Recurrent miscarriage is associated with low concentrations of mannan-binding lectin (MBL), but it is not known below which value relative MBL deficiency becomes a significant risk factor. The sera of 397 patients (male and female) suffering from recurrent miscarriage and 376 controls were assayed for MBL and the data analysed. It was found that the lower the cut-off value, the greater the statistical strength of the association. It was concluded that only MBL concentrations </=0.1 microg/ml were clinically significant in this context. A corollary of this conclusion is that genotyping for point mutations in the structural gene for MBL would be a much less sensitive means of identifying couples at risk of experiencing recurrent miscarriage.

Mannan binding protein (MBP) may be important for host defence particularly in infancy. MBP concentration was measured in 237 umbilical cord blood samples from singleton pregnancies and compared to those of 352 blood donors. Both data sets yielded a bimodal frequency distribution, consisting of a log-normal peak and a long tail of lower values. The range (0-23 U/ml) and median (7.2 U/ml) of cord blood values were significantly lower than those of blood donors (range 0-43 U/ml; median 8.3 U/ml). MBP was also measured in the cord blood samples of 8 pairs of twin siblings. Discordant values in two pairs of twins suggest that cord blood MBP is derived from the fetoplacental unit and not from the maternal circulation by transplacental passage.

The distribution of mannan binding protein (MBP) in blood donor sera was determined by enzyme-linked immunosorbent assay to establish normal concentrations. Abnormally low MBP concentrations were found in 16% (21 out of 135) of female partners and 14% (15 out of 108) of male partners of couples experiencing recurrent miscarriage, compared with < 5% of obstetrically normal controls (P < 0.005). This relationship was even stronger (9.5 versus 1.0%) and more significant (P < 0.002) when only subjects presumed to be homozygous for the mutant allele responsible for MBP deficiency were considered. By immunohistochemistry, MBP could be demonstrated in first trimester placenta. We suggest that low concentrations of MBP within the feto-placental unit increase susceptibility to fetal loss, possibly via an infection-induced placental cytokine imbalance.

SummarySummaryThe records of 143 patients presenting with recurrent miscarriage were reviewed. Most (84 per cent) were classified as having idiopathic recurrent abortion and the clinical and laboratory characteristics of this majority subgroup were analysed. Most clinical features were unremarkable, although some degree of subfertility was reported by a third, and 5 per cent had some form of heart disorder. Laboratory findings were generally unremarkable, although elevated IgM was found in a sizeable minority. Anti-lymphocyte antibodies were strongly and significantly associated with parity, and therefore absence of lymphocytotoxins in such patients is not a consequence of a defective antibody response to paternal HLA.

OBJECTIVE:Our purpose was to investigate the putative association between immunoglobulin G antibodies to Chlamydia trachomatis and recurrent spontaneous abortions.STUDY DESIGN:Sera from 106 idiopathic recurrent aborters and 81 of their partners were tested for immunoglobulin G antichlamydial antibodies by whole inclusion immunofluorescence and compared with 3890 sera from a general antenatal population. Positive sera were further investigated by microimmunofluorescence to determine species (Chlamydia trachomatis, Chlamydia pneumoniae, Chlamydia psittaci) specificity.RESULTS:Twenty-six (24.5%) of women with recurrent spontaneous abortions had immunoglobulin G antichlamydial antibodies compared with 28 (34.6%) of their partners (chi 2 2.25, p < 0.05) and 788 (20.3%) of the general antenatal population (chi 2 1.16, p < 0.05), and the incidence of antibody positivity showed no trend with increasing number of previous abortions. Fourteen women with recurrent spontaneous abortions had antibodies to Chlamydia trachomatis, 12 to Chlamydia pneumoniae. The prevalence of antibodies to C. trachomatis did not differ significantly between women with recurrent spontaneous abortions and their partners, but the male partners had a significantly (p = 0.005) higher prevalence of Chlamydia pneumoniae antibodies. Chlamydial antibody seropositivity did not correlate with subfertility or subsequent pregnancy outcome.CONCLUSION:There is no association between immunoglobulin G antibodies to Chlamydia trachomatis and recurrent spontaneous abortion.

The 14 kD S-type lectin from human placenta may have a role in regulating the maternal immune response to fetal antigens. In this study, an immunoperoxidase technique was used to determine the distribution of the lectin at the human maternofetal interface. Tissue obtained during the first trimester of pregnancy and at term was used. The lectin was not detectable in either the villous syncytiotrophoblast or the underlying cytotrophoblast in first-trimester tissue, although some cells of the cytotrophoblast columns were reactive. It was also not detectable in villous or extravillous trophoblast populations at term. In contrast, strong reactivity was found in decidual stromal cells throughout gestation, and endometrial stromal cells were also positive. The lectin is, therefore, not a component of the immunosuppressive factors associated with syncytiotrophoblast membranes, but may have a role in either the decidual control of trophoblast migration or some functions unrelated to pregnancy, or both.

Twenty-eight recurrent miscarriage patients who had live births after leukocyte immunization were followed. Sixteen had subsequent pregnancies (without further treatment). One pregnancy was terminated, and five others were spontaneously aborted. The live birth rate was therefore 10 of 15 (67%). Any benefit from leukocyte immunotherapy does not appear to be long lasting.

The mean birthweight of babies eventually born to couples with a history of recurrent spontaneous abortion (RSA) is allegedly lower if the parents have a high degree of HLA antigen sharing (Reznikoff-Etievant et al., 1991), but this relationship has not been independently confirmed. We have re-investigated this question by analysing data from 36 families. In 22 instances, we were able to relate birthweight directly to feto-maternal HLA compatibility for the first time in such families. We were unable to confirm any appreciable influence of paternal or feto-maternal HLA sharing on birthweight or placental weight and conclude that RSA families do not differ markedly from normal families in this respect.

SummaryPatients with a history of recurrent miscarriage were studied. Some were immunised with their partners' lymphocytes, while others immunised with their own (autologous) cells served as controls. Lymphocyte antibodies were detected in the first trimester of pregnancy in only one out of nine control patients immunised with autologous cells. Approximately half of the patients immunised with their partners' cells had lymphocyte antibodies of some kind during pregnancy, but neither lymphocyte antibodies in general or any particular specificity correlated with good or bad pregnancy outcome. Any beneficial effect of leucocyte immunotherapy is therefore not the result of lymphocyte antibody production.

A total of 108 couples with recurrent spontaneous abortion (RSA) were studied to investigate the possible influence of histocompatibility antigens (HLA) on their condition and its management. HLA-B18 was shown to be at a higher frequency in RSA women, but not significantly so after statistical correction. Just over half the RSA women shared two or more HLA-A, B or DR antigens with their partners (P < 0.01), but this group did not differ from the others in clinical or laboratory features, nor in subsequent pregnancy success rate. Leukocyte immunotherapy in which the donor shared at least one HLA-DR antigen with his partner was not associated with a significant improvement in subsequent pregnancy outcome compared with HLA-DR mismatched immunotherapy. HLA antibody production following leukocyte immunotherapy was influenced by both inoculum size and degree of HLA incompatibility, but had no effect on birthweight. Tissue-typing investigations are not indicated for individual RSA patients seeking advice or treatment.

A cellular enzyme-linked immunospecific assay is described which is ideally suited to measuring the IgG antibody response to donor-specific lymphocyte infusions. The procedure is simple, sensitive, specific and objective.

Objective: To confirm that leukocyte immunotherapy stimulates the production of cardiolipin antibodies and to relate changes to pregnancy outcome.Patients: Fifty patients with idiopathic recurrent abortion were studied. Thirty-six patients received injections of their partners' leukocytes; 14, injected with their own cells, served as controls.Design: Cardiolipin antibodies were measured a month before and after leukocyte immunization. Patients who became pregnant were immunized a second time in early pregnancy, and cardiolipin antibodies were again measured a month later.Results: Thirty-six patients immunized with their partners' leukocytes showed no appreciable change in cardiolipin antibody levels a month after vaccination. Twenty-nine of them subsequently became pregnant and were immunized again in early pregnancy: again, no change in cardiolipin antibody level was observed. There was no difference between the minority who aborted again and the majority who subsequently had successful pregnancies nor between those who responded to immunotherapy by producing cytotoxic antilymphocyte antibodies and those who did not.Conclusion: Leukocyte immunotherapy does not stimulate cardiolipin antibody production in women with normal pretreatment levels of the autoantibody.

SummaryForty consecutive cardiolipin antibody positive sera were identified and the case notes of the corresponding patients reviewed. Thirteen of these patients were females without systemic lupus erythematosus who had been pregnant at least once. Five (38 per cent) of them had had proteinuric pre-eclampsia in first pregnancy. There was also a high frequency of history of miscarriage in these patients. A review of their known clinical and laboratory features failed to reveal characteristics which could be confidently used to predict cardiolipin antibody positivity in apparently healthy patients, and we suggest that routine antenatal screening for cardiolipin antibodies may be justified.

Tissue-typing for HLA-A, B, and DR antigens was carried out on 53 babies, 47 of them unrelated, born to mothers known to be HIV-infected from intravenous drug usage or sexual contact with drug users. These babies were followed up to assess whether HLA phenotype was associated with vertical transmission of HIV infection or disease progression. Of the 47 unrelated babies, eight became infected with HIV. The frequency of HLA-DR3 was three times higher in the HIV-positive infants compared to the HIV-negative infants (43 per cent vs 15 per cent) in our study population. Conversely, HLA-A3 was three times less common in the HIV-positive infants (12.5 per cent vs 42 per cent). A comparison of HLA antigens between our study group babies and babies born to healthy mothers unselected for HIV status revealed higher proportions of HLA-B18, B7, and DR2 in the study group. Moreover, the combination, A3, B7, DR2 was four times commoner in our study population relative to controls (RR = 3.9; p less than 0.003), but was found only in babies who were not HIV infected. The combination A1, B8, DR3, in contrast, was found less often than expected in our study group (RR = 0.39) and was disproportionately represented amongst the infected babies. We have observed an unexpectedly low (6 per cent) mother-to-infant transmission rate of HIV among prospectively studied intravenous drug users. We speculate that the unusually high ratio of the common antigen combinations (often halotypes), A3, B7, DR2 to A1, B8, DR3 in this population may be contributory.

We present here anticardiolipin antibody data in endometriosis patients in relation to fertility status that was not considered in previous reports. Our results, although largely negative, are consistent with previous data, but we propose an alternative view of their significance