Placental pathology associated with lenticulostriate vasculopathy (LSV) in preterm infants

Our aim was to examine the frequency and type of placental abnormalities in neonates with LSV. We prospectively reviewed cranial ultrasounds (cUS) from neonates born at ≤32 weeks of gestation at Parkland Hospital between 2012 and 2014. Our cohort included neonates with LSV and gestational age and sex matched controls with normal cUS. We retrieved placental pathology reports retrospectively and compared placental abnormalities in both groups. We reviewed 1351 cUS from a total of 407 neonates. Placental pathology evaluations were complete for 64/65 (98%) neonates with LSV and 68/70 (97%) matched controls. There were no significant differences for any type of placental abnormities between LSV and control groups. However, infants with highest stage LSV were more likely to have large for gestational age (LGA) placentas (p = 0.01). The association between LSV and LGA placenta may indicate a shared vascular response to an adverse prenatal environment.


INTRODUCTION
Lenticulostriate vasculopathy (LSV) is diagnosed by the cranial ultrasound (cUS) based on appearance of linear branching echogenicities in the basal ganglia and/or thalamus. LSV is one of the least understood cUS findings despite decades of research. Although congenital infections were initially considered the primary cause [1,2], several other non-infectious etiologies are also associated with LSV [3]. Recently, we demonstrated that LSV is strongly associated with inflammatory diseases such as bronchopulmonary dysplasia (BPD) and necrotizing enterocolitis (NEC) in preterm infants [4]. However, the factors that precede and contribute to the development of LSV remain unclear.
There is increasing evidence supporting an association between placental inflammation and development of adverse neurodevelopmental outcomes in neonates [5]. Placental pathology is an important and underutilized tool to learn about the prenatal environment, which is particularly relevant for infants born preterm. Adherence to accepted standards for placental sampling and definitions like those described in the Amsterdam Placental Workshop Group Consensus Statement [6] improves the quality of such investigations.
Previous small studies have evaluated the relationship between LSV and clinical and acute histopathologic chorioamnionitis and failed to demonstrate an association [7,8]. However, no studies to date have explored the prevalence of other gross and histologic placental abnormalities in preterm neonates diagnosed with LSV. Placental examination, particularly lesions that involve chronic inflammation and/or vascular anomalies, may add insight into the timing and etiology of LSV in preterm neonates. The objective of this study was to determine the frequency and type of placental pathologic lesions in neonates with LSV using the Amsterdam Placental Workshop Group Consensus Statement guidelines. We hypothesized that preterm neonates with LSV would have a higher prevalence of histologic placental abnormalities, particularly those involving chronic inflammation.

METHODS
This is a retrospective study of prospectively collected data from our previous case-control study of neonates ≤32 wks gestational age (GA) admitted to the Neonatal Intensive Care Unit (NICU) at Parkland Hospital, Dallas, TX between January 2012 and June 2014 [4]. The study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center and Parkland Health and Hospital Systems. Parental consent was obtained prior to enrollment.
Neonates born at ≤32 wks GA admitted to the Neonatal Intensive Care (NICU) undergo cUS at 3-5 d (and/or 7-10 d), at 30 d postnatally, and at discharge, as previously described [9]. All of the cUS were prospectively reviewed for the presence of LSV by a team of pediatric radiologists assigned to the NICU. The diagnosis of LSV was based on the recognition of a lesion defined as "hyperechogenic lines within the basal ganglia and/ or thalamus on both coronal and parasagittal views on cranial ultrasound". All neonates diagnosed with LSV by two independent radiologists were included in the study. In order to establish a more standardized and consistent method for determining the presence of LSV and assessing its extent, we utilized a validated grading system, as we previously reported [4]. Briefly, Stage 0 has no hyperechoic lines/vessels; Stage 1 includes the presence of lenticulostriate vessels that are considered thin and faintly seen; Stage 2 includes vessels that are thin, but hyperechogenic; and Stage 3 includes vessels that are considered thick and hyperechogenic (i.e., the echogenicity resembled that seen in the Sylvian fissure). For this study, the highest stage LSV agreed on by two radiologists was used to subsequently group the neonates by severity. We enrolled the next born GA and sexmatched neonate without evidence of LSV, intracranial hemorrhage (ICH), periventricular leukomalacia (PVL), or ventriculomegaly (VM) on any cUS during their hospitalization as controls. We then retrieved placental pathology reports and evaluated for placental abnormalities using the guidelines described by the Amsterdam Placental Workshop Group Consensus Statement. Associations between LSV stages and placental abnormalities were examined using χ 2 or Fisher exact test. A 2-tailed significance level of p < 0.05 was used.

