Advertisement

Placental Histological Features and Neurodevelopmental Outcomes at Two Years in Very-Low-Birth-Weight Infants

      Abstract

      Background

      We evaluated the rates of placental pathologic lesions and their relationship with two-year neurodevelopmental outcomes in very-low-birth-weight (VLBW) infants.

      Methods

      This is a cohort observational study comprising 595 VLBW infants during 2007 to 2015. Neurodevelopmental assessment was carried out at 24 months corrected age.

      Results

      In univariate analysis the rates of survival with normal neurodevelopmental outcomes were lower in pregnancies with severe histologic chorioamnionitis (38 of 43, 88.4% when compared with 305 of 450, 67.8%), severe maternal vascular malperfusion (MVM) (17 of 37, 45.9% when compared with 326/492, 66.3%), and intravillous hemorrhage (37 of 82, 45.1% when compared with 306 of 449, 68.1%). In logistic models, severe MVM (adjusted odds ratio [adj. OR] = 0.45, 95% confidence interval [CI] = 0.22 to 0.92), severe fetal vascular malperfusion (FVM) (adj. OR = 0.46, 95% CI = 0.22 to 0.45), and intravillous hemorrhage (adj. OR = 0.38, 95% CI = 0.22 to 0.62) were associated with lower rates of infant survival with normal neurodevelopmental outcome. FVM (adj. OR = 0.46, 95% CI = 0.21 to 0.97) and intravillous hemorrhage (adj. OR = 0.37, 95% CI = 0.22 to 0.62) were also the only placental lesions that were independent predictors of a lower rate of intact survival in stepwise analysis for prognostic factors of the entire cohort.

      Conclusions

      Placental pathologic findings such as severe MVM, FVM, and intravillous hemorrhage are significant predictors of neonatal survival and subsequent adverse neurodevelopmental outcomes. Data on the placental pathology could be useful in the neurodevelopmental follow-up of VLBW infants.

      Keywords

      Introduction

      Very-low-birth-weight infants (VLBW) account for approximately 1% to 1.5% of all births but are responsible for a considerable portion of neurodevelopmental impairments among survivors.
      • Blencowe H.
      • Cousens S.
      • Chou D.
      • et al.
      Born too soon: the global epidemiology of 15 million preterm births.
      Meta-analytic reviews suggest that among VLBW survivors, the pooled prevalence of cerebral palsy and cognitive and motor delays are 6.8%, 16.9%, and 20.6%, respectively.
      • Pascal A.
      • Govaert P.
      • Oostra A.
      • Naulaers G.
      • Ortibus E.
      • Van der Broeck C.
      Neurodevelopmental outcome in very preterm and very-low-birthweight infants born over the past decade: a meta-analytic review.
      Although lower gestational age and birth weight are by far the most predictive factors of neonatal death and abnormal infant neurodevelopmental outcomes, other antenatal and postnatal variables play a key role in the neonatal prognosis.
      • Chevallier M.
      • Debillon T.
      • Pierrat V.
      • et al.
      Leading causes of preterm delivery as risk factors for intraventricular hemorrhage in very preterm infants: results of the EPIPAGE 2 cohort study.
      Among these, etiology of VLBW, antenatal steroid and magnesium sulfate use, pregnancy, and neonatal complications have been associated with adverse infant outcomes.
      • Delorme P.
      • Kayem G.
      • Lorthe E.
      • et al.
      Neurodevelopment at age 2 and umbilical artery Doppler in cases of preterm birth after prenatal hypertensive disorder or suspected fetal growth restriction: the EPIPAGE 2 prospective population-based cohort study.
      • Mukhopadhyay S.
      • Puopolo K.M.
      • Hansen N.I.
      • et al.
      Impact of early-onset sepsis and antibiotic use on death or survival with neurodevelopmental impairment at 2 years of age among extremely preterm infants.
      • Gardella B.
      • Iacobone A.D.
      • Bogliolo S.
      • et al.
      Obstetric risk factors and time trends of neurodevelopmental outcome at 2 years in very-low-birthweight infants: a single institution study.
      • Bauer S.E.
      • Schneider L.
      • Lynch S.K.
      • Malleske D.T.
      • Shepherd E.G.
      • Nelin L.D.
      Factors associated with neurodevelopmental impairment in bronchopulmonary dysplasia.
      Most of the pathophysiologic events leading to VLBW births occur at the placental level; choriodecidual inflammation, maternal and FVM, loss of placental integrity, and villitis are placental pathologic lesions involved in the pathogenesis of preterm delivery and fetal growth restriction (FGR).
      • Raghavan R.
      • Helfrich B.B.
      • Cerda S.R.
      • et al.
      Preterm birth subtypes, placental pathology findings, and risk of neurodevelopmental disabilities during childhood.
      ,
      • Sun C.
      • Groom K.M.
      • Oyston C.
      • Chamley L.W.
      • Clark A.R.
      • James J.L.
      The placenta in fetal growth restriction: what is going wrong?.
      Surprisingly, with the only possible exception of chorioamnionitis, studies on placental lesions of VLBW infants are mostly incomplete, represented by small case series or case-control investigations.
      • Roescher A.M.
      • Timmer A.
      • Erwich J.J.
      • Bos A.F.
      Placental pathology, perinatal death, neonatal outcome, and neurological development: a systematic review.
      The role of chorioamnionitis as an adverse factor in developmental outcomes of VLBW is controversial,
      • Ylijoki M.
      • Lehtonen L.
      • Lind A.
      • et al.
      Chorioamnionitis and five-year neurodevelopmental outcome in preterm infants.
      • Xing L.
      • Wang G.
      • Chen R.
      • Ren J.
      • Qian J.
      • Huang Y.
      Is chorioamnionitis associated with neurodevelopmental outcomes in preterm infants? A systematic review and meta-analysis following PRISMA.
      • Bierstone D.
      • Wagenaar N.
      • Gano D.L.
      • et al.
      Association of histologic chorioamnionitis with perinatal brain injury and early childhood neurodevelopmental outcomes among preterm neonates.
      whereas FVM and abruptio placentae have been associated with early markers of infant brain damage.
      • Saleemuddin A.
      • Tantbirojn P.
      • Sirois K.
      • et al.
      Obstetric and perinatal complications in placentas with fetal thrombotic vasculopathy.
      ,
      • Vik T.
      • Redline R.
      • Nelson K.B.
      • et al.
      The placenta in neonatal encephalopathy: a case-control study.
      The recognition of placental features associated with neurodevelopmental impairment of VLBW could be very useful to identify newborns requiring proper neurodevelopmental follow-up due to an increased risk of developing neurodevelopmental impairments. The purpose of this study was to evaluate the association between placental pathologic lesions, neonatal mortality, and neurodevelopmental outcomes at age two years in a cohort of VLBW infants.

      Methods

      The study population included all VLBW infants delivered consecutively at our institution between 2007 and 2015. The trial was approved by ethical committee of Foundation IRCCS Policlinico San Matteo. The eligibility criteria included (1) absence of fetal malformations, known chromosomal anomalies, or congenital infections; (2) enrollment for prenatal care during the first trimester of pregnancy with ultrasound-confirmed gestational age; and (3) delivery of a viable fetus ≤1500 g and ≥24 weeks of pregnancy. Demographic data of participants were collected upon enrollment in the study. Clinical data were collected at discharge and were stored together with the pathology data in a computer database.

