In clinical genetics, as in all of medicine, accurate diagnosis is the most important first step in patient care.

The long list of syndromes associated with congenital malformations can be overwhelming to the clinician. More than 400 conditions are listed in Smith’s Recognizable Patterns of Human Malformations, and more than 1000 are accessible through the POSSUM or London Dysmorphology computerized databases (see Internet Resources at the end of this chapter). This number imparts a sense that the diagnosis of a malformation syndrome lies in the arena of academic trivia. However, this is not the case.

Consider, for example, the child who is large for gestational age and has a number of physical abnormalities: omphalocele (intestinal protrusion at the umbilicus), large tongue, facial angioma, flank mass, and asymmetrical limb length ( Fig. 15.1 ). His family has questions such as, “What does he have?,” “How will he do?,” “Will he look different?,” “Will he have developmental delays?,” “What is the chance of his condition occurring again in another child?”

An infant with Beckwith-Wiedemann syndrome. Note the prominent eyes and large, protruding tongue.

By putting these features together and making the pattern recognition diagnosis of the Beckwith–Wiedemann syndrome (BWS), the clinician is able to answer all of the parents’ questions fairly precisely. Most cases of Beckwith–Wiedemann syndrome occur sporadically, but some are inherited. In addition, the genes that are involved in causing the condition exhibit imprinting effects (see Chapter 5 ). If the child has the typical methylation alteration that causes the majority of cases, the sibling recurrence risk is low and less than 1%. If the results of the methylation analysis are normal, the recurrence risk can be higher, and linkage or mutation analysis can provide a more precise risk estimate. In future pregnancies, prenatal diagnosis using sonogram can test for an omphalocele in the second trimester and for large size for gestational age, excessive amniotic fluid (polyhydramnios), and large tongue. If a fetus is thought to have BWS, then the delivery plan would change and the baby should be born in a tertiary care center.

Children with the syndrome do not usually have cognitive disabilities. Although the large tongue can cause orthodontic problems, speech difficulties, and occasionally upper airway problems, these conditions usually improve as the child gets older. The facial appearance is not strikingly abnormal in later childhood.

Chromosome analysis using a cytogenomic microarray (see Chapter 6 ) should be considered, although most Beckwith–Wiedemann patients do not have the chromosome 11p duplication that has been reported in a small number of cases.

The main foci of the medical care plan include regular tumor surveillance because children with BWS have a 5% to10% risk of developing specific embryonal tumors, and the neoplasms are treatable if detected early. The surveillance consists of abdominal sonograms to look for intraabdominal malignancies, especially Wilms tumor and hepatoblastoma, every 3 months, along with serum alpha-fetoprotein (AFP), until age 4 years; thereafter the recommended screening consists of renal sonograms until age 7 years.

In this example, it was important to diagnose the Beckwith–Wiedemann syndrome. The correct label led to precise information for genetic counseling, prediction of natural history (including reassurance), organization of appropriate laboratory studies, a health supervision plan, and referral to the specific genetic support group. Accurate diagnosis was helpful to the parents, the family physician, and the child.

The universe gives special-needs children to regular everyday parents, who choose to step up and be strong for their child! That special child teaches you life lessons that you never could have anticipated, and throughout your journey together, you learn how to become the parent that your child needs. In closing, together, through awareness, love, and community, we can all keep this possibility alive for future families.

One of the common discussions on ward rounds is the notation that a person’s family history is negative or “noncontributory.” This is often thought to exclude a genetic disorder. However, the majority of persons who have a genetic disease do not have a positive family history. A quick review of the mechanisms of Mendelian, chromosomal, and multifactorial disease inheritance illustrates that a lack of other affected persons in the family is common and does not by any means exclude the presence of a genetic disease. For example, the sibling recurrence risk is 25% for diseases with autosomal recessive inheritance. Thus a significant number of families with multiple offspring have only one affected child and no family history. Even some well-established autosomal dominant disorders often present a “negative” family history because of high proportions of new (“ de novo ”) mutations (examples include Marfan syndrome, neurofibromatosis type 1 [NF1], and achondroplasia, in which the percentages of cases caused by new mutations are 30%, 50%, and 80%, respectively). Chromosomal syndromes usually have a low recurrence risk. Even when a parent carries a balanced chromosome rearrangement, the recurrence risk among the offspring is usually less than 15%. The sibling recurrence risks for multifactorial conditions are usually 5% or less.

