G Model PRP-51743; No. of Pages 4
ARTICLE IN PRESS Pathology – Research and Practice xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original article
Placental findings in feto-maternal hemorrhage in livebirth and stillbirth Sanjita Ravishankar a,b,1 , Alison Migliori a , Judy Struminsky a , Phinnara Has a , C. James Sung a,b , Mai He a,b,∗,2 a b
Women & Infants Hospital of Rhode Island, Providence, RI, United States Warren Alpert Medical School of Brown University, Providence, RI, United States
a r t i c l e
i n f o
Article history: Received 6 January 2017 Keywords: Placenta Fetomaternal hemorrhage Kleihauer-Betke test
a b s t r a c t Feto-maternal hemorrhage (FMH) is not an uncommon event during pregnancy with important clinical implications for both maternal and fetal outcomes. The diagnosis is often made using Kleihauer-Betke (KB) test. As FMH occurs transplacentally, examination of the placenta may contribute to the diagnosis of FMH. This retrospective case-control study aims to examine the placental features associated FMH in patients with known positive KB test results. When compared with KB negative placentas (n = 88), KB positive placentas (n = 49) had significantly higher incidence of pallor (6/49 vs 0/88, p = 0.0017), IVT (11/49 vs. 5/88, p = 0.0032) and nRBCs (12/49 vs. 4/88, p = 0.0008). Autopsy cases with fetal or neonatal death due to FMH, (n = 13) compared to a cohort of 162 placentas associated with other, non-FMH causes of death also had significantly higher frequency of pallor (5/13 vs 0/162, p < 0.0001), IVT (6/13 vs 24/162, p = 0.011) and nRBCs (11/13 vs 67/162, p = 0.003). Pallor and nRBC were also associated with higher volume of FMH. Placental parenchymal pallor, intervillous thrombi and presence of nRBCs are significantly associated with documented FMH in both normal pregnancies and pregnancies associated with fetal or neonatal death. The presence of these findings, especially in combination, may suggest the need for maternal KB testing to rule out FMH and neonatal monitoring and/or intervention. © 2017 Elsevier GmbH. All rights reserved.
1. Introduction Fetomaternal hemorrhage (FMH) is part of a natural transfer of fetal blood cells into the maternal circulation, and occurs in approximately 50–75% of pregnancies without any adverse effects [1,2]. Depending on the volume, FMH can lead to a spectrum of maternal, fetal, and neonatal complications, some of which may affect long-term outcomes. FMH is known to be responsible for alloimmunization in Rh negative mothers. FMH may cause poor growth and cognitive deficits in neonates with mild anemia [3]. Moderate to severe FMH is a less common event, occurring 1–3 per 1000 live births, but carries more severe consequences, including fetal anemia or demise [2].
∗ Corresponding author at: Department of Pathology & Immunology, Washington University School of Medicine, 660 South Euclid, Campus Box 8118, St Louis, Missouri 63110, United States. E-mail address:
[email protected] (M. He). 1 Present address: University Hospitals Cleveland Medical Center and Case Western Reserve University School of Medicine, Cleveland, Ohio, United States. 2 Present address: Washington University School of Medicine, St. Louis, Missouri, United States.
Presenting signs and symptoms of significant FMH usually include decreased or absent fetal movements, hydrops, and nonreassuring fetal tracing [4]. However, most FMH is clinically silent. Currently known risk factors include hypertensive disorders, abruptio, trauma, substance abuse, and others. However, one study found that these risk factors were not predicative of massive FMH [5], while another demonstrated that the risk of FMH might be increased in conditions with placental permeability dysfunction [6]. FMH may be underdiagnosed as it is not routinely tested for [1]. FMH can be confirmed and quantitatively assessed by detecting the presence of fetal blood cells and fetal antigens in maternal circulation. The Kleihauer-Betke (KB) acid elution test is regarded as a definitive test for FMH; it detects fetal red blood cells (RBCs) in a maternal blood sample [7]. Fetal red blood cells can be detected in maternal blood for approximately 70 days [8]. Flow-cytometry detection using antibodies that recognize neonatal antigens such as anti-HbF, anti-I Ag, and anti-carbonic anhydrase are an alternate technique for detecting FMH. Additionally, detection of anti-D (Rh) can be used in flow cytometry when the mother is Rh negative. Detection of FMH is clinically important for determining the cause and appropriate treatment of neonatal anemia. Since most FMH occurs transplacentally, the placenta is thought to demon-
http://dx.doi.org/10.1016/j.prp.2017.02.005 0344-0338/© 2017 Elsevier GmbH. All rights reserved.