      Obstetrics and fetal variables

      FGR was diagnosed based on longitudinal evaluation of fetal abdominal circumference, estimated fetal weight, and Doppler pulsatility index of umbilical artery according to standard criteria.
      • Lees C.C.
      • Stampalija T.
      • Baschat A.A.
      • et al.
      ISUOG Practice Guidelines: diagnosis and management of small-for-gestational-age fetus and fetal growth restriction.
      When feasible, a conservative management plan for FGR was adopted according to a predefined protocol, which included antenatal visits, ultrasonographic surveillance, and cardiotocographic monitoring. Ultrasonographic evaluation included weekly monitoring of amniotic fluid volume and fetal biometry fortnightly. Doppler studies of fetal circulation included weekly or biweekly measurements of blood flow velocity waveforms of the umbilical artery, middle cerebral artery, and ductus venosus depending on the severity of blood flow abnormalities. Amniotic fluid volume was expressed using the amniotic fluid index and measured sonographically as the sum of the deepest vertical pool in each of the four maternal abdominal quadrants. Clinical chorioamnionitis was defined as the presence of maternal fever (>38 C) in addition to two other signs such as maternal (>100/min) or fetal tachycardia (>160/min), uterine tenderness, or foul/purulent amniotic fluid.
      • Tita A.T.
      • Andrews W.W.
      Diagnosis and management of clinical chorioamnionitis.
      During the period of the study, preterm labor and preterm premature rupture of membranes (PPROM) were treated according to a definite protocol including the maternal use of atosiban and nifedipine for 48 hours as tocolytic drugs among women with intact membranes, betamethasone up to 36 weeks of pregnancy irrespective of the cause of prematurity, and magnesium sulfate for neonatal neuroprotection up to 34 weeks of pregnancy. For women diagnosed with PPROM without other complications, a standard approach with initial hospitalization, expectant management, and erythromycin or penicillin administration for at least 10 days was used. When feasible, women were followed weekly as outpatients up to 36 weeks of pregnancy. Both outpatient and in-hospital ultrasonographic evaluations were carried out by the same group of obstetricians who were experienced in perinatal medicine. To evaluate the severity of the growth deficit, the birth weight z score was computed for each case by comparing the actual birth weight with the standard Italian birth weight for gestational age, adjusted for maternal parity and fetal sex. Preeclampsia was diagnosed according to standard criteria.
      ACOG Practice Bulletin No
      202 summary: gestational hypertension and preeclampsia.

      Pathologic variables

      Pathologists were unaware of FGR or other maternal diseases; only gestational age was known at the time of evaluation/revision of the specimens.
      The guidelines used for the release of placentas to the pathology unit were those suggested by Langston et al.
      • Langston C.
      • Kaplan C.
      • Macpherson T.
      • et al.
      Practice guideline for examination of the placenta: developed by the Placental Pathology Practice Guideline Development Task Force of the College of American Pathologists.
      Histologic lesions assessed and classified according to the Amsterdam Placental Workshop Group Consensus statement
      • Khong T.Y.
      • Mooney E.E.
      • Ariel I.
      • et al.
      Sampling and definitions of placental lesions: Amsterdam placental workshop group consensus statement.
      ,
      • Redline R.W.
      Classification of placental lesions.
      were placental inflammatory immune processes including maternal and fetal acute inflammatory responses (chorioamionitis stage and grade); noninfectious chronic villitis (villitis of unknown origin) distinguishing between low grade (more than one focus, each involving < 10 contiguous villi, either focal if seen on only one slide or multifocal if seen on two or more slides) and high grade (patchy in case of multiple foci, at least one affecting >10 contiguous villi; diffuse when more than 30% distal villi were involved); FVM including obstructive lesions of the umbilical cord and proximal (chorionic or stem vessels) villous thrombosis; avascular villi, intramural fibrin deposition, and karyorrhexis reserving the term high-grade FVM for severe cases (≥45 avascular villi in three disk sections, two or more thrombi in the proximal fetal vessels, or multiple nonocclusive thrombi); maternal vascular malperfusion (MVM) lesions including accelerated villous maturation and distal villous hypoplasia and decidual vasculopathy (persistence of arterial smooth muscle, fibrinoid necrosis, atherosis, absence of spiral artery remodeling); and loss of integrity of placenta, which is a subcategory of MVM and includes infarcts; acute and chronic abruptio and marginal abruptio was classified separately by the other causes of MVM. Finally, MVM was defined as severe in the case of multiple lesions involving >30% of placental parenchyma. Given its recently evidenced strong relationship with fetal death,
      • Jaiman S.
      • Romero R.
      • Pacora P.
      • et al.
      Disorders of placental villous maturation in fetal death.
      a histologic diagnosis of intravillous hemorrhage whether or not being associated with placental abruption, was evaluated as a single pathologic category. The fixed placental weight was recorded after trimming extraplacental membranes and removing the umbilical cord. Placental volumes were derived from measurements of major and minor diameters and thickness (largest diameter × smallest diameter × thickness). Placental area was calculated by assuming an elliptical surface (largest diameter × smallest diameter × π/4). To evaluate placental efficiency (grams of the fetus produced per gram of placenta) we used fetal/placental weight ratio.

      Neonatal data and follow-up

      Perinatal and neonatal data were recorded for each infant according to the Vermont Oxford Network database collection criteria.
      Vermont Oxford Network
      Vermont Oxford Network Database Manual of Operations: Release.
      In addition, for each newborn, the clinical risk index for babies score was calculated, designed to quantify the clinical severity based on six variables evaluated during the first 12 hours of life (birth weight, gestational age, minimum and maximum FiO2 to keep a saturation between 88% and 95%, worst bases excess, and presence of congenital abnormalities).
      • Gagliardi L.
      • Cavazza A.
      • Brunelli A.
      • et al.
      Assessing mortality risk in very low birthweight infants: a comparison of CRIB, CRIB-II, and SNAPPE-II.
      Each newborn underwent a serial head sonographic examination. Periventricular and intraventricular hemorrhage (IVH) and other neonatal head ultrasound findings were classified and grouped according to Rademaker
      • Rademaker K.J.
      • Uiterwaal C.S.P.M.
      • Beek F.J.A.
      • et al.
      Neonatal cranial ultrasound versus MRI and neurodevelopmental outcome at school age in children born preterm.
      as normal, when there were no abnormalities or only minor abnormalities, (Group 1); slightly abnormal, in the presence of grade I/II IVH, periventricular leukomalacia grade I, germinal layer necrosis, or a combination of these features, or ventricular dilatation as the sole feature (Group 2); or severely abnormal, when one or more of the following features were present: grade III/IV IVH, cystic periventricular leukomalacia grade II/III, thalamic lesions, focal infarctions, posthemorrhagic ventricular dilatation needing therapeutic intervention (Group 3).
      Each newborn was examined by a pediatric neuropsychiatrist, at term corrected age; neonatal neurological examinations were classified as normal, in the absence of any abnormal neurological sign, or abnormal, in the presence of pathologic neurological signs of varying degrees. Every infant was afterward followed up with neurological assessments every three months during the first year of life and every six months in the second year.
      • Gosselin J.
      • Amiel-Tison C.
      Neurological Assessment from Birth to Six Years-Evaluation Neurologique De La Naissance À 6 Ans.
      Ophthalmologic and audiometric examinations were performed periodically to exclude sensorial abnormalities.
      At 24 months corrected age every infant underwent a complete neurological examination and Griffiths' Mental Developmental Scales-Extended Revised
      • Griffiths R.
      • Huntley M.
      The Griffiths Mental Development Scales-Revised Manual: from Birth to 2 Years.
      to obtain the General Quotient (GQ) and five subquotient scores corresponding to five neurodevelopmental areas (locomotor, personal-social, hearing and speech, eye-hand coordination, performance). Neurodevelopmental outcomes were classified according to the neurological examination and the GQ into (1) normal (no pathologic signs on neurological examination and GQ ≥ 88); (2) minor sequelae (tone or reflex abnormalities or asymmetry without functional deficits and presence of at least one sign from the triad described by Amiel-Tison and Gosselin
      • Gosselin J.
      • Gahagan S.
      • Amiel-Tison C.
      The Amiel-Tison neurological assessment at term: conceptual and methodological continuity in the course of follow-up.
      : 76≤ GQ ≤ 87, squint or refractive errors, mild hypoacusia); and (3) major sequelae (one or more of the following: disabling or nondisabling cerebral palsy, GQ ≤ 75, sensorineural hearing loss requiring active intervention, severe visual impairment of peripheral or central origin).