Case

A family comes to clinic with a 6-year-old boy who has 10 café-au-lait spots exceeding 0.5 cm in diameter widely distributed on the trunk and limbs and an optic glioma detected on a brain MRI ( Fig. 15.2 ). The family has questions about the diagnosis and the recurrence risk in future pregnancies. On initial telephone contact it is learned that there is no history of a family member with similar features.

A 6-year-old boy with multiple café-au-lait spots.

From Burger P, Scheithauer B, Vogel FS: Surgical Pathology of the Nervous System and Its Coverings , 4th ed. Philadelphia: Churchill Livingstone; 2002.

There are several possible explanations for this finding. Exploring them underscores the implications of the so-called “negative family history”:

• ▪
  

  New mutation of the NF1 gene. Because of the relatively high percentage of new mutations for this disorder, this is the most likely explanation.

  
  
• ▪
  

  Variable expression. It is also possible that one of the parents carries the gene but has mild expression of the phenotype. Occasionally a parent has multiple café-au-lait spots and a few unrecognized neurofibromas, but a diagnosis of NF1 has never been made. Thus it is important to evaluate the parents for a mild expression of NF1.

  
  
• ▪
  

  Incomplete penetrance. This is a possibility; however, it is unlikely for NF1, in which penetrance is close to 100%. If a family has two children with NF1 and neither parent has the gene, germline mosaicism would be the more likely explanation.

  
  
• ▪
  

  Incorrect diagnosis. One of the assumptions and basic principles of medical genetics is accurate diagnosis. This patient meets the National Institutes of Health established criteria for NF1 (see Chapter 4 ). However, if this patient had only café-au-lait spots, then the certainty of diagnosis would have been an issue. One would need to know the differential diagnosis of multiple café-au-lait spots.

  
  
• ▪
  

  False paternity. Although it is relatively unlikely, this possibility must be kept in mind.

We began with an individual who had a classic autosomal dominant disorder with no family history. This can be explained in a number of ways. The statement that there is “a negative family history” should not be considered conclusive evidence against the presence of a heritable condition.

The birth of a newborn with Down syndrome presents many challenges. Typically, the infant is not acutely ill, and unless prenatal diagnosis has been carried out (see Chapter 13 ), the parents will not be aware of the diagnosis before the birth. Thus the practitioner must approach the parents, often strangers, with unexpected and sometimes difficult news. The family can experience a series of emotions that are somewhat similar to the reactions after a loss: anger, denial, sadness, and then usually reorganization and adaptation. Families face these situations with markedly different backgrounds; varying attitudes toward crisis, different demographic and socioeconomic circumstances, and a wide range of differences in the cultural meaning of a disability or birth defect all play a role in this setting. All of these variables, plus the fact that physicians are often not trained in being the bearers of unexpected news, can make this a challenging situation. Parents remember in detail the way the news is presented. The practitioner has both the opportunity and the challenge to help and guide the family through these events.

A number of practical suggestions have come from studies investigating the recommendations of parents who have experienced this event:

• ▪
  

  Prepare yourself. Set up the interview scenario, and think about how you will begin the discussion. Think about the script. Starting with “Congratulations on the birth “ of their son or daughter establishes a caring tone from the outset.

  
  
• ▪
  

  Talk to both parents together whenever possible. This is sometimes not practical, but when it can be accomplished, it is critical.

  
  
• ▪
  

  Communicate the diagnosis as soon as possible. All studies of parental interviews show that they prefer early communication of the diagnosis.

  
  
• ▪
  

  Choose a place that is private and quiet where both the parents and the professionals can sit down. Avoid standing up with the parents seated. Always be sure to introduce yourself. Structure the interview from the beginning.

