Objective
Fetal infection with human parvovirus B19 (hParvo-B19) has been associated mainly with fetal anemia, although data regarding other fetal hematologic effects are limited. Our aim was to assess the rate and consequences of severe fetal thrombocytopenia after fetal hParvo-B19 infection.
Study Design
We conducted a retrospective study of pregnancies that were complicated by fetal hParvo-B19 infection that underwent fetal blood sampling (FBS). The characteristics and outcomes of fetuses with severe thrombocytopenia (<50 × 10 9 /L) were compared with those of fetuses with a platelet concentration of ≥50 × 10 9 /L (control fetuses). Fetuses in whom 3 FBSs were performed (n = 4) were analyzed to assess the natural history of platelet levels after fetal hParvo-B19 infection.
Results
A total of 37 pregnancies that were affected by fetal hParvo-B19 infection were identified. Of the 29 cases that underwent FBS and had information regarding fetal platelets, 11 cases (38%) were complicated by severe fetal thrombocytopenia. Severely thrombocytopenic fetuses were characterized by a lower hemoglobin concentration (2.6 ± 0.9 g/dL vs 5.5 ± 3.6 g/dL; P = .01), lower reticulocyte count (9.1% ± 2.8% vs 17.3% ± 10.6%; P = .02), and lower gestational age at the time of diagnosis (21.4 ± 3.1 wk vs 23.6 ± 2.2 wk; P = .03). Both the fetal death rate within 48 hours of FBS (27.3% vs 0%; P = .02) and the risk of prematurity (100.0% vs 13.3%; P < .001) were higher in fetuses with severe thrombocytopenia. Fetal thrombocytopenia was more common during the second trimester but, in some cases, persisted into the third trimester. Intrauterine transfusion (IUT) of red blood cells resulted in a further mean decrease of 40.1% ± 31.0% in fetal platelet concentration.
Conclusion
Severe fetal thrombocytopenia is relatively common after fetal hParvo-B19 infection, can be further worsened by IUT, and may be associated with an increased risk of procedure-related fetal loss after either FBS or IUT.
Human parvovirus B19 (hParvo-B19) infection complicates up to 1-2% of pregnancies and is associated with a fetal vertical transmission rate of 30-50%. HParvo-B19 infection is 1 of the most common infectious causes of fetal anemia, hydrops, and death.
The most studied clinical consequence of fetal hParvo-B19 infection is the induction of apoptosis of erythroid precursors cells, which lead to fetal anemia that, in severe cases, can result in high output fetal heart failure, hydrops, and death. Recent data suggest that fetal hParvo-B19 infection can also be associated with fetal thrombocytopenia in 15-54% of cases, which is thought to be the result of a direct cytotoxic effect of hParvo-B19 on megakaryocytes. This is of major importance, because it can lead to fetal hemorrhagic complications that include hemorrhage and even exsanguination after fetal blood sampling (FBS) and/or intrauterine transfusion (IUT) of red blood cells (RBCs).
Unfortunately, data on hParvo-B19–induced fetal thrombocytopenia are relatively sparse and have been reported in few studies, most of which included only a small number of cases (n = 8-13). In only 2 of those studies was fetal thrombocytopenia the main focus of the study. Furthermore, data on factors that are predictive of severe thrombocytopenia, the impact of severe thrombocytopenia on perinatal outcome, the natural history of fetal thrombocytopenia, and the possible impact of RBC IUT on the severity of fetal thrombocytopenia are limited even more.
The aims of this study were to address several questions related to severe fetal thrombocytopenia in pregnancies that are complicated by fetal hParvo-B19 infection: (1) the rate of severe thrombocytopenia, (2) the possible impact of severe thrombocytopenia on perinatal death, (3) the factors predictive of severe thrombocytopenia, (4) the natural history of such thrombocytopenia, and (5) the effect of RBC IUT on platelet concentration.
Methods
This was a retrospective study of all pregnancies that were complicated by fetal hParvo-B19 infection that underwent FBS ± IUT at a single tertiary care fetal medicine referral center from August 1992 to January 2014 ( Figure 1 ). The study was approved by the institutional research ethics board.
Cases were identified with the use of the fetal medicine unit database; the following data were retrieved: demographics, medical and obstetric history, characteristics of the hParvo-B19 infection, gestational age (GA) at diagnosis, maternal symptoms, maternal serology, fetal manifestations (ultrasound and laboratory), and management.