Placental pathology
Board-certified pathologists who have expertise in placental pathology routinely examine placentas from all preterm births at our institute for gross and histologic findings according to a specific protocol [10,11]. The umbilical cord, membranes, and placental disc are inspected for any gross abnormalities. The placenta is weighed after removing the umbilical cord, fetal membranes, and non-adherent blood clots. Placentas weighing below the 10th percentile for estimated GA were considered small for GA, and those weighing >90th percentile for GA were considered large for GA [12]. The placental disk is then serially sectioned at 1-2-cm intervals and examined for intraparenchymal lesions. Representative sections of the umbilical cord, fetal membranes, placental parenchyma, and any abnormalities seen on gross examination are submitted for standard histological examination. We previously reported 100% concordance between the routine placental pathology report and independent review of placentas by a blinded placental pathologist [11].
Based on the Amsterdam criteria, the histopathologic abnormalities assessed were divided into the following subcategories [6].
Routine clinical placental pathology reports were reviewed and histopathologic abnormalities were categorized using the Amsterdam classifications by a neonatologist with placental pathology expertise (INM) and who was blinded to the clinical history and outcomes. Chronic lesions included maternal vascular malperfusion, fetal vascular malperfusion (or fetal thrombotic vasculopathy), chronic villitis, other chronic inflammation (e.g., chronic deciduitis), SGA placentas, LGA placentas and chronic villous edema, and acute placental abnormalities included acute chorioamnionitis with or without fetal inflammatory response [13,14]. Dichorionic twin placentas were reviewed and categorized separately for each infant, but the same placental report and subcategory was used for twin infants who had monochorionic placentas. Since a placenta may contain more than one abnormality, we also evaluated placental lesions with multiple chronic lesions and combined acute and chronic lesions [15].

RESULTS
A total of 1351 cUS were reviewed from 407 preterm neonates revealing 65 neonates with LSV and 70 matched controls with normal cUS. Maternal and neonatal characteristics associated with the different stages of LSV in this cohort and our cohort of matched controls were published previously [4]. No significant differences were observed in any of the variables examined including birthweight, birthweight <10th percentile, birthweight <25th percentile, maternal diabetes, maternal hypertension, or rate of clinical chorioamnionitis. Complete placental pathologic evaluations were available for 64/65 (98%) neonates with LSV and 68/70 (97%) matched controls. There were 12 sets of twins in the study, three had monochorionic placentas.