      Statistical analysis

      Chi-square analysis was used to compare categorical variables. Comparisons of continuous variables between multiple groups were performed by Kruskal-Wallis ANOVA with Bonferroni correction for pairwise comparisons.
      StataCorp
      Stata Statistical Software: Release 13.
      Partitioning of chi-square statistics with Bonferroni correction was used to compare categories in multiway contingency tables. Penalized logistic regression analysis was used to test the association of placental pathologic variables with survival and two-year neurodevelopmental outcomes correcting for potential confounders. Models included neonatal outcomes as dependent variables and gestational age, birth weight, fetal sex, type of delivery (vaginal versus Caesarean), and type of placental lesion as explanatory variables. To study the relative importance of the placental lesions across the various etiologies of VLBW, we included in the model an interaction term between the type of placental lesion and the type of etiology of VLBW. Finally, to evaluate the best independent pathologic predictors of two-year intact survival in the whole population of VLBW, we used a stepwise logistic regression including two-year infant outcomes as dependent variables, and all pathologic variables studied, gestational age, birth weight, type of delivery, and sex of the newborn, as initial explanatory variables.

      Results

      During the period of the study, of 595 (518 mothers) VLBW infants delivered at our institution, 21 subjects were excluded due to the lack of maternal or histologic placental data or unexpected severe fetal malformations, leaving 574 (503 mothers) newborns for analysis. The characteristics of the population under study is reported in Table 1. Major determinants of VLBW were FGR (52%), spontaneous prematurity (43.7), twins (28%), preeclampsia (27%), and PPROM (12.5%).
      TABLE 1.Maternal and Obstetrics Characteristics of Population Enrolled
      N = 574Median (IQR)
      Maternal age33 (29-37)
      BMI23 (21-26)
      Gestational age29.4 (27-31.3)
      Birth weight1100 (854-1354)
      Birth weight percentile6 (1-48)
      Maternal ethnic origin
       Europe441 (76.8)
       Africa49 (8.5)
       Asia15 (2.6)
       North America16 (2.8)
       South America53 (9.2)
      Maternal education ≤8 years181 (31.5)
      Nulliparous378 (65.9)
      First trimester BMI ≥3045 (7.8)
      Previous preterm or LBW47 (8.2)
      Smoking during pregnancy111 (19.3)
      Reproductive failures ≥318 (3.1)
      Twins161 (28)
      FGR298 (52)
      Preeclampsia155 (27)
      Spontaneous prematurity251 (43.7)
      PPROM72 (12.5)
      Abbreviations:
      BMI = Body mass index
      FGR = Fetal growth restriction
      IQR = Interquartile range
      LBW = Low birth weight
      PPROM = Premature preterm rupture of membranes