The diagnosis of fetal hParvo-B19 infection was made during the investigation of fetal anemia and/or ascites or hydrops that was detected either on routine ultrasound examination or as a result of fetal monitoring because of maternal exposure to and/or infection with hParvo-B19. Doppler ultrasound evaluation of the middle cerebral artery peak systolic velocity (MCA-PSV) was used to assess for fetal anemia. Fetal hParvo-B19 infection was confirmed either by positive polymerase chain reaction for DNA on amniotic fluid or fetal ascites or the presence of immunoglobulin M antibodies in fetal blood. All women with fetal hParvo-B19 infection in whom fetal anemia was suspected were offered FBS. The most common sampling route was the intrahepatic segment of the umbilical vein (83.7%) or the umbilical vein (18.6%) at the placental cord root. Rarely, if fetal vascular access was difficult, such as at a very early GA, an intracardiac (9.3%) or intraperitoneal (2.3%) approach was used. Fetal blood was sent for rapid analysis of hemoglobin (gram/deciliter), platelets (×10 9 /liter), and reticulocyte count (percentage). If fetal anemia was confirmed, an IUT was performed at the same session of the FBS, as described previously. The level of thrombocytopenia at which platelets were transfused was at the discretion of the fetal medicine staff physician. Ultrasound examinations usually were repeated on the next day and then weekly thereafter. Subsequent FBS ± IUTs were determined by MCA-PSV values. Fetal thrombocytopenia was defined as a platelet concentration <150 × 10 9 /L and was classified as mild (100-149 × 10 9 /L), moderate (50-99 × 10 9 /L) or severe (<50 × 10 9 /L).
Data analysis was performed with IBM SPSS software (Statistical Package for the Social Sciences, version 19.0; SPSS Inc, Chicago, IL). Assuming that it is the presence of severe fetal thrombocytopenia that has the most important clinical implications, we compared the characteristics and outcome of severely thrombocytopenic fetuses (<50 × 10 9 /L) with those with platelet concentrations ≥ 50 × 10 9 /L (control fetuses). The Student t test and Mann-Whitney U test were used to compare continuous variables with and without a normal distribution. The χ 2 and Fisher exact tests were used for categoric variables. Spearman’s correlation coefficient was used to assess the correlation between platelet concentration and other continuous variables, such as fetal hemoglobin concentration and GA at diagnosis. Fetuses in whom 3 FBSs were performed were analyzed to assess the natural history and rate of recovery of platelets. Comparison of platelet levels before and immediately after RBC IUT was performed to assess the dilutional effect of RBC transfusion on fetal platelets. Differences with a probability value of < .05 were considered to be significant.
Results
Characteristics of the study and control groups
Of 37 fetuses with hParvo-B19 infection, 31 underwent FBS, in 29 of whom fetal platelets results were available ( Figure 1 ). Of these, 11 fetuses (37.9%) were severely thrombocytopenic and constituted the study group ( Figure 1 ). The control fetuses (n = 18) included 10 fetuses (34.5%) who were moderately thrombocytopenic, 1 fetus (3.4%) who were mildly thrombocytopenic, and 7 fetuses (24.1%) with a normal platelet count ( Figure 1 ).
The characteristics of the severely thrombocytopenic and control fetuses are presented in Table 1 . Women whose fetuses were severely thrombocytopenic were more likely to be >35 years old, and GA at diagnosis of fetal hParvo-B19 infection was earlier. There were no differences between the severely thrombocytopenic and control groups with respect to parity, fetal sex, maternal symptoms of hParvo-B19 infection, or fetal ultrasound findings ( Table 1 ).
