Prenatal risk stratification of severe small-for

http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–5 ! 2015 Informa UK Ltd. DOI: 10.3109/14767058.2015.1049147

J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by National Center for Child Health & Development on 06/04/15 For personal use only.

ORIGINAL ARTICLE

Prenatal risk stratification of severe small-for-gestational-age infants: a Japanese multicenter study Aya Yoshida1, Nagayoshi Umehara1, Jun Sasahara2, Katsusuke Ozawa3, Kiyotake Ichizuka4, Kei Tanaka5, Tomohiro Tanemoto6, Hiroshi Ishikawa7, Takeshi Murakoshi8, Kenji Kiyoshi9, Mari S. Oba10, Keisuke Ishii2, and Haruhiko Sago1 1

Center of Maternal-Fetal, Neonatal and Reproductive Medicine, National Center for Child Health and Development, Tokyo, Japan, 2Department of Maternal and Fetal Medicine, Osaka Medical Center and Research Institute for Maternal and Child Health, Osaka, Japan, 3Department of Maternal and Fetal Medicine, Miyagi Children’s Hospital, Sendai, Japan, 4Department of Obstetrics and Gynecology, Showa University, Tokyo, Japan, 5 Department of Obstetrics and Gynecology, Kyorin University, Tokyo, Japan, 6Department of Obstetrics and Gynecology, The Jikei University School of Medicine, Tokyo, Japan, 7Department of Obstetrics and Gynecology, Kanagawa Children’s Medical Center, Yokohama, Japan, 8Department of Obstetrics, Seirei Hamamatsu General Hospital, Hamamatsu, Japan, 9Department of Obstetrics, Hyogo Prefectural Kobe Children’s Hospital, Kobe, Japan, and 10Department of Biostatics, Yokohama City University, Yokohama, Japan Abstract

Keywords

Objective: To establish a prenatal prognostic classification of severe small-for-gestational-age (SGA) infants based on gestational age and fetal findings. Methods: A retrospective cohort study of 366 singleton infants (birth weight 53rd percentile) delivered between 22 and 34 weeks’ gestation at nine tertiary perinatal centers. A decision tree model was developed for the prediction of death or severe morbidity. Results: There were 35 infants with poor outcome. Prematurity was the most powerful factor in those born before 27.9 weeks’ gestation, while oligohydramnios was the most powerful factor in those born at 27.9 weeks or after. The rate of poor outcome in infants born before 25.1 weeks, between 25.1 and 27.9 weeks, at 27.9 weeks or after with oligohydramnios, at 27.9 weeks or after without oligohydramnios, was 53.9%, 18.2%, 13.6% and 3.2%, respectively. Conclusions: Risk stratification based on gestation of 25 weeks, 28 weeks and oligohydramnios may aid in prognosis of severe SGA infants.

Oligohydramnios, prognosis, risk stratification, SGA

Introduction Small-for-gestational-age (SGA) infants are associated with neonatal mortality and morbidity, such as neurodevelopmental complications [1,2]. Severe SGA infants (birth weight53 rd percentile) are usually a result of fetal growth restriction (FGR) [3]. Such infants have a ninefold increased risk of neonatal mortality compared to those with birth weight within the 25th–50th percentile [4]. There are several studies on the association between intrauterine factors, such as abnormal ultrasonography findings, and mortality in severe SGA infants [3,5]. The outcome of SGA infants due to placental insufficiency could be improved by appropriate diagnosis and timing of delivery [6]. However, prenatal factors that are critical for predicting perinatal mortality and morbidity remain undetermined [7,8]. While the outcomes of extremely preterm and very low birth weight infants have no doubt been improved with Address for correspondence: Haruhiko Sago, M.D., PhD, Head, Center of Maternal-Fetal, Neonatal and Reproductive Medicine, National Center for Child Health and Development, 2-10-1 Okura, Setagaya-ku, Tokyo 157-8535, Japan. Tel: +81-3-3416-0181. Fax: +81-3-3416-2222. E-mail: [email protected]

History Received 3 December 2014 Accepted 5 May 2015 Published online 4 June 2015

advances in medical care [9,10], gestational age can significantly affect perinatal mortality [11]. It is essential to assess the value of gestational age and prenatal prognostic factors, such as Doppler flow parameters, amniotic fluid volume and cardiotocography, in the prenatal management of severe SGA infants. In addition, the clarification of the outcomes of severe SGA infants who are under perinatal care will help in the area of prenatal counseling. We conducted a retrospective cohort study of severe SGA infants to establish a prenatal prognostic classification of severe FGR based on a combination of gestational age and fetal ultrasonography findings.

