BACKGROUND:
Miscarriage occurs in approximately 13% to 20% of pregnancies.1 In recurrent spontaneous miscarriage, which is the most thoroughly studied, parental karyotyping reveals chromosomal abnormalities in < 5% of couples.1 Anatomic uterine cavity abnormalities can be found in up to 20% of women.1 Recent studies suggest that the great majority of other first trimester miscarriages are caused by inherited and/or acquired thrombophilia or hypofibrinolysis, or both.2-23 However, all published studies have focused on women with recurrent pregnancy loss ≥ 3 consecutive miscarriages before 20 weeks, and none have focused on women with a single first trimester miscarriage.
Historically, known disease and environmental risk factors for clinically serious complications of pregnancy have included juvenile onset diabetes, gestational diabetes, cocaine-heroin-crack abuse, alcohol abuse, and cigarette smoking. Historically, women with otherwise unexplained recurrent first trimester miscarriages have usually been evaluated for disease and environmental risk factors (as above), and for the antiphospholipid antibody syndrome.
Pre-conception knowledge of heritable thrombophilic or hypofibrinolytic risk factors2-23 in women and with recurrent miscarriage is important. We have shown that Lovenox thromboprophylaxis is effective in women with recurrent pregnancy loss.3 Bick et al21 have shown that preconception treatment with low dose aspirin and within-pregnancy treatment with low molecular weight heparin will protect against the pregnancy complications of thrombophilia. Fragmin (a low molecular weight heparin), 5000 units/day, is effective in thromboprophylaxis of pregnancy. It is also likely that preconception treatment with low dose aspirin and within pregnancy treatment with low molecular weight heparin (Fragmin, Lovenox) will protect against very early pregnancy loss (EPL).
SPECIFIC AIM:

We will assess thrombophilia and hypofibrinolysis to determine whether, and to what degree coagulation disorders are pathoetiologic for first trimester miscarriages. Two groups of women will be studied, as follows:
- > 1First trimester miscarriage: Women having sustained > 1 first trimester miscarriage will be eligible for participation. Up to 100 women will be studied.
- > 1 normal pregnancy without any complication of pregnancy: Women with > 1 naturally occurring pregnancy with no first trimester miscarriage and without complications of pregnancy will be matched by women with = 1 complication of pregnancy (as above), by are, race, and where possible, body mass index. Up to 100 women will be studied.
PREVIOUS STUDIES FROM THE JEWISH HOSPITAL CHOLESTEROL CENTER:

We have now documented that genetic thrombophilia, particularly the G1691A Factor V Leiden mutation is a major etiology of recurrent pregnancy loss, and that thromboprophylaxis with low molecular weight heparin will, to a large degree, prevent pregnancy loss in women heterozygous for the Factor V Leiden mutation.3
As the placenta rapidly grows, it must quickly establish arterio-venous anastomoses with the maternal circulation in the endometrium and myometrium, which, of necessity must involve hemorrhage and thrombosis. It is our hypothesis, and that of Kupferminc et13 that thrombophilia and/or hypofibrinolysis, leading to increased placental thrombosis, produce placental insufficiency and complications of pregnancy.
Abstracts 2003-2004
Glueck CJ, Wang P, Bornovali S, Sieve L. Pregnancy loss, polycystic ovary syndrome, thrombophilia, hypofibrinolysis, low molecular weight heparin. Abstract, J Invest Med, 2004;52 (Suppl 2), S349 Presented, Chicago, IL, national Clinical Research Meetings, 2004.
Bornovali S, Glueck CJ, Wang P, Goldenberg N, Sieve-Smith L. Recurrent pregnancy loss, thrombophilia, hypofibrinolysis, and polycystic ovary syndrome. J Invest Med 2003, suppl 2, S364., Presented, National Clinical Research Meetings, Baltimore MD, 3/14/-16/03.
Publications 2003-2004
542. Glueck CJ, Wang P, Bornovali S, Goldenberg N, Sieve L. Polycystic ovary syndrome, the G1691A factor V Leiden mutation, and plasminogen activator inhibitor activity: Associations with recurrent pregnancy loss. Metabolism 2003;Dec 52:1627-32.
543. Glueck CJ, Wang P, Goldenberg N, Sieve L. Pregnancy loss, polycystic ovary syndrome, thrombophilia, hypofibrinolysis, enoxaparin, metformin. Clinical and Applied Thrombosis/Hemostasis, In Press, 12/18/03
EXPERIMENTAL DESIGN AND METHODS:

