2024 Volume 12 Issue 2 Pages 44-49
Recommended treatment for patients at high risk for recurrent preeclampsia is low-dose aspirin before 16 weeks’ gestation. Cases in which preeclampsia is suspected in the absence of hypertension, proteinuria, or organ damage, and cases of suspected preeclampsia before 20 weeks’ gestation are reported as atypical preeclampsia. We treated a patient with a history of early-onset preeclampsia in two previous pregnancies. Despite treatment with low-dose aspirin and low-molecular-weight heparin initiated in the first trimester, intrauterine fetal death occurred at 24 weeks’ gestation. Hypertension or proteinuria was not present, but the soluble Fms-like tyrosine kinase-1/placental growth factor ratio was high; considering her medical history, a diagnosis of atypical preeclampsia was rendered. Diagnosis and treatment at an early stage in patients at high risk for atypical preeclampsia must be considered. Additionally, for aspirin non-responders such as our patient, the establishment of treatments other than low-dose aspirin for effective preeclampsia prophylaxis is required.
Preeclampsia (PE) develops when insufficient spiral artery remodeling and placental dysmaturity results in the production of anti-angiogenic factors such as soluble Fms-like tyrosine kinase 1 (sFlt-1), which causes endothelial damage leading to maternal hypertension and fetal growth restriction (FGR). In patients who developed PE in a first pregnancy, there is reportedly a high risk of its recurrence in a subsequent pregnancy.1,2) It is recommended that the risk be evaluated in terms of the mother’s medical history, uterine artery velocity waveform, and sFlt-1 level,3) and that high-risk patients be started on low-dose aspirin (LDA) before 16 weeks’ gestation.4) PE may be suspected in cases of hypertensive disorders during pregnancy; however, when PE is suspected based on clinical findings despite not meeting the clinical definition, such cases are reported as atypical PE.5,6) Reports of atypical PE have been increasing in recent years due to the wider clinical use of the sFlt-1/pregnancy growth factor (PlGF) ratio. Early-onset PE is associated with a particularly high risk of maternal complications and fetal death.7) For high-risk patients, it is important to conduct a risk assessment and initiate prophylactic intervention during the first trimester to suppress the pathophysiological progression of PE. However, no method of prophylaxis other than LDA has yet been established, and the establishment of effective methods for aspirin non-responders such as our patient is required. The basic cause of PE is immune tolerance failure, and if there was a method of prophylaxis available that is focused on immune tolerance failure, PE may be preventable with greater certainty.
A 23-year-old gravida 3 para 1 woman with the general comorbidity of latent hypothyroidism had been taking levothyroxine sodium hydrate only while pregnant. Her family history was unremarkable. She had developed PE in both her previous pregnancies and had undergone preterm emergency cesarean section.
In the course of her first pregnancy (Figure 1), systolic blood pressure had remained at approximately 130 mmHg after the first trimester. However, FGR was identified at 19 weeks’ gestation, and blood pressure increased to 171/98 mmHg at 30 weeks’ gestation. An emergency cesarean section for non-reassuring fetal status (NRFS) was conducted at 30 weeks 6 days’ gestation. No proteinuria was noted. The infant female weighed 459 g, had an Apgar score of 7/8, and umbilical artery (UA) pH of 7.309. However, she died at age 4 months due to intestinal perforation of the ileocecal junction. The infant’s karyotype was 46XX, and no chromosomal abnormality was apparent. The placenta weighed 137 g, and the results of pathological testing were consistent with hypertensive disorders of pregnancy (HDP). Postpartum, the patient’s blood pressure improved, and neither proteinuria nor impaired renal function had been evident during pregnancy. The final diagnosis was early-onset PE. Further investigations conducted 3 months postpartum were negative for antiphospholipid syndrome (APS)-related antibodies and antinuclear antibodies.