Methods This was a retrospective cohort study of severe SGA singleton infants (less than the 3rd percentile according to the Japanese standard reference for birth weight [12]) without congenital abnormality delivered from 22 weeks to 34 weeks of gestation between January 1, 2006 and December 31, 2010. This study was conducted in nine tertiary perinatal centers in Japan and approved by the Ethics Committee of the National Center for Child Health and Development.


2

A. Yoshida et al.

We enrolled 366 SGA infants who met our inclusion criteria. Three infants were excluded from the analysis because of lost to follow-up. Finally, 363 infants were analyzed in the present study. A ‘‘poor’’ outcome was defined as death or complication with severe morbidity at 28 days after birth. Severe morbidity was defined as severe brain damage, such as intraventricular hemorrhage Grades 3–4 and/ or periventricular leukomalacia detected by ultrasonography or magnetic resonance imaging. Otherwise, an infant was deemed as ‘‘good’’ outcome. Between infants with poor and good outcomes, we compared maternal baseline characteristics (age, parity and type of assisted reproductive technology pregnancy), medical complications (diabetes mellitus, hypertension and autoimmune disease), medication during pregnancy (steroid, ritodrine and MgSO4), pregnancy complications (pregnancyinduced hypertension, clinical chorioamnionitis, gestational diabetes mellitus, premature rupture of membranes, abruptio placenta and placenta previa), fetal findings (oligohydramnios, abnormal umbilical artery (UA) Doppler flow, growth arrest and abnormal fetal heart rate pattern) and birth outcome (gestational age at delivery, birth weight, method of delivery and Apgar score at 1 min, 5 min). Clinical chorioamnionitis was diagnosed using Lencki’s criteria (maternal fever, clinical signs and leucocytosis) [13]. Oligohydramnios was defined as an amniotic fluid index 5.0 or maximum vertical pocket of 2.0 cm. Abnormal UA Doppler flow was defined as absent or reversed end-diastolic flow. Growth arrest was defined as an increase of estimated fetal weight of less than 5% over a period of two weeks. Abnormal fetal heart rate pattern was defined as fetal bradycardia, loss of variability, repeated severe variable deceleration, repeated late deceleration or prolonged deceleration in cardiotocography. To assess the association of prenatal factors and the outcomes in infants, we performed univariate analyses. Continuous variables were compared by Wilcoxon’s rank sums test. Categorical variables were compared by Chisquared test or Fisher’s exact test if appropriate. We performed recursive partitioning analyses using a partition procedure of the JMP software, version 11 (SAS Institute Inc., Cary, NC) for the development of a decision tree model for the prediction of poor infant outcome. Patients in each node were partitioned into two descendant based on the Chi-squared statistics until the statistically significant difference disappeared. Clinically relevant factors found to be associated with poor outcome (p50.10) by univariate analyses were used as predictors. The rate of poor infants for each group classified according to the decision tree was calculated. All analyses were conducted using JMP. A value of p 0.05 was considered statistically significant.

Results Figure 1 is a schematic description of the study population. There were 328 infants with good outcome. Among the 35 poor outcome infants (9.6%), 20 had severe morbidity and 15 died. Figure 2 shows the outcome by birth weight and gestational age at delivery. There is a strong correlation between birth weight and gestational age at delivery.

J Matern Fetal Neonatal Med, Early Online: 1–5

Figure 1. Schematic description of study population.

Figure 2. Outcome by birth weight and gestational age at delivery. There is a strong correlation between birth weight and gestational age at delivery. Correlation coefficient is 0.88.

