and severe preeclampsia was not made. In pregnancies complicated by preeclampsia or suspected intrauterine growth restriction a vaginal delivery was primarily attempted in cases without signs of fetal distress. The diagnosis of chorioamnionitis represented pathologic findings on placental histology with inflammation of the placental membranes and the chorionic plate with polymorphonuclear leukocyte infiltration. Clinical chorioamnionitis was diagnosed in the presence of maternal temperature N38 °C, white blood cell count at least 15,000 cells/mm3 or elevated CRP levels N9. White blood cell counts and CRP levels were routinely performed in all pregnancies. During the study period PPROM and preterm labor were treated expectantly before 34 weeks with screening for infection and fetal distress. Women were treated intravenously with antibiotics (e.g., penicillin, ampicillin/clavulanate, cephalosporine) at the time of initial presentation. Beta-sympathomimetic tocolytic agents were given prophylactically in case of PPROM or in the setting of preterm labor in the absence of clinical signs of chorioamnionitis at the discretion of the attending physician, while two intramuscular doses of corticosteroids (12 mg betamethasone) were given 24 h apart to achieve lung maturation. Serial laboratory and ultrasound evaluations were performed to detect signs of chorioamnionitis, abruptio placentae, fetal growth and well-being during the surveillance period. Birth was induced and delivery expedited if 34 weeks of gestation were
Table 1 Maternal factors and antenatal complications among cases of PVL and controls Factor
Number of
OR (95% CI)
Cases (n = 95)
Controls (n = 245)
Maternal factor Age b 20 years Mean maternal age Primigravidity Parity N 3 Previous preterm delivery Multiple pregnancies Artificial reproduction techniques TTTS Amniocentesis performed
9/95 (9.5%) 28.4 ± 6.2 31/95 (32.6%) 4/95 (2.1%) 6/95 (6.3%) 22/95 (23%) 4/95 (4.2%) 2/22 (9%) 6/89 (6.7%)
7/244 (2.9%) 28.8 ± 5.7 98/245 (40%) 21/245 (2.4%) 18/245 (7.3%) 54/245 (22%) 6/245 (2.4%) 9/54 (18%) 32/239 (13.3%)
Antenatal complication PPROM Mean gestational age at PPROM Preeclampsia Hypertension at booking Mean gestational age at birth in Preeclampsia Antepartal hemorrhage Placental abruption Diabetes mellitus Cervical incompetence
55/95 (57.9%) 29.8 ± 3.7 5/95 (5.3%) 6/95 (6.3%) 33.8 ± 1.3 40/95 (42.1%) 3/95 (3.2%) 6/95 (6.3%) 8/95 (8.4%)
97/230 (42.2%) 31.1 ± 3.4 33/245 (13.5%) 42/245 (17.1%) 32.1 ± 2.2 95/245 (38.8%) 19/245 (7.8%) 16/245 (6.5%) 19/245 (7.8%)
P-value
.019 .555 .209 .167 .82 .885 .474 .734 .120
2.1 (1.3–3.4) 0.4 (0.1–0.9)
1.1 (0.7–1.8)
.003 .029 .034 .009 .094 .62 .3 1 .82
PVL = periventricular leukomalacia; OR = odds ratio; CI = confidence interval; TTTS = twin–twin transfusion syndrome; PPROM = preterm premature rupture of membranes.
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx Table 2 Gestational age at PPROM among cases of PVL and controls Gestational age at PPROM
Number of
≤27 weeks 28–30 weeks 31–34 weeks N 34 weeks
Table 4 Medical treatments among cases of PVL and controls P-value
Cases (n = 55)
Controls (n = 97)
14/55 14/55 24/55 3/55
18/97 (18.6%) 16/97 (16.5%) 51/97 (52.6%) 12/97 (12.3%)
(25.4%) (25.4%) (43.6%) (5.4%)
3
.025⁎
Medical treatment
Antibiotics Tocolytic agents Tocolysis N 24 h Betamethasone
Number of Cases (n = 95)
Controls (n = 245)
46/95 65/94 31/95 65/94
79/243 (32.5%) 131/238 (55%) 59/242 (24.4%) 148/245 (60.4%)
(48.4%) (69%) (32.6%) (69%)
OR (95% CI)
P-value
1.9 (1.2–3.1) 1.8 (1.1–3.0)
.008 .019 .008 .16
1.4 (0.8–2.4)
PPROM = preterm premature rupture of membranes. ⁎Jonckheere–Terpstra-test.
