Initial excitement surrounding the treatment of antiphospholipid syndrome (APS) to improve pregnancy outcomes, particularly to avoid pregnancy loss, was understandable. In the mid-1980s there was no proven “treatment” for recurrent miscarriage or otherwise unexplained fetal death . Early case series included patients with lupus anticoagulant (LA) and involved treatment during pregnancy with glucocorticoids and low-dose aspirin (LDA). Heparin was used in hopes of improving placental blood flow. Both prednisone and heparin seemed to improve pregnancy outcomes in uncontrolled case series. A multicenter randomized trial found that heparin and LDA were as effective as glucocorticoid and LDA and were associated with a better safety/adverse event profile [1]. This trial, however, was limited in terms of small sample size and uncertainties regarding patient selection. By the early 1990s, the most commonly used treatment of pregnant patients with APS was established as heparin and LDA, and so it remains.

This chapter has two sections concerning pregnant aPL-positive patients : assessment and treatment. Treatment is discussed separately for two different populations: (a) those with recurrent early miscarriage; and (b) those with a history of thrombosis , mid-second or third trimester fetal death, or early delivery because of severe preeclampsia or placental insufficiency . (The term “obstetric APS” used in this text refers to both forms.)

Given the relative infrequency of fetal death (loss of a fetus ≥10 weeks of gestation) relative to early pregnancy loss (miscarriage, before 10 weeks), proper assessment should document the gestational age of the pregnancy death (not, as is commonly done, the gestational age at diagnosis, when clinical symptoms or serendipitous diagnosis occur). Thus, an embryo that dies at eight menstrual weeks of gestation , but is first diagnosed at 11 weeks, is often wrongly considered to be a fetal death [2].

Several objective criteria can be used to confirm gestational age of a loss. Crown-rump length measured on ultrasound ≥3.0 cm documents 10-week gestational age. Other methods include having heard fetal heart tones with a handheld Doppler device at a specified time or assessment of fetal size directly after delivery or uterine evacuation [2].

The fetal death should be unexplained to qualify as an APS criterion . Other known causes of pregnancy loss should be excluded. This is not always done for a variety of reasons, including an assumption that an evaluation “will not bring my baby back”, fears or misconceptions about testing, avoidance of uncomfortable or unpleasant issues by the clinician, and financial concerns.

Considering early spontaneous abortion, chromosomal abnormality is the most frequent reason in patients with two or more miscarriages, a frequency similar to that found in patients with a single sporadic spontaneous abortion [3]. Investigation of chromosomal abnormalities in abortions is infrequently performed in clinical practice because of cost, accessibility of the test, and technical limitations. Another genetic cause of recurrent miscarriage is parental balanced translocations, detectable through analysis of parental karyotype [4].

Uterine malformations occur in up to 15% of women with recurrent pregnancy losses. Evaluation with three-dimensional ultrasound, hysterosalpingogram, diagnostic hysteroscopy, and/or magnetic resonance imaging (MRI) may be considered if suspected. Although more related to second trimester losses, some authors advise that uterine cavity evaluation should be part of an evaluation. Uncontrolled endocrine abnormalities, such as diabetes and hypothyroidism, may also cause early abortion; treatment can improve gestational outcome [4].

Recent well-designed studies have questioned whether inherited thrombophilias, infections, and alloimmune factors, previously associated with recurrent miscarriages, are worth investigating. The current answer is no, because of lack of established association and/or effective treatment [4].

The optimal evaluation for potential causes of fetal death is uncertain. Considering cost, inconvenience, and potential yield, it is reasonable to focus on tests that have a high chance of providing useful information, on the most frequent conditions and on those, such as APS, with implications for subsequent pregnancies. Although many tests are recommended in the evaluation of stillbirth [5], the most consistently useful tests are perinatal autopsy, placental histology, and genetic testing (either karyotype or chromosomal microarray). Testing for aPL is helpful in cases in which fetal growth restriction, placental insufficiency, or severe preeclampsia has occurred. Other testing is best limited to cases in which clues from the clinical history or results of autopsy or other findings suggest specific diagnoses.

Determining a cause of death can be difficult, even after a complete diagnostic evaluation. There are many different causes of fetal death, and there may be uncertainties regarding a cause. There may be more than one cause; many potential etiologies may be risk factors rather than causes, since they often occur in live births, and some conditions, for instance, diabetes, may contribute to fetal death when severe but not when mild.

Over 60 classification systems have been developed to catalog potential causes of stillbirth. None is uniformly accepted worldwide. A major hurdle is a lack of a gold standard to validate the classification system. The Stillbirth Collaborative Research Network (SCRN) , using rigorous definitions and the best available evidence to create a classification system named INCODE [6], conducted a multicenter case-control study of stillbirths and live births in the USA in five regions. In a cohort of over 500 stillbirths, INCODE identified a probable cause of death in 60.9% and a possible or probable cause in 76.2% [7].

