Neil S. Pattison

Antiphospholipid antibodies in pregnancy: prevalence and clinical associations

Objective To determine prevalence, clinical association and predictive power of antiphospholipid antibodies in pregnancy.

Recurrent miscarriage--outcome after supportive care in early pregnancy.

Summary: One hundred and thirty three couples were investigated at a recurrent miscarriage clinic In their next pregnancy 42 women (Group 1) with unexplained recurrent miscarriage were managed with a programme of formal emotional support and close supervision at an early pregnancy clinic Two women were seen in 2 pregnancies (44 supervised pregnancies); 86% (38 of 44) of these pregnancies were successful. Four of the 6 miscarriages had an identifiable causal factor. Nine women (Group 2), also with unexplained recurrent miscarriage, acted as a control group. After initial investigation they were reassured and returned to the care of their family practitioner and did not receive formal supportive care in their subsequent pregnancy; 33% (3 of 9) of these pregnancies were successful (p = 0.005; Fishers Exact Test). Whilst acknowledging that there is a significant spontaneous cure rate in this condition, emotional support seems to be important in the prevention of unexplained recurrent miscarriage, giving results as good as any currently accepted therapy.

Antiphospholipid Antibodies in Pregnancy

Summary: We assessed the relationship between antiphospholipid antibodies and recurrent miscarriage, fetal deaths, and the pregnancy complications ‐ placental abruption, fetal growth retardation and preeclampsia. The subjects were 81 women with a history of 3 or more miscarriages, 62 with a history of fetal death in the index pregnancy, 105 with a poor obstetric history or pregnancy complications and 13 with systemic lupus erythematosus. Antiphospholipid antibodies were found in 41% of women with a history of recurrent miscarriages, 29% with a history of recent intermediate fetal death or stillbirth, 19% with a poor obstetric history and 69% with systemic lupus erythematosus. There is a high incidence of antiphospholipid antibodies in complicated pregnancies. Patients presenting with the above pregnancy disorders should be tested for antiphospholipid antibodies because of the risk conferred on a fetus by their presence and to expand the treatment options.

Fetal liver length in normal and isoimmunized pregnancies.

The length of the right lobe of the fetal liver was measured by means of ultrasound examined on 53 occasions in 21 isoimmunized pregnancies. Fetal blood samples were taken in all patients within 24 hours. Comparisons were made with 350 measurements of liver length in normal pregnancies. A good correlation was found between liver length and fetal hemoglobin level (r = 0.794, p less than 0.001) and between liver length and reticulocyte count (r = 0.721, p less than 0.001). At the time of first sample, all fetuses with a hemoglobin of less than 100 gm/L had a liver length that was greater than the ninetieth percentile. Liver length measurement seems to be a useful indicator of the degree of fetal anemia in isoimmunized pregnancies.

Inhibition of heparin/antithrombin III cofactor activity by anticardiolipin antibodies: a mechanism for thrombosis.

Anticardiolipin antibodies (aCL) are autoantibodies which react with negatively charged phospholipids and are associated with thrombotic disease and recurrent fetal loss. We have found that 11% of aCL are cross-reactive with the glycosaminoglycans heparin and heparan sulphate. One of these antibodies was studied in detail and was found to inhibit the heparin dependent activation of antithrombin III by up to 80%. The inhibition of heparin dependent antithrombin III activation represents a new mechanism by which anticardiolipin antibodies may induce thrombosis or fetal loss in some patients with these antibodies.

Cofactor dependent and cofactor independent anticardiolipin antibodies

Two distinct types of anticardiolipin antibodies are described, one of which requires the presence of a serum factor (cofactor) to bind cardiolipin in ELISA and liposome affinity systems. The second type does not require this cofactor. The requirement of a cofactor for the binding of some but not all anticardiolipin antibodies provides an explanation for the confounding variability of the results of assays for these antibodies. It also explains why blocking agents such as BSA and gelatin do not produce consistent results in this assay.

Elution of anticardiolipin antibodies and their cofactor β2-glycoprotein 1 from the placentae of patients with a poor obstetric history

Anticardiolipin antibodies (aCL) were eluted from the placentae of four women with elevated serum levels of aCL, demonstrating that these antibodies are bound to affected placentae. Anticardiolipin antibodies bound to affected placentae were only of the IgG isotype and the level of aCL in placental eluates did not reflect serum levels. Anticardiolipin antibodies were not isolated from placental eluates of control normal pregnancies. beta 2-Glycoprotein 1, the anticardiolipin antibody cofactor, was present in the placental eluates from both control and antiphospholipid antibody (aPL) affected pregnancies and was localised in the syncytiotrophoblast by immunohistochemical analysis. Antinuclear antibodies were present in the placental eluates of 3 of the 4 patients with antiphospholipid antibody syndrome and were absent from the placental eluates of control pregnancies. The authors propose that anticardiolipin antibody binds directly to placental tissue, disrupting uteroplacental blood flow and/or transport through the villi.

The effect of human anticardiolipin antibodies on murine pregnancy

Anticardiolipin antibodies (aCL) were affinity purified or isolated in the IgG fraction of serum from 6 patients with antiphospholipid antibody syndrome. Anticardiolipin antibodies from one patient consistently compromised murine pregnancy. However in 92% (45 of 49) of cases injection of human anticardiolipin antibodies had no adverse effect on murine pregnancy, regardless of whether affinity purified aCL or IgG fractions were used. It is concluded that in most cases human anticardiolipin antibodies alone do not induce murine fetal loss.

Separation of Lupus Anticoagulant from Anticardiolipin Antibodies by Ion-Exchange and Gel Filtration Chromatography

Separation of lupus anticoagulant from anticardiolipin antibodies in the serum of a patient containing both antibodies is described. A simple two-step procedure utilizing diethylaminoethylcellulose ion-exchange chromatography followed by Sepharose CL-4B gel filtration chromatography allowed the separation of IgM from IgG isotype. Lupus anticoagulant was found to be exclusively IgM, whilst anticardiolipin antibodies were IgG. This is apparently the first report that anticardiolipin antibodies and lupus anticoagulant can be different isotypes and adds to the increasing evidence that they are separate entities.

Transdermal oestrogen for postmenopausal women: a double blind crossover comparative study with ethinyl oestradiol.

EDITORIAL COMMENT: The importance of postmenopausal symptoms and whether or not they can be safely controlled is undeniable — the 3 main considerations are patient compliance, complications and cost. This paper shows that transdermal administration of oestradiol was a clear winner in comparison with oral ethinyl oestradiol, although the important entity of osteoporosis was not addressed in this study. The findings reported in this paper are of more than academic interest; the trial warrants verification with a larger number of patients, and other oral oestrogen preparations (oestriol, oestrone, oestradiol) should be compared with transdermal and subcutaneous oestradiol therapy. We will be pleased to publish the findings of such studies in this journal. The pattern of uterine bleeding with different treatment regimens warrants careful documentation, since the return of menstruation is often the reason patients give for discontinuation of hormone replacement therapy.