- Email: [email protected]
Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia
Brain & Development 27 (2005) 17–21 www.elsevier.com/locate/braindev
Original article
Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia Yasutaka Murataa, Atsuo Itakurab,*, Katsuji Matsuzawac, Akihisa Okumurad, Kenji Wakaie, Shigehiko Mizutania a
Department of Obstetrics and Gynecology, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan b Maternity and Perinatal Care Center, Nagoya University Hospital, Nagoya, Japan c Department of Obstetrics and Gynecology, Anjo Kosei Hospital, Aichi, Japan d Department of Pediatrics, Nagoya University Graduate School of Medicine, Nagoya, Japan e Division of Epidemiology and Prevention, Aichi Cancer Center Research Institution, Nagoya, Japan Received 7 October 2003; received in revised form 9 February 2004; accepted 19 February 2004
Abstract Cystic periventricular leukomalacia (cPVL), the principal ischemic brain injury in premature infants, is characterized by necrosis of the white matter in the periventricular region and the major neuropathology for spastic motor deficits in cerebral palsy or epilepsy. Recent reports strongly suggest that the brain injury associated with cPVL may have already occurred in utero. In this study we searched retrospectively for possible clinical situations related to cPVL to facilitate assessment of optimal management. A total of 201 babies born at gestational ages from 24 to 33 weeks were entered into the study (1992 – 1997) and examined for involvement of 18 factors in cPVL retrospectively. And psychomotor development was examined at least until 18 months of corrected age. Among 201 premature babies 35 cases were diagnosed as cPVL later developed spastic diplegia. There are 23 cases of preeclampsia, no infant suffering from cPVL. In the univariate analysis, exposure to antenatal indomethacin, cord length $ 40 cm, and a low Apgar score were significantly associated with a 2 – 3 risk increased of cPVL occurrence, while antenatal magnesium sulfate reduced the risk. Chorioamnionitis was positively correlated with the risk, but did not reach statistical significance. In the multivariate analysis we found the statistical significance in exposure to antenatal indomethacin, a low Apgar score, and antenatal magnesium sulfate. Our results suggested that preeclampsia and antenatal exposure of magnesium sulfate reduced the risk while antenatal exposure of indomethacin and low Apgar score associated with the occurrence of cPVL. These findings support a growing consensus that cPVL is often the result of maternal and fetal factors as well as antenatal treatment. q 2004 Elsevier B.V. All rights reserved. Keywords: Periventricular leukomalacia; Perinatal factors; Preeclampsia; Indomethacin; Magnesium sulfate; Cord factor; Apgar score; Case– control study
1. Introduction Preterm birth, which occurs in about 5.3% of all pregnancies in Japan, is a major problem in obstetrics [1]. Although the survival rate has dramatically increased recently with developments in obstetric and neonatal care, neurological sequelae are a subject of increasing interest in perinatal medicine. Although the survival rate of low birth weight (LBW) infants is now over 85% at week 24 or more in hospitals having neonatal intensive care unit in Japan, the morbidity * Corresponding author. Tel.: þ 81-52-744-2261; fax: þ81-52-744-2268. E-mail address: [email protected] (A. Itakura). 0387-7604/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2004.02.011
rate with very LBW (VLBW) infants still remains at about 20% [2]. Two imaging patterns in ultrasonography, intraventricular hemorrhage (IVH) and cystic periventricular leukomalacia (cPVL) have significant sensitivity and specificity in the prediction of neurological alteration [3 – 5]. The principal ischemic brain injury in premature infants; cPVL is characterized by necrosis of the white matter in the periventricular region. It is suggested to be the major neuropathology for spastic motor deficits in cerebral palsy (CP) and is reported to occur in 5 – 17% of all infants born weighing , 1500 g [6]. In addition 66 –100% of infants affected with cPVL are destined to suffer from CP. Although almost all preterm birth infants suffering from CP were once believed to be injured during delivery and/or
18
Y. Murata et al. / Brain & Development 27 (2005) 17–21
in postnatal period, recent reports strongly suggest that the brain injury associated with cPVL may have already occurred in utero [7 – 10]. It is known that both neonatal and maternal conditions affect the risk of cPVL and/or developing CP in VLBW cases. The LBW neonate has an inherent gestational age dependent vulnerability as a consequence of anatomic and hemodynamic immaturity. However, little information is available as to antenatal management and/or clinical conditions of patients’ impact on the risk of cPVL and/or developing CP with VLBW. Although the early changes of periventricular leukomalacia may be apparent histologically within hours of insult, the lesions take at least 2 –6 weeks to before they can be visualized ultrasonographically. Therefore, if seen within 7 days of life, the origins must be intrauterine. This time lag makes determining the cause of the disease complicated [11]. In the present study we searched retrospectively for possible clinical situations related to cPVL to facilitate assessment of optimal management.
2. Subjects and methods We reviewed the maternal neonatal medical records for all appropriate for date babies born at gestational ages from 24 to 33 weeks at Anjo Kosei Hospital, a tertiary center in the central Japan area, between January 1992 and December 1997. The study population included maternal transports with predominant indications for maternal transfers and then all inborn neonates ðn ¼ 240Þ were candidates except for babies ðn ¼ 39Þ with major anomalies, with grade 3 or 4 IVH, or which died within 2 weeks after birth. A total of 201 babies were entered into the study and examined for involvement of antenatal and intrapartum factors in cPVL retrospectively. The antenatal and intrapartum factors were as follows: gestational age, birth weight, cord length, Apgar scores after 1 and 5 min, presence or absence of prenatal genital bleeding, preterm rupture of membranes (PROM), chorioamnionitis (CAM), multiple pregnancy, preeclampsia, and maternal serum CRP on the day of delivery, mode of delivery, whether cerclage was performed or not, kind of tocolytic agent administered (hydrochloric ritodorine, magnesium sulfate, indomethacin), and exposure to drugs for pulmonary maturation (betamethasone, TRH). The gestational ages were all determined by ultrasonographic examination in the first trimester. All of the babies underwent detailed antenatal and intrapartum cardiotocographic monitoring. Cervical cerclage was performed when repeated midtrimester abortion or fetal membranes were bulging without maternal findings for infection or labor pains. Antepartum administration of steroids was recorded for each case. In our series hydrochloric ritodorine was used for preterm labor at first intravenously up to 400 mg/min. When uterine contraction was not controlled or side effects were intolerable, magnesium sulfate and/or indomethacin
were selected at random and administered in addition. Magnesium sulfate was applied intravenously, with a loading dose of 1 –4 g/h and maintenance infusion of about 0.5– 2 g/h, maintaining the maternal serum level in a therapeutic window about 4– 8 mg/dl. Indomethacin was applied as rectal suppository, 25– 50 mg every 6 h for a maximum dose of 200 mg daily. Preeclampsia was diagnosed using the criteria of the American College of Obstetricians and Gynecologists. After delivery all placentas were examined by a pathologist who was blind to all clinical features. CAM was defined by the presence of an acute inflammatory cell infiltration in both layers of the membranes. Neonatal cranial transfontanelle scans were evaluated by an experienced pediatrician. Standard sagital and coronal scans were serially performed at least three times 1 month of life. Ultrasonograms were obtained with a 7.5 MHz transducer (Sonos 1000, Hewlett Packard), and cPVL was defined as formation of cyst more than 3 mm in diameter in periventricular areas on coronal and sagittal views, and included cPVL (þ ). Abnormal scans after these time limits were considered to be related to neonatal events and therefore the cases were not included in the cPVL (þ ) group. The remaining neonates showing no abnormal findings and excluded from cPVL (þ ) were included in the cPVL (– ) group. MRI was performed in all patients with cPVL during late infancy through early childhood. All patients had ventriculomegaly with an irregular margin accompanied by periventricular abnormal high intensity areas on T2-weighed imaging. Psychomotor development was examined every 3 months after discharge at least until 18 months of corrected age. We paid close attention to spasticity of the lower extremities, such as a tight popliteal angle, foot joint tightness, and hyperreflexia in the deep tendon reflex. Spastic diplegia was diagnosed when an infant, displaying spastic gait or inability to walk had signs of spasticity of the lower extremities. Cranial magnetic resonance imaging was performed during late infancy to early childhood, if an infant exhibited psychomotor retardation or spasticity. 2.1. Statistical analysis We used statistical methods for case – control studies, comparing cPVL (þ ) (cases) and cPVL (2 ) (controls) infants. All baseline and pregnancy-related characteristics were assessed as potential risk factors for cPVL during pregnancy or intra-partum with Chi-square, or Fisher’s exact tests. Odds ratios (OR) were estimated to approximate relative risk for cPVL along with 95% confidence intervals. Finally, multivariate analysis was performed including variables that were significant in the univariate analysis, with birth weight and gestational week by the logistic procedure of the Statistical Analysis System (SAS) (SAS Institute Inc, Cary, NC). All P values were two-sided and P less than 0.05 were considered as significant.
Y. Murata et al. / Brain & Development 27 (2005) 17–21 Table 1 Clinical characteristics of the study population
Gestational weeks Birth weight (g) Apgar score after 1 min Apgar score after 5 min Maternal serum CRP Cord length (cm) Cesarean section (%) Genital bleeding (%) Premature rupture of membranes (%) Multiple pregnancies (%) Preeclampsia (%) Cerclage (%) Betamethasone exposed (%) TRH exposed (%) Ritodorine exposed (%) Magnesium exposed (%) Indomethacin exposed (%) Chorioamnionitis (%)
cPVL (þ) ðn ¼ 35Þ
cPVL (2) ðn ¼ 166Þ
29.4 ^ 2.1 1324 ^ 324 5^2 7^1 2.3 ^ 2.4 44.8 ^ 12.4 17 37 40 39 0 29 11 3 80 3 46 68
29.6 ^ 2.1 1304 ^ 371 6^2 8^2 2.1 ^ 3.5 42.1 ^ 10.6 20 25 39 27 14 19 16 6 72 17 24 47
Values are shown with ^ SD.
3. Results During the study period, 240 infants were born at gestational ages from 24 to 34 weeks. Of these, 39 infants were excluded because of major anomaly, grade 3 or 4 IVH, or neonatal death within 2 weeks, or insufficient clinical records. The remained 201 were entered into the study and their obstetrical factors were examined retrospectively (Table 1). All of them had survived over the study period. Thirty-five cases diagnosed as cPVL later developed spastic diplegia by transfrontanel ultrasound scanning (mean gestational age 29.4 ^ 2.1 weeks, range from 24.4 to 32.3
19
weeks). A tight popliteal angle, foot joint tightness, and hyperreflexia in the deep tendon reflex were observed in all of them. There are 23 cases of preeclampsia, in which preterm delivery was inevitable because of maternal complications, or growth arrest or distress of the fetuses. The group included no infant suffering from cPVL, which was statistically significant. Further, in the univariate analysis, exposure to antenatal indomethacin, cord length $ 40 cm, and a low Apgar score (, 5 after 1 min of birth or , 7 after 5 min) were significantly associated with a 2– 3 risk increased of cPVL occurrence, while antenatal magnesium sulfate reduced the risk. Chorioamnionitis was positively correlated with the risk, but did not reach statistical significance (Table 2). In the multivariate analysis we also found exposure to antenatal indomethacin, and a low Apgar score (, 7 after 5 min of birth) to be independently associated with cPVL and administration of antenatal magnesium sulfate reduced the risk (Table 3). The remaining factors including cord length $ 40 cm did not reach statistical significance. Preeclampsia could not be included in the multivariate model because no infant suffering from preeclampsia developed cPVL. Apgar score after 1 min of birth was excluded from this analysis since it was strongly related to the score after 5 min and the OR for 1 min score was smaller than that for 5 min score in the univariate analysis.