  
  
• ▪
  

  Humanize the situation as much as possible. Learn the baby’s first name if it has been decided on, and always know the baby’s gender. Refer to the infant by name or as a son or daughter, and be aware of the use of all language. Phrases such as “mental retardation” have great impact, and currently that term is not considered the appropriate designation by professionals in pediatrics, psychology, and medical genetics (it has been replaced by intellectual disability or cognitive disability). Terms such as “mongolism” are not appropriate because they are stigmatizing, pejorative, and incorrect.

  
  
• ▪
  

  Develop a sense of realistic positivism. It is important to discuss the developmental challenges in a person with Down syndrome, but it is also important to have an optimistic and positive attitude. Phrases or language conveying a loss of hope are not appropriate. This suggestion comes from the advocacy and parents’ organizations that have developed in the last six decades.

  
  
• ▪
  

  Answer the parents’ questions, but avoid technical overload. It is important to be accurate and current on the biological and medical aspects of the condition under discussion. When an answer is not known, mention that the question can be reviewed or referred to a consultant.

  
  
• ▪
  

  Listen actively. Assume that almost all feelings are natural and that parents will be wrestling with their own guilt and shame. Validate all feelings that arise. Most parents can meet this challenge quite effectively and do not require psychiatric consultation.

  
  
• ▪
  

  Refer the family to the appropriate resources early. This would include early intervention programs, parent advocacy groups, and even individual parents who have a child with Down syndrome. Share available written material and online resources, but make certain that they are accurate and current.

Above all, be aware of the unique plight of families in such a situation, and make an effort to spend time with them. Although it is difficult to present in written form how one can develop attributes such as kindness and empathy, it is important for physicians in training to learn from their mentors and use their own individual communication style as a strength. Clearly, the recommendations provided here apply not only to genetic counseling, but also to any situation in which difficult information is presented to patients or families. (See Suggested Readings at the end of the chapter.)

The estimation of recurrence risks was treated at some length in Chapter 4, Chapter 5 . A typical example of recurrence risk estimation is a case in which a man with hemophilia A, an X-linked recessive disorder, produces a daughter (individual II-1 in the following diagram). Because the man can transmit only the X chromosome carrying the hemophilia A mutation to his daughter, she must be a carrier. The carrier’s daughter, individual III-6, has a 50% chance of receiving the X chromosome carrying the mutation and being herself a carrier. Even though the daughter in generation III has five normal brothers, her risk remains 50% because we know that the mother in generation II is a carrier.

Suppose now that the woman in generation III produces three sons (generation IV), none of whom has hemophilia A. Intuitively, we might begin to suspect that she is not a carrier after all. How can we incorporate this new information into our recurrence risk estimate?

A statistical principle that allows us to make use of such information is called Bayes theorem (the application of Bayes theorem is often termed Bayesian analysis or Bayesian inference). The table below summarizes the basic steps involved in Bayesian analysis. We begin with the prior probability that the woman in generation III is a carrier. As its name suggests, the prior probability denotes the probability that she is a carrier before we account for the fact that she has produced three normal sons. Because we know her mother is a carrier, this woman’s prior probability must be 1/2. Then the prior probability that she is not a carrier is 1−1/2, or 1/2.

Next we take into account the woman’s three normal sons by estimating the probability that all three of them would be normal given that she is a carrier. Because this probability is conditioned on her carrier status, it is termed a conditional probability. If she is a carrier, the conditional probability that all three of her sons are normal would be (1/2) ³ , or 1/8. We also estimate the probability that all of her sons would be normal given that she is not a carrier. This conditional probability is, of course, very close to 1.

Next we want to find the probability that the woman is a carrier and that she is a carrier with three normal sons. To obtain the probability of the cooccurrence of these two events, we multiply the prior probability times the conditional probability to derive a joint probability (i.e., the probability of both events occurring together, a concept discussed in Chapter 4 ). The joint probability that she is a carrier is then 1/2 × 1/8 = 1/16. Similarly, the joint probability that she is not a carrier is 1/2 × 1 = 1/2. These joint probabilities indicate the woman is 8 times more likely not to be a carrier than to be a carrier.