      Obstetrics, and neonatal characteristics and placental lesions

      The rates of maternal body mass index ≥30 or maternal age ≥40 years were 8.9% (45 of 503) and 9.5% (48 of 503), respectively. Table 2 reports the main obstetric and neonatal variables according to the main pathologic features. Severe MVM, FVM, and chorioamnionitis (stage≥2 and grade = 2) were recorded in 6.4%, 7.4%, and 15.5% of the cases, respectively. Given the different VLBW etiologies, birth weight z scores (z score was computed for each case by comparing the actual birth weight with the standard Italian birth weight for gestational age, adjusted for maternal parity and fetal sex) were lower, and the rates of FGR and preeclampsia were higher among pregnancies complicated by severe FVM than those without this lesion. On the other hand, severe histological chorioamnionitis was associated with lower gestational age, birth weight, placental weight, and higher rates of spontaneous delivery, PPROM, and clinical chorioamnionitis. Overall, the rate of neonatal death was 13.9% (80 of 574), and in univariate analysis, severe MVM, FVM, and chorioamnionitis were associated with increased rates of neonatal death. After adjustment for gestational age and birth weight, severe MVM (adjusted odds ratio [adj. OR] = 3.98, 95% confidence interval [CI] = 1.73 to 9.16) and severe FVM (adj. OR = 3.28, 95% CI = 1.37 to 7.85), but not chorioamnionitis (adj. OR = 0.77, 95% CI = 0.4 to 1.49), were still associated with an increased risk of neonatal death. Loss of placental integrity was associated with lower gestational age and birth weight z scores and was more common among FGR pregnancies (Table 3). Intravillous hemorrhage was more common among FGR pregnancy, in spontaneous prematurity, and in pregnancies with a clinical diagnosis of abruptio, whereas villitis of unknown origin was associated with increased rates of FGR and of umbilical artery Doppler abnormalities. After adjustment for gestational age and birth weight, loss of placental integrity (adj. OR = 1.9, 95% CI = 1.02 to 3.66) and intravillous hemorrhage (adj. OR = 2.2, 95% CI = 1.16 to 4.11) were still significantly associated with an increased risk of neonatal death.
      TABLE 2.Obstetrics, Hystologic, and Neonatal Characteristics According to Placental Lesions Indicating FVM, MVM, and Severe Chorioamnionitis
      Severe MVMSevere FVMSevere Chorioamnionitis
      Yes (n = 37)No (n = 537)Yes (n = 41)No (n = 533)Yes (n = 89)No (n = 485)
      Gestational age28 (26-31.5)29 (26-31)30 (27-34)29 (26-31)26 (24-27)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      29 (27-31)
      Birth weight939 (662-1261)1098 (798-1354)1031 (759-1398)1092 (793-1350)815 (687-1022)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      1130 (859-1376)
      z-score birth weight−1.82 (−3.7-0.05)−1.57 (−3.3-0.09)−3.05 (−3.8-1.3)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      −1.4 (−3.1-0.03)0.22 (−0.6 + 0.88)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      −2 (−3.6-0.41)
      Placental weight240 (204-316)260 (211-321)274 (215-329)260 (210-320)250 (208-335)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      260 (207-334)
      Fetal/placental ratio4.1 (3.19-4.8)4.1 (3.1-4.9)4.1 (2.7-4.8)4.1 (3.1-4.9)3.4 (2.9-4.1)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      4.3 (3.1-5)
      Twins7 (18.9)154 (28.7)15 (36.6)146 (27.4)16 (18)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      145 (29.9)
      FGR21 (56.8)277 (51.6)29 (70.7)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      269 (50.5)13 (14.6)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      285 (58.8)
      Preeclampsia15 (40.5)140 (26.1)18 (43.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      137 (25.7)0 (-)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      155 (32)
      Spontaneous preterm birth16 (43.2)235 (43.8)7 (17.1)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      244 (45.8)76 (85.4)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      175 (36.1)
      PPROM2 (5.4)70 (13)8 (19.5)64 (12)30 (33.7)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      42 (8.7)
      ARED8 (21.6)82 (15.3)9 (22)81 (15.2)2 (2.2)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      88 (18.1)
      Caesarean section29 (78.4)392 (73)47 (52.8)374 (77.1)
      CTG abnormalities18 (48.6)229 (42.6)21 (51.2)226 (42.4)20 (22.5)227 (46.8)
      Antenatal diagnosis of abruptio3 (8.1)50 (9.3)6 (14.6)47 (8.8)10 (11.2)43 (8.9)
      Clinical chorioamnionitis3 (8.1)26 (4.8)1 (2.4)28 (5.3)11 (12.4)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      18 (3.7)
      Antenatal steroids33 (89.2)473 (88.1)37 (90.2)469 (88)76 (85.4)430 (88.7)
      Magnesium sulfate7 (18.9)52 (9.7)9 (22)50 (9.4)11 (12.4)48 (9.9)
      CRIB II score
       <525 (67.6)418 (77.8)31 (75.6)412 (77.3)55 (61.8)388 (80)
       5-103 (8.1)78 (14.5)7 (17.1)74 (13.9)20 (22.5)61 (12.6)
       >109 (24.3)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      41 (7.6)3 (7.3)47 (8.8)14 (15.7)36 (7.4)