4. Discussion Our present data showed Apgar score, cord length, indomethacin and magnesium sulfate to be significantly associated with cPVL on univariable analysis of 18 items (Table 2), while multivariate analysis revealed significance
Table 2 Univariate analysis of neonatal and maternal characteristics for cPVL
Gestational week (under 28 weeks) Birth weight (less than 1200 g) Apgar score after 1 min (under 5) Apgar score after 5 min (under 5) CRP (above 3 ng/ml) Cord length (cm) (more than 40 cm) Mode of delivery (transvaginal) Genital bleeding Premature rupture of membranes Multiple pregnancies Preeclampsia Cerclage Steroid exposed TRH exposed Ritodorine exposed Magnesium exposed Indomethacin exposed Chorioamnionitis OR, Odds ratio; CI, confidence intervals.
cPVL (þ ) ðn ¼ 35Þ
cPVL (2) ðn ¼ 166Þ
OR
95% CI
P
8 10 21 21 10 27 6 13 14 6 0 10 4 1 28 1 16 15
39 69 58 51 30 91 33 42 65 45 23 29 29 28 119 28 40 36
0.96 0.56 2.51 3.06 1.81 2.63 0.83 1.74 1.04 0.56 0.00 1.68 0.61 0.46 1.58 0.14 2.65 2.38
0.41–2.30 0.25–1.25 1.18–5.31 1.44–6.51 0.79–4.17 1.13–6.15 0.32–2.17 0.81–3.77 0.49–2.18 0.22–1.43 Not calculable 0.73–3.84 0.20–1.86 0.057– 3.71 0.65–3.86 0.040– 0.95 1.25–5.64 0.873– 6.50
0.94 0.15 0.01 0.003 0.16 0.02 0.71 0.15 0.93 0.22 0.02 0.22 0.38 0.45 0.31 0.03 0.01 0.09
20
Y. Murata et al. / Brain & Development 27 (2005) 17–21
Table 3 Multivariate analysis of neonatal and maternal characteristics for cPVL
Gestational week (under 28 weeks) Birth weight (less than 1200 g) Apgar score after 5 min (under 7) Cord length (cm) ($40 cm) Magnesium exposed Indomethacin exposed
OR
SE
95% CI
1.44 0.31 3.71 1.99 0.058 5.34
0.73 0.66 0.44 0.49 1.1 0.51
0.34–6.041 0.084–1.11 1.58–8.75 0.77–5.16 0.007–0.498 1.96–14.6
OR, Odds ratio; SE, standard error; CI, confidence intervals.
for only Apgar score, indomethacin, and magnesium sulfate. In contrast to previous reports, we could not confirm chorioamnionitis [12], genital bleeding [7], multiple pregnancies, antenatal steroid administration, including other general- and perinatal-related characteristics, as factors associated with cPVL in this series. It is well known that birth asphyxia is related to CP in term births. There may also be a relation to the Apgar score through change in fetal cerebral and systemic circulation, since a long umbilical cord may be compressed easily. Indeed, previous animal experiments and pathological findings of placenta demonstrated that repetitive cord occlusion is associated with the damage of cerebral white matter [13,14]. In contrast to our result for indomethacin, it has been suggested that infants treated with this agent, a prostaglandin synthetase inhibitor used as a tocolytic agent for preterm labor, demonstrate low rates of cranial ultrasonographic abnormalities and IVH [15,16]. However, there have been some case reports of indomethacin association with fetal and neonatal complications such as oligohydroamnios, hyperbilirubinemia, necrotizing enterocolitis and preterm contraction of ductus arteriosus. The mechanism by which indomethacin affects the incidence of intracranial disease is uncertain. The effects of magnesium sulfate on cPVL in cases of preterm labor are controversial [17,18]. In 1995, Nelson et al. [19] reported that fetal exposure to magnesium sulfate reduced the risk of subsequent development of CP among prematurely born infants. However, Mittendorf et al. [20] recently demonstrated adverse effects of antenatal magnesium sulfate in a dose-dependent fashion in a randomized study. In our series, magnesium sulfate was administered at a lower dose (,48 g/day) than conventionally applied and furthermore it was stopped at least 4 h before delivery. A possible mechanism that explains the difference may relate to the observation that pretreatment of relatively low dose magnesium sulfate in utero might reduce the N-methyl-D -aspartate-mediated brain injury [21], but high concentration of magnesium ion might interfere with the coagulation systems in immature newborn brain resulting in development of IVH. Our present data did not show any significant link between any of the remaining clinical characteristics and the development of cPVL. Many published reports in the literature addressing prognosis of preterm birth showed
clinical evidence of CAM was strongly associated with adverse outcomes. We diagnosed CAM by the presence of acute inflammatory cell infiltration of both layers of the membranes. Since many authors categorized clinical CAM including more severe cases than ours, the contradiction between data set might be due to differences in diagnosis criteria. In our present study no cases of preeclampsia developed cPVL, in line with previous studies showing that VLBW infants born to preeclamptic women have a low rate of cerebral hemorrhage [22,23]. Often such cases result in preterm delivery due to maternal complications, growth arrest and/or fetal distress. In our present study none of the preeclamptic mothers was treated with magnesium sulfate. Furthermore, all cases were delivered by cesarean section. It should be noted that the incidence and severity of cPVL are significantly lower in cases where delivery is thus determined by the physician for maternal or fetal reasons. The pathophysiologic mechanisms of cPVL involve acute fluctuation in cerebral blood flow in the preterm fetus and neonate, where autoregulation is impaired. An increase in cerebral blood flow, irrespective of the etiology and timing, predisposes to IVH, whereas a marked reduction can cause infarction and ischemic necrosis, leading to cPVL. Therefore it is probable that one of causes of cPVL is sudden decrease of tonus in the cerebral circulation after exclusion of feto-placental circulation. It is well known that the fetoplacental circulation is mainly autoregulated, and fetal cerebral circulation increases in cases of preeclampsia. A potent candidate for responsible hormone to regulate the fetal circulation is angiotensin II [24] and we have recently confirmed to activate the renin –angiotensin system in the feto-placental unit in preeclampsia [25]. Therefore the mechanism by which preeclampsia exerts a positive impact on the incidence of cPVL, might associated with continuous activation of the rennin – angiotensin system in fetoplacental unit. Further extensive studies for clarification of the mechanisms underlying fetal circulation in preeclampsia will contribute to reduce cPVL. In conclusion, this study clarified possible antenatal and perinatal related factors in development of cPVL. Together with other similar research, our results will help to develop techniques to identify fetuses who are at high risk of developing cPVL, which contribute to reduce LBW infants suffered from CP.