Systolic blood pressure was approximately 130 mmHg during the first trimester, but FGR was evident from 19 weeks’ gestation. Blood pressure subsequently increased, and an emergency cesarean for NRFS was conducted at 30 weeks 6 days’ gestation.
In the course of the patient’s second pregnancy (Figure 2), because her history of PE placed her at high risk, LDA was initiated at 6 weeks’ gestation. However, FGR was identified at 22 weeks’ gestation, and subcutaneous injection of heparin calcium 10,000 units/day was initiated. At 26 weeks’ gestation, hypertension (151/94 mmHg) and proteinuria (24-hour urine protein, 990 mg/day) were evident. Additionally, serious maternal organ damage, including hydrops and impaired renal function (Cr 1.4 mg/dl) became apparent, and an emergency cesarean section was performed at 28 weeks 1 day’s gestation. The infant female weighed 459 g, had an Apgar score of 7/8, and UA pH of 7.309. The child is currently growing normally with no developmental or other concerns. The placenta weighed 158 g, and the results of pathological testing were consistent with HDP. Postpartum, blood pressure and renal function both improved, and proteinuria was negative. The final diagnosis was early-onset PE.
As the patient was at high risk for PE, LDA was started at 16 weeks’ gestation. At 22 weeks’ gestation, FGR was identified and subcutaneous injections of heparin calcium 10,000 units/day were started. Hypertension developed at 26 weeks’ gestation, and as evidence of maternal organ damage (hydrops, aggravated renal dysfunction) became apparent, an emergency cesarean section was performed at 28 weeks’ gestation.
Herein, we report the third pregnancy in which the patient conceived naturally and was started on LDA 100 mg/day at 5 weeks’ gestation. Subcutaneous injections of heparin calcium 10,000 units/day were initiated at 7 weeks’ gestation. At 14 weeks’ gestation, the bilateral uterine artery resistive index (RI) was 0.8, and bilateral uterine artery notching was evident (Figure 3A). At 18 weeks’ gestation, the estimated fetal weight was 147 g (−1.86 SD), and FGR was identified. The sFlt-1/PlGF ratio was high at 527, and as the patient was at extremely high risk for PE due to her previous history, we explained to her and her husband that continuing the pregnancy would be risky for the mother and entail the possibility of intrauterine fetal death (IUFD). They elected to continue the pregnancy. An ultrasound at 23 weeks 3 days’ gestation revealed retrograde flow in the ductus venosus flow waveform (Figure 3B), and flow redistribution was also evident. The RI of the umbilical artery was 0.85 (Figure 3C), and the RI of the right middle cerebral artery was 0.69 (Figure 3D). Flow distribution was evident in comparison with the RI of the umbilical artery. After consulting with a neonatologist, the decision was made to admit the patient for in-hospital management when the estimated fetal weight reached 300 g, and she was admitted at 24 weeks 3 days’ gestation. Despite careful inpatient management, IUFD was confirmed at 24 weeks 5 days’ gestation. Because the patient had undergone two previous cesarean sections, she was at high risk for uterine rupture; therefore, delivery by cesarean section was performed at 24 weeks 6 days’ gestation. The infant male weighed 318 g, and the placenta weighed 120 g. Pathological investigation of the placenta revealed an infarction (Figure 4) and changes consistent with HDP. Neither hypertension nor proteinuria had been evident from the beginning of pregnancy to 12 weeks postpartum, and clinically the patient had not been diagnosed with HDP.
H&E staining reveals widespread infarction.
In this patient with recurrent PE, IUFD occurred at 24 weeks’ gestation. In terms of the underlying mechanism of PE, a two-stage disorder theory has been proposed. According to this theory, immune tolerance failure results in reduced placental blood flow and insufficient spiral artery remodeling (first stage). Subsequently, placental hypoxia and ischemia promote the production of antiangiogenic factors and inflammatory substances, causing both maternal and fetal circulatory insufficiency and the appearance of hypertension and FGR as clinical symptoms (second stage).8) In PE, vascular resistance remains high as a result of insufficient spiral artery remodeling, and high RI of the uterine artery flow wave and notching are evident. These findings are particularly apparent in early-onset PE and are used as predictors of this condition.9) The antiangiogenic factor sFlt-1 is used to screen for PE with levels increasing before the onset of PE, while the level of the angiogenic factor PIGF decreases, leading to an increase in the sFlt-1/PlGF ratio. This ratio has been found to be useful in a range of studies, and it is recommended that patients be managed in a higher-level medical institution when the ratio value exceeds 38.10) Our patient was at high risk for PE, considering her history of the condition, notching of the uterine artery, and a high sFlt-1/PlGF ratio. Pathological investigations of the placenta revealed a placental infarction, suggesting that uteroplacental insufficiency had occurred. It is thus probable that insufficient spiral artery remodeling and placental dysmaturity in the first trimester caused the overproduction of angiogenic factors, and uteroplacental insufficiency led to FGR and IUFD. Because the patient did not develop hypertension or proteinuria in this pregnancy, HDP could not be diagnosed; however, given her high risk for PE and the clinical findings, it is likely that this was a case of atypical PE, with the condition progressing to the second stage according to the two-stage disorder theory.
Studies have found that taking aspirin ≥100 mg/day before 16 weeks’ gestation can suppress the development of PE at <37 weeks’ gestation.2 When the Fetal Medical Foundation algorithm was applied to our patient to calculate the probability of her developing PE, she was found to be at high risk, and starting LDA before 16 weeks’ gestation was recommended.11) Our patient was prescribed 100 mg/day LDA, which is the standard dose of aspirin tablets available in Japan; however, studies in other countries have demonstrated the value of a dose of 150 mg/day.13) In aspirin non-responders such as our patient, the recommendation is to change the dose of aspirin14); thus, increasing the aspirin dose should have been considered in the present case. However, low-dose administration is recommended because aspirin not only inhibits the generation of thromboxane A2 (TXA2), which has a platelet-aggregating action by blocking COX, but it also inhibits the generation of the vasodilator prostaglandin I2 (PGI2). The TAX2 synthetase inhibitor ozagrel directly blocks TAX2 without affecting PGI2 and is reportedly effective in preventing PE15); however, clinical studies of ozagrel have not progressed since the start of the 21st century. We are considering using ozagrel to treat patients at high risk for PE who are resistant to LDA. However, both LDA and ozagrel are said to be ineffective in approximately half of patients, and although studies have also described calcium supplementation and statin therapies, these have not been established as prophylaxis. In hypocalcemia, bone resorption is increased by parathyroid hormone (PTH) and the intercellular calcium concentration rises, causing vasoconstriction. Some studies have found that calcium or vitamin D supplementation is useful for pregnant women with a low calcium intake, including those in some developing countries, but specific indications and supplement doses have yet to be determined.16,17) In our patient, the serum calcium concentration was within normal limits at 8.7 mg/dl; thus, calcium supplementation was not considered. However, we could have considered its administration after measuring intact PTH (iPTH) and other parameters to evaluate parathyroid function. Statins are the first-choice drug for the treatment of low-density-lipoprotein (LDL) hypercholestoleremia, but their actions also include suppressing vascular endothelial damage and the inflammatory response. It is now thought that they may be helpful in cases of PE, and large-scale clinical trials have yet to be conducted.18) We did not measure our patient’s LDL-cholesterol (LDL-C) level, but as there is no set LDL-C criterion for statin treatment as PE prophylaxis, its administration would have been feasible. We also treated our patient with low-molecular-weight heparin during both her second pregnancy and this pregnancy. Anticoagulant therapy with low molecular weight heparin reduces the risk of placental thrombosis and infarction and is reportedly useful for preventing placental abruption and the exacerbation of obstetric abnormalities with placental involvement, including PE and FGR.12) Because no method of PE prophylaxis other than LDA is currently available, our patient provided informed consent to receive low-molecular-weight heparin with the aim of preventing the progression of PE.
The use of sFlt-1 apheresis therapy as treatment for HDP has been reported. Apheresis therapy is a type of plasma replacement therapy that can be used to remove sFlt-1 by creating an electrolytic.19,20) In terms of the removal of antiangiogenic factors, this method of treatment is closer to a fundamental treatment for HDP in comparison with previous treatments and helps stabilize blood pressure and improve proteinuria. In early-onset PE, the sFlt-1 level increases significantly due to insufficient spiral artery remodeling,21) and as more effective treatment is desirable, apheresis therapy might have been effective for our patient. However, although one study reported an 8–24% decrease in sFlt-1 in PE patients after apheresis, pregnancy is only extended by a few days to approximately 2 weeks; thus, apheresis therapy is regarded as the PE treatment of last resort.22) A variety of studies are currently underway with the aim of increasing the efficiency of apheresis, and magnetic blood purification is one such method.23) If apheresis can be made more efficient, its therapeutic effectiveness would improve and make it a treatment option. We hope that more efficient methods of decreasing sFlt-1 will be investigated in the future.
Immune tolerance in the maternal body of the fetus, which possesses paternal antigens, is essential for the establishment of a normal pregnancy. In the immune response in the first trimester, NK cells, macrophages, and T cells present in the decidua interact to promote immune tolerance and spiral artery remodeling. The root cause of PE is immune tolerance failure in which regulatory T cells (Tregs) play a major role. Paternal antigen-specific Tregs increase through exposure to seminal plasma, and reduced Treg expression and T cell differentiation in PE reportedly differ from those in normal pregnancy.24) Immune memory cells are also involved during pregnancy, recognizing paternal antigens and differentiating into Tregs so that the paternal antigens can still be remembered after pregnancy.25) After the first pregnancy, if the father is the same in subsequent pregnancies, the expression of fetus-specific Tregs is promoted, resulting in immune tolerance. Primipara and women who become pregnant by a new partner do not have this memory of paternal antigens, which may make immune tolerance failure more likely and put them at higher risk for PE. For PE to recur, it may be that paternal antigens are remembered from the first pregnancy as substances that cannot be immunologically tolerated, making immune tolerance failure more likely to occur in subsequent pregnancies. Immune checkpoint receptors on the surfaces of the macrophages and T cells in the decidua, including PD-1, Tim-3, RANK, and CTLA, are also involved in immune tolerance, and studies have found low PD-L1 expression in patients who have suffered recurrent miscarriage.26,27) The abnormal immune response mechanisms in PE and miscarriage associated with impaired immune tolerance are gradually becoming clearer, and studies targeting immune checkpoints are currently underway.28) We are considering the use of ozagrel or statins in the future to treat aspirin non-responders at high risk for PE such as this patient. If PE develops in a pregnancy that has continued to 28 weeks’ gestation, sFlt-1 apheresis therapy may be one option. However, these methods are of limited use, and the ideal prophylaxis would improve immune tolerance failure, which is the root cause of PE.
We treated a patient at high risk for PE in whom IUFD occurred despite LDA treatment. This case was considered “atypical PE,” which is when PE is suspected from clinical findings even in the absence of hypertension. Additionally, atypical PE may occur before 20 weeks’ gestation, and it is important to conduct an early PE risk evaluation in terms of uterine artery velocity waveform and sFlt-1 level. No method of PE prophylaxis other than LDA has yet been established, and new prophylaxis and treatment that are effective for aspirin non-responders such as our patient need to be developed. The elucidation of the mechanisms of immune tolerance failure and prophylaxis focused on immune tolerance are needed if PE is to be prevented with greater certainty.
We did not receive any funding for this study. We acknowledge the contributions of the doctors in our department for their support in patient management during this study and the doctors in our district for their prompt referral and initial management.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this study.