All abbreviations as in Table 1.
completed or if clinical and laboratory signs of infection or fetal distress were noted. A neonatologist attended all deliveries and all infants were transferred to the neonatal intensive care unit. Placentas were submitted to histopathology. Cranial ultrasound was performed and multiple images were obtained in the coronal and sagittal planes through the anterior fontanelle in all infants on days 1, 4, 7 and, thereafter, once a week in case of pathology. Periventricular echodensities were defined as confluent areas of increased echogenicity comparable to that of the choroid plexus in both the coronal and the sagittal planes. Cystic transformation was visualized as small cavities in the anterior, anterior to the frontal horn of the lateral ventricle, parietal, lateral to the body of the lateral ventricle, or occipital, adjacent and lateral to the occipital horn of the lateral ventricle. All infants were followed up clinically until their corrected age of 2 years at the Department of Pediatrics, Medical University Graz.
PVL (P = .019). The mean maternal age was 28.4 ± 6.2 and 28.8 ± 5.7 years in cases and controls (P = .555), respectively. Primigravidity (P = .209), parity N3 (P = .167) and a history of previous preterm deliveries (P = .82) revealed no differences between mothers of cases and controls. There was no difference in the occurrence of multiple pregnancies between the groups (P = .885) and in the application of artificial reproduction techniques (P = .474). In twin pregnancies twin–twin transfusion syndrome (TTTS) occurred in 2/22 (9%) cases and in 9/54 (18%) controls (P = .734). Amniocentesis was performed in 6/89 (6.7%) cases compared to 32/239 (13.3%) controls (P = .120). Among antenatal complications (Table 1), PPROM (odds ratio 2.1 [95% CI 1.3–3.4], P = .003) was identified more frequently in pregnancies complicated by PVL. Gestational age at PPROM (Table 2) was significantly lower in cases than in controls (P = .025). The mean gestational age at PPROM was 29.8 ± 3.7 weeks in cases and 31.1 ± 3.4 in controls (P = .029). Prolonged rupture of membranes (Table 3) was diagnosed significantly more often in cases than in controls (P = .0001), while preeclampsia (0.4 [0.1–0.9], P = .034) and hypertension at booking (P = .009) were not (Table 1). In preeclampsia the mean gestational age at birth was 33.8 ± 1.3 weeks in cases and 32.1 ± 2.2 in controls (P = .094). Antepartal hemorrhage (1.1 [0.7–1.8], P = .62), placental abruption (P = .3), diabetes mellitus (P = 1) and cervical incompetence (P = .82) occurred without statistical difference between cases and controls. When looking at medical treatment (Table 4), antibiotics and betasympathomimetic tocolytic agents were applied to mothers of cases significantly more often (1.9 [1.2–3.1], P = .008 and 1.8 [1.1–3.0], P = .019, respectively). The administration of beta-sympathomimetic tocolytic agents N24 h to prolong pregnancy was also found significantly more often in cases (P = .008), while no difference was found for the application of betamethasone (1.4 [0.8–2.4], P = .16). No significant differences were observed regarding parameters indicating clinical and laboratory signs of chorioamnionitis (Table 5). A prolonged latency between the increase in maternal leukocyte count
3. Statistical methods In a univariate analysis maternal, categorical variables were compared between infants with PVL and infants without brain injuries using Pearson's Chi-square test or Fisher's exact test depending on the expectation counts, and the Jonckheere–Terpstra-test for categorized variables such as interval PPROM-birth, gestational age at PPROM and duration of labor. Continuous variables were compared with twosample-t-tests and/or Wilcoxon-test, depending on the normality of the data. For data analysis we used SPSS 14 (SPSS Inc, Chicago, Il) and StatXact 5.0 (Cytel, Boston, MA). For all univariate tests, P-values b .05 were considered statistically significant. Logistic regression was performed with a backward elimination procedure and an inclusion level of P b.10. The results of the logistic regression can only be used to support our univariate analysis because of the structure of the missing values (158/ 343 cases and controls were enrolled) and the resulting loss of power. 4. Limitations of the study Since this was a retrospective study comprising a time interval of 14 years the charts were not complete for all data, resulting in changes of the denominator between analyses.
Table 5 Clinical and laboratory signs of chorioamnionitis and placental histology among cases of PVL and controls Factor
When we compared maternal factors (Table 1), maternal age b20 years was found more frequently in pregnancies complicated by Table 3 Latency of rupture of membranes to birth among cases of PVL and controls Time interval in h
b 11.9 12–23.9 24–47.9 ≥48 ⁎Jonckheere–Terpstra-test.