References [1] Maternal and Child Health Division, Children and Families Bureau, Ministry of Health and Welfare, Japan. Maternal and child health statistics of Japan, Tokyo: Kaneda I, Mothers and Children and Family Health Education Group Representative; 2003. [2] Sakamoto S, Terao T. How to lower perinatal mortality? Perinatal care in Japan. Croat Med J 1998;39:197–207. [3] Weindling M. Periventricular haemorrhage and periventricular leukomalacia. Br J Obstet Gynaecol 1995;102:278–81.
Y. Murata et al. / Brain & Development 27 (2005) 17–21 [4] Verma U, Tejani N, Klein S, Reale MR, Beneck D, Figueroa R, et al. Obstetric antecedents of intraventricular hemorrhage and periventricular leukomalacia in the low-birth-weight neonate. Am J Obstet Gynecol 1997;176:275 –81. [5] Volpe JJ. Neurobiology of periventricular leukomalacia in the premature infant. Pediatr Res 2001;50:553 –62. [6] Hagberg B, Hagberg G, Beckung E, Uvebrant P. Changing panorama of cerebral palsy in Sweden. VIII. Prevalence and origin in the birth year period 1991–94. Acta Paediatr 2001;90:271–7. [7] Itakura A, Kurauchi O, Hayakawa F, Matsuzawa K, Mizutani S, Tomoda Y. Timing of periventricular leukomalacia using neonatal electroencephalography. Int J Gynaecol Obstet 1996;55:111–5. [8] Okamura M, Itakura A, Kurauchi O, Hayakawa F, Mizutani S, Tomoda Y. Fetal heart rate pattern associated with periventricular leukomalacia. Int J Gynaecol Obstet 1997;56:13–18. [9] Hayakawa F, Okumura A, Kato T, Kuno K, Watanabe K. Determination of timing of brain injury in preterm infants with periventricular leukomalacia with serial neonatal electroencephalography. Pediatrics 1999;104:1077–81. [10] Okumura A, Hayakawa F, Kato T, Itomi K, Maruyama K, Ishihara N, et al. Hypocarbia in preterm infants with periventricular leukomalacia: the relation between hypocarbia and mechanical ventilation. Pediatrics 2001;107:469 –75. [11] Kubota T, Okumura A, Hayakawa F, Kato T, Itomi K, Kuno K, et al. Relation between the date of cyst formation observable on ultrasonography and the timing of injury determined by serial encephalography in preterm infants with periventricular leukomalacia. Brain Dev 2001;23:390–4. [12] Wu YW, Colford Jr JM. Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. J Am Med Assoc 2000;284:1417– 24. [13] Clapp JF, Peress NS, Wesley M, Mann LI. Brain damage after intermittent partial cord occlusion in the chronically instrumented fetal lamb. Am J Obstet Gynecol 1988;159:504– 9. [14] Kumazaki K, Nakayama M, Sumida Y, Ozono K, Mushiake S, Suehara N, et al. Placental features in preterm infants with periventricular leukomalacia. Pediatrics 2002;109:650– 5. [15] Vohr BR, Allan WC, Westerveld M, Schneider KC, Katz KH, Makuch RW, et al. School-age outcomes of very low birth weight
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
21
infants in the indomethacin intraventricular hemorrhage prevention trial. Pediatrics 2003;111:e340 –6. Ment LR, Vohr B, Allan W, Westerveld M, Sparrow SS, Schneider KC, et al. Outcome of children in the indomethacin intraventricular hemorrhage prevention trial. Pediatrics 2000;105:485 –91. FineSmith RB, Roche K, Yellin PB, Walsh KK, Shen C, Zeglis M. Effect of magnesium sulfate on the development of cystic periventricular leukomalacia in preterm infants. Am J Perinatol 1997;14: 303–7. Canterino JC, Verma UL, Visintainer PF, Figueroa R, Klein SA, Tejani NA. Maternal magnesium sulfate and the development of neonatal periventricular leukomalacia and intraventricular hemorrhage. Obstet Gynecol 1999;93:396–402. Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of cerebral palsy in very low birth weight infants? Pediatrics 1995;95: 263–9. Mittendorf R, Dambrosia J, Pryde PG, Lee KS, Gianopoulos JG, Besinger RE, et al. Association between the use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am J Obstet Gynecol 2002;186:1111–8. Nowak L, Bregestovski P, Ascher P, Herbelt A, Prochiantz A. Magnesium gates glutamate-activated channels in mouse central neurons. Nature 1984;307:462 –5. Leviton A, Kuban KC, Pagano M, Brown ER, Krishnamoorthy KS, Allred EN. Maternal toxemia and neonatal germinal matrix hemorrhage in intubated infants less than 1751 g. Obstet Gynecol 1988;72:571 –6. Gray PH, O’Callaghan MJ, Mohay HA, Burns YR, King JF. Maternal hypertension and neurodevelopmental outcome in very preterm infants. Arch Dis Child Fetal Neonatal Ed 1998;79:F88–F93. Cooper AC, Robinson G, Vinson GP, Cheung WT, Broughton Pipkin F. The localization and expression of the renin-angiotensin system in the human placenta throughout pregnancy. Placenta 1999;20:467– 74. Ito M, Itakura A, Ohno Y, Nomura M, Senga T, Nagasaka T, et al. Possible activation of the renin-angiotensin system in the fetoplacental unit in preeclampsia. J Clin Endocrinol Metab 2002;87: 1871– 8.