Subjects:
Through our website (http://www.jewishhospitalcincinnati.com/glueck/cholesterol.html) and through obstetricians, gynecologists, and family practice physicians in the tristate area, we plan to recruit two groups of 100 women, one having had ≥ 1 unexplained miscarriage and 100 women (matched to the first group by age, race, and BMI) with ≥ 1 uncomplicated pregnancy.
Inclusion criteria will be as follows, by group:
- ≥ 1 unexplained 1st trimester pregnancy loss: Women with ≥1 unexplained pregnancy first trimester pregnancy loss. These will be women with no known uterine anatomical abnormalities or known karyotypic abnormalities.
- Women with ≥1 uncomplicated pregnancy without any pregnancy loss.
Exclusion criteria
Alcoholism, juvenile onset diabetes, known cocaine, heroin, or crack use. Women with known anatomic defects (bicornuate uterus, etc) will also be excluded. Cigarette smoking will not be an exclusion criterion.
PROTOCOL:

Study Entry
- Inclusionary criteria for ≥ 1 first trimester miscarriage in the absence of known uterine anatomic abnormality or karyotypic abnormality will be verified by review of hospital and physician records, by history, and by brief physical examination. Inclusionary criteria for women with ≥ 1 uncomplicated pregnancy without 1st trimester miscarriage will be verified by review of hospital and physician records, by history, and by brief physical examination
- History: Information will be obtained on demographic details: age, race, height, weight, systolic and diastolic blood pressures and a detailed health history.
- Laboratory measures: The following measures will be obtained in a blood sample:
PCR measures: G1691A Factor V Leiden mutation, G20210A Prothrombin gene mutation, C677T and A1298C MTHFR mutations, 4G4G mutation of the PAI-1 gene:
Serologic measures: Homocysteine, anticardiolipin antibodies, lupus anticoagulant, Factor VIII, Factor XI, resistance to activated protein C, protein C, protein S, plasminogen activator inhibitor activity (PAI-Fx).
DATA ANALYSIS :