Please cite this article in press as: S. Ravishankar, et al., Placental findings in feto-maternal hemorrhage in livebirth and stillbirth, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.02.005
G Model PRP-51743; No. of Pages 4
ARTICLE IN PRESS S. Ravishankar et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
2
Table 1 Placental findings in fetomaternal hemorrhage compared to controls. Liveborns Placental finding
KB positive (n = 49)
KB negative (n = 88)
p
Placental pallor Intervillous thrombi Increased nucleated red blood cells
6 (12%) 11 (22%) 12 (24%)
0 (0%) 5 (5.6%) 4 (4.6%)
0.0017 0.0052 0.0008
Fetal/neonatal demise
Placental pallor Intervillous thrombi Increased nucleated red blood cells
KB positive (n = 13)
KB negative (n = 162)
p
5 (38.5%) 6 (46.2%) 11 (84.6%)
0 (0%) 24 (14.8%) 67 (41.4%)
<0.0001 0.011 0.003
KB: Kleihauer-Betke.
strate gross and microscopic changes that are associated with FMH [9,10]. Features suggestive of FMH include intervillous thrombi (IVT), retroplacental hematomas, and placental infarcts [9]. Retroplacental hematoma is often associated with placental abruption, an event in which FMH is expected to occur. Fetal pallor, placental IVT, and increased nucleated RBCs have been described in fetal and neonatal deaths due to FMH [10–12]. While pallor of the fetus has been observed in cases of FMH resulting in demise, pallor of the placental parenchyma may also be observed in placentas associated with FMH at the time of gross examination, a parameter which has not been frequently evaluated or reported in the literature. The aim of this study is to determine (and validate) whether the above placental features in both routine pregnancies and in fetal/neonatal deaths may aid in the diagnosis of FMH. Identification of such characteristic gross and microscopic findings may impact patient care by suggesting FMH and prompting confirmative testing, possibly even prior to recognition of clinical symptoms. 2. Methods This is a retrospective case-control study. With IRB approval (14-0067), placenta pathology reports associated with positive KB test results (≥0.1%) and autopsy cases with FMH were searched in the departmental archives. A cohort of placentas associated with negative KB test results (0%) was identified as a control group. The gestational age, placental weight and frequency of placental parenchymal pallor, IVT and nucleated red blood cells (nRBCs) were analyzed in all groups. Increased circulating fetal nRBCs is generally defined as the presence of at least 10 nucleated red blood cells seen in 10 high power (40x) fields of villous parenchyma [13]. Statistical analysis was performed using students t-test, Fisher’s exact test and linear regression with p < 0.05 as significant. The Kleihauer-Betke test is usually ordered by the obstetrician when there is suspicion for abruption; it is also ordered on all Rh negative mothers if the baby is Rh positive, in order to calculate dosage for RhIg. It is also part of the stillbirth workup. The test is performed routinely in the Hematology Laboratory following standard procedure (The Sure-Tech Fetal Hemoglobin testing kit, Procedure No. 101, Sure Tech Diagnostic Associates. Inc. St. Louis, MO). Calculation of feto-maternal hemorrhage volumes was % fetal performed using the following formula [14]: cells × 5000 ml = ml of fetal blood in maternal circulation. The blood volume of the infant can be estimated to be approximately 80 ml per kilogram of body weight. 3. Results Forty nine placentas from women with a positive KB and 88 placentas from women with a negative KB were identified. There were no significant differences in gestational age (34.3 weeks vs.