Number of Cases (n = 94)
5. Results
Number of
P-value
Cases (n = 93)
Controls (n = 245)
45/93 (46.3%) 8/93 (8.6%) 9/93 (9.7%) 33/93 (35.5%)
177/245 (72.8%) 9/245 (3.7%) 17/245 (7.0%) 40/245 (16.5%)
OR (95% CI)
P-value
Controls (n = 245)
Signs of chorioamnionitis Maternal 6/93 (6.4%) temperature N 38 °C Maternal leukocyte 33/91 (36.3%) count N 15,000 12/54 (22.2%) Elevated maternal CRP levels Increase in leukocyte 33/69 (47.8%) count Increase in CRP 13/49 (26.5%) Fetal tachycardia 4/94 (4.2%) Placental histology Chorioamnionitis Hypoxic changes
54/145 (37.2%) 30/145 (20.7%)
12/243 (4.9%)
.59
58/210 (27.6%)
.13
18/122 (14.7%)
.27
89/179 (49.7%)
.88
36/100 (36%) 17/245 (6.9%)
.27 .45
.0001⁎ 33/72 (45.8%) 13/72 (18.1%)
1.4 (0.8–2.5)
.24 .44
CRP = C-reactive protein; all other abbreviations as in Table 1.
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS 4
M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx
Table 6 Latency between the increase in maternal leukocyte count and birth among cases of PVL and controls Time interval in hours
Number of
No latency 24 N 24
Table 8 Duration of labor among cases of PVL and controls Time interval in hours
P-value
Cases (n = 90)
Controls (n = 210)
36/90 (29.8%) 33/90 (27.3%) 52/90 (43%)
90/210 (42%) 89/210 (42.4%) 31/210 (14.8%)
.034⁎
≤6 N6 to ≤12 N12
Number of
P-value
Cases (n = 48)
Controls (n = 152)
30/48 (62.5%) 10/48 (20.8%) 8/48 (16.6%)
101/152 (66.4%) 35/152 (23%) 16/152 (10.5%)
.074⁎
⁎Jonckheere–Terpstra-test.
⁎Jonckheere–Terpstra-test.
and birth (Table 6) was the only parameter which was recorded significantly more often in cases (P = .034). A histological diagnosis of the placenta (Table 5) was not available in all records, but the diagnosis of chorioamnionitis was recorded at a higher rate in pregnancies of cases (45.8%) compared to controls (37.2%), but the difference did not reach statistical significance (1.4 [0.8–2.5], P = .24). Hypoxic changes on placental histology were described more frequently in the control group, but again the difference did not reach statistical significance (P = .44). The data on antenatal monitoring (Table 7) revealed no significant differences in terms of non-reassuring fetal heart rates and the sonographic diagnosis of oligo-anhydramnios between cases and controls (P = .10 and P = .079, respectively). Doppler studies of the umbilical artery were performed in 35/94 (37.2%) cases and in 85/245 (34.7%) controls. Abnormal umbilical blood flow was diagnosed significantly more often in controls (P = .032), as was the sonographic diagnosis of IUGR (P = .020). Among intrapartum characteristics (Table 7), the mode of delivery was significantly different between cases and controls. Cesarean section without labor was performed significantly more often in controls (0.5 [0.3–0.8], P = .006) whereas cesarean section in labor (1.2 [0.7–2.2], P = .54), spontaneous onset of labor (1.8 [1.0–2.9], P = .026), and vaginal delivery (1.7 [1.1–2.8], P = .029) were found significantly more frequently in cases. Vertex presentation (P = .8) and nonreassuring fetal heart rates in labor (1.0 [0.6–1.6], P = 1.0) were recorded with a similar frequency between the groups. The time interval from onset of labor to birth (Table 8) revealed no statistical differences between cases and controls (P = .074).
Table 7 Antenatal monitoring and intrapartum characteristics among cases of PVL and controls Factor
Number of Cases (n = 95)
Antenatal monitoring Non-reassuring fetal heart rates Oligo-anhydramnios Doppler studies performed Abnormal umbilical blood flow Sonographic diagnosis of IUGR
OR (95% CI)
P-value
Controls (n = 245)
15/95 (15.8%)
59/245 (24%)
.10
42/93 (45%) 35/94 (37.2%)
84/243 (34.6%) 85/245 (34.7%)
.079
6/35 (17%)
32/85 (37.6%)
.032
4/95 (4.2%)
37/245 (15.1%)
.020
Intrapartum characteristic Cesarean section 20/95 (21%) without labor Cesarean section 21/95 (22.1%) in labor Spontaneous onset 67/95 (70.5%) of labor Vaginal delivery 52/95 (54.7%) Vertex presentation 70/95 (73.7%) Non-reassuring fetal 43/95 (45.2) heart rates
Among the infant characteristics (Table 9) there were no significant differences in mean gestational age at birth (P = .66), mean birth weight (P = .52), Apgar score at 5 min ≤5 (P = .82) and umbilical artery pH of 7.10 or less (P = .53) between cases and controls. The mean pH of umbilical artery was 7.28 ± 0.08 for cases and 7.27 ± 0.08 for controls (P = .088). At birth 27 babies were small for gestational age, 4/95 (4.2%) cases and 23/245 (9.4%) controls (P = .17). Male gender was recorded significantly more often in cases (1.5 [1.0–2.0], P = .04).There were 58/ 95 (61%) boys in cases and 118/245 (48%) in controls. In a backward stepwise logistic regression analysis we included all perinatal parameters in a chronological order they occur during pregnancy and birth. Prolonged rupture of the membranes (P = .748), preeclampsia (P = .973), sonographic diagnosis of IUGR (P = .677), the use of antibiotics (P = .617) and beta-sympathomimetic tocolytic agents (P = .563) lost statistical significance. Maternal age (P = .006), PPROM (P = .007), mean gestational age at PPROM (P = .025), cesarean section without labor (P = .003), cesarean section in labor (P = .029), spontaneous onset of labor (P = .004), and male gender (P =.005) were found to keep their significance. Additionally, birth weight (P = .036) became significant. 6. Discussion This study has shown that PPROM and prolonged rupture of membranes are significantly associated with the development of PVL in premature infants. Consistent with our data, prolonged rupture of membranes was identified as a risk factor for PVL and cerebral palsy in recent studies [5,13–18]. Neonatal morbidity was significantly reduced after a short latency period [19]. We and others [16] showed that gestational age at PPROM was significantly related to the development of PVL. In the PVL group 50.9% of PPROM occurred ≤30 weeks of gestational age compared to 35.1% in the control group. With increasing latency between PPROM and birth the fetus is more likely exposed to ascending infections or hypoxic episodes which pose him at risk for adverse neurologic outcome. A number of studies found increasing rates of ascending infections during the course of prolonged rupture of membranes [2,17,20], but the effect on fetal brains remained inconclusive. Maternal fever, leukocytosis [2,17] and histologic chorioamnionitis [2,21,22] were found more frequently among mothers who delivered more than 24 h after membrane rupture. The association between chorioamnionitis and neonatal brain lesions is well established
Table 9 Infant characteristics among cases of PVL and controls 90/245 (36.7%)
0.5 (0.3–0.8)
.006
46/245 (18.7%)
1.2 (0.7–2.2)
.54
140/245 (57.1%)
1.8 (1.0–2.9)
.026
101/245 (41.2%) 177/245 (72.2%) 110/245 (44.9%)
1.7 (1.0–2.8)
.029 .8 1.0
1.0 (0.6–1.6)
IUGR = intrauterine growth restriction; all other abbreviations as in Table 1.
Infant characteristic
Mean gestational age at birth Mean birth weight in gram Apgar score at 5 min ≤ 5 pH of umbilical artery ≤7.10 Mean pH of umbilical artery SGA Male gender
Number of
OR (95% CI) P-value
Cases (n = 95)
Controls (n = 245)
31.1 ± 2.5 1518 ± 425 3/95 (3.1%) 2/79 (2.5%) 7.28 ± 0.08 4/95 (4.2%) 58/95 (61%)
31.3 ± 2.5 1492 ± 518 9/244 (3.3%) 11/199 (5.5%) 7.27 ± 0.08 23/245 (9.4%) 118/245 (48%)
.66 .52 .82 .53 .088 .17 1.5 (1.0–2.0) .04
SGA = small for gestational age; all other abbreviations as in Table 1.
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx
[2,3,14,15,23–26]. A meta-analysis of 30 studies confirmed chorioamnionitis as a risk factor for PVL and cerebral palsy [27]. Histologic chorioamnionitis increased the risk of PVL to 60–70% [9]. In the presence of intrauterine infection the rate of PVL was increasing when there was an antecedent PPROM [23]. In our study placentas of cases were slightly more often affected by chorioamnionitis, but in concordance with others the difference did not reach statistical significance [28–30]. Previously, neither histological nor clinical nor neonatal infection was found to be associated with white matter disease [31]. We found no differences in parameters indicating clinical chorioamnionitis between cases and controls. The only marker we identified to be associated with PVL was maternal leukocytosis in terms of a prolonged latency between the increase in leukocyte count and birth. This finding, however, is difficult to interpret in clinical practice since the administration of corticosteroids also results in maternal leukocytosis [32,33] which is, therefore, often neglected. However, this effect was found to be transient and any leukocytosis persisting N3 days was not due to betamethasone administration [34]. It is well established that antenatal treatment with betamethasone reduces the incidence of PVL significantly [15], even in the presence of histologic chorioamnionitis [35]. Our data could not confirm this protective effect since mothers of infants with and without PVL were treated with betamethasone at a similar frequency, which was about 60%. A similar result was published by Murphy et al. [5], but in their study corticosteroid administration was less than 10% for cases and controls. In concordance with Baud et al. [15], we found that mothers of cases received tocolytic agents significantly more often. In particular, the administration of beta-sympathomimetic agents N24 h was associated with an increased risk of PVL, which was also shown by Zupan et al. [23]. We and others found antibiotics more often applied to mothers of cases [24,27,36]. This might reflect that in our study population PPROM was more frequent in cases, and mothers were, therefore, treated with antibiotics. Hypoxia has been suggested as another important stimulus for the development of PVL. Macroscopic and histologic features in placentas from preterm infants exhibited ischemic changes in the majority of PVL cases, contributing to disturbed placental circulation [28]. It was also reported that histologic chorioamnionitis together with placental perfusion defects increased the risk of abnormal neurologic outcome compared to histologic chorioamnionitis alone [37]. Our analysis on impaired circulation in the placental histology provided no differences between the groups. Obstetric episodes that may induce fetal hypoxemia such as antepartal hemorrhage and placental abruption were registered without statistical differences between the groups. Antenatal hemorrhage, which was found as a risk factor in a previous study [38], was recorded at a similar frequency in cases and controls, but, interestingly, placental abruption which was a rare event, was found more often in