Original article
Possible antenatal and perinatal related factors in development of cystic periventricular leukomalacia Yasutaka Murataa, Atsuo Itakurab,*, Katsuji Matsuzawac, Akihisa Okumurad, Kenji Wakaie, Shigehiko Mizutania a
Department of Obstetrics and Gynecology, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan b Maternity and Perinatal Care Center, Nagoya University Hospital, Nagoya, Japan c Department of Obstetrics and Gynecology, Anjo Kosei Hospital, Aichi, Japan d Department of Pediatrics, Nagoya University Graduate School of Medicine, Nagoya, Japan e Division of Epidemiology and Prevention, Aichi Cancer Center Research Institution, Nagoya, Japan Received 7 October 2003; received in revised form 9 February 2004; accepted 19 February 2004
Abstract Cystic periventricular leukomalacia (cPVL), the principal ischemic brain injury in premature infants, is characterized by necrosis of the white matter in the periventricular region and the major neuropathology for spastic motor deficits in cerebral palsy or epilepsy. Recent reports strongly suggest that the brain injury associated with cPVL may have already occurred in utero. In this study we searched retrospectively for possible clinical situations related to cPVL to facilitate assessment of optimal management. A total of 201 babies born at gestational ages from 24 to 33 weeks were entered into the study (1992 – 1997) and examined for involvement of 18 factors in cPVL retrospectively. And psychomotor development was examined at least until 18 months of corrected age. Among 201 premature babies 35 cases were diagnosed as cPVL later developed spastic diplegia. There are 23 cases of preeclampsia, no infant suffering from cPVL. In the univariate analysis, exposure to antenatal indomethacin, cord length $ 40 cm, and a low Apgar score were significantly associated with a 2 – 3 risk increased of cPVL occurrence, while antenatal magnesium sulfate reduced the risk. Chorioamnionitis was positively correlated with the risk, but did not reach statistical significance. In the multivariate analysis we found the statistical significance in exposure to antenatal indomethacin, a low Apgar score, and antenatal magnesium sulfate. Our results suggested that preeclampsia and antenatal exposure of magnesium sulfate reduced the risk while antenatal exposure of indomethacin and low Apgar score associated with the occurrence of cPVL. These findings support a growing consensus that cPVL is often the result of maternal and fetal factors as well as antenatal treatment. q 2004 Elsevier B.V. All rights reserved. Keywords: Periventricular leukomalacia; Perinatal factors; Preeclampsia; Indomethacin; Magnesium sulfate; Cord factor; Apgar score; Case– control study
1. Introduction Preterm birth, which occurs in about 5.3% of all pregnancies in Japan, is a major problem in obstetrics [1]. Although the survival rate has dramatically increased recently with developments in obstetric and neonatal care, neurological sequelae are a subject of increasing interest in perinatal medicine. Although the survival rate of low birth weight (LBW) infants is now over 85% at week 24 or more in hospitals having neonatal intensive care unit in Japan, the morbidity * Corresponding author. Tel.: þ 81-52-744-2261; fax: þ81-52-744-2268. E-mail address: [email protected] (A. Itakura). 0387-7604/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2004.02.011
rate with very LBW (VLBW) infants still remains at about 20% [2]. Two imaging patterns in ultrasonography, intraventricular hemorrhage (IVH) and cystic periventricular leukomalacia (cPVL) have significant sensitivity and specificity in the prediction of neurological alteration [3 – 5]. The principal ischemic brain injury in premature infants; cPVL is characterized by necrosis of the white matter in the periventricular region. It is suggested to be the major neuropathology for spastic motor deficits in cerebral palsy (CP) and is reported to occur in 5 – 17% of all infants born weighing , 1500 g [6]. In addition 66 –100% of infants affected with cPVL are destined to suffer from CP. Although almost all preterm birth infants suffering from CP were once believed to be injured during delivery and/or
18
Y. Murata et al. / Brain & Development 27 (2005) 17–21
in postnatal period, recent reports strongly suggest that the brain injury associated with cPVL may have already occurred in utero [7 – 10]. It is known that both neonatal and maternal conditions affect the risk of cPVL and/or developing CP in VLBW cases. The LBW neonate has an inherent gestational age dependent vulnerability as a consequence of anatomic and hemodynamic immaturity. However, little information is available as to antenatal management and/or clinical conditions of patients’ impact on the risk of cPVL and/or developing CP with VLBW. Although the early changes of periventricular leukomalacia may be apparent histologically within hours of insult, the lesions take at least 2 –6 weeks to before they can be visualized ultrasonographically. Therefore, if seen within 7 days of life, the origins must be intrauterine. This time lag makes determining the cause of the disease complicated [11]. In the present study we searched retrospectively for possible clinical situations related to cPVL to facilitate assessment of optimal management.
