Placental Pathology and Neonatal Thrombocytopenia: Lesion Type Is Associated with Increased Risk
Total Page:16
File Type:pdf, Size:1020Kb
Journal of Perinatology (2014) 34, 914–916 © 2014 Nature America, Inc. All rights reserved 0743-8346/14 www.nature.com/jp ORIGINAL ARTICLE Placental pathology and neonatal thrombocytopenia: lesion type is associated with increased risk JS Litt1 and JL Hecht2 OBJECTIVE: To investigate the association between thrombocytopenia and placental lesions. STUDY DESIGN: Cases included singleton infants admitted to the intensive care unit (2005 to 2010) with platelet counts o100 000 μl−1. We selected a contemporaneous control group matched for gestational age: 49 cases and 63 controls. The frequency of thrombosis in fetal vessels, fetal thrombotic vasculopathy, acute chorioamnionitis, chronic villitis, infarcts, hematomas, cord insertion and increased circulating nucleated red blood cells were identified on retrospective review of placental histology. Logistic regression models were used to test for associations. RESULT: Placental lesions associated with poor maternal perfusion (odds ratio (OR) 3.36, 95% confidence interval (CI) 1.38, 8.15) or affecting fetal vasculature (OR 2.75, 95% CI 1.05, 7.23), but not inflammation, were associated with thrombocytopenia. A Pearson Chi-Square Test for Independence for fetal and maternal lesions indicated that the two are independent factors. CONCLUSION: Poor maternal perfusion and fetal vascular lesions are independently associated with thrombocytopenia in the newborn. Journal of Perinatology (2014) 34, 914–916; doi:10.1038/jp.2014.117; published online 19 June 2014 INTRODUCTION placental lesions affecting the fetal circulation, such as fetal Isolated neonatal thrombocytopenia is a common condition and vascular thrombosis, are associated with neonatal thrombocyto- affects between 22% and 35% of all infants admitted to the penia. A secondary aim is to assess whether placental efficiency, a neonatal intensive care unit.1 A wide spectrum of etiologies has measure of villous maturation and branching, mediates the been reported in the literature, ranging from infection, immune- relationship between placental vascular abnormalities and neo- mediated processes and derangements of the feto-placental unit.2 natal thrombocytopenia. More severe thrombocytopenia is often seen in conditions, such as neonatal alloimmune thrombocytopenia, disseminated intra- MATERIALS AND METHODS vascular coagulation and sepsis. Milder transient forms may also occur in the first few days of life. These are often attributed to Population placental insufficiency with resulting decreased fetal platelet The sample was drawn from all infants admitted to the Beth Israel production and associated with conditions such as preeclampsia Deaconess Medical Center neonatal intensive care unit between 2005 and and fetal growth restriction.3 2010. Infant products of twin gestation pregnancies were excluded. Distinguishing cases of thrombocytopenia caused by infection or immune-mediated processes from those related to problems Identification of cases with the feto-placental unit is difficult. The timing of thrombocy- With permission from our institutional review board, we searched topenia can be a helpful guide. Thrombocytopenia occurring electronic medical records for cases of thrombocytopenia, defined as − within the first 72 h of life is largely related to fetal and placental any instance of platelets o100 000 μl 1. We collected clinical data, pathology and that occurring after 72 h is associated with including gestational age, platelet count, birth weight and the main fi postnatal infection. However, there is some degree of overlap.4 diagnosis at the time of platelet count. We identi ed a control group sequentially from all admissions to the neonatal intensive care unit in the This is an important clinical concern, as the latter is currently a − same time period whose initial platelet counts were ⩾100 000 μl 1.We diagnosis of exclusion, with thrombocytopenia in the neonate matched cases and controls by gestational age group, defined categori- necessitating a diagnostic work-up for infection and hematologi- cally as 23 0/7 to 29 6/7 weeks, 30 0/7 to 36 6/7 weeks and 37 0/7 to 42 0/7 cal, metabolic and immune derangements. Furthermore, the weeks. All platelet values were measured within 72 h of birth. thrombocytopenia associated with placental insufficiency is 5,6 typically asymptomatic and self-limited. Evaluation of pathology specimens To our knowledge, there is no study linking specific histopatho- logical placental lesions with neonatal thrombocytopenia. We aim Placentas are routinely sent to pathology after delivery for any fetal or maternal indication, including prematurity. One author (JLH) reviewed the to quantify the relative incidence of placental lesions associated pathology slides and pathology reports of the corresponding placentas by with poor maternal perfusion (for example, infarcts, increased retrospectively for evidence of abnormality, with