Intravenous Immunoglobulin for the Treatment of Recurrent Pregnancy Loss , C.B. Coulam, L. Kyrsa, J. J. Stern, M. Bustillo, Genetics & IVF Institute, Fairfax VA.
Objective: To evaluate the efficacy of intravenous immunoglobulin for treatment of individuals experiencing unexplained recurrent pregnancy loss.
Design: Prospective randomized, placebo-controlled clinical trial.
Materials and Methods: 95 women experiencing 2 or more consecutive spontaneous abortions (SA) with no known cause were randomized and received either intravenous immunoglobulin (IVIG) 500/mg/kg/mo or placebo (albumin)
Results: Of 95 women participating in he study, 47 received IVIG and 48 received placebo. Medication was discontinued in 34 women who failed to conceive within 4 cycles. The remaining 61 women achieved pregnancy. Pregnancy outcomes included 29 delivery and 32 recurrent SA. Among women delivering live births 18 (62%) received IVIG and 11 (33%) received placebo. By contrast 21 (67%) women experiencing SAs received placebo and 11 (33%) received IVIG. Among 61 women who conceived, 29 received IVIG and 32 received placebo. Of the 29 women who conceived and received IVIG 18 (62%) delivered live births and 11 (33%) experienced recurrent SA. Of the 33 women who conceived and received placebo, 11 (33%) delivered live births and 22 (67%) had recurrent SA. The difference in live birth rates between women receiving IVIG and placebo was significant
(P = 0.01, odd ratio 0.2)
Conclusion: IVIG is effective in enhancing the percentage of live births among women experiencing unexplained recurrent SA.
Recurrent spontaneous abortion is a common complication of pregnancy for which there is no known cure.1 An immunologic cause has been suggested for more than 80% of otherwise unexplained recurrent spontaneous abortion and various immunotherapies have been proposed as treatment for these couples.2 White blood cell immunization has been the most widely used immunotherapy.3 The efficacy of this treatment is low, with an absolute reduction of risk of another abortion between 8% and 10%. Thus, the number of cases needing to be treated in order to prevent a single pregnancy failure was between 9 and 13.4 Because of the low treatment effect, alternative treatments for recurrent spontaneous abortion have been sought. Among alternative treatments reported to result in successful pregnancies is intravenous immunoglobulin.7,15 None of the studies reporting successful pregnancies after treatment with intravenous immunoglobulin included control subjects. We therefore undertook a prospective randomized placebo-controlled double-blinded clinical trial to define the efficacy of intravenous immunoglobulin in the treatment of recurrent spontaneous abortion. We now report the results of this randomized placebo-controlled trial.
Materials and Methods
Women experiencing two or more consecutive spontaneous abortions with the same partner were offered the opportunity of participating in an Institutional Review Board (IRB) approved randomized placebo controlled trial using intravenous gammaglobulin (IVIg) or albumin (placebo) from January 1991 to March 1994. The obstretrical histories of each of the women were obtained and the number of total pregnancies, live births stillbirths, abortions, ectopic pregnancies, and hydatidiform moles and the number of partners for each pregnancy were recorded. All couples were investigated with chromosome analysis, hysterosalpinography, and hysteroscopy, luteal phase endometrial biopsy, and serum progesterone timed with ovulation documented by ultasonic monitoring of folliculogenesis, anticardiolipin antibody (ACA) and activated partial thromboplastin time. All couples with a diagnosis of chromosomal, anatomic, endocrinologic and autoimmunologic etiology of recurrent pregnancy loss were excluded from the study. Also excluded were women less than 18 years or greater than 45 years of age, and women with a history of IgA deficiency or hypersensitivity to immunoglobulin. Each woman had blood screened for the presence of HIV antibodies and the Hepatitis B antigen.
