during the pregnancy.
Given the heterogeneous nature of genetic disease, being alert to alternative mechanisms of causation will on occasion be rewarding. For example, during a consultation with a patient who had previously delivered a child with the autosomal recessive Meckel–Gruber syndrome, preparatory discussions about establishing the specific mutation from each parent could reveal that the father is not a carrier of a mutation in the culprit gene. Although nonpaternity is more likely, a judicious approach would also include consideration of uniparental disomy.430, 431 This mode of inheritance, in which an offspring can inherit two copies – part or all of a chromosome from one parent and no copy from the other parent – has been seen in a number of disorders, including Prader–Willi syndrome and Angelman syndrome (see discussion later and Chapter 14). About 25 percent of cases of Prader–Willi syndrome are caused by maternal uniparental disomy.432 Involvement of chromosomes 7, 11, 14, and 15 have been notable. Uniparental disomy is caused primarily by meiotic nondisjunction events and followed by trisomy or monosomy “rescue.” Most cases described have been associated with advanced maternal age and have been detected primarily in the process of prenatal genetic studies.433, 434
Recognition of the molecular basis of a disorder from which a previous child died may provide a couple with an opportunity for prenatal diagnosis in a subsequent planned pregnancy. A caveat would be the availability of analyzable tissue from the deceased child. In the recent past this was mostly not done, but with the escalation of new discoveries in genetics, tissues should now be frozen for potential future DNA analysis. The establishment of the molecular basis of recognized syndromes, previously undetectable prenatally, now provides new opportunities for couples seeking prenatal diagnosis. Examples abound and include some of the craniosynostosis syndromes, certain skeletal dysplasias, and many other disorders.
In one of our cases, a father with metaphyseal dysplasia of Schmid, troubled by the indignities and hurts of growing up with severe short stature, elected prenatal diagnosis at a preconception visit. Subsequent mutation analysis of conceived twins yielded a normal prenatal diagnosis result confirmed postnatally.435
Heterogeneity and pleiotropism also require consideration in the context of a previous child's disorder and anticipation of future prenatal diagnosis. For example, a previous child with tuberous sclerosis or a fetus with a cardiac rhabdomyoma would prompt molecular analysis of the TSC1 and TSC2 genes for more precise future prenatal diagnosis.436
A parent with a genetic disorder
Physicians are now advised to determine whether a culprit gene has been found for a specific genetic disorder under discussion, since prenatal diagnosis would then be available for that couple or their children. Adult‐onset genetic disorders (breast/ovarian cancer, colon cancer, hypertrophic cardiomyopathy, long QT syndrome) serve as examples where prenatal diagnosis is an option. The long‐established prenatal diagnoses for both presymptomatic and symptomatic neurodegenerative disorders437 continue to be expanded to include disorders such as amyotrophic lateral sclerosis and frontotemporal dementia by analysis of the C9orf72 gene.438 In prenatal diagnosis discussions for all adult‐onset disorders, there is a natural focus on the tortured questions of personal existence and self‐extinction. One example is that of a young father with CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy) who, faced with our prenatal diagnosis of this disorder, by mutation analysis of the Notch3 gene, with his wife, elected termination.439 Mutation analysis in a subsequent pregnancy assured an unaffected fetus.440
These consultations may invoke deep personal emotional conflict, especially when pleiomorphic genes are concerned. For example, a parent with tuberous sclerosis and normal intelligence could not be certain that an affected child would not have intellectual disability. This was especially evident in our series of 50 couples having prenatal diagnosis for tuberous sclerosis.436 Discovery of fetal cardiac rhabdomyoma led to sequencing of both the TSC1 and TSC2 genes in the fetus and diagnosis in one of the asymptomatic parents. Parental decisions are neither simple nor predictable. In a UK study441 of 644 deaf individuals and 143 with hearing impairment, 2 percent opined that they would prefer to have deaf children and would consider an elective abortion if the fetus was found to be hearing!
Prospective mothers with insulin‐dependent diabetes mellitus (IDDM) could find their disorder harder to control during pregnancy. Diabetes should be well controlled before pregnancy. The better the control, the lower the risk of having a child with congenital defects.442, 443 An Australian study noted that with good preconception care of type 1 IDDM, the major congenital malformation rate decreased from a high of 14 percent to 2.2 percent.444 Notwithstanding extant knowledge about IDDM and pregnancy, a report of 273 women noted rates of stillbirth (1.85 percent), perinatal mortality (2.78 percent), and congenital anomalies (6 percent).445 An important Stockholm study of 1,089 stillbirths usefully separated causes in preterm and term/post‐term births.446 Infection and intrauterine growth restriction/placental insufficiency accounted for over 44 percent of cases in about equal proportion.
The genetics of diabetes is complex with multiple types, both polygenic, multifactorial, syndromic, and monogenic in origin. The polygenic type 1 diabetes (T1DM) and type 2 diabetes (T2DM) have over 40 and 90 genes implicated, respectively. Between 1 and 5 percent of diabetes is monogenic and symptoms overlap with T1DM and T2DM diabetes.447, 448 Affected monogenic type patients mostly do not have islet autoantibodies, often have endogenous insulin production, and are frequently misdiagnosed.449, 450 Both T2DM and monogenic diabetes are often not insulin‐dependent, have a family history of diabetes, and can occur in the young. Usually, insulin resistance does not occur, nor does acanthosis nigricans in monogenic diabetics, who are mostly not obese.449
Diabetes diagnosed in the first year of life is monogenic and due to KATP channel mutations.451 There are multiple types of monogenic autosomal dominant maturity‐onset diabetes of the young (MODY), four subtypes predominating with mutations in HNFIA (52 percent), GCK (32 percent), HNF4A (10 percent), and HNF1B (6 percent).452
A precise preconception molecular diagnosis is important so as to direct appropriate treatment. No pharmacologic treatment is indicated for the GCK‐MODY type, low dose sulfonylureas are prescribed for HNF1A‐MODY and HNF4A‐MODY, with high‐dose sulfonylureas for KATP channel‐related diabetes.451
Pregestational T1DM and T2DM are associated with poorer pregnancy outcomes, including up to a fourfold higher rate of perinatal mortality.453 The poorer glycemic control at the time of conception and the first trimester, the higher the frequency of stillbirths, congenital abnormalities, perinatal morbidity and mortality, macrosomia, dystocia in labor, and maternal mortality.454–458 Obesity, with its burden of obstetric complications and congenital anomalies457 (as discussed earlier), compounds all the problems in the diabetic mother.
Pregnant women with the chronic multifactorial autoimmune disease systemic lupus erythematosus (SLE) face a host of complications. This disorder, with its predilection for women of childbearing age, is more prevalent in non‐white populations and is characterized by involvement that includes renal, cardiovascular, musculoskeletal, neurological, rheumatological, and cutaneous systems.459 Adverse pregnancy outcomes include fetal death, preterm births, intrauterine growth restriction, and neonatal lupus.460 Women with anti‐Ro/anti‐La antibodies, the latter being specific for the diagnosis of SLE and Sjögren syndrome,461 can be asymptomatic. Anti‐Ro antibodies may precede the clinical manifestations of SLE by an average of 3.6 years.462 Note, however, these antibodies are found in up to 3 percent of the general population.463
The prime consequences of having anti‐Ro antibodies is the risk of fetal/neonatal heart block and neonatal lupus. In a study of 325 children with second‐ or