The final step is to standardize the joint probabilities so that the two probabilities under consideration (i.e., being a carrier versus not being a carrier) sum to 1. To do this, we simply divide the joint probability that the woman is a carrier (1/16) by the sum of the two joint probabilities (1/16 + 1/2). This yields a posterior probability of 1/9 that she is a carrier and 8/9 that she is not a carrier. Notice that this standardization process allows us to provide a risk estimate (1/9, or 11%), while preserving the odds of noncarrier versus carrier status indicated by the joint probabilities.

Having worked through the Bayesian analysis, we see that our intuition was confirmed: The fact that the woman in question produced three normal sons reduced her risk of being a carrier substantially, from an initial estimate of 50% to a final probability of only 11%.

Another common application of Bayesian analysis is illustrated in part A of the diagram that follows. The male in generation II is affected with Duchenne muscular dystrophy (DMD), a lethal X-linked recessive disease (see Chapter 5 ). Either his unaffected mother is a carrier of the mutation, or he received a new mutation on the X chromosome transmitted by his mother. It is important to determine whether the mother is a carrier or not, because this fact will influence recurrence risks for DMD in her subsequent offspring. If the mother has only one affected offspring, the probability that she is a carrier can be evaluated directly, because one-third of all cases of X-linked lethal recessive disorders arise as a result of new mutations. (To understand this, consider the fact that because females have two X chromosomes and males have only one, 2/3 of all X-linked disease-causing mutations in a population must be found in females. For a lethal X-linked recessive, all of the male X chromosomes are eliminated from the population in each generation. Yet the frequency of the mutation remains the same, generation after generation. This is because new disease-causing mutations arise at the same rate as the loss of mutation-containing X chromosomes. Because one-third of the mutation-containing X chromosomes are lost each generation, it follows that one-third of the mutations in the population must occur as the result of new mutation.) If the probability that the affected son received a new mutation is 1/3, then the probability that the mother is a carrier—the alternative possibility—must be 1−1/3, or 2/3.

In the table below, we use Bayesian analysis to evaluate the probability that the mother is a carrier. As in the previous example, we derive a prior probability that she is a carrier, assuming no knowledge that she has produced an affected son. This probability is given by 4μ, where μ is the mutation rate for the DMD locus (i.e., the probability, per generation, that a disease-causing mutation arises at this locus in an individual). The derivation of the probability, 4μ, is beyond the scope of this text, but it can be found elsewhere (see Suggested Readings, Hodge, 1998). Because the prior probability that the mother is a carrier is 4μ, the prior probability that she is not a carrier is 1−4μ, which is approximately equal to 1 because μ is very small. The conditional probability that the woman transmits the mutation given that she is a carrier is 1/2 (there is also a very small probability that she transmits her normal allele, which is then mutated, but this can be ignored). The conditional probability that she transmits a mutation given that she is not a carrier (i.e., the probability that a new mutation arises in the gamete she transmits) is μ. We then multiply the prior probability that she is a carrier, 4μ, by the corresponding conditional probability, 1/2, to obtain a joint probability of 2μ. The same procedure produces a joint probability of μ that she is not a carrier. Finally, we standardize the joint probabilities to get the posterior probabilities. The posterior probability that she is a carrier is 2μ ÷ (2μ + μ) = 2/3, and the posterior probability that she is not a carrier is μ ÷ (2μ + μ) = 1/3. As expected, these probabilities correspond to the ones we obtained by simple direct observation.