2. Subjects and methods We reviewed the maternal neonatal medical records for all appropriate for date babies born at gestational ages from 24 to 33 weeks at Anjo Kosei Hospital, a tertiary center in the central Japan area, between January 1992 and December 1997. The study population included maternal transports with predominant indications for maternal transfers and then all inborn neonates ðn ¼ 240Þ were candidates except for babies ðn ¼ 39Þ with major anomalies, with grade 3 or 4 IVH, or which died within 2 weeks after birth. A total of 201 babies were entered into the study and examined for involvement of antenatal and intrapartum factors in cPVL retrospectively. The antenatal and intrapartum factors were as follows: gestational age, birth weight, cord length, Apgar scores after 1 and 5 min, presence or absence of prenatal genital bleeding, preterm rupture of membranes (PROM), chorioamnionitis (CAM), multiple pregnancy, preeclampsia, and maternal serum CRP on the day of delivery, mode of delivery, whether cerclage was performed or not, kind of tocolytic agent administered (hydrochloric ritodorine, magnesium sulfate, indomethacin), and exposure to drugs for pulmonary maturation (betamethasone, TRH). The gestational ages were all determined by ultrasonographic examination in the first trimester. All of the babies underwent detailed antenatal and intrapartum cardiotocographic monitoring. Cervical cerclage was performed when repeated midtrimester abortion or fetal membranes were bulging without maternal findings for infection or labor pains. Antepartum administration of steroids was recorded for each case. In our series hydrochloric ritodorine was used for preterm labor at first intravenously up to 400 mg/min. When uterine contraction was not controlled or side effects were intolerable, magnesium sulfate and/or indomethacin
were selected at random and administered in addition. Magnesium sulfate was applied intravenously, with a loading dose of 1 –4 g/h and maintenance infusion of about 0.5– 2 g/h, maintaining the maternal serum level in a therapeutic window about 4– 8 mg/dl. Indomethacin was applied as rectal suppository, 25– 50 mg every 6 h for a maximum dose of 200 mg daily. Preeclampsia was diagnosed using the criteria of the American College of Obstetricians and Gynecologists. After delivery all placentas were examined by a pathologist who was blind to all clinical features. CAM was defined by the presence of an acute inflammatory cell infiltration in both layers of the membranes. Neonatal cranial transfontanelle scans were evaluated by an experienced pediatrician. Standard sagital and coronal scans were serially performed at least three times 1 month of life. Ultrasonograms were obtained with a 7.5 MHz transducer (Sonos 1000, Hewlett Packard), and cPVL was defined as formation of cyst more than 3 mm in diameter in periventricular areas on coronal and sagittal views, and included cPVL (þ ). Abnormal scans after these time limits were considered to be related to neonatal events and therefore the cases were not included in the cPVL (þ ) group. The remaining neonates showing no abnormal findings and excluded from cPVL (þ ) were included in the cPVL (– ) group. MRI was performed in all patients with cPVL during late infancy through early childhood. All patients had ventriculomegaly with an irregular margin accompanied by periventricular abnormal high intensity areas on T2-weighed imaging. Psychomotor development was examined every 3 months after discharge at least until 18 months of corrected age. We paid close attention to spasticity of the lower extremities, such as a tight popliteal angle, foot joint tightness, and hyperreflexia in the deep tendon reflex. Spastic diplegia was diagnosed when an infant, displaying spastic gait or inability to walk had signs of spasticity of the lower extremities. Cranial magnetic resonance imaging was performed during late infancy to early childhood, if an infant exhibited psychomotor retardation or spasticity. 2.1. Statistical analysis We used statistical methods for case – control studies, comparing cPVL (þ ) (cases) and cPVL (2 ) (controls) infants. All baseline and pregnancy-related characteristics were assessed as potential risk factors for cPVL during pregnancy or intra-partum with Chi-square, or Fisher’s exact tests. Odds ratios (OR) were estimated to approximate relative risk for cPVL along with 95% confidence intervals. Finally, multivariate analysis was performed including variables that were significant in the univariate analysis, with birth weight and gestational week by the logistic procedure of the Statistical Analysis System (SAS) (SAS Institute Inc, Cary, NC). All P values were two-sided and P less than 0.05 were considered as significant.
Y. Murata et al. / Brain & Development 27 (2005) 17–21 Table 1 Clinical characteristics of the study population
Gestational weeks Birth weight (g) Apgar score after 1 min Apgar score after 5 min Maternal serum CRP Cord length (cm) Cesarean section (%) Genital bleeding (%) Premature rupture of membranes (%) Multiple pregnancies (%) Preeclampsia (%) Cerclage (%) Betamethasone exposed (%) TRH exposed (%) Ritodorine exposed (%) Magnesium exposed (%) Indomethacin exposed (%) Chorioamnionitis (%)
cPVL (þ) ðn ¼ 35Þ
cPVL (2) ðn ¼ 166Þ
29.4 ^ 2.1 1324 ^ 324 5^2 7^1 2.3 ^ 2.4 44.8 ^ 12.4 17 37 40 39 0 29 11 3 80 3 46 68
29.6 ^ 2.1 1304 ^ 371 6^2 8^2 2.1 ^ 3.5 42.1 ^ 10.6 20 25 39 27 14 19 16 6 72 17 24 47
Values are shown with ^ SD.
3. Results During the study period, 240 infants were born at gestational ages from 24 to 34 weeks. Of these, 39 infants were excluded because of major anomaly, grade 3 or 4 IVH, or neonatal death within 2 weeks, or insufficient clinical records. The remained 201 were entered into the study and their obstetrical factors were examined retrospectively (Table 1). All of them had survived over the study period. Thirty-five cases diagnosed as cPVL later developed spastic diplegia by transfrontanel ultrasound scanning (mean gestational age 29.4 ^ 2.1 weeks, range from 24.4 to 32.3
19
weeks). A tight popliteal angle, foot joint tightness, and hyperreflexia in the deep tendon reflex were observed in all of them. There are 23 cases of preeclampsia, in which preterm delivery was inevitable because of maternal complications, or growth arrest or distress of the fetuses. The group included no infant suffering from cPVL, which was statistically significant. Further, in the univariate analysis, exposure to antenatal indomethacin, cord length $ 40 cm, and a low Apgar score (, 5 after 1 min of birth or , 7 after 5 min) were significantly associated with a 2– 3 risk increased of cPVL occurrence, while antenatal magnesium sulfate reduced the risk. Chorioamnionitis was positively correlated with the risk, but did not reach statistical significance (Table 2). In the multivariate analysis we also found exposure to antenatal indomethacin, and a low Apgar score (, 7 after 5 min of birth) to be independently associated with cPVL and administration of antenatal magnesium sulfate reduced the risk (Table 3). The remaining factors including cord length $ 40 cm did not reach statistical significance. Preeclampsia could not be included in the multivariate model because no infant suffering from preeclampsia developed cPVL. Apgar score after 1 min of birth was excluded from this analysis since it was strongly related to the score after 5 min and the OR for 1 min score was smaller than that for 5 min score in the univariate analysis.