      BPD6 (16.2)118 (22)7 (17.1)117 (22)33 (37.1)91 (18.8)
      NEC1 (2.7)31 (5.8)1 (2.4)31 (5.8)7 (7.9)25 (5.2)
      Early sepsis1 (2.7)12 (2.2)0 (-)13 (2.4)4 (4.5)9 (1.9)
      Late sepsis5 (13.5)84 (15.6)6 (14.6)83 (15.6)27 (30.3)62 (12.8)
      IVH > 21 (2.7)18 (3.4)2 (4.9)17 (3.2)7 (7.9)12 (2.5)
      PVLC > 31 (2.7)10 (1.9)1 (2.4)10 (1.9)7 (7.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      4 (0.8)
      Neonatal death13 (35.1)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      67 (12.5)11 (26.8)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      69 (12.9)20 (22.5)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      60 (12.4)
      Abbreviations:
      ARED = Absent or reverse end diastolic flow
      BPD = Biparietal diameter
      CRIB = Clinical risk index for babies
      CTG = Cardiotocography
      FGR = Fetal growth restriction
      FVM = Fetal vascular malperfusion
      IVH = Intraventricular hemorrhage
      MVM = Maternal vascular malperfusion
      NEC = Necrotizing enterocolitis
      PPROM = Premature preterm rupture of membranes
      PVCL = Periventricular cystic leukomalacia
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      TABLE 3.Obstetrics, Hystologic, and Neonatal Characteristics According to Placental Lesions
      Loss of IntegrityIntravillous HemorrhageVillitis Unknown Origin
      Yes (n = 83)No (n = 491)Yes (n = 85)No (n = 489)Yes (n = 83)No (n = 491)
      Gestational age28 (26-30)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      29 (26-31)29 (26-30)29 (26-31)30 (28-32)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      29 (26-31)
      Birth weight978 (778-1279)1095 (794-1360)1037 (783-13079)1090 (795-1354)1138 (841-1350)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      1072 (787-1353)
      z-score birth weight−0.61 (−2.7 + 0.46)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      −1.73 (−3.4-0.17)−0.8 (−2.4 + 0.41)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      −0.16 (−1.7-0.16)−2.6 (−4-1.1)−1.3 (−3.1- 0.01)
      Placental weight256 (207-323)260 (210-321)262 (208-333)260 (210-319)260 (210-315)260 (210-321)
      Fetal/placental ratio3.9 (3.2-4.6)4.1 (3-5)3.9 (3-4.6)4.1 (3.1-5)4.3 (3-5.1)4.1 (3.1-4.9)
      Twins19 (22.9)142 (28.9)31 (36.5)130 (26.6)22 (26.5)139 (28.3)
      FGR24 (28.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      264 (53.8)33 (38.8)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      264 (54)56 (67.5)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      241 (49.1)
      Preeclampsia17 (20.5)138 (28.1)11 (12.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      143 (29.2)28 (33.7)126 (25.7)
      Spontaneous preterm birth44 (53)207 (42.2)49 (57.6)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      201 (41.1)23 (27.7)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      227 (46.2)
      PPROM9 (10.8)63 (12.8)9 (10.6)63 (12.9)7 (8.4)65 (13.2)
      ARED14 (16.9)76 (15.5)13 (15.3)77 (15.7)20 (24.1)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      70 (14.3)
      Caesarean64 (77.1)357 (72.7)62 (72.9)359 (73.4)62 (74.7)359 (73.1)
      CTG abnormalities32 (38.6)215 (43.8)33 (38.8)214 (43.8)39 (47)208 (42.4)
      Antenatal diagnosis of abruptio15 (18.1)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      38 (7.7)18 (21.2)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      35 (7.2)2 (2.4)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      51 (10.4)
      Clinical chorioamnio-nitis6 (7.2)23 (4.7)11 (12.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      18 (3.7)4 (4.8)25 (5.1)
      Antenatal steroids69 (83.1)437 (89)79 (92.9)427 (87.3)75 (90.4)431 (87.8)
      Magnesium sulfate9 (10.8)50 (10.2)7 (8.2)52 (10.6)11 (13.3)48 (9.8)
      CRIB II score
       <558 (69.9)387 (78.8)63 (74.1)380 (77.7)71 (85.5)372 (75.8)
       5-1014 (16.9)67 (13.6)13 (15.3)68 (13.9)9 (10.8)72 (14.7)
       >1011 (13.3)39 (7.9)9 (10.6)41 (8.4)3 (3.6)47 (9.6)
      BPD16 (19.3)108 (22)19 (22.4)105 (21.5)15 (18.1)109 (22.2)
      NEC3 (3.6)29 (5.9)4 (4.7)27 (5.5)5 (6)26 (5.3)
      Early sepsis3 (3.6)10 (2)2 (2.4)11 (2.2)1 (1.2)12 (2.4)
      Late sepsis18 (21.7)71 (14.5)20 (23.5)69 (14.1)18 (21.7)71 (14.5)
      IVH > 23 (3.6)16 (3.3)2 (2.4)17 (3.5)3 (3.6)16 (3.3)
      PVLC > 32 (2.4)9 (1.8)2 (2.4)9 (1.8)1 (1.2)10 (2)
      Neonatal death19 (22.9)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      61 (12.4)20 (23.5)
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.
      60 (12.3)8 (9.6)72 (14.7)
      Abbreviations:
      ARED = Absent or reverse end diastolic flow
      BPD = Biparietal diameter
      CRIB = Clinical risk index for babies
      CTG = Cardiotocography
      FGR = Fetal growth restriction
      IVH = Intraventricular hemorrhage
      NEC = Necrotizing enterocolitis
      PPROM = Premature preterm rupture of membrane
      PVCL = Periventricular cystic leukomalacia
      P < 0.05 Partitioning of chi-square statistics with Bonferroni correction.