Table 1 shows the comparison of maternal characteristics, medication during pregnancy, pregnancy complication, prenatal findings, pregnancy outcomes and neonatal outcomes between infants with poor outcomes and those with good outcomes. There was no significant difference in maternal characteristics, medication during pregnancy and pregnancy complications between the two groups. In the prenatal findings, the prevalence of oligohydramnios and abnormal UA Doppler flow was significantly higher in the poor outcome group than in the good outcome group, and there was a marginally significant difference in the prevalence of abnormal fetal heart rate pattern between the two groups; however, there was no significant difference in the prevalence

Small-for-gestational-age risk stratification

DOI: 10.3109/14767058.2015.1049147

3

Table 1. Maternal demographics, perinatal characteristics and birth outcomes. Variables Maternal characteristics Age (yrs, mean ± SD) Multi-parity (n, %) Assisted reproductive technology pregnancy (n, %)


Medical complications Diabetes mellitus (n/total N, %) Hypertension (n, %) Autoimmune disease (n, %) Medication during pregnancy Maternal steroid administration (n, %) Maternal ritodrine administration (n/total N, %) Maternal MgSO4 administration (n, %)

Poor outcome N ¼ 35

Good outcome N ¼ 328

p value

33.0 ± 5.2 15 (42.9) 2 (5.7)

33.2 ± 5.1 120 (36.6) 35 (10.7)

0.93 0.47 0.56

0 (0.0) 6 (17.1) 1 (2.9)

5/327 (1.5) 42 (12.8) 10 (3.0)

1.00 0.47 1.00

16 (45.7) 8 (22.9) 9 (25.7)

178 (54.3) 74/326 (22.7) 115 (35.1)

0.33 0.98 0.27

Pregnancy complications Pregnancy-induced hypertension (n/total N, %) Clinical chorioamnionitis (n/total N, %)* Gestational diabetes mellitus (n, %) Premature rupture of membranes (n/total N, %) Abruptio placenta (n, %) Placenta previa (n/total N, %)

14 1/34 0 3 3 0/34

(40.0) (2.9) (0.0) (8.6) (8.6) (0.0)

154/327 2/323 2 17/326 20 3/325

(47.1) (0.6) (0.6) (5.2) (6.1) (0.9)

0.42 0.26 1.00 0.43 0.48 1.00

Prenatal findings Oligohydramniosy (n/total N, %) Abnormal UA Doppler flowz (n/total N, %) Growth arrestô (n/total N, %) Abnormal fetal heart rate patternx (n/total N, %)

16 22/34 7/32 26

(45.7) (64.7) (21.9) (74.3)

77/319 132/307 85/306 183/320

(24.1) (43.0) (27.8) (57.2)

0.006 0.02 0.48 0.051

Pregnancy outcomes Gestational age at delivery (wks, median range) Cesarean section (n, %)

27.7 (24.1–33.7) 35 (100)

30.4 (23.6–33.9) 322 (98.2)

50.001 1.00

Neonatal outcomes Birth weight (g, median range) Neonatal sex (male, %) Apgar score 1 min (median range) Apgar score 5 min (median range)

600 15 3 7

898 167 6 8

50.001 0.36 50.001 0.001

(218–1184) (42.9) (0–8) (1–10)

(332–1558) (50.9) (1–9) (1–10)

*Diagnosed by Lencki’s criteria. yDiagnosed by amniotic fluid index 5.0 or maximum vertical pocket 2.0 cm. zDiagnosed by absent or reversed end-diastolic flow in the UA. ôDiagnosed by estimated fetal weight increasing less than 5% for 2 weeks. xDiagnosed by fetal bradycardia, loss of variability, repeated severe variable deceleration, late deceleration or prolonged deceleration. UA, umbilical artery.

of growth arrest. In terms of pregnancy outcomes, the gestational age at delivery was significantly earlier in the poor outcome group than in the good outcome group, but there was no significant difference for cesarean section. In the neonatal outcome, birth weight was significantly lower in the poor outcome group than in the good outcome group. Both Apgar scores at 1 min and 5 min were significantly lower in the poor outcome group than in the good outcome group. Figure 3 shows the decision tree model for the prediction of poor outcome. Population was partitioned interactively and terminated in cases of statistically insignificant p values. Gestational age at delivery was the most powerful factor; at 28 days after birth the rate of poor outcome in infants born before 27.9 weeks’ gestation was 24.1%, while that in infants born 27.9 weeks or after was 5.6%. In those who were born at 27.9 weeks of gestation or after, oligohydramnios was the most powerful factor. Within this group, the rate of poor outcome in those with normal amniotic fluid volume was 3.2%. Before 27.9 weeks of gestation, prematurity was the most powerful factor. The rate of poor infant outcome with delivery gestation age below 25.1 weeks was 53.9%.