DISCUSSION
The primary finding of this study was a higher rate of LGA placentas in neonates with LSV, particularly in those with highest stage (Stage 3) LSV. As LGA placenta frequently occurs in pregnancies complicated by maternal diabetes, importantly, our cohort did not have an increased incidence of maternal diabetes ensuring that this did not have a confounding effect on our results. In fact, none of the infants who developed Stage 3 LSV were exposed to maternal diabetes. This unexpected association between LSV and LGA placenta raises questions regarding the possibility of a common pathophysiologic pathway contributing to placental growth and development of LSV, which may involve specific prenatal exposures and/or risk factors.
Investigations linking prenatal exposures such as placental structural abnormalities and inflammation, to neonatal outcomes, known as "fetal programming," has gained significant attention in recent years [16][17][18][19]. In response to prenatal perturbations, the placenta may undergo functional adaptations such as advanced villous maturation or compensatory growth [16]. The association between disturbances in fetal environment during pregnancy resulting in small placenta are well established [20,21], but surprisingly, studies investigating compensatory placental growth and large placenta are limited.
Adaptive placental hypertrophy has been described in several types of pregnancy complications, suggesting that the physiologic drivers of this response are likely multifactorial. Analyses of offspring born in the time of Dutch famine, a period of extreme starvation from 1944-1945, demonstrated that early gestation exposure to famine following high food intake later in mid or late pregnancy resulted in increased placental weight but no change in birthweight [22], indicating a possible role for maternal nutritional status in compensatory placental growth. Alternatively, this placental growth may be mediated by pregnancy-associated plasma protein A (PAPP-A), which can be stimulated by stress and regulates the insulin-like growth factor (IGF) system, the major driver of placental growth [23]. Another landmark cohort study including 449 men and women born during 1935-1943 in Preston, England showed that elevated blood pressure was strongly and independently related to increased placental weight and low birthweight [24], which may signal an increased susceptibility to vascular diseases in pregnancies with mismatched placental and fetal growth.
Placental-to-birth weight ratio (PWR) > 90 th percentile is also associated with maternal obesity, smoking, placenta previa, and partial placental abruption, conditions which all lead to some degree of ischemic placental disease, another potential inducer of placental growth [25][26][27][28]. Similarly, in a prospective study investigating placental growth using MRI in pregnancies complicated by fetal congenital heart disease (CHD), PWR was greatest in those with cyanotic CHD, suggesting a possible response to fetal hypoxia that modulates placental growth [29]. Together, these studies indicate that the mechanisms regulating adaptive placental growth may involve multiple pathways and as yet, are not well defined.
In addition to its relevance to fetal programming, our study contributes to a growing body of literature investigating placental connections to neonatal neurologic disorders in vulnerable populations, an emerging field known as neuroplacentology [21]. The relationship between placental disturbances and echogenic changes in white matter on neonatal cUS have been well documented [30], but to our knowledge there has been only  [4], suggesting that LSV may represent the neurologic manifestation of a systemic inflammatory disease in this vulnerable population. However, our data did not show higher rates of chronic placental inflammation associated with LSV as we hypothesized.
Although we report on a large cohort with LSV and a GA and sex-matched control group using a standardized approach to placental pathology, our study has some limitations. One important limitation is the lack of a healthy low-risk control group to determine baseline rates of placental pathologies. Given the design of our study, all our enrolled controls were preterm infants ≤32 weeks GA. Prior studies of healthy control infants have reported a 10% prevalence of any of the listed placental abnormalities [13] which is significantly less than the 63% rate of placental pathology in our control group. Although, a retrospective cohort study of all neonates <29 weeks born in our institution demonstrated 75% prevalence of at least one placental pathology, and 34% multiple pathologies [15]. In addition, due to our reliance on placental pathologic examination, our study does not provide details on the evolution of placental growth during pregnancy. Similarly, we do not have information on other metrics of placental microstructure such as nutrient transport densities or functional placental imaging from these pregnancies.
The strengths of our study include the size of our cohorts and rigorous application of our LSV staging classification that has been previously validated [4]. In addition, due to the importance placed on placental pathology at our center, we had placental pathology completed in 97.5% of study subjects. A placental expert reviewed all reports and consistently applied the Amsterdam definitions of pathologic lesions to clearly delineate our results for statistical comparison.

CONCLUSIONS
In conclusion, neonates with highest stage LSV appear to have increased risk for LGA placenta without associated maternal diabetes. Given that LGA placentas could reflect an adaptation to an adverse intrauterine environment, LSV may represent the sequelae of intrauterine stress, nutritional deficit, hypoxia, and/ or inflammation. Although LSV is a radiologic finding with variable pathologic correlation, it represents an understudied neurologic abnormality that warrants continued focus to elucidate its etiology and significance. Additional studies are needed to confirm this association between LGA placenta and LSV, and to explore potential pathophysiologic underpinnings.

DATA AVAILABILITY
The datasets analyzed during the current study are available from the corresponding author on reasonable request and may require institutional data agreements.