      Placental lesions and infant neurodevelopmental outcomes

      Of the 494 surviving infants, 43 (8.7%) subjects who had been judged neurologically normal at discharge from hospital were lost during follow-up, leaving 451 infants with complete two-year neurodevelopmental outcome measures. Of these, 153 infants (33.9%) were clinically considered as neurologically abnormal at the time of discharge. Overall, 41.7% (20 of 48) with severe and 46.7% (28 of 60) with mild two-year neurodevelopmental impairments had no abnormal neurological findings at discharge from the hospital. Infants delivered after histologically severe chorioamnionitis probably as a result of a lower gestational age had a lower general quotient, eye-hand coordination and performance subscale scores, and higher rates of severe neurodevelopmental sequelae compared with patients without placental lesion (Table 4).
      TABLE 4.Placental Characteristics According to Neurodevelopmental Outcomes
      Severe MVMSevere FVMSevere Chorioamionitis
      Yes (n = 24)No (n = 427)Yes (n = 28)No (n = 423)Yes (n = 61)No (n = 390)
      Median (IQR)Median (IQR)Median (IQR)Median (IQR)Median (IQR)Median (IQR)
      GQ102 (95-109)102.5 (83.5-110.5)103 (89.2-106.5)102 (95-109)100 (82.5-106)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      102 (95-110)
      Locomotor105 (96-111)102 (91.6-110.5)100 (89.4-1069)105 (96-112)100 (88-106.5)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      105 (96-112)
      Personal-social100 (69.5-1049)107 (94.4-112)105 (95.4-112)107 (93-112)105 (88-110)107 (95-112)
      Hearing-speech100.5 (69.5-104)96 (81-104)98 (76.5-101.89)96 (80-104)95 (75.5-100)96 (83-104.5)
      Eye-hand coordination102 (94.2-112)103 (94-109)103 (94.7-111.5)103 (94-109)99 (89.5-104)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      104 (96-110.5)
      Performance100 (86.5-103)105 (93.5-110)101 (83.3-107)105 (94-110)101.5 (86-105.3)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      105 (94-111)
      N. (%)N (%)N. (%)N (%)N. (%)N. (%)
      Neurologically suspected or abnormal at discharge6 (25)147 (34.4)4 (14.3)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      149 (35.2)29 (47.5)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      124 (31.8)
      Sequelae:
       Intact17 (70.8)326 (76.3)20 (71.4)323 (76.4)38 (62.3)305 (78.2)
       Mild abnormal4 (16.7)56 (13.1)3 (10.7)57 (13.5)14 (23)46 (11.8)
       Severe3 (12.5)45 (10.5)5 (17.9)43 (10.2)9 (14.8)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      39 (10)
      GQ < 753 (12.5)38 (8.9)5 (17.9)36 (8.5)8 (13.1)33 (8.5)
      Survival without major sequelae21/37 (56.7)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      382/494 (77.3)23/39 (59)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      380/492 (77.23)52/81 (64.2)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      351/450 (78)
      Intact survival17/37 (45.9)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      326/494 (66)20/39 (51.2)323/492 (65.6)38/43 (88.37)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      305/450 (67.8)
      Loss of IntegrityIntravillous HemorrhageVillitis Unknown Origin
      Yes (n = 61)No (n = 390)Yes (N = 62)No (N = 389)Yes (N = 71)No (N = 380)
      GQ104.5 (91.4-111)102 (95-108)98 (76-110)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      102 (96-109)102 (97-107)102 (92-109)
      Locomotor105 (96-110.5)105 (95.2-111)100 (87.5-107.5)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      105 (96-111.5)105 (98-111)105 (93-111)
      Personal-social107 (95-116)107 (94-112)107 (96-112)100 (87.5-107.5)107 (100-112)106 (93-112)
      Hearing-speech99.5 (78.5-105)95 (80-104)94 (63.5-105)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      96 (84-104)95.5 (88-104)96 (78-104)
      Eye-hand coordination103 (98-112)103 (94-109)100 (86-111.5)104 (96-109)104 (96-107)103 (94-110)
      Performance103 (98.5-111)105 (93-110)100 (84-110)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      105 (94-110)105 (93-110)104 (93-110)
      N. (%)N. (%)N. (%)N. (%)N. (%)N. (%)
      Neurologically suspected or abnormal at discharge16 (26.2)137 (35.1)17 (27.4)136 (35)21 (29.6)132 (34.7)
      Sequelae:
       Intact45 (73.8)298 (76.4)37 (59.7)306 (78.7)59 (83.1)290 (76.3)
       Mild abnormal10 (16.4)50 (12.8)11 (17.7)49 (12.6)7 (9.9)53 (13.9)
       Severe6 (9.8)42 (10.8)14 (22.6)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      34 (8.7)5 (7)43 (11.3)
      GQ < 754 (6.6)37 (9.5)13 (21)28 (7.2)5 (7)36 (9.5)
      Survival without major sequelae55/80 (68.7)348/451 (77.2)48/82 (58.5)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      305/449 (67.9)66/79 (83.5)337/452 (74.5)
      Intact survival45/80 (56.2)298/451 (66)37/82 (45.1)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      306/449 (67.9)59/79 (74.7)
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      284/452 (62.8)
      Abbreviations:
      GQ = General quotient, Griffiths' Mental Developmental Scales
      IQR = Interquartile range
      P < 0.05 Partitioning of chi-squared statistics with Bonferroni correction.
      In crude analysis, MVM, FVM, and chorioamnionitis were associated with lower rates of survival without major neurodevelopmental abnormalities, whereas MVM and chorioamnionitis were associated with lower survival rates with intact neurodevelopmental outcomes. Intravillous hemorrhage was associated with a two-year lower general quotient and locomotor, hearing and speech, and performance scores and with an excess rate of infant severe neurodevelopmental sequelae compared with pregnancies without these lesions. As a consequence, this placental lesion was also associated with a lesser likelihood of intact survival or survival without major developmental sequelae.
      The odds ratios of intact survival and survival without major neurodevelopmental sequelae associated with placental lesions adjusted for the confounding effect of gestational age, birth weight, fetal sex, and type of delivery are reported in Table 5. Severe MVM and FVM and intravillous hemorrhage were associated with an overall reduced intact survival and survival without major sequelae. There were several interactions between placental lesions and etiology of VLBW suggesting that some lesions were strongly associated with some causes and less in others. Severe chorioamnionitis was absent among pregnancies delivered as a result of preeclampsia. The effect of intravillous hemorrhage on two-year survival with intact neurodevelopmental outcomes was significant among twins (nine of 31 when compared with 87 of 123, adj. OR = 0.12, 95% CI = 0.05 to 0.32) but not in singleton pregnancies (28 of 51 when compared with 217 of 326, adj. OR = 0.66, 95% CI = 0.35 to 1.25, P for interaction = 0.005) and more marked among preeclamptic (three of 11 when compared with 105 of 132, adj. OR = 0.04, 95% CI = 0.006 to 0.25) than normotensive pregnancies (34 of 71 when compared with 199 of 317, adj. OR = 0.52, 95% CI = 0.3 to 0.9, P for interaction = 0.015).
      TABLE 5.Odds Ratios of Intact Survival and Survival Without Major Neurodevelopmental Sequelae Associated With Placental Lesions Adjusted for the Confounding Effect of Gestational Age, Birth Weight, Fetal Sex, and Type of Delivery
      Survival Without Major Neurodevelopmental ImpairmentsSurvival with Normal Neurodevelopmental Outcome
      OR (95% CI)OR (95% CI)
      Maternal vascular underperfusion0.38 (0.18-0.82)0.45 (0.22-0.92)
      Fetal vascular underperfusion0.32 (0.15-0.69)0.46 (0.22-0.45)
      Chorioamnionitis1.1 (0.6-2)0.75 (0.43-1.29)
      Loss of placental integrity0.72 (0.41-1.27)0.73 (0.44-1.21)
      Intravillous hemorrhage0.35 (0.2-0.6)0.38 (0.22-0.62)
      Villitis unknown origin1.48 (0.75-2.91)1.54 (0.86-2.75)
      Abbreviations:
      CI = Confidence interval
      OR = Odds ratio as obtained by penalized logistic regression
      Finally, in the whole population of VLBW, and after correction for gestational age, birth weight, sex of the newborns, and type of delivery, the best independent predictors of two-year infant intact neurodevelopmental outcomes corresponded to severe FVM (adj. OR = 0.46, 95% CI = 0.21 to 0.97) and intravillous hemorrhage (adj. OR = 0.37, 95% CI = 0.22 to 0.62).