controls. Oligo-anhydramnios, which may contribute to umbilical cord compression, in particular after PPROM, was more frequently diagnosed in cases, but the differences did not reach statistical significance. Data from previous studies showed that Apgar scores were significantly lower in infants developing PVL [5,30], which is indicative for hypoxic episodes during labor. Moreover, frequent moderate variable decelerations were observed in infants developing PVL, and were associated with abnormalities of the umbilical cord, indicating cord compression [39]. However, they found no signs of fetal distress since umbilical artery blood gases were similar in cases and controls. A meta-analysis of 13 randomized controlled trials led to the result that electronic fetal heart rate monitoring was not able to predict cerebral white matter injury or cerebral palsy [40]. This study showed no differences regarding signs of intrapartum hypoxemia such as non-reassuring fetal heart rates, low Apgar scores and cord blood gas estimates between the groups.
5
There is still an ongoing debate whether the mode of delivery affects the outcome of very low-birth-weight infants. We found that caesarean section without labor was performed significantly more often in controls, while spontaneous onset of labor, cesarean section in labor and vaginal delivery were found significantly more frequently in PVL cases. This result was independent of potentially confounding factors such as preeclampsia. Our finding is consistent with the results of other authors [5,24,41] who observed a dramatic decrease in the incidence of PVL and cerebral palsy in infants delivered by cesarean section. Unfortunately our data did not provide the specific indications for the caesarean sections between the groups, hence we cannot estimate to which extent preeclampsia influences this result. Vaginal delivery was reported to increase the risk of PVL in preterm infants weighing less than 1251 g [42] as was spontaneous labor for infants delivered between 26 and 28 weeks [3]. In univariate analysis, cesarean section in labor [43] and vaginal delivery [44] increased the risk of PVL in low-birth-weight infants, but both results lost significance in multivariate analysis. However, these results need a careful interpretation, since they were obtained from retrospective studies. The mechanism which may protect fetal brains from PVL during cesarean section is not clear. During vaginal delivery and cesarean section in labor, hypoxic events together with preexisting inflammatory reactions may contribute to worse neonatal outcome. Vaginal birth is expedited more likely in women presenting with preterm labor and favorable cervix, which are both triggered by inflammation. However, interestingly, it was shown that duration of labor does not increase the risk of white matter disease even in the presence of clinical, laboratory or histological evidence of intrauterine infection, a finding that strongly weakens a preventive effect of caesarean section without labor [31]. In concordance our study did not show any difference in duration of labor between the groups for infants delivered vaginally or by cesarean section after the onset of labor. In our study preeclampsia, hypertension at booking and sonographic diagnosis of IUGR were less frequently observed in the PVL group, which has already been shown by others [5,30,45]. One protective effect might be the increase in cerebral blood flow in severe preeclampsia. Unfortunately, our data provided no information on cerebral blood flow, but the umbilical blood flow was impaired in controls significantly more often. Hypertensive mothers were found to be less exposed to intrauterine infections compared to mothers with preterm labor and PPROM [46]. In addition, fetal growth restriction was suggested to accelerate pulmonary and neurologic maturation reducing the risk of cerebral disorders [47]. A recent animal study revealed that pre-exposure to hypoxia induces adaptive mechanisms and, therefore, is protective against subsequent hypoxic– ischemic injury [48]. This might explain the better neurologic outcome in preeclampsia, but contradicts the general finding that hypoxia is a trigger for PVL. It has been reported that preterm male infants are at higher risk for morbidity and mortality. Early genderrelated differences were found for the need of ventilatory and circulatory support [49]. In this study boys were affected by PVL significantly more often than controls. This is in agreement with previous studies in which boys, particularly when born before term, were found to be at increased risk of cerebral palsy [26,50]. A similar finding was reported for intraventricular hemorrhage, but not for PVL [5,16,46]. It is of interest, that male infants were significantly more likely to have positive placental membrane cultures than female infants [51]. The higher risk of male infants might be related to polymorphisms of cytokine genes which have been proposed in infants susceptible to PVL [52]. Since preterm birth, following antecedent preterm labor or PPROM, is strongly related to maternal infection, screening and therapy of maternal genital infections might be the most effective preventive strategy against PVL. Infection screening programmes in routine antenatal care were successful in the reduction of preterm births [53].