Sample Size and power estimates:
Given the prevalence of V Leiden heterozygosity of 5%, prothrombin gene heterozygosity of 6%, MTHFR C677T homozygosity or C677T-A1298C compound heterozygosity of 11%, and PAI-1 gene 4G4G homozygosity of 20%, in order to declare a significant difference against a 3 fold increase in gene mutations in the miscarriage group, at a p<.05, 123, 91, 29, and 13 patients would need to be studied. Hence, our patient group sizes of 100 should be sufficient to allow declaration of significant differences versus healthy normal controls.
Data Management:
The data management system for the study will consist of four major components; i. a baseline data entry system at the clinic; ii. a laboratory data entry system for the urine assays; iii. a computer mainframe data merging and management system; iv. a statistical analysis and summarization system.
Data Collection:
The baseline data entry system will be facilitated on a Pentium-PC using display screens, which resemble paper forms. These data screens will be windows-menu driven and will prompt data entry in an orderly sequence. Baseline data will be cumulated weekly on diskette and transferred to the mainframe computer. Duplicate data files will be maintained at the clinic and at the mainframe computer.
The laboratory data system will be maintained on a Pentium-PC using a spreadsheet program. Data will be cumulated on a weekly basis and transferred to the mainframe computer for merging with the baseline data. Duplicate data files will be maintained at both locations.
Data Editing and Documentation:
The task of editing the baseline data will occur at three levels. First, a validity check will be performed by the computer program for each data item as it is entered. Inappropriate characters or out-of-range values will be challenged. Second, when the entry of the baseline interview data has been completed, a consistency check will be performed by the computer, and prompt notification of any problems will be made. Third, when the merging of the baseline and laboratory data is made in the mainframe computer, further consistency reviews will be made.
The laboratory values will also be scrutinized at entry into the spreadsheet by the computer for validity and consistency.
Any corrections to the data base which are made subsequent to the initial entry will be noted and kept in a separate data file to provide an "audit trail" of data corrections.
Quality Control:
Duplicate assays will be prepared on a random basis and compared for consistency with their counterpart assays. Serious discrepancies will be noted and appropriate remedial action will be taken.
Data Confidentiality and Security
Several steps will be taken to ensure the privacy and security of the study data. Each user of the data entry system will be issued a password by the supervisor, or Principal Investigator. Data diskettes and duplicate diskettes will be kept in a locked file or other secure location when not in use. Merged data files at the mainframe computer will be password protected, and duplicate files will be maintained in a secure fireproof vault.
Data Analysis
Summary statistics will be prepared quarterly (monthly) and patient accrual will be monitored on a weekly basis. The rates of occurrence of the several polymorphism's of interest in the study will be noted both for maternal and fetal patients.
At a convenient intermediate time point in the study a more complete analysis of the genetic factors and their influence will be made using maximum likelihood methods and Bayesian and likelihood ratio tests. Poisson regression analysis will also be used so that ancillary baseline data can be incorporated into the analysis. To ensure the absence of bias in the final significance levels for the study, the method described in Pocock will be used for the intermediate analysis.
First: The frequency of abnormalities in the measurements of thrombophilia and hypofibrinolysis between women having at least one 1st trimester miscarriage those with ≥1 pregnancy and no miscarriage will be compared using non-parametric tests of difference and X2 analyses. Comparisons will also be made against the population distribution of wild type normal genes and thrombophilic-hypofibrinolytic gene mutations in 234 healthy normals previously studied by our group.
Second: simple correlations between the measurements of coagulation and pregnancy outcomes will be done in the 2 groups of women whose pregnancy outcomes are known.
Third: stepwise logistic regression will be done for each pregnancy outcome with the dependent variable being outcome yes (for example first trimester miscarriage) vs outcome no (no first trimester miscarriage), and explanatory variables including age, sex, number of pregnancies, Quetelet Index, fasting blood sugar, and coagulation disorders. This logistic regression will be done separately for each coagulation disorder and categorically for women having ≥1 coagulation disorder to try to dissect out the independent contributions of coagulation disorders to complications of pregnancy.
Setting:
The study will be carried out in the outpatient Alliance Cholesterol Center on the fifth floor of Burnet Ave pavilion, ABC building.
Laboratory methods and facilities
The cDNA PCR assays will be carried out in the coagulation laboratory of the Health Alliance or at the Coagulation Laboratory of Children’s Hospital Medical Center. Similarly, all serologic assays will be carried out at these laboratories.
RISKS AND BENEFITS:

Documentation of coagulation disorders of a heritable nature might, were they known to medical insurance companies, be identified as a pre-existing risk for thrombosis. However, the data for this study will be processed following strict confidentiality rules and will be released only with signed patient consent. Benefits include ability to preventatively treat severe coagulation disorders with targeted anticoagulant therapy.
Payment: There will be no financial remuneration. Parking will be free in the Alliance ABC garage.
Subject costs: There are no anticipated costs. The study is entirely free to participants
LITERATURE CITED:

The Jewish Hospital of Cincinnati, Inc. follows a policy of making all decisions concerning financial compensation or medical care for injuries occurring during or caused by participation in biomedical or behavioral research on a case-by-case basis. If you believe you have been injured as a result of any procedure of research, contact Dr. Steven Goldberg, Chairperson, IRB, 513/686-5446.