      Discussion

      The results of this study suggest that placental pathologic lesions in VLBW deliveries are rather heterogeneous and that their frequencies follow the etiologic causes of prematurity or fetal growth defects. Inflammatory lesions were more common among spontaneous prematurity and PPROM, whereas severe FVM, loss of placental integrity, intravillous hemorrhage, and villitis of unknown origin were more common among FGR and pregnancies with preeclampsia. After correction for the confounding effect of gestational age and birth weight, which are the most powerful predictors of adverse neonatal outcomes, severe MVM, FVM, and intravillous hemorrhage were significantly associated with lower rates of two-year intact infant neurodevelopmental outcomes. The relative size of the risk of adverse neonatal outcomes associated with intravillous hemorrhage was modulated by other etiologic VLBW factors. In the whole population, only severe FVM and intravillous hemorrhage were independent predictors of a lower rate of intact neurodevelopmental outcomes. Literature data on the relationship between placental pathologic features and subsequent infant neurodevelopmental outcomes among VLBW are limited and mostly confined to the relationship between clinical and histological chorioamnionitis, early infant brain damage, and subsequent infant neurodevelopmental outcomes.
      • Raghavan R.
      • Helfrich B.B.
      • Cerda S.R.
      • et al.
      Preterm birth subtypes, placental pathology findings, and risk of neurodevelopmental disabilities during childhood.
      ,
      • Roescher A.M.
      • Timmer A.
      • Erwich J.J.
      • Bos A.F.
      Placental pathology, perinatal death, neonatal outcome, and neurological development: a systematic review.
      Although in our data prematurity-associated placental inflammation correlated strongly with reduced survival and increased rates of neurodevelopmental sequelae in crude analysis, this association disappeared after correction for gestational age and birth weight. Our data largely confirm data from previous population studies obtained in very preterm infants suggesting that the adverse effects of histological chorioamnionitis on infant neurodevelopmental outcomes are mainly related to the complications of prematurity rather than to a direct effect of histologically diagnosed placental inflammation on the brain of preterm fetuses.
      • Ylijoki M.
      • Lehtonen L.
      • Lind A.
      • et al.
      Chorioamnionitis and five-year neurodevelopmental outcome in preterm infants.
      ,
      • Xing L.
      • Wang G.
      • Chen R.
      • Ren J.
      • Qian J.
      • Huang Y.
      Is chorioamnionitis associated with neurodevelopmental outcomes in preterm infants? A systematic review and meta-analysis following PRISMA.
      According to these suggestions, Bierstone et al.
      • Bierstone D.
      • Wagenaar N.
      • Gano D.L.
      • et al.
      Association of histologic chorioamnionitis with perinatal brain injury and early childhood neurodevelopmental outcomes among preterm neonates.
      in a cohort multicenter study of neonates ≤32 weeks' gestational age confirmed that gestational age-related neonatal complications attenuate the apparent adverse effect of chorioamnionitis on infant neurodevelopmental outcome.
      MVM is characterized by placental vascular lesions such as infarcts and retroplacental hemorrhage, by villous lesions such as villous hypoplasia or accelerated villous maturation, or by lesions suggesting a defective spiral artery modeling. All these lesions could lead to a malperfusion of the placental bed and subsequent placenta and fetal hypoxia.
      • Stanek J.
      Placental hypoxic overlap lesions: a clinicoplacental correlation.
      This type of lesion has been associated with an increased risk of abnormal developmental outcomes in a cohort, uncontrolled study suggesting that a hypoxic placental pattern could interfere with fetal growth and infant brain development.
      • Raghavan R.
      • Helfrich B.B.
      • Cerda S.R.
      • et al.
      Preterm birth subtypes, placental pathology findings, and risk of neurodevelopmental disabilities during childhood.
      On the other hand, animal studies have found that both acute and mild intrauterine hypoperfusion similar to MVM of the human placenta are able to produce neurodevelopmental disorders typical of FGR and prematurity.
      • Ohshima M.
      • Coq J.O.
      • Otani K.
      • et al.
      Mild intrauterine hypoperfusion reproduces neurodevelopmental disorders observed in prematurity.
      FVM is a placental pathologic lesion involving the thrombosis of the chorionic plate or umbilical cord vessels, which is uncommon in term but diagnosed in up to 25% of premature deliveries.
      • Heider A.
      Fetal vascular malperfusion.
      This lesion is more frequently found in FGR and preeclamptic pregnancies or in pregnancies complicated by a maternal hypercoagulable state but has also been observed in cases of ascending uterine infection.
      • Khong T.Y.
      • Mooney E.E.
      • Ariel I.
      • et al.
      Sampling and definitions of placental lesions: Amsterdam placental workshop group consensus statement.
      ,
      • Redline R.W.
      Classification of placental lesions.
      ,
      • Heider A.
      Fetal vascular malperfusion.
      In several retrospective or case-control studies, FVM has been associated with fetal demise, neonatal encephalopathy, or cerebral palsy.
      • Raghavan R.
      • Helfrich B.B.
      • Cerda S.R.
      • et al.
      Preterm birth subtypes, placental pathology findings, and risk of neurodevelopmental disabilities during childhood.
      • Sun C.
      • Groom K.M.
      • Oyston C.
      • Chamley L.W.
      • Clark A.R.
      • James J.L.
      The placenta in fetal growth restriction: what is going wrong?.
      • Roescher A.M.
      • Timmer A.
      • Erwich J.J.
      • Bos A.F.
      Placental pathology, perinatal death, neonatal outcome, and neurological development: a systematic review.
      ,
      • Saleemuddin A.
      • Tantbirojn P.
      • Sirois K.
      • et al.
      Obstetric and perinatal complications in placentas with fetal thrombotic vasculopathy.
      Stepwise logistic regression in our study confirms that FVM is an independent predictor of a low two-year intact survival in the whole population of VLBW infants.
      Intravillous hemorrhage from a placental lesion is often seen in immature villi characterized by the extravasation of intact red blood cells in villous stroma leading to villous coagulative necrosis.
      • Stanek J.
      Hypoxic patterns of placental injury: a review.
      The rupture of the villous vascular integrity with subsequent infarct is considered to be a lesion suggestive for hypoxic-ischemic placental injury.
      • Stanek J.
      Hypoxic patterns of placental injury: a review.
      The lesion is most commonly, but not exclusively, associated with placental abruption, preceding clinical manifestations of placental detachment by a median of eight hours.
      • Stanek J.
      Hypoxic patterns of placental injury: a review.
      Other significant placental features and perinatal complications associated with intravillous hemorrhage are chorioamnionitis, intervillous thrombosis, FGR, and fetal death.
      • Genest D.R.
      • Ringer S.
      Placental findings correlate with neonatal death in extremely premature infants (24–32 weeks): a study of 150 cases.
      In a recent case-control study of fetal death
      • Jaiman S.
      • Romero R.
      • Pacora P.
      • et al.
      Disorders of placental villous maturation in fetal death.
      intravillous hemorrahge was present in 7% of the cases and 0.1% of the controls confirming old data associating intravillous hemorrhage with fetal death.
      • Genest D.R.
      • Ringer S.
      Placental findings correlate with neonatal death in extremely premature infants (24–32 weeks): a study of 150 cases.
      In our study intravillous hemorrhage was an independent predictor of a lower rate of two-year intact infant survival in the whole population of VLBW. The extent of the adverse effect was modulated by other clinical factors being higher in twins than in singleton and in preeclamptic compared with normotensive pregnancies.
      Villitis of unknown origin was more common among FGR pregnancies and was associated with better two-year intact survival in univariate analysis, but its effect disappeared after correction for potential confounders, suggesting that the role of this lesion on neurodevelopmental outcomes of VLBW infants is limited.
      From a biological viewpoint placental lesions suggestive for FVM have been associated with reduced blood movement in the intervillous space in diffusion-weighted magnetic resonance imaging studies of dysfunctional placentas, suggesting that this type of lesion correlates to the severity of growth failure.
      • Kristi B.A.
      • Ditte N.H.
      • Caroline H.
      • et al.
      Placental diffusion-weighted MRI in normal pregnancies and those complicated by placental dysfunction due to vascular malperfusion.
      Intravillous hemorrhage leads to coagulation necrosis of villi and to the formation of intervillous thrombi, both lesions impairing fetomaternal exchanges in FGR pregnancies.
      • Stanek J.
      Placental hypoxic overlap lesions: a clinicoplacental correlation.
      ,
      • Stanek J.
      Hypoxic patterns of placental injury: a review.
      The main strengths of the study involve the type of the study and the methods used. Data on neurodevelopmental outcomes of VLBW infants including placental pathology are lacking. Most data derive from single series without controls or are mainly aimed at the evaluation of the effect of chorioamnionitis on subsequent neurodevelopmental outcomes in preterm deliveries.
      • Roescher A.M.
      • Timmer A.
      • Erwich J.J.
      • Bos A.F.
      Placental pathology, perinatal death, neonatal outcome, and neurological development: a systematic review.
      ,
      • Xing L.
      • Wang G.
      • Chen R.
      • Ren J.
      • Qian J.
      • Huang Y.
      Is chorioamnionitis associated with neurodevelopmental outcomes in preterm infants? A systematic review and meta-analysis following PRISMA.
      To avoid the so-called survival bias,
      • Sy R.W.
      • Bannon P.G.
      • Bayfield M.S.
      • Brown C.
      • Kritharides L.
      Survivor treatment selection bias and outcomes research: a case study of surgery in infective endocarditis.
      which most studies of subsequent infant outcomes have not taken into account, we have included among the outcomes both neonatal death and infant neurodevelopmental outcomes. Intravillous hemorrhage, when basal oriented, could be the result of mechanical placental separation after surgical procedures such as dilatation and curettage or Caesarean section.
      • Stanek J.
      Placental hypoxic overlap lesions: a clinicoplacental correlation.
      ,
      • Stanek J.
      Hypoxic patterns of placental injury: a review.
      Although pathologists involved in the study were well aware of this possibility, to avoid the potential confounding effects of Caesarean sections we included the type of delivery (vaginal versus Caesarean) in the study of the relationship between placental intravillous hemorrhage and neonatal outcomes. As this was a single-center study, the results should be interpreted cautiously and cannot be readily generalized for other institutions. Although unlikely, given the cohort approach and the method used, potential biases could have been introduced in the type of maternal population (e.g., prevalence of obesity, maternal age, racial differences) enrolled, the treatments used (protocols of maternal and neonatal care), and placental histological interpretations. Finally, another possible bias is the proportion (8.7%) of infants without brain lesions and who had been judged neurologically normal at hospital discharge who were lost to follow-up. Although a subsequent severe motor handicap seems unlikely in these infants, we cannot exclude subsequent minor disabilities.