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS 6
M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx
Our data clearly showed that PPROM and prolonged latency are closely linked to the development of PVL even after close surveillance. It was pointed out in 3 randomized clinical trials that expectant management of PPROM at 30 to 36 weeks gestation increased the incidence of chorioamnionitis compared to immediate delivery [54– 56]. The absence of clinical signs of chorioamnionitis is no guarantee for optimal intrauterine conditions. Two thirds of cases with histologic chorioamnionitis were missed when looking at clinical evidence of chorioamnionitis [57]. CRP and maternal leukocyte count were not useful in the prediction of histologic chorioamnionitis following PPROM [58]. For the prevention of cystic PVL expeditious delivery after treatment with corticosteroids for 48 h might be an option for pregnancies complicated by PPROM beyond 30 weeks of gestation, but a benefit has to be proven in prospective studies. However, when we compared the gestational age at PPROM between cases and controls, 49.1% of PPROM occurred ≥31 weeks in the PVL group compared to 64.9% in the control group, which means that in a high number of healthy controls an impetuous preterm birth would have been induced with all consequences of prematurity. We and others found that cesarean section without labor was less likely associated with the diagnosis of PVL in preterm infants. Caution has to be taken regarding a protective effect of caesarean section without labor on fetal brains since all results were obtained from retrospective studies.
References [1] Volpe JJ. Brain injury in the premature infant. Clin Perinatol 1997;29:574–87. [2] Leviton A, Paneth N. White matter damage in preterm newborns — an epidemiologic perspective. Early Hum Dev 1990;24:1–22. [3] Resch B, Vollaard E, Maurer U, Haas J, Rosegger H, Müller W. Risk factors and determinants of neurodevelopmental outcome in cystic periventricular leukomalacia. Eur J Pediatr 2000;159:663–70. [4] Murphy DJ, Squier MV, Hope PL, Sellers S, Johnson A. Clinical associations and time of onset of cerebral white matter damage in very preterm babies. Arch Dis Child Fetal Neonatal Ed 1996;75:F27–32. [5] Murphy DJ, Sellers S, MacKenzie IZ, Yudkin PL, Johnson AM. Case–control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995;346:1449–54. [6] Grether JK, Nelson KB, Emery III ES, Cummins SK. Prenatal and perinatal factors and cerebral palsy in very low birth weight infants. J Pediatr 1996;128:407–14. [7] Mallard C, Welin AK, Peebles D, Hagberg H, Kjellmer I. White matter injury following systemic endotoxemia or asphyxia in the fetal sheep. Neurochem Res 2003;28:215–23. [8] Mallard EC, Williams CE, Gunn AJ, Gunning MI, Gluckman PD. Frequent episodes of brief ischaemia sensitize the fetal sheep brain to neuronal loss and induce striatal injury. Pediatr Res 1993;33:61–5. [9] Dammann O, Leviton A. Maternal intrauterine infection, cytokines and brain damage in the preterm newborn. Pediatr Res 1997;42:1–8. [10] Vigneswaran R. Infection and preterm birth: evidence of a common causal relationship with bronchopulmonary dysplasia and cerebral palsy. J Paediatr Child Health 2000;36:293–6. [11] Kadhim H, Khalifa M, Deltenre P, Casimir G, Sebire G. Molecular mechanisms of cell death in periventricular leukomalacia. Neurology 2006;67:293–9. [12] Rezaie P, Dean A. Periventricular leukomalacia, inflammation and white matter lesions within the developing nervous system. Neuropathology 2002;22:106–32. [13] Spinillo A, Capuzzo E, Stronati M, Ometto A, Orcesi S, Fazzi E. Effect of preterm premature rupture of membranes on neurodevelopmental outcome: follow up at two years of age. Br J Obstet Gynaecol 1995;102:882–7. [14] Perlman MJ, Risser R, Broyles RS. Bilateral cystic periventricular leukomalacia in the preterm infant: associated risk factors. Pediatrics 1996;97:822–7. [15] Baud O, Foix-L'Helias L, Kaminski M, Audibert F, Jarreau PH, Papiernik E, et al. Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190–6. [16] Locatelli A, Ghidini A, Paterlini G, Patane L, Doria V, Zorloni C, et al. Gestational age at preterm premature rupture of membranes: a risk factor for neonatal white matter damage. Am J Obstet Gynecol 2005;193:947–51. [17] Ramsey PS, Lieman JM, Brumfield CG, Carlo W. Chorioamnionitis increases neonatal morbidity in pregnancies complicated by preterm premature rupture of membranes. Am J Obstet Gynecol 2005;192:1162–6. [18] Livinec F, Ancel PY, Marret S, Arnaud C, Fresson J, Pierrat V, et al. Epipage Group. Prenatal risk factors for cerebral palsy in very preterm singletons and twins. Obstet Gynecol 2005;105:1341–7. [19] Mehdi A, Collet F, Aiguier M, Miras T, Teyssier G, Seffert P. Premature rupture of the membranes between 28 and 34 weeks of amenorrhea. Retrospective study apropos of 71 cases. J Gynecol Obstet Biol Reprod (Paris) 2000;29:599–606. [20] Ustun C, Kokcu A, Cil E, Kandemir B. Relationship between endomyometritis and the duration of premature membrane rupture. J Matern Fetal Med 1998;7:243–6.