Discussion In this study, we showed that gestational age is the most significant factor for poor outcomes in severe SGA infants born before 27.9 weeks. For those born at 27.9 weeks or after, oligohydramnios, however, was the strongest indicator. Risk stratification based on gestation of 25 weeks, 28 weeks and oligohydramnios would be useful in the prognosis of severe SGA infants. Birth weight and gestational age at delivery were associated with an increased risk of neonatal mortality and morbidity [9]. A recent large Japanese cohort study of very low birth weight (51500 g) infants reported the mortality rate of those with a gestational age 22–24 weeks, 25–27 weeks and 28 weeks or over was 30%, 10% and less than 5%, respectively [10]. On the other hand, very preterm (32 weeks gestation) SGA infants with birth weight less than the fifth percentile have higher neonatal morbidity and mortality compared to infants of similar gestational age without SGA [14]. In the present study, the rate of poor outcome for those with gestational age below 25.1 weeks, 25.1–27.9 weeks and


4

A. Yoshida et al.

J Matern Fetal Neonatal Med, Early Online: 1–5

Figure 3. Risk stratification of poor infant outcome obtained by decision tree analysis.

28 weeks or over was about 50%, 20% and 5%, respectively. In severe SGA infants who were delivered before 27.9 weeks’ gestation, the rate of poor outcome was higher than infants of similar gestational age without SGA. However, in severe SGA infants who were delivered at 27.9 weeks of gestation or after, the rate of poor outcome was no different from that of similar gestational age infants without SGA. We found that oligohydramnios was strongly associated with perinatal mortality and morbidity among severe SGA infants who were delivered after at 27.9 weeks of gestation or after. The amniotic fluid volume reflects fetal renal perfusion and is an indirect measure of fetal circulation status [15]. The FGR fetus responds to sustained hypoxemia by selective distribution of cardiac output, with preferential flow directed to the brain, heart, adrenals and placenta at the expense of all other organ systems [16]. The reduction in renal blood flow is thought to cause oligohydramnios [15]. In term infants and preterm infants without FGR, oligohydramnios is not a predictor of morbidity and mortality [17,18]. In infants with FGR, however, oligohydramnios was reported to be associated with perinatal mortality [19–21]. Oligohydramnios associated with FGR implies a further deterioration of intrauterine environment under placental dysfunction. The reversed end-diastolic flow in UA is associated with adverse perinatal outcome, and the measurement of the UA Doppler flow is used for the management of high-risk pregnancies to improve perinatal outcomes [3]. The cause of FGR, which is suspected to be placental insufficiency, is characterized by abnormal UA flow [22]. In the present study, the abnormal UA Doppler flow defined as absent or reversed end-diastolic flow was associated with poor infant outcomes in univariate analysis, but was not associated with the poor infant outcomes in decision tree analysis. Abnormal UA flow

may be a background character of severe SGA arising from FGR [3]. There are several limitations in this study. The Doppler parameters (middle cerebral artery Doppler, cerebroplacental ratio, ductus venosus (DV) Doppler and aortic isthmus Doppler) could be useful in determining the timing of delivery [3,5]. In a recent study, reverse DV flow allowed stratification between high- and low-risk perinatal mortality between 26 and 28 weeks of gestation [5]. However, we could not evaluate these Doppler parameters because we could not obtain sufficient data due to the retrospective nature of this study. A prospective study that includes measurement of these Doppler parameters is required to contribute to the knowledge of management of severe SGA. In this study, intrauterine fetal death (IUFD) was not considered, even though the risk of IUFD in severe SGA fetuses was reported to be more than 15 times higher compared to non-SGA fetuses [23]. To consider IUFD when assessing prenatal risk factors of FGR, information on timing of death and actual birth weight without factoring in in utero edema is required. However, it is difficult to determine either. In conclusion, this retrospective cohort study showed that prematurity was the most significant factor for neonatal mortality and morbidity in severe SGA infants born before 27.9 weeks’ gestation. In those who were born at 27.9 weeks’ gestation or after, oligohydramnios was the strongest indicator.

Acknowledgements We would like to thank Dr Julian Tang of the Department of Education for Clinical Research, National Center for Child Health and Development, for proofreading, editing and rewriting parts of this manuscript.

DOI: 10.3109/14767058.2015.1049147

Declaration of interest The authors report no declarations of interest. This work was supported by the Grant of National Center for Child Health and Development 26-30 of Japan.