      Conclusions

      In conclusion, the results of this study have demonstrated that among VLBW infants, placental pathologic findings such as severe MVM, FVM, and intravillous hemorrhage are significantly associated with neonatal survival and subsequent neurodevelopmental outcomes. Etiologic factors of VLBW can modulate the effect of placental pathologic findings on infants' neurodevelopmental outcomes. Data on placental pathology results should be taken into consideration in the follow-up of VLBW infants at risk for neurodevelopmental impairment.

      Acknowledgments

      The authors acknowledge the collaboration with the Neurodevelopment Study Group: Davide Tonduti, Giada Ariaudo, Nina Tritto, Alice Decio, Sara Olivotto, Silvia Spairani, Gianfranco Perotti, and Giulia Vittoria Carletti.

      References

        • Blencowe H.
        • Cousens S.
        • Chou D.
        • et al.
        Born too soon: the global epidemiology of 15 million preterm births.
        Reprod Health. 2013; 10: S2
        • Pascal A.
        • Govaert P.
        • Oostra A.
        • Naulaers G.
        • Ortibus E.
        • Van der Broeck C.
        Neurodevelopmental outcome in very preterm and very-low-birthweight infants born over the past decade: a meta-analytic review.
        Dev Med Child Neurol. 2018; 60: 342-355
        • Chevallier M.
        • Debillon T.
        • Pierrat V.
        • et al.
        Leading causes of preterm delivery as risk factors for intraventricular hemorrhage in very preterm infants: results of the EPIPAGE 2 cohort study.
        Am J Obstet Gynecol. 2017; 216: 518.e1-518.e12
        • Delorme P.
        • Kayem G.
        • Lorthe E.
        • et al.
        Neurodevelopment at age 2 and umbilical artery Doppler in cases of preterm birth after prenatal hypertensive disorder or suspected fetal growth restriction: the EPIPAGE 2 prospective population-based cohort study.
        Ultrasound Obstet Gynecol. 2020; 56: 557-565
        • Mukhopadhyay S.
        • Puopolo K.M.
        • Hansen N.I.
        • et al.
        Impact of early-onset sepsis and antibiotic use on death or survival with neurodevelopmental impairment at 2 years of age among extremely preterm infants.
        J Pediatr. 2020; 221: 39-46.e5
        • Gardella B.
        • Iacobone A.D.
        • Bogliolo S.
        • et al.
        Obstetric risk factors and time trends of neurodevelopmental outcome at 2 years in very-low-birthweight infants: a single institution study.
        Dev Med Child Neurol. 2015; 57: 1035-1041
        • Bauer S.E.
        • Schneider L.
        • Lynch S.K.
        • Malleske D.T.
        • Shepherd E.G.
        • Nelin L.D.
        Factors associated with neurodevelopmental impairment in bronchopulmonary dysplasia.
        J Pediatr. 2020; 218: 22-27.e2
        • Raghavan R.
        • Helfrich B.B.
        • Cerda S.R.
        • et al.
        Preterm birth subtypes, placental pathology findings, and risk of neurodevelopmental disabilities during childhood.
        Placenta. 2019; 83: 17-25
        • Sun C.
        • Groom K.M.
        • Oyston C.
        • Chamley L.W.
        • Clark A.R.
        • James J.L.
        The placenta in fetal growth restriction: what is going wrong?.
        Placenta. 2020; 96: 10-18
        • Roescher A.M.
        • Timmer A.
        • Erwich J.J.
        • Bos A.F.
        Placental pathology, perinatal death, neonatal outcome, and neurological development: a systematic review.
        PLoS One. 2014; 9: e89419
        • Ylijoki M.
        • Lehtonen L.
        • Lind A.
        • et al.
        Chorioamnionitis and five-year neurodevelopmental outcome in preterm infants.
        Neonatology. 2016; 110: 286-295
        • Xing L.
        • Wang G.
        • Chen R.
        • Ren J.
        • Qian J.
        • Huang Y.
        Is chorioamnionitis associated with neurodevelopmental outcomes in preterm infants? A systematic review and meta-analysis following PRISMA.
        Medicine. 2019; 98: e18229
        • Bierstone D.
        • Wagenaar N.
        • Gano D.L.
        • et al.
        Association of histologic chorioamnionitis with perinatal brain injury and early childhood neurodevelopmental outcomes among preterm neonates.
        JAMA Pediatr. 2018; 172: 534-541
        • Saleemuddin A.
        • Tantbirojn P.
        • Sirois K.
        • et al.
        Obstetric and perinatal complications in placentas with fetal thrombotic vasculopathy.
        Pediatr Dev Pathol. 2010; 13: 459-464
        • Vik T.
        • Redline R.
        • Nelson K.B.
        • et al.
        The placenta in neonatal encephalopathy: a case-control study.
        J Pediatr. 2018; 202: 77-85.e3
        • Lees C.C.
        • Stampalija T.
        • Baschat A.A.
        • et al.
        ISUOG Practice Guidelines: diagnosis and management of small-for-gestational-age fetus and fetal growth restriction.
        Ultrasound Obstet Gynecol. 2020; 56: 298-312
        • Tita A.T.
        • Andrews W.W.
        Diagnosis and management of clinical chorioamnionitis.
        Clin Perinatol. 2010; 37: 339-354
        • ACOG Practice Bulletin No
        202 summary: gestational hypertension and preeclampsia.
        Obstet Gynecol. 2019; 133: 211-214
        • Langston C.
        • Kaplan C.
        • Macpherson T.
        • et al.
        Practice guideline for examination of the placenta: developed by the Placental Pathology Practice Guideline Development Task Force of the College of American Pathologists.
        Arch Pathol Lab Med. 1997; 121: 449-476
        • Khong T.Y.
        • Mooney E.E.
        • Ariel I.
        • et al.
        Sampling and definitions of placental lesions: Amsterdam placental workshop group consensus statement.
        Arch Pathol Lab Med. 2016; 140: 698-713
        • Redline R.W.
        Classification of placental lesions.
        Am J Ostet Gynecol. 2015; 213: S21-S28
        • Jaiman S.
        • Romero R.
        • Pacora P.
        • et al.
        Disorders of placental villous maturation in fetal death.
        J Perinat Med. 2020;
        • Vermont Oxford Network
        Vermont Oxford Network Database Manual of Operations: Release.
        Vermont Oxford Network, Burlington, VT2019
        • Gagliardi L.
        • Cavazza A.
        • Brunelli A.
        • et al.
        Assessing mortality risk in very low birthweight infants: a comparison of CRIB, CRIB-II, and SNAPPE-II.
        Arch Dis Child Fetal Neonatal Ed. 2004; 89: 419-422
        • Rademaker K.J.
        • Uiterwaal C.S.P.M.
        • Beek F.J.A.
        • et al.
        Neonatal cranial ultrasound versus MRI and neurodevelopmental outcome at school age in children born preterm.
        Arch Dis Child Fetal Neonatal Ed. 2005; 90: 489-493
        • Gosselin J.
        • Amiel-Tison C.
        Neurological Assessment from Birth to Six Years-Evaluation Neurologique De La Naissance À 6 Ans.
        2nd ed. Diffusion exclusive pour le Canada: Éditions du CHU Sainte-Justine; Diffusion exclusive pour l’Europe: Elsevier Masson, Montreal2007
        • Griffiths R.
        • Huntley M.
        The Griffiths Mental Development Scales-Revised Manual: from Birth to 2 Years.
        ARICD, High Wycombe1996
        • Gosselin J.
        • Gahagan S.
        • Amiel-Tison C.
        The Amiel-Tison neurological assessment at term: conceptual and methodological continuity in the course of follow-up.
        Ment Retard Dev Disabil Res Rev. 2005; 11: 34-51
        • StataCorp
        Stata Statistical Software: Release 13.
        StataCorp LP, College Station, TX2013
        • Stanek J.
        Placental hypoxic overlap lesions: a clinicoplacental correlation.
        J Obstet Gynaecol Res. 2015; 41: 358-369
        • Ohshima M.
        • Coq J.O.
        • Otani K.
        • et al.
        Mild intrauterine hypoperfusion reproduces neurodevelopmental disorders observed in prematurity.
        Sci Rep. 2016; 6: 39377
        • Heider A.
        Fetal vascular malperfusion.
        Arch Pathol Lab Med. 2017; 141: 1484-1489
        • Stanek J.
        Hypoxic patterns of placental injury: a review.
        Arch Pathol Lab Med. 2013; 137: 706-720
        • Genest D.R.
        • Ringer S.
        Placental findings correlate with neonatal death in extremely premature infants (24–32 weeks): a study of 150 cases.
        Lab Invest. 1993; 68: 126A
        • Kristi B.A.
        • Ditte N.H.
        • Caroline H.
        • et al.
        Placental diffusion-weighted MRI in normal pregnancies and those complicated by placental dysfunction due to vascular malperfusion.
        Placenta. 2020; 91: 52-58
        • Sy R.W.
        • Bannon P.G.
        • Bayfield M.S.
        • Brown C.
        • Kritharides L.
        Survivor treatment selection bias and outcomes research: a case study of surgery in infective endocarditis.
        Circ Cardiovasc Qual Outcomes. 2009; 2: 469-474