35.9 weeks, p = 0.1216) or placental weight (376.5 g vs. 406.9 g, p = 0.3209) between the KB positive and negative groups. Placental pathology is summarized in Table 1. Placental pallor, as seen in Fig. 1A, was significantly more common in patients with positive KB (6/49 vs. 0/88, p = 0.0017), as were intervillous thrombi (see Fig. 1B, 11/49 vs. 5/88, p = 0.0032) and nRBCs (see Fig. 1C, 12/49 vs. 4/88, p = 0.0008). Linear regression analysis with log-transformation of KB as an outcome showed that a categorical change from absence to presence of placental pallor is associated with a 2.18 fold increase in KB (95% confidence interval 1.19–3.18, p < 0.001). Similarly, a categorical change from absence to presence of nRBCs is associated with a 1.81 fold increase in KB (95% confidence interval 1.09–2.54, p < 0.001). Analysis of intervillous thrombi did not reach statistical significance. For the patients with placental pallor, the KB percentage was higher (mean 4.05%, median 2.35%), as compared to the overall KB positive cohort (mean 0.96%, median 0.1%), p = 0.0113. The same held true for patients with increased nRBCs (mean 3.28% vs. 0.96%, p = 0.0135). However, for intervillous thrombi, there was no significant difference in the KB percentage (mean 1.89% vs. 0.96%, p = 0.3147). Thirteen placentas associated with fetal or neonatal death due to FMH with positive KB were identified and related findings are summarized in Table 2. Gestational age (GA) in 12 cases (12/13, 92.3%) were within the third trimester, with most at term (8/12), 3 late preterm (36 weeks), and 1 preterm (33 weeks). Placental findings were compared to a cohort of placentas associated with fetal or neonatal death due to other causes (n = 162). The frequency of placental parenchymal pallor (5/13 vs 0/162, p < 0.0001), intervillous thrombi (6/13 vs 24/162, p = 0.011), and nRBC (11/13 vs 67/162, p = 0.003) were all significantly higher in the FMH group (Table 1). Besides these features targeted in the current study, five cases had features related to hydrops; one intrauterine fetal demise (IUFD) case demonstrated avascular villi involving 30% of the parenchyma, suggestive of fetal thrombotic vasculopathy and one IUFD case had a chorangioma comprising 40% of the placental weight.
4. Discussion The current study demonstrated that some macro- and microscopic placental pathology parameters, such as parenchymal pallor, intervillous thrombi, and increased nRBCs, are significantly higher in placentas with FMH that was confirmed by KB test, in both liveborns and stillborns. Pallor of the placental parenchyma is associated with decreased perfusion of the chorionic villi as a result of fetal placental blood loss or decreased blood flow, such as in cases of FMH, torn or avulsed cord vessels, or unprotected blood vessels. This study further indicates that pallor tends to occur in cases of FMH in which a large
Please cite this article in press as: S. Ravishankar, et al., Placental findings in feto-maternal hemorrhage in livebirth and stillbirth, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.02.005
KB (%)
Est. fetal blood loss (mL)
Estimated fetal blood loss (%)
Fetal weight (g)
Placental weight (g)
Placental Growth
Placental pallor
IVT
Increased nRBCs
Other Findings
1 2
18 35
38 37
0.175 4
8.8 200
3.5% 80%
2696 3162
355 340
SGA SGA
Yes Yes
No No
No Yes
3
33
36
2.4
120
75%
2070
619
LGA
No
No
No
4 5 6 7 8 9
30 35 32 33 32 31
36 37 40 36 37 38
6.8 6.9 5.8 6 8 3.5
340 345 290 300 400 175
165% 173% 104% 60% 172% 55%
2562 2520 3489 2416 2900 3965
354 341 387 368 465 567
SGA SGA SGA SGA AGA LGA
Yes Yes Yes No Yes NA
Yes No No No No Yes
Yes Yes Yes Yes Yes Yes
10
34
33
3.33
167
87%
2410
571
LGA
Yes
Yes
Yes
11 12 13
33 34 36
37 23 39
0.55 0.1 “Positive”
30 5 N/A
12% 21.3% N/A
3235 294 3080
415 111 372
AGA SGA SGA
No No No
Yes No Yes
No No Yes
N/A Hydrops, avascular villi involving 30% of the parenchyma Chorangioma (40% of placental weight), pre-eclampsia Generalized fetal pallor Fetal pallor Hydrops Fetal hydrops N/A Pericardial and pleural effusions and placental hydrops Pleural effusion and placental hydrops Abruption N/A N/A
KB: Kleihauer-Betke; GA: gestational age; IVT: intervillous thrombi; N/A: not applicable; nRBCs: nucleated red blood cells.