[21] McElrath TF, Allred EN, Leviton A, for the developmental epidemiology network investigators. Prolonged latency after preterm premature rupture of membranes: an evaluation of histologic condition and intracranial ultrasonic abnormality in neonate born b 28 weeks of gestation. Am J Obstet Gynecol 2003;189:794–8. [22] Dexter SC, Pinar H, Malee MP, Hogan J, Carpenter MW, Vohr BR. Outcome of very low birth weight infants with histopathologic chorioamnionitis. Obstet Gynecol 2000;96:172–7. [23] Zupan V, Gonzalez P, Lacaze-Masmonteil T, Boithias C, d'Allest AM, Dehan M, et al. Periventricular leukomalacia: risk factors revisited. Dev Med Child Neurol 1996;38:1061–7. [24] O'Shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol 1998;147:362–9. [25] O'Shea TM, Kothadia JM, Roberts DD, Dillard RG. Perinatal events and the risk of intraparenchymal echodensity in very-low-birthweight neonates. Paediatr Perinat Epidemiol 1998;12:408–21. [26] Costantine MM, How HY, Coppage K, Maxwell RA, Sibai BM. Does peripartum infection increase the incidence of cerebral palsy in extremely low birthweight infants? Am J Obstet Gynecol 2007;196:e6–8. [27] Wu YW, Colford Jr JM. Chorioamnionitis as a risk factor for cerebral palsy. A meta analysis. JAMA 2000;284:1417–24. [28] Kumazaki K, Nakayama M, Sumida Y, Ozono K, Mushiake S, Suehara N, et al. Placental features in preterm infants with periventricular leukomalacia. Pediatrics 2001;109:650–5. [29] Wharton KN, Pinar H, Stonestreet BS, Tucker R, McLean KR, Wallach M, et al. Severe umbilical cord inflammation—a predictor of periventricular leukomalacia in very low birth weight infants. Early Human Dev 2004;77:77–87. [30] Murata Y, Itakura A, Matsuzawa K, Okumura A, Wakai K, Mizutani S. Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia. Brain Dev 2005;27:17–21. [31] Locatelli A, Vergani P, Ghidini A, Assi F, Bonardi C, Pezzullo JC, et al. Duration of labor and risk of cerebral white-matter damage in very preterm infants who are delivered with intrauterine infection. Am J Obstet Gynecol 2005;193:928–32. [32] Diebel ND, Parsons MT, Spellacy WN. The effects of betamethasone on white blood cells during pregnancy with PPROM. J Perinat Med 1998;26:204–7. [33] Vaisbuch E, Levy R, Hagay Z. The effect of betamethasone administration to pregnant women on maternal serum indicators of infection. J Perinat Med 2002;30:287–91. [34] Kadanali S, Ingec M, Kucukozkan T, Borekci B, Kumtepe Y. Changes in leukocyte, granulocyte and lymphocyte counts following antenatal betamethasone administration to pregnant women. Int J Gynaecol Obstet 1997;58:269–74. [35] Elimian A, Verma U, Beneck D, Cipriano R, Visintainer P, Tejani N. Histologic chorioamnionitis, antenatal steroids and perinatal outcomes. Obstet Gynecol 2000;96:333–6. [36] Leviton A, Paneth N, Reuss ML, Susser M, Allred EN, Dammann O, et al. Maternal infection, fetal inflammatory response, and brain damage in very low birth weight infants. Developmental Epidemiology Network Investigators. Pediatr Res 1999;46: 566–75. [37] Kaukola T, Herva R, Perhomaa M, Paakko E, Kingsmore S, Vainionpaa L, et al. Population cohort associating chorioamnionitis, cord inflammatory cytokines and neurologic outcome in very preterm, extremely low birth weight infants. Pediatr Res 2006;59:478–83. [38] Weindling AM, Wilkinson AR, Cook J, Calvert SA, Fok TF, Rochefort MJ. Perinatal events which precede periventricular hemorrhage and leukomalacia in the newborn. Br J Obstet Gynecol 1985;92:1218–23. [39] Ito T, Kadowaki K, Takahashi H, Nagata N, Makio A, Terakawa N. Clinical features of and cardiotocographic findings for premature infants with antenatal periventricular leukomalacia. Early Hum Dev 1997;47:195–201. [40] Graham EM, Petersen SM, Christo DK, Fox HE. Intrapartum electronic fetal heart rate monitoring and the prevention of perinatal brain injury. Obstet Gynecol 2006;108:656–66. [41] Baud