special focus on vascular numbers of trophoblastic syncytial knots) among newborns with pathology. We identified several lesion types, including: (1) recent and without thrombocytopenia. We also aim to test whether thrombosis of fetal vessels, defined by the presence of an organizing 1Department of Neonatology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA and 2Department of Pathology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA. Correspondence: Dr JL Hecht, Department of Pathology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215, USA. E-mail: [email protected] Received 15 March 2014; revised 16 April 2014; accepted 5 May 2014; published online 19 June 2014 Placenta and thrombocytopenia JS Litt and JL Hecht 915 fibrin thrombus in the wall, (2) fetal thrombotic vasculopathy, defined as The distribution of lesion types is reported in Table 2. clusters of 415 sclerotic villi with or without an associated fetal thrombus, Proportions of individual lesion types do not differ significantly such villi often also show stromal karyorrhexis or basement membrane between cases and controls. However, the combined fetal vascular calcification; and (3) hemorrhagic endovasculitis, grouped with fetal fi (28.6% vs 12.7%) and maternal vascular (38.9% vs 15.9%) thrombotic vasculopathy in our analysis, de ned as vascular wall injury fi with extravasated fragmented red cells and sometimes villous hemor- categories were signi cantly greater in the cases compared rhage. We considered other common lesions as well, including weight with controls. above the 90th or below the 10th percentile for age, acute chorioamnio- Results from the logistic regression analyses are presented in nitis with fetal inflammatory response, chronic villitis of uncertain etiology, Table 3. Estimates for the individual lesion types did not reach infarcts, parenchymal hematomas, increased circulating nucleated red thresholds for statistical significance, but both fetal vascular (odds blood cells and abnormal cord insertion. Although the presence of ratio (OR) 2.75, 95% confidence interval (CI) 1.05, 7.23) and trophoblastic syncytial knots on terminal villi and chorangiosis was noted, we did not include these data in the analysis as their diagnosis is poorly maternal vascular (OR 3.36, 95% CI 1.38, 8.15) combined lesions reproduced between pathologists. predicted neonatal thrombocytopenia. Evidence of infection on placental pathology was not associated with increased risk of Data analysis thrombocytopenia. Having any type of lesion was also associated We described sample characteristics using simple means and proportions. We then reported the percentage of cases and control case with each placental lesion. We tested whether the specific pathological findings predict thrombocytopenia using separate logistic regression models. We Table 2. Placental lesion type by group then tested the effect of placental lesion category: fetal vascular (fetal thrombotic vasculopathy, nucleated red blood cells, thrombosis), maternal Cases Controls vascular (atherosis, hematoma, infarct), and infection (acute chorio- amnionitis, villitis of uncertain etiology). Next, we tested for potential N % N % mediation by ‘placental functional efficiency’,defined as the ratio of birth- weight/placental weight.7 Finally, we performed a Pearson Chi-Square Test Fetal vascular for Independence to test whether the fetal and maternal lesion types Fetal thrombotic vasculopathy 2 4.1 2 3.2 represented a common pathway to neonatal thrombocytopenia. Nucleated RBC 5 10.2 3 4.7 Thrombosis chorion 7 14.3 3 4.8 Thrombosis stem 6 12.2 2 3.2 RESULTS Combineda 14 28.6 8 12.7 There were 112 infants in our sample, 49 cases and 63 controls Maternal vascular (Table 1). The groups were balanced with respect to very preterm 0 6 Atherosis 5 10.2 2 3.2 birth (23 to 29 ). Mean platelet count was 52.3 (s.d. 24.8) in the Hematoma 4 8.2 2 3.2 cases and 236.3 (s.d. 67.5) in the controls. The mean birth weight, Infarct 13 26.5 8 12.7 placental weight and placental functional efficiency were the Combineda 19 38.9 10 15.9 same between groups. Infection Acute chorioamnionitis 6 12.2 15 23.8 Villitis of uncertain etiology 1 2.0 1 1.6 Combined 7 14.3 16 25.4 Table 1. Sample description by group Any lesiona 34 69.4 46 63 Cases Controls Abbreviation: RBC, red blood cells. aPo0.05. N % N % Total 49 100 63 100 Table 3. Regression models predicting thrombocytopenia by lesion Gestational age type 230–296 19 42.2 28 44.4 300–366 20 44.4 18 28.6 OR 95% CI P-value 370–416 10 20.4 17 27.0 Fetal vascular Clinical indication for placental path Fetal thrombotic vasculopathy 1.30 0.18, 9.56 0.8 NAIT 1 2.2 0 0 Nucleated RBC 2.27 0.52, 10.0 0.3 Infection 2 4.4 4 6.3 Thrombosis chorion 3.33 0.82, 13.6 0.09 IUGR 6 13.3 5 6.3 Thrombosis stem 4.26 0.82, 22.1 0.08 Preeclampsia 8 17.8 11 17.5 Combined 2.75 1.05, 7.23 0.04 Abruption 1 2.2 10 30.3 PTL 7 15.6 5 7.9 Maternal vascular PPROM 0 0 10 30.3 Atherosis 3.47 0.64, 18.7 0.1 Other 15 33.3 19 30.2 Hematoma 2.71 0.48, 15.5 0.3 Mean s.d.