4. Discussion Our present data showed Apgar score, cord length, indomethacin and magnesium sulfate to be significantly associated with cPVL on univariable analysis of 18 items (Table 2), while multivariate analysis revealed significance
Table 2 Univariate analysis of neonatal and maternal characteristics for cPVL
Gestational week (under 28 weeks) Birth weight (less than 1200 g) Apgar score after 1 min (under 5) Apgar score after 5 min (under 5) CRP (above 3 ng/ml) Cord length (cm) (more than 40 cm) Mode of delivery (transvaginal) Genital bleeding Premature rupture of membranes Multiple pregnancies Preeclampsia Cerclage Steroid exposed TRH exposed Ritodorine exposed Magnesium exposed Indomethacin exposed Chorioamnionitis OR, Odds ratio; CI, confidence intervals.
cPVL (þ ) ðn ¼ 35Þ
cPVL (2) ðn ¼ 166Þ
OR
95% CI
P
8 10 21 21 10 27 6 13 14 6 0 10 4 1 28 1 16 15
39 69 58 51 30 91 33 42 65 45 23 29 29 28 119 28 40 36
0.96 0.56 2.51 3.06 1.81 2.63 0.83 1.74 1.04 0.56 0.00 1.68 0.61 0.46 1.58 0.14 2.65 2.38
0.41–2.30 0.25–1.25 1.18–5.31 1.44–6.51 0.79–4.17 1.13–6.15 0.32–2.17 0.81–3.77 0.49–2.18 0.22–1.43 Not calculable 0.73–3.84 0.20–1.86 0.057– 3.71 0.65–3.86 0.040– 0.95 1.25–5.64 0.873– 6.50
0.94 0.15 0.01 0.003 0.16 0.02 0.71 0.15 0.93 0.22 0.02 0.22 0.38 0.45 0.31 0.03 0.01 0.09
20
Y. Murata et al. / Brain & Development 27 (2005) 17–21
Table 3 Multivariate analysis of neonatal and maternal characteristics for cPVL
Gestational week (under 28 weeks) Birth weight (less than 1200 g) Apgar score after 5 min (under 7) Cord length (cm) ($40 cm) Magnesium exposed Indomethacin exposed
OR
SE
95% CI
1.44 0.31 3.71 1.99 0.058 5.34
0.73 0.66 0.44 0.49 1.1 0.51
0.34–6.041 0.084–1.11 1.58–8.75 0.77–5.16 0.007–0.498 1.96–14.6
OR, Odds ratio; SE, standard error; CI, confidence intervals.
for only Apgar score, indomethacin, and magnesium sulfate. In contrast to previous reports, we could not confirm chorioamnionitis [12], genital bleeding [7], multiple pregnancies, antenatal steroid administration, including other general- and perinatal-related characteristics, as factors associated with cPVL in this series. It is well known that birth asphyxia is related to CP in term births. There may also be a relation to the Apgar score through change in fetal cerebral and systemic circulation, since a long umbilical cord may be compressed easily. Indeed, previous animal experiments and pathological findings of placenta demonstrated that repetitive cord occlusion is associated with the damage of cerebral white matter [13,14]. In contrast to our result for indomethacin, it has been suggested that infants treated with this agent, a prostaglandin synthetase inhibitor used as a tocolytic agent for preterm labor, demonstrate low rates of cranial ultrasonographic abnormalities and IVH [15,16]. However, there have been some case reports of indomethacin association with fetal and neonatal complications such as oligohydroamnios, hyperbilirubinemia, necrotizing enterocolitis and preterm contraction of ductus arteriosus. The mechanism by which indomethacin affects the incidence of intracranial disease is uncertain. The effects of magnesium sulfate on cPVL in cases of preterm labor are controversial [17,18]. In 1995, Nelson et al. [19] reported that fetal exposure to magnesium sulfate reduced the risk of subsequent development of CP among prematurely born infants. However, Mittendorf et al. [20] recently demonstrated adverse effects of antenatal magnesium sulfate in a dose-dependent fashion in a randomized study. In our series, magnesium sulfate was administered at a lower dose (,48 g/day) than conventionally applied and furthermore it was stopped at least 4 h before delivery. A possible mechanism that explains the difference may relate to the observation that pretreatment of relatively low dose magnesium sulfate in utero might reduce the N-methyl-D -aspartate-mediated brain injury [21], but high concentration of magnesium ion might interfere with the coagulation systems in immature newborn brain resulting in development of IVH. Our present data did not show any significant link between any of the remaining clinical characteristics and the development of cPVL. Many published reports in the literature addressing prognosis of preterm birth showed
clinical evidence of CAM was strongly associated with adverse outcomes. We diagnosed CAM by the presence of acute inflammatory cell infiltration of both layers of the membranes. Since many authors categorized clinical CAM including more severe cases than ours, the contradiction between data set might be due to differences in diagnosis criteria. In our present study no cases of preeclampsia developed cPVL, in line with previous studies showing that VLBW infants born to preeclamptic women have a low rate of cerebral hemorrhage [22,23]. Often such cases result in preterm delivery due to maternal complications, growth arrest and/or fetal distress. In our present study none of the preeclamptic mothers was treated with magnesium sulfate. Furthermore, all cases were delivered by cesarean section. It should be noted that the incidence and severity of cPVL are significantly lower in cases where delivery is thus determined by the physician for maternal or fetal reasons. The pathophysiologic mechanisms of cPVL involve acute fluctuation in cerebral blood flow in the preterm fetus and neonate, where autoregulation is impaired. An increase in cerebral blood flow, irrespective of the etiology and timing, predisposes to IVH, whereas a marked reduction can cause infarction and ischemic necrosis, leading to cPVL. Therefore it is probable that one of causes of cPVL is sudden decrease of tonus in the cerebral circulation after exclusion of feto-placental circulation. It is well known that the fetoplacental circulation is mainly autoregulated, and fetal cerebral circulation increases in cases of preeclampsia. A potent candidate for responsible hormone to regulate the fetal circulation is angiotensin II [24] and we have recently confirmed to activate the renin –angiotensin system in the feto-placental unit in preeclampsia [25]. Therefore the mechanism by which preeclampsia exerts a positive impact on the incidence of cPVL, might associated with continuous activation of the rennin – angiotensin system in fetoplacental unit. Further extensive studies for clarification of the mechanisms underlying fetal circulation in preeclampsia will contribute to reduce cPVL. In conclusion, this study clarified possible antenatal and perinatal related factors in development of cPVL. Together with other similar research, our results will help to develop techniques to identify fetuses who are at high risk of developing cPVL, which contribute to reduce LBW infants suffered from CP.