O, Ville Y, Zupan V, Boithias C, Lacaze-Masmonteil T, Gabilan JC, et al. Are neonatal brain lesions due to intrauterine infection related to mode of delivery? Br J Obstet Gynaecol 1998;105:121–4. [42] Deulofeut R, Sola A, Lee B, Buchter S, Rahman M, Rogido M. The impact of vaginal delivery in premature infants weighing less than 1,251 grams. Obstet Gynecol 2005;105:525–31. [43] Wadhawan R, Vohr BR, Fanaroff AA, Perritt RL, Duara S, Stoll BJ, et al. Does labor influence neonatal and neurodevelopmental outcomes of extremely-low-birthweight infants who are born by cesarean delivery? Am J Obstet Gynecol 2003;189:501–6. [44] Hansen A, Leviton A. Labor and delivery characteristics and risks of cranial ultrasound abnormalities among very-low-birth-weight infants. Am J Obstet Gynecol 1999;181:997–1006. [45] Xiong X, Saunders LD, Wang FL, Davidge ST, Buekens P. Preeclampsia and cerebral palsy in low-birthweight and preterm infants: implications for the current “ischemic model” of preeclampsia. Hypertens Pregnancy 2001;20:1–13. [46] Ancel PY, Marret S, Larroque B, Arnaud C, Zupan-Simunek V, Voyer M, et al. The Epipage Study Group. Are maternal hypertension and small-for-gestational age risk factors for severe intraventricular hemorrhage and cystic periventricular leukomalacia? Results of the EPIPAGE cohort study. Am J Obstet Gynecol 2005;193:178–84. [47] Hadi HA. Foetal cerebral maturation in hypertensive disorders of pregnancy. Obstet Gynecol 1984;63:214–9. [48] Miller BA, Perez RS, Shah AR, Gonzales ER, Park TR, Gidday JM. Cerebral protection by hypoxic preconditioning in a murine model of focal ischemia–reperfusion. Neuroreport 2001;12:1663–9.
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
ARTICLE IN PRESS M. Bauer et al. / Early Human Development xxx (2008) xxx–xxx [49] Elsmen E, Hansen Pupp I, Hellstrom-Westas L. Preterm male infants need more initial respiratory and circulatory support than female infants. Acta Paediatr 2004;93:529–33. [50] Thorngren-Jerneck K, Herbst A. Perinatal factors associated with cerebral palsy in children born in Sweden. Obstet Gynecol 2006;108:1499–505. [51] Goldenberg RL, Andrews WW, Faye-Petersen OM, Goepfert AR, Cliver SP, Hauth JC. The Alabama Preterm Birth Study: intrauterine infection and placental histologic findings in preterm births of males and females less than 32 weeks. Am J Obstet Gynecol 2006;195:1533–7. [52] Baier RJ. Genetics of perinatal brain injury in the preterm infant. Front Biosci 2006;11:1371–87. [53] Kiss H, Petricevic L, Husslein P. Prospective randomised controlled trial of an infection screening programme to reduce the rate of preterm delivery. BMJ 2004;329:371–6. [54] Mercer BM, Crocker LG, Boe NM, Sibai BM. Induction versus expectant management in premature rupture of the membranes with mature amniotic fluid at 32 to 36 weeks: a randomized trial. Am J Obstet Gynecol 1993;169:775–82.
7
[55] Cox SM, Leveno KJ. Intentional delivery versus expectant management with preterm ruptured membranes at 30–34 weeks gestation. Obstet Gynecol 1995;86:875–9. [56] Naef RW, Albert JR, Ross EL, Weber BM, Martin RW, Morrisin JC. Premature rupture of the membranes at 34–37 weeks gestation: aggressive versus conservative management. Am J Obstet Gynecol 1998;178:126–30. [57] Richardson BS, Wakim E, daSilva O, Walton J. Preterm histologic chorioamnionitis: impact on cord gas and pH values and neonatal outcome. Am J Obstet Gynecol 2006;195:1357–65. [58] Sereepapong W, Limpongsanurak S, Triratanachat S. The role of maternal C-reactive protein and white blood cell count in the prediction of chorioamnionitis in women with premature rupture of membranes. J Med Assoc Thail 2001;84(Suppl 1):S360–6.
Please cite this article as: Bauer M, et al, Cystic periventricular leukomalacia in preterm infants: An analysis of obstetric risk factors, Early Hum Dev (2008), doi:10.1016/j.earlhumdev.2008.07.007