ARTICLE IN PRESS
GA (wks)
G Model
PRP-51743; No. of Pages 4
Maternal age
S. Ravishankar et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
Case
3
Fig. 1. A) Placental (not PAlcental). B) Intervillous (not IBntervillous). C) Nucleated (not NCucleated).
Please cite this article in press as: S. Ravishankar, et al., Placental findings in feto-maternal hemorrhage in livebirth and stillbirth, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.02.005
Table 2 KB testing and placental findings in fetal/neonatal demise.
G Model PRP-51743; No. of Pages 4
ARTICLE IN PRESS S. Ravishankar et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
4
volume of blood was lost. Determining the presence and degree of pallor requires gross evaluation by an experienced individual. IVT were among the first placental histologic findings described in association with FMH [9,15]. Fetal red blood cells have been demonstrated within IVT in an unselected group of placentas, further supporting this association [16]. The current study suggests that IVT is not associated with FMH volume. In addition, IVT have been noted in association with other, relatively common and potentially confounding conditions, including diabetes (both as a group and with gestational diabetes in particular) [17] and obesity [18]. The detection of IVT in some placentas with negative KB results (5/88, 5.7%) may suggest that there are other mechanisms leading to IVT, such as increased levels of prothrombotic factors or increased viscosity. The presence of fetal normoblastemia in the placenta is correlated with the presence of increased nucleated red blood cells in cord blood [13]. This finding has been previously described in association with fetal deaths due to FMH [12]. The results of this study again suggest that the presence of fetal nRBCs tends to increase in prevalence in cases with higher volumes of blood loss. Increased fetal nRBCs are also associated with neonatal anemia and with the presence of infectious organisms in the intrauterine environment, such as parvovirus. In 1968, Devi et al. compared certain microscopic features in placentas of liveborns who had positive versus negative KB tests, including parenchymal infarct, retroplacental hematoma, Kline’s hemorrhage (fluid blood), and intervillous thrombus (coagulated blood) [9]. These features were found to be positively correlated with FMH. A few recent studies looked into the placental and fetal characteristics seen in fetal deaths due to FMH, with one, five, and 17 cases, respectively [11,12,10]. The current study was an investigation of certain macroscopic and microscopic placental findings in FMH in both liveborns and stillborns. The larger number of cases and controls in this study made statistical analysis possible with correlation between placental features and blood loss also being investigated. Particularly noteworthy is the case with a chorangioma comprising 40% of the placental weight associated with FMH (case 3); which has only rarely been reported in the literature [19,20]. Another autopsy case of FMH (case 2) revealed avascular villi involving 30% of the placental parenchyma with a hypocoiled cord in a small for gestational age placenta. Association between FMH and fetal vascular malperfusion is rare in the literature, with only one case report of FMH associated with a clot in the umbilical vein [21]. The diversity of placental features that can be seen in FMH further conveys the importance of the few key features investigated in the current study. The current study and many previous studies have relied on the KB test to diagnose FMH. However, not all positive KB test results are diagnostic of FMH. Conditions such as hereditary persistence of fetal hemoglobin and some types of thalassemia can result in positive KB results without FMH. These caveats apply to alternate testing methods such as flow cytometry as well. Hemoglobin variant analysis, ethnic background, hematological parameters such as mean corpuscular volume (MCV), and autopsy findings such as hepatosplenomegaly can be used to aid in resolving these potential false positive test results [22].