References 1. O’Callaghan ME, MacLennan AH, Gibson CS, et al. Epidemiologic associations with cerebral palsy. Obstet Gynecol 2011;118:576–82. 2. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8. 3. Figueras F, Gratacos E. Stage-based approach to the management of fetal growth restriction. Prenat Diagn 2014:34:655–9. 4. Grisaru-Granovsky S, Reichman B, Lerner-Geva L, et al. Mortality and morbidity in preterm small-for-gestational-age infants: a population-based study. Am J Obstet Gynecol 2012;206:150 e1–7. 5. Cruz-Lemini M, Crispi F, Van Mieghem T, et al. Risk of perinatal death in early-onset intrauterine growth restriction according to gestational age and cardiovascular Doppler indices: a multicenter study. Fetal Diagn Ther 2012;32:116–22. 6. Baschat AA. Pathophysiology of fetal growth restriction: implications for diagnosis and surveillance. Obstet Gynecol Surv 2004;59: 617–27. 7. Group GS. A randomised trial of timed delivery for the compromised preterm fetus: short term outcomes and Bayesian interpretation. BJOG 2003;110:27–32. 8. Walker DM, Marlow N, Upstone L, et al. The Growth Restriction Intervention Trial: long-term outcomes in a randomized trial of timing of delivery in fetal growth restriction. Am J Obstet Gynecol 2011;204:34 e31–9. 9. Fanaroff AA, Stoll BJ, Wright LL, et al. Trends in neonatal morbidity and mortality for very low birthweight infants. Am J Obstet Gynecol 2007;196:147 e141–8. 10. Kusuda S, Fujimura M, Uchiyama A, et al.. Trends in morbidity and mortality among very-low-birth-weight infants from 2003 to 2008 in Japan. Pediatr Res 2012;72:531–8.

Small-for-gestational-age risk stratification

5

11. Baschat AA, Cosmi E, Bilardo CM, et al. Predictors of neonatal outcome in early-onset placental dysfunction. Obstet Gynecol 2007;109:253–61. 12. Itabashi K, Miura F, Uehara R, Nakamura Y. New Japanese neonatal anthropometric charts for gestational age at birth. Pediatr Int 2014;56:702–8. 13. Lencki SG, Maciulla MB, Eglinton GS. Maternal and umbilical cord serum interleukin levels in preterm labor with clinical chorioamnionitis. Am J Obstet Gynecol 1994;170:1345–51. 14. Aucott SW, Donohue PK, Northington FJ. Increased morbidity in severe early intrauterine growth restriction. J Perinatol 2004;24:435–40. 15. Galan HL. Timing delivery of the growth-restricted fetus. Semin Perinatol 2011;35:262–9. 16. Cohn HE, Sacks EJ, Heymann MA, Rudolph AM. Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Am J Obstet Gynecol 1974;120:817–24. 17. Rossi AC, Prefumo F. Perinatal outcomes of isolated oligohydramnios at term and post-term pregnancy: a systematic review of literature with meta-analysis. Eur J Obstet Gynecol Reprod Biol 2013;169:149–54. 18. Melamed N, Pardo J, Milstein R, et al. Perinatal outcome in pregnancies complicated by isolated oligohydramnios diagnosed before 37 weeks of gestation. Am J Obstet Gynecol 2011;205:241 e241–6. 19. Apel-Sarid L, Levy A, Holcberg G, Sheiner E. Placental pathologies associated with intra-uterine fetal growth restriction complicated with and without oligohydramnios. Arch Gynecol Obstet 2009;280:549–52. 20. Chauhan SP, Taylor M, Shields D, et al. Intrauterine growth restriction and oligohydramnios among high-risk patients. Am J Perinatol 2007;24:215–21. 21. Scifres CM, Stamilio D, Macones GA, Odibo AO. Predicting perinatal mortality in preterm intrauterine growth restriction. Am J Perinatol 2009;26:723–8. 22. Savchev S, Figueras F, Sanz-Cortes M, et al. Evaluation of an optimal gestational age cut-off for the definition of earlyand late-onset fetal growth restriction. Fetal Diagn Ther 2014;36: 99–105. 23. Pilliod RA, Cheng YW, Snowden JM, et al. The risk of intrauterine fetal death in the small-for-gestational-age fetus. Am J Obstet Gynecol 2012;207:318 e311–16.