References [1] Maternal and Child Health Division, Children and Families Bureau, Ministry of Health and Welfare, Japan. Maternal and child health statistics of Japan, Tokyo: Kaneda I, Mothers and Children and Family Health Education Group Representative; 2003. [2] Sakamoto S, Terao T. How to lower perinatal mortality? Perinatal care in Japan. Croat Med J 1998;39:197–207. [3] Weindling M. Periventricular haemorrhage and periventricular leukomalacia. Br J Obstet Gynaecol 1995;102:278–81.
Y. Murata et al. / Brain & Development 27 (2005) 17–21 [4] Verma U, Tejani N, Klein S, Reale MR, Beneck D, Figueroa R, et al. Obstetric antecedents of intraventricular hemorrhage and periventricular leukomalacia in the low-birth-weight neonate. Am J Obstet Gynecol 1997;176:275 –81. [5] Volpe JJ. Neurobiology of periventricular leukomalacia in the premature infant. Pediatr Res 2001;50:553 –62. [6] Hagberg B, Hagberg G, Beckung E, Uvebrant P. Changing panorama of cerebral palsy in Sweden. VIII. Prevalence and origin in the birth year period 1991–94. Acta Paediatr 2001;90:271–7. [7] Itakura A, Kurauchi O, Hayakawa F, Matsuzawa K, Mizutani S, Tomoda Y. Timing of periventricular leukomalacia using neonatal electroencephalography. Int J Gynaecol Obstet 1996;55:111–5. [8] Okamura M, Itakura A, Kurauchi O, Hayakawa F, Mizutani S, Tomoda Y. Fetal heart rate pattern associated with periventricular leukomalacia. Int J Gynaecol Obstet 1997;56:13–18. [9] Hayakawa F, Okumura A, Kato T, Kuno K, Watanabe K. Determination of timing of brain injury in preterm infants with periventricular leukomalacia with serial neonatal electroencephalography. Pediatrics 1999;104:1077–81. [10] Okumura A, Hayakawa F, Kato T, Itomi K, Maruyama K, Ishihara N, et al. Hypocarbia in preterm infants with periventricular leukomalacia: the relation between hypocarbia and mechanical ventilation. Pediatrics 2001;107:469 –75. [11] Kubota T, Okumura A, Hayakawa F, Kato T, Itomi K, Kuno K, et al. Relation between the date of cyst formation observable on ultrasonography and the timing of injury determined by serial encephalography in preterm infants with periventricular leukomalacia. Brain Dev 2001;23:390–4. [12] Wu YW, Colford Jr JM. Chorioamnionitis as a risk factor for cerebral palsy: a meta-analysis. J Am Med Assoc 2000;284:1417– 24. [13] Clapp JF, Peress NS, Wesley M, Mann LI. Brain damage after intermittent partial cord occlusion in the chronically instrumented fetal lamb. Am J Obstet Gynecol 1988;159:504– 9. [14] Kumazaki K, Nakayama M, Sumida Y, Ozono K, Mushiake S, Suehara N, et al. Placental features in preterm infants with periventricular leukomalacia. Pediatrics 2002;109:650– 5. [15] Vohr BR, Allan WC, Westerveld M, Schneider KC, Katz KH, Makuch RW, et al. School-age outcomes of very low birth weight
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
21
infants in the indomethacin intraventricular hemorrhage prevention trial. Pediatrics 2003;111:e340 –6. Ment LR, Vohr B, Allan W, Westerveld M, Sparrow SS, Schneider KC, et al. Outcome of children in the indomethacin intraventricular hemorrhage prevention trial. Pediatrics 2000;105:485 –91. FineSmith RB, Roche K, Yellin PB, Walsh KK, Shen C, Zeglis M. Effect of magnesium sulfate on the development of cystic periventricular leukomalacia in preterm infants. Am J Perinatol 1997;14: 303–7. Canterino JC, Verma UL, Visintainer PF, Figueroa R, Klein SA, Tejani NA. Maternal magnesium sulfate and the development of neonatal periventricular leukomalacia and intraventricular hemorrhage. Obstet Gynecol 1999;93:396–402. Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of cerebral palsy in very low birth weight infants? Pediatrics 1995;95: 263–9. Mittendorf R, Dambrosia J, Pryde PG, Lee KS, Gianopoulos JG, Besinger RE, et al. Association between the use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am J Obstet Gynecol 2002;186:1111–8. Nowak L, Bregestovski P, Ascher P, Herbelt A, Prochiantz A. Magnesium gates glutamate-activated channels in mouse central neurons. Nature 1984;307:462 –5. Leviton A, Kuban KC, Pagano M, Brown ER, Krishnamoorthy KS, Allred EN. Maternal toxemia and neonatal germinal matrix hemorrhage in intubated infants less than 1751 g. Obstet Gynecol 1988;72:571 –6. Gray PH, O’Callaghan MJ, Mohay HA, Burns YR, King JF. Maternal hypertension and neurodevelopmental outcome in very preterm infants. Arch Dis Child Fetal Neonatal Ed 1998;79:F88–F93. Cooper AC, Robinson G, Vinson GP, Cheung WT, Broughton Pipkin F. The localization and expression of the renin-angiotensin system in the human placenta throughout pregnancy. Placenta 1999;20:467– 74. Ito M, Itakura A, Ohno Y, Nomura M, Senga T, Nagasaka T, et al. Possible activation of the renin-angiotensin system in the fetoplacental unit in preeclampsia. J Clin Endocrinol Metab 2002;87: 1871– 8.