anemia in the neonate, thus providing guidance toward appropriate and timely care. References [1] V. Stefanovic, Fetomaternal hemorrhage complicated pregnancy: risks, identification and management, Curr. Opin. Obstet. Gynecol. 28 (2016) 86–94. [2] A. Stroustrup, C. Plafkin, D.A. Savitz, Impact of physician awareness on diagnosis of fetomaternal hemorrhage, Neonatology 105 (2014) 250–255. [3] M. Kadooka, H. Kato, A. Kato, S. Ibara, H. Minakami, Y. Maruyama, Effect of neonatal hemoglobin concentration on long-term outcome of infants affected by fetomaternal hemorrhage, Early Hum. Dev. 90 (2014) 431–434. [4] G.P. Giacoia, Severe fetomaternal hemorrhage: a review, Obstet. Gynecol. Surv. 52 (6) (1997) 372–380. [5] R. Samadi, J.S. Greenspoon, I. Gviazda, R.H. Settlage, T.M. Goodwin, Massive fetomaternal hemorrhage and fetal death: are they predictable? J. Perinatol. 19 (3) (1999) 227–229. [6] T. Umazume, T. Yamada, M. Morikawa, S. Ishikawa, T. Kojima, K. Cho, N. Masauzi, H. Minakami, Occult fetomaternal hemorrhage in women with pathological placenta with respect to permeability, J. Obstet. Gynaecol. Res. 42 (6) (2016) 632–639. [7] Y.A. Kim, R.S. Makar, Detection of fetomaternal hemorrhage, Am. J. Hematol. 87 (2012) 417–423. [8] H.A. Pearson, Life span of the fetal red blood cell, J. Pediatr. 70 (2) (1967) 166–171. [9] B. Devi, R.F. Jennison, F.A. Langley, Significance of placental pathology in transplacental haemorrhage, J. Clin. Pathol. 21 (1968) 322–331. [10] D. Carles, G. Andre, F. Pelluard, O. Martin, F. Sauvestre, Pathological findings in feto-maternal hemorrhage, Pediatr. Dev. Pathol. 17 (2014) 102–106. [11] L.F. Akanli, N.E. Cohen-Addad, N.V. Malabanan, F. Margono, M.A. Krilov, Massive fetomaternal hemorrhage, Am. J. Perinatol. 14 (5) (1997) 271–273. [12] S.A. Biankin, S.M. Arbuckle, N.S. Graf, Autopsy findings in a series of five cases of fetomaternal haemorrhages, Pathology 35 (4) (2003) 319–324. [13] R. Redline, Elevated circulating fetal nucleated red blood cells and placental pathology in term infants who develop cerebral palsy, Hum. Pathol. 39 (2008) 1378–1384. [14] R.N. Baergen, Manual of Benirschke and Kaufmann’s Pathology of the Human Placenta, first ed., Springer Science + Business Media, Inc, New York, 2005, pp. 378–381. [15] R.W. Redline, Classification of placental lesions, Am. J. Obstet. Gynecol. 213 (Suppl. 4) (2015) S21–8. [16] C. Kaplan, W.A. Blanc, J. Elias, Identification of erythrocytes in intervillous thrombi: a study using immunoperoxidase identification of hemoglobins, Hum. Pathol. 13 (1982) 554–557. [17] K. Basnet, R. Bentley-Lewis, D.J. Wexler, F. Kilic, D.J. Roberts, The prevalence of intervillous thrombi is increased in placentas from pregnancies complicated by diabetes, Pediatr. Dev. Pathol. (2015), http://dx.doi.org/10.2350/15-111734-OA.1 (in press). [18] M. He, P. Curran, C. Raker, S. Martin, L. Larson, G. Bourjeily, Placental findings associated with maternal obesity at early pregnancy, Pathol. Res. Pract. 212 (4) (2016) 282–287. [19] R. Kawano, S. Takemoto, K. Shimamatsu, D. Hori, T. Kamura, Fetomaternal hemorrhage with intraplacental chorioangioma, J. Obstet. Gynaecol. Res. 39 (2) (2013) 583–587. [20] C.A. Brandt, C. Ryom, A. Grove, Chorioangioma placentae and feto-maternal transfusion; a report of two cases, Eur. J. Obstet. Gynecol. Reprod. Biol. 33 (1) (1989) 95–98. [21] A. Manoura, E. Hatzidaki, M. Christodoulaki, E. Korakaki, H. Galanaki, Z. Repapinou, C. Giannakopoulou, Anemia due to massive chronic foetomaternal hemorrhage, Haematologia 29 (4) (1999) 319–321. [22] D.L. Weaver, J.C. Barthold, B. Hamill, G.H. Sharp, B.H. Tindle, Hereditary persistence of fetal hemoglobin presenting as fetal-maternal hemorrhage, Am. J. Clin. Pathol. 93 (2) (1990) 277–280.
5. Conclusion Placental parenchymal pallor, intervillous thrombi, and nucleated RBCs are associated with documented FMH in both liveborn pregnancies and those associated with fetal or neonatal death due to FMH. The combination of these findings may prompt the pathologist to suggest confirmatory testing of the mother and/or testing for Please cite this article in press as: S. Ravishankar, et al., Placental findings in feto-maternal hemorrhage in livebirth and stillbirth, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.02.005