The major determinant of sample size for this type of study is the expected proportion of subsequent pregnancies to end in a spontaneous abortion among the nonintervention group. Estimates provided in the literature suggest 40% as a reasonable expected proportion of third to sixth spontaneous abortions.1,14,16 The next major consideration is the level of reduction to be achieved by the intervention (therapy). If the causes of recurrent spontaneous abortion differ from the causes of isolated spontaneous abortions and can be eliminated by the intervention, the base line risk for subsequent abortion would be about 12%.16Thus the maximum effect of the treatment would be a decrease in risk of abortion from 0.40 to 0.12. According to traditional parameters for sample size computations, including type 1 error of 0.05, type 2 of 0.2 and a one sided test, 25 pregnancies would be required in the treated and untreated groups. This sample size requires that all of the patients conceive. Previous data indicate 73% of women with three spontaneous abortions and no viable pregnancies have a subsequent pregnancy.16Therefore both the treated and untreated groups should consist of a minimum of 43 patients. Since the reduction of the excess risk of spontaneous abortion due to "recurrent causes" might not be complete due to heterogeneity of the causes of spontaneous abortion and /or effectiveness of the intervention, a sample size of 45 in each group would ensure reliable data (Graph PAD, InStat Version 1.12a, Graph PAD Software, 1990.)
A total of 95 women were randomized using computer-generated random number, on half receiving intravenous immunoglobulin (IVIg) and the remaining one half albumin infusions. Each patient received an intravenous infusion of the follicular phase of the cycle when pregnancy was desired. Patients were randomized in a double-blinded fashion to receive either (IVIg 500 mg/kg per month of albumin 0.5% in an intravenous infusion. The patient received the infusion every 28 days until pregnant or for 4 months. If the patient was not pregnant in 4 months, she was dropped from the study and replaced with another patient. Once conception occurred, the patient received an infusion every 28 days until delivery or until 28-32 weeks gestation.
Ninety-five women participated in the study. Their mean age was 35 years (range 27-44) and gravidity 5.7 (range 2-18). Forty-six women experienced recurrent spontaneous abortion after a previous live birth (secondary recurrent spontaneous abortion), and 49 women lost two or more pregnancies with no pregnancy progressing beyond 20 weeks of gestation (primary recurrent spontaneous abortion). Of 95 women participating in the study 47 received IVIG and 48 received placebo. The mean age of women receiving IVIg was 35 years (range 27-44), mean gravidity was 5 (range 2-11), 27 women were primary aborters, and 20 were secondary aborters. No differences in age, gravidity, parity, or proportion of primary and secondary aborters existed between the group receiving IVIg and placebo. Medication was discontinued in 34 women (18 receiving IVIg and 16 placebo) because of lack of conception.
Among 61 women who conceived, 29 received IVIg and 32 received placebo. Of the 29 women who conceived and received IVIg, 18 (62%) delivered live births and 11 (38%) experienced recurrent spontaneous abortion. Of 32 women who conceived and received placebo, 11 (34%) delivered live births and 21 (66%) had recurrent spontaneous abortion. The difference in live birth rates between women receiving IVIg and placebo was significant (P=0.04, odds ratio 3.1)
Pregnancy outcomes included 29 deliveries and 32 spontaneous abortions (Table 1). Among women delivering live births, 18 (62%) received IVIg and 11 (38%) received placebo. By contrast, 21 (66%) women experiencing recurrent spontaneous abortions received placebo and 11 (34%) received IVIg.
TABLE I. Out come of 61 pregnancies randomized to receiving intravenous immunoglobulin (IVIg) or placebo (albumin)
Pregnancy Outcome n IVIg Placebo P value
# (%) # (%)
Delivery 29 18 (62) 11 (38) 0.04
Abortion 32 11 (34) 21 (66) 0.04
Blighted Ovum 15 8 (53) 7 (47) NS
Intrauterine Death 17 3 (18) 14 (82) 0.004
Total 61 29 (48) 32 (52) NS
Thirty-two women experienced recurrent spontaneous abortion (Table I.) Ultrasonographic finding of the 32 pregnancy losses included 15 (47%) empty embryonic gestational sacs or blighted ova and 17 (53%) intrauterine deaths after establishment of cardiac activity in the first trimester of pregnancy. Eight of the blighted ova occurred in women receiving IVIg and 17 in those receiving placebo. Of 17 intrauterine embryonic deaths, 3 (18%) occurred in women receiving IVIg and 14 (82%) in women receiving placebo. Among the 11 pregnancy losses occurring in women receiving IVIg, 8 (73%) were blighted ova, and 3 (27%) were intrauterine embryonic deaths. Twenty-one pregnancy losses occurred in women receiving placebo; 7 (33%) were blighted ova, and 14 (67%) were intrauterine embryonic deaths. The difference in intrauterine embryonic deaths between women receiving IVIg and placebo was significant (P<0.004, odds ratio 0.1)
No reactions to study medications occurred. One infant was born with Mosaic Down syndrome and one pregnancy was complicated by an umbilical cord accident at 30 weeks gestation. Both women received IVIg.