| Characteristic | Severe thrombocytopenia: platelets <50 × 10 9 /L (n = 11) | Control fetuses: platelets ≥50 × 10 9 /L (n = 18) | P value |
|---|---|---|---|
| Maternal age, y a | 32.5 ± 6.1 | 30.3 ± 5.3 | .3 |
| >35 y | 5 (45.5) | 2 (11.1) | .04 |
| Parity b | 2 (1-2) | 1 (0-2) | .3 |
| Nulliparity, n (%) | 1 (9.1) | 7 (38.9) | .08 |
| Fetal male sex, n (%) | 6 (54.5) | 11 (61.1) | .7 |
| Characteristics of infection | |||
| Gestational age at diagnosis, wk a | 21.4 ± 3.1 | 23.6 ± 2.2 | .03 |
| ≤22 wk, n (%) | 8 (72.7) | 7 (38.9) | .08 |
| Maternal symptoms, n (%) | 4 (36.4) | 6 (33.3) | .9 |
| Fetal findings, n (%) | |||
| Hydrops | 10 (90.9) | 15 (83.3) | .6 |
| Ascites | 10 (90.9) | 17 (94.4) | .7 |
| Pericardial effusion | 6 (54.5) | 10 (55.6) | .9 |
| Pleural effusion | 2 (18.2) | 3 (16.7) | .9 |
| Subcutaneous edema | 5 (45.5) | 6 (33.3) | .5 |
| Placentomegaly/thickening | 3 (27.3) | 3 (16.7) | .5 |
| Echogenic bowel | 2 (18.2) | 5 (27.8) | .6 |
| Cardiomegaly | 7 (63.6) | 5 (27.8) | .06 |
| Degree of thrombocytopenia, n (%) | N/A | ||
| Severe (<50 × 10 9 /L) | 11 (100.0) | ||
| Moderate (50-99 × 10 9 /L) | 10 (55.6) | ||
| Mild (100-149 × 10 9 /L) | 1 (5.6) | ||
| Normal (≥150 × 10 9 /L) | 7 (38.9) | ||
| Management | |||
| Total fetal blood samplings, n | 17 | 28 | .9 |
| Fetuses who underwent 2, n (%) | 2 (18.2) | 4 (22.2) | .8 |
| Fetuses who underwent 3, n (%) | 2 (18.2) | 3 (16.7) | .9 |
| Platelet count (×10 9 /L) a | 31.6 ± 10.2 | 144.2 ± 120.6 | .005 |
| Platelet transfusion, n (%) | 8 (72.7) | 5 (27.8) | .02 |
| Hemoglobin (g/dL) a | 2.6 ± 0.9 | 5.5 ± 3.6 | .01 |
| Hemoglobin <4.0 g/dL, n (%) | 10 (90.9) | 9 (50.0) | .03 |
| Reticulocytes (%) a | 9.1 ± 2.8 | 17.3 ± 10.6 | .02 |
a Data are presented as mean ± SD
Severely thrombocytopenic fetuses had significantly lower hemoglobin and reticulocyte levels compared with control fetuses (2.6 vs 5.5 g/dL, and 9.1% vs 17.3%, respectively; Table 1 ). Platelets were transfused to 8 severely thrombocytopenic fetuses (72.7%) compared with 5 control fetuses (27.8%; Table 1 ).
There was a significant correlation between fetal platelet and hemoglobin concentrations (r = 0.67; P < .001; Figure 2 ), such that fetuses with lower hemoglobin levels were more likely to have lower platelet concentrations. There was also a significant correlation between fetal platelets and GA at diagnosis (r = 0.51; P = .006; Figure 3 ). Mean fetal platelet concentration was significantly lower in cases who were diagnosed with hParvo-B19 infection at ≤22 weeks’ gestation compared with those who were diagnosed later in pregnancy (52.7 vs 152.1 × 10 9 /L, respectively; P = .015; Figure 3 ).
Perinatal outcome in study and control groups
Severely thrombocytopenic fetuses were more likely to die in-utero, and this difference was significant for fetal deaths within 48 hours of the procedure ( Table 2 ). The rate of preterm delivery was significantly higher for fetuses who were severely thrombocytopenic compared with control fetuses ( Table 2 ). There were no statistically significant differences in neonatal hemoglobin or platelet concentrations between the groups ( Table 2 ).
| Outcome | Severe thrombocytopenia: platelets <50 × 10 9 /L (n = 11) | Control fetuses: platelets ≥50 × 10 9 /L (n = 18) | P value |
|---|---|---|---|
| Gestational age at first transfusion, wk a | 21.7 ± 3.3 | 23.7 ± 2.2 | .07 |
| Termination of pregnancy, n (%) | 1 (9.1) | 0 | .2 |
| Fetal death, n (%) | 4 (36.4) | 2 (11.1) | .1 |
| Within 48 hr of procedure | 3 (27.3) | 0 | .02 |
| Neonatal death, n (%) | 0 | 1 (5.6) | .4 |
| Live birth, n (%) | 6 (54.5) | 15 (83.3) | .09 |
| Gestational age at delivery, wk a , b | 32.9 ± 2.6 | 37.8 ± 2.8 | .003 |
| <37 wk | 6 (100.0) | 2 (13.3) | < .001 |
| <34 wk | 4 (66.7) | 1 (6.7) | .004 |
| Neonatal hemoglobin (g/dL) a | 20.3 ± 1.4 | 18.4 ± 5.7 | .6 |
| Neonatal platelet count (×10 9 /L) a | 115.0 ± 70.7 | 164.5 ± 111.5 | .6 |
a Data are presented as mean ± SD
Table 3 presents the details of the cases who were complicated by fetal death. In severely thrombocytopenic fetuses, 2 deaths occurred because of hemorrhage from the puncture site in the umbilical cord at the time of the FBS procedure. The 2 other cases of fetal death were diagnosed 2 and 3 days after the procedure ( Table 3 ). The 2 cases of fetal death in the control group occurred 5 and 7 days after the procedure, 1 of which was attributed to chorioamnionitis ( Table 3 ).