Suppose, however, that the woman has had an affected son and an unaffected son (see part B of the figure below). This gives us additional information, and intuitively it increases the possibility that she is not a carrier (i.e., that the one affected offspring is the result of a new mutation). In the table below, we incorporate this new information. The prior probabilities remain the same as before (i.e., we assume no knowledge of either of her offspring). But the conditional probability of transmission, given that she is a carrier, changes to account for the fact that she now has two offspring: 1/2 × 1/2 = 1/4 (i.e., the probability that she did not transmit the mutation to one offspring times the probability that she did transmit the mutation to the other offspring). The conditional probability that she transmitted a new mutation to the affected offspring is μ, and the probability that she did not transmit a mutation to the unaffected offspring is 1−μ. Thus the probability of both events, given that she is not a carrier, is μ × (1−μ) ≈ μ. The joint and posterior probabilities are obtained as before, and we see that the woman’s chance of being a carrier is now reduced from 2/3 to 1/2. Again, this confirms (and quantifies) our expectation.

Before the advent of disease diagnosis through linked markers or mutation detection, Bayesian analysis was often the only way to derive a risk estimate in situations such as these. Now, of course, an attempt would be made to identify the factor VIII or DMD mutation that causes hemophilia A or DMD in these families directly, or, failing that, linked markers would be used. This is a much more direct and accurate approach for determining carrier status. However, as discussed in Chapter 13 , it is not always possible to identify the responsible mutation, particularly when a large number of mutations can cause the disorder (as is the case for hemophilia A, DMD, or cystic fibrosis). Bayesian inference can be used in such cases to incorporate the sensitivity of the genetic test (e.g., if a standard mutation analysis of the CFTR gene reveals 85% of the mutations [see Chapter 13 ], there is a 15% probability that the person in question has the mutation even though the test did not reveal it). In addition, linkage analysis is not always informative. Thus Bayesian analysis is still sometimes a useful tool for refining risk estimates.

The additional information incorporated in Bayesian analysis is not confined to the assessment of health status in relatives, as was shown in these examples. Another type of information is a biochemical assay, such as factor VIII activity level, that could help to indicate carrier status. Because there is usually overlap between carriers and normal homozygotes for such tests, the assay cannot determine carrier status with certainty, but it does provide a probability estimate for incorporation into Bayesian analysis. In diseases with delayed age of onset, such as adult polycystic kidney disease, the probability of being affected at a certain age can be used in a Bayesian analysis. Here, one considers the fact that the at-risk person is less and less likely to possess the disease gene if he or she remains unaffected beyond a certain age.

The majority of geneticists subscribe to the principle of nondirectiveness in reproductive counseling. Information about risks, natural history, treatment, and outcome is presented in a balanced and neutral manner, and decisions about reproduction are left to the family.

Our daughter Juliett arrived on a beautiful summer afternoon. Late in my pregnancy, an ultrasound had showed an enlarged heart, dilated left kidney, and possible malformation of the cerebellum. During labor, our daughter’s heart rate decelerated significantly, and we were given the option of an emergency C-section. Without hesitation, we opted for the C-section. A drape was hung so that I could not see, and I only knew that the baby was born when the pediatrician ran out of the room with something in his arms. My husband quickly followed, and then I waited for what seemed an eternity.

Juliett weighed 4 pounds, 6 ounces and was 18 inches long. I graduated as an RN just before Juliett was conceived and had worked in a pediatric ICU. I had just enough experience to pick up a few of her obvious problems, but many escaped me. She was clearly much too thin, and her rib cage looked too short and prominent. But compared to the mental images I had formed after the ultrasound, I was relieved to see how beautiful she was. Her most striking feature was her incredible blue eyes, which were wide open and very alert. Her nose and mouth were beautifully formed and very petite. As my husband and I sat in awe over her, a neonatologist entered the room. He pointed out several characteristics, the only one of which I clearly remember was the clenched fist with the index finger lying over the middle finger. He concluded that she probably had trisomy 18. Of the grim things he rattled off, the only thing I remembered was that he said she would be a vegetable and that she would most likely die within the next couple of days. He then walked away, and we sat there, stunned. In this state of grief and turmoil I tried to understand how this clenched fist could lead to death, and how these bright, alert eyes could belong to a vegetable.

In the days that followed, I often opened up her fist and laid her fingers straight, hoping that the blood tests would not confirm the doctor’s suspicions. Our bonding with Juliett had been instant, and our great desire was to be able to take her home before she died. As she began to tolerate feedings and was weaned away from oxygen, our pleas to take her home were granted.