Analysis of results from this randomized, double blinded placebo-controlled clinical trial suggests that IVIg us efficacious in the treatment of recurrent spontaneous abortion. Another randomized, placebo-controlled trial has been performed in Germany as a multicenter study.17 A significant specific effect of IVIg on live birth rate could not be demonstrated. However , success rates for both IVIg and albumin were in the same range as allogeneic leukocytes.17 The difference in interpretation of results in the current study and the German experience 17 has at least three explanations. The first explanation involves patient selection. The women included in the studies could represent different populations with different risk factors for pregnancy loss. More sensitive and specific markers are needed to identify individuals most likely to respond to immunotherapy before differences in study populations can be prepared. The second explantation involves the differences in the study design between the two studies. In the current study, therapy was begun before conception, whereas the German study instituted all treatment after a positive pregnancy test was obtained.17 Preconception treatment for recurrent spontaneous abortion using various forms of immunotherapy has been shown to be more effective than postconception treatment.17 The third exception for differences in success rate is co-intervention by the control treatment. The reason the German study showed no treatment effect of IVIg is that the effect seen was the same as that of albumin. Both effects were the same as the treatment effect seen by IVIg in the current study and by leukocyte immunization in the worldwide prospective collaborative study.4 The concentration of albumin in the German study17 was a 5% solution in contrast to the current study in which 0.5% albumin was used. Little is known about the immunomodulating effects of albumin. Recently, soluble HLA molecules have been detected, not only in IVIg19 but also in smaller amounts, in albumin preparations.20
Intravenous immunoglobulin therapy has been previously reported to be effective in prevention of recurrent spontaneous abortion (5-13). The mechanism of this antiabortive effect is not known. Immune modulation by IVIg has been speculated to result from passively transferred blocking or antiidiotypic antibodies,22 blockage of Fc receptors,23 enhancement of suppressor T-cell function,24 down regulation of B-cell function,25 and/or reduction of activation of complement components, 13,26,27 natural killer cell function, and cytokine production.28
Whatever the mode of action, the mechanism does not maintain pregnancies associated with blighted ova29 but does maintain pregnancies that are lost as a result of intrauterine demise after the establishment of embryonic cardiac activity (Table I.).
The majority of pregnancies (73%) lost after treatment with intravenous immunoglobulin are blighted ova (Table I.). Limited data in the literature suggest that ultrasonographic demonstration of an empty sac is associated with an abnormal analysis of the chorionic villus sampling .30,31 If these observations can be confirmed in women experiencing recurrent pregnancy loss, then pregnancies lost after treatment with intravenous immunoglobulin will be those with abnormal chromosome complements. IVIg is effective in enhancing the percentage of live births among women experiencing unexplained recurrent spontaneous abortion. Since IVIg preparations are free of cells and can be quarantined for prolonged periods, IVIg provides a safer alternative for treatment of recurrent pregnancy loss than white blood cell immunization.
The authors should like to thank the following physicians who provided patients who participated in the study: S. Alexander, D. Fein, J. Langley, A. Toofanian, G. Janneck, M. Liptak, D. Koepping, G. Shuster-Haynes, M. Hinton, K.Duprey, A. Haney, J. Eberhardy, A. Hough, R. Zold, R. Nehls, D. Mullaney, R. Suarez, M. Maloney, M. Aiken, M. Jones, M. Freedman, J. Davidson, P. Taylor, C. Whitworth, D. Ross, S. Marynick, B. Wassell, L. Beard, M. Utley, S. Reager, R. Chopyck, S. Collins, C. Calvello, M. Turner, T. Markus, A. West, A. Gonzales, J. Thompson, L Underwood, J. Jones, K. Fischer, R. Reinsch, D. Bewall, A. Peters, and R. Lloyd.