The results of testing for antibodies to cardiolipin (aCL) and to β₂GPI-glycoprotein-I (aβ₂GPI) in the SCRN study underscore the importance of a thorough evaluation for fetal death. A higher proportion of women with stillbirth (excluding fetal anomalies) had positive tests for IgG aCL than did controls (5.0% vs 1.0%; OR 5.30; 95% CI 2.29–11.76) [8]. IgM aCL and IgG aβ₂GPI were also associated with stillbirth. Although 56 women had positive tests for aPL, only 14% of these had APS as a probable cause based on INCODE [8]. Several cases with aPL had major genetic abnormalities.

The number of requirements for valid evaluation of pregnancy loss highlights a glaring oversight of many published trials: failure to meticulously characterize [2] and stratify subjects according to the gestational age of prior pregnancy losses, exclusion of other causes, and number of prior losses required (or allowed) for inclusion in the trial. Studies are difficult to compare because the definition of aPL positivity varies [9, 10], many reported subjects having indeterminate or low levels. In the majority of published trials, many subjects meet neither the 1999 [11] nor the 2006 [12] criteria for APS. Most published trials are plagued by small sample size, lack of blinding, and lack of a placebo control.

The associations between aPL and preeclampsia and placental insufficiency are addressed in Chap. 6. Two systematic reviews emphasize that the association between aPL and preeclampsia rests on the link to preeclampsia with severe features, typically in women who are delivered preterm [13, 14].

Although the review by do Prado et al. included 12 primary studies in a meta-analysis, confirming the strong association of aPL with severe preeclampsia (OR 11.15, 95% CI 2.66–46.75), only half of the studies specified criteria for severe preeclampsia.

The review by Abou-Nassar et al. included 28 studies. LA positivity was associated with preeclampsia in case-control (OR 2.34, 95% CI 1.18–4.64) but not in cohort studies (OR 5.17, 95% CI 0.60–44.56) [14]; aCL was associated with preeclampsia in case-control studies (OR 1.52, 95% CI 1.05–2.20) but not in cohort studies (OR 1.78, 95% CI 0.39–8.16); and, paradoxically, aβ₂GPI was associated with preeclampsia in cohort studies (OR 19.14, 95% CI 6.34–57.77) but not in case-control studies. This review also evaluated preeclampsia and placental insufficiency, defined by late fetal loss, intrauterine growth restriction (IUGR), or placental abruption and found an association of IUGR with LA in case-control studies (OR 4.65, 95% CI 1.18–4.64) and with aβ₂GPI in cohort studies (OR 20.03, 95% CI 4.59–87.43) [14].

Both systematic reviews note that existing studies have high variability in their definitions of preeclampsia and placental insufficiency , often including women who deliver at term, and inclusion inconsistent with the revised Sapporo APS Classification Criteria [12]. Existing studies also include widely variable definitions of aPL positivity, including considering low thresholds of aPL, and not testing for all three aPL (LA, aCL, and aβ₂GPI). Only a handful of studies report confirmatory testing of aPL at 12 or more weeks after initial test.

An abstract presented by Gibbins and colleagues during the 15th International Congress on Antiphospholipid Antibodies used strict cutoffs for aPL positivity (>40 GPL or MPL), tested all three aPL (LA, aCL, and aβ₂GPI), and only considered a patient a case if she was delivered before 36 weeks. The investigators found an association between aPL and severe, preterm preeclampsia or placental insufficiency [15]. In this study, 10.5% of patients with preeclampsia or placental insufficiency tested positive for aPL, compared to 1.5% of controls (OR 7.59, 95% CI 1.63–35.42). Only 52.6% of women returned for repeat confirmatory testing.

In many trials, it is difficult to determine what proportion of enrolled subjects had only pregnancy losses prior to 10 weeks of gestation versus fetal deaths at/beyond 10 weeks of gestation , a distinction required by the criteria for APS [12]. Subjects are more frequently diagnosed with APS because of recurrent early miscarriage than because of prior thrombosis, fetal death, or early delivery for severe preeclampsia or placental insufficiency. Among women with recurrent early miscarriage who also test positive for aPL, the frequency of history of thrombosis, stillbirth, or pregnancy complicated by severe preeclampsia is small. These characteristics, greater numbers, and lack of comorbid conditions make a population of recurrent early miscarriage patients more amenable to treatment trials. Also, women diagnosed with APS because of prior thrombosis, fetal death, or severe preeclampsia are difficult to enroll in trials that include a placebo arm [9]. Because women with prior pregnancy losses and women with preterm birth due to preeclampsia or placental insufficiency desperately seek a modifiable cause, heparin and LDA are commonly prescribed when a positive test for aPL is found, even though the patient does not meet clinical or laboratory criteria for APS.

Many experts note the marked trial heterogeneity caused by variable entry criteria [9, 16]. In one review of association between aPL and recurrent miscarriage, almost one third of 46 studies analyzed patients with first trimester losses together with patients with second and/or third trimester losses, and 15 studies did not mention the gestational age. This is problematic, as pathogenesis of abortion varies with duration of gestation and chromosomal abnormalities being more common in early losses [9].