We left with no follow-up or care plan. Each time she went to sleep, we prayed she would wake up again and that we could complete another feeding. At 3 months of age, she started to smile at us, and our hopes brightened. We have been fortunate to see her outlive the grim statistics, and we have learned that there is no clear explanation why some children with this condition live longer than others. Juliett’s heart was enlarged because of a defect similar to a tetralogy of Fallot. Mild scoliosis at birth has now progressed to a 100-degree lumbar curve and a 90-degree thoracic curve. Despite her many physical challenges, Juliett has continued at her slower pace to learn and develop new skills. Her personality is delightful, and people are often surprised to see how responsive and interactive she is.

We have often been asked if we were afraid to have more children. Perhaps we were crazy, but we felt that another like Juliett would be great. We have had four more girls. To everyone’s surprise, our fifth child, Camille, was born with Down syndrome. With Juliett, the grieving process had been covered up with the gratitude we felt that she was even alive. With Camille, we experienced the more typical grieving process.

On the day of Juliett’s birth, a pediatrician came forward, put his arms around us, and told us he thought she was beautiful and to love her for as long as she could be with us. He turned her into a human being with a life to be highly valued. In the 13 years that have followed, Juliett has seen many doctors. Most of them, although they could not cure her problems, gave us the most important thing we needed: to know that our daughter’s life was of great worth and that, if they could, they would do anything to help her.

Tommy was born via an emergency cesarean, because 1 week before his delivery date his fetal movements markedly decreased. At birth, he weighed only 4 pounds, and the first time that I saw him he was in an incubator connected to all sorts of tubes. He spent his first month of life in the neonatal intensive care unit so that his weight gain could be closely monitored. Because he was so small, he was fed through a feeding tube for many months, and as a consequence, he refused to drink from a bottle. Eventually, he overcame his aversion to using his mouth to eat but only after substantial training. Nevertheless, despite our care, Tommy remained small for his age.

The following summer, Tommy developed dark red marks on his cheeks and under his eyes. Our pediatrician referred us to a dermatologist, who suspected that the marks on Tommy’s face were related to his growth failure. We were very surprised. How could these two findings be related? That is when we were told that Tommy might have a genetic disorder called Bloom syndrome. We hoped that the doctor was wrong, but soon thereafter Tommy had a genetic test that measured the number of sister chromatid exchanges per cell (see Chapter 2 ). This test confirmed that Tommy had Bloom syndrome. Although I insisted that it was a false-positive result, I learned to accept that our son had a very rare cancer syndrome.

We were barraged by questions from family, friends, and doctors. As a result, we became very protective of our son and his privacy. Nevertheless, there was only so much we could do to protect him because he is such a social little boy, loving to play with family and friends. This also made choosing an appropriate elementary school a very difficult decision for us. We expected that children would pick on him because of his small size. However, to our surprise, he easily developed friendships and adjusted well to his classmates. In fact, the problems that he did develop were largely because of his misbehavior. Thus we struggled to find a balance between protecting Tommy while not permitting him special privileges because of his small stature.

In our home we try to treat Tommy like any of our other children. One challenge for us is that because of Tommy’s small size, people wrongly perceive that he is much younger than his chronological age. This is very frustrating for Tommy, yet we occasionally reinforce this image because of our concern for his safety. For example, although Tommy is 6 years old, he weighs only 21 pounds. Thus he must sit in an infant seat when he travels in a car, and we explain to Tommy’s friends that it helps him see through the windows. Another safety problem is that many of the sensors for the automatic doors at supermarkets cannot detect his presence and easily slam into him.

Overall, Tommy has adapted well. He climbs or jumps in order to reach things. To keep up with his peers, he often runs, hops, or jumps instead of walking. We constantly worry about his safety, but we cannot control all that happens to him. To date, he has been healthy, and although it seems like we have been riding an emotional roller-coaster, we wouldn’t trade our experiences for anything.