natural history of trisomy 18 and trisomy 13: i. growth...

14
American Journal of Medical Genetics 49:175-188 (1994) Natural History of Trisomy 18 and Trisomy 13: I. Growth, Physical Assessment, Medical Histories, Survival, and Recurrence Risk Bonnie J. Baty, Brent L. Blackburn, and John C. Carey Division of Medical Genetics, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, Utah The natural history of trisomy 18 and trisomy 13 was investigated using data derived from parent questionnaires and medical records from 98 families with an index case of trisomy 18 and 32 families with an index case of tri- somy 13. Data are presented on pregnancy, delivery, survival, medical complications, im- munizations, growth, cause of death, cyto- genetics, and recurrence risk. Half of the tri- somy 18 babies were delivered by C-section. Fetal distress was a factor in half, and the only reason in a third of C-section deliveries. One minute Apgar scores were significantly lower in C-section and breech deliveries. There were more small for gestational age babies than in the general population, but most of the low birth weight newborns were small for ges- tational age, unlike the general population. Survival in this group of children was better than in other studies due to ascertainment bias. There were more girls than boys at all ages for both conditions, and the sex ratio de- creased with time. Growth curves for length, weight, head circumference, and weight vs height are provided. Long-term survival did not appear to be due to mosaicism. We found no adverse reactions attributable to immuniz- ations. At age 1 year there was an average of approximately 2 operations per living child. We report the second case of successful major cardiac surgery in a trisomy 18 child. Almost 70% of deaths were attributed to cardiopulmo- nary arrest. The sibling recurrence risk for trisomy 18 or trisomy 13 was 0.55%. Q 1994 Wiley-Liss, Inc. ~ Received for publication March 5,1993; revision received August 16, 1993. Address reprint requests to Bonnie Jeanne Baty, Department of Pediatrics, University of Utah Medical Center, Salt Lake City, UT 84112. 0 1994 Wiley-Liss, Inc. KEY WORDS: trisomy 18, trisomy 13, chro- mosome 18, chromosome 13, natural history INTRODUCTION Trisomy 18 and trisomy 13 are two well known chro- mosomal disorders which have been reviewed exten- sively [Finley et al., 1963; Hodes et al., 1978; Lewis, 1964; Rohde et al., 1964; Smith, 1964; Smith, 1969; Taylor, 1967, 1968; Warkany et al., 1966; Zellweger et al., 19641. In spite of the hundreds of medical articles describing trisomy 18 and trisomy 13, there is still a lack of precise data about their natural history. There also tends to be a skewing of information toward the extreme negative, with families often being told that their child has no chance for survival, and, more impor- tantly, little chance for meaningful interaction with their families. Although it is clear that these conditions are severely handicapping and the rate of early death is high, it is crucial that accurate information about the range of the natural history of these conditionsbe avail- able to health care personnel. Accurate information is especially important to pediatricians and obstetricians who may help families make important decisions involv- ing care and interruption of pregnancy, help the family adjust to the birth ofa child with trisomy 18 or 13, and in some cases provide health care for many years. The objectivesof this study were to help delineate the natural history of trisomy 18 and trisomy 13, and to obtain better information to answer parents’ questions in a clinical setting. Some of the important goals were (1) to gather data for construction ofgrowth curves, (2) to assess whether immunizations are as safe for children with trisomy 18 and 13 as the rest of the pediatric popu- lation, (3) to develop a profile of psychomotor develop- ment, and (4) to learn more about what congenital de- fectdphysical handicaps and morbidity were occurring. The developmental information is presented in a sepa- rate publication [Baty et al., 19941. MATERIALS AND METHODS Questionnaires were sent to 483 families nationwide who were members of S.O.F.T. (Support Organization for Trisomy 18,13,and Related Disorders) and who were

Upload: hoanghanh

Post on 13-Apr-2018

230 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

American Journal of Medical Genetics 49:175-188 (1994)

Natural History of Trisomy 18 and Trisomy 13: I. Growth, Physical Assessment, Medical Histories, Survival, and Recurrence Risk Bonnie J. Baty, Brent L. Blackburn, and John C. Carey Division of Medical Genetics, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, Utah

The natural history of trisomy 18 and trisomy 13 was investigated using data derived from parent questionnaires and medical records from 98 families with an index case of trisomy 18 and 32 families with an index case of tri- somy 13. Data are presented on pregnancy, delivery, survival, medical complications, im- munizations, growth, cause of death, cyto- genetics, and recurrence risk. Half of the tri- somy 18 babies were delivered by C-section. Fetal distress was a factor in half, and the only reason in a third of C-section deliveries. One minute Apgar scores were significantly lower in C-section and breech deliveries. There were more small for gestational age babies than in the general population, but most of the low birth weight newborns were small for ges- tational age, unlike the general population. Survival in this group of children was better than in other studies due to ascertainment bias. There were more girls than boys at all ages for both conditions, and the sex ratio de- creased with time. Growth curves for length, weight, head circumference, and weight vs height are provided. Long-term survival did not appear to be due to mosaicism. We found no adverse reactions attributable to immuniz- ations. At age 1 year there was an average of approximately 2 operations per living child. We report the second case of successful major cardiac surgery in a trisomy 18 child. Almost 70% of deaths were attributed to cardiopulmo- nary arrest. The sibling recurrence risk for trisomy 18 or trisomy 13 was 0.55%. Q 1994 Wiley-Liss, Inc.

~

Received for publication March 5,1993; revision received August 16, 1993.

Address reprint requests to Bonnie Jeanne Baty, Department of Pediatrics, University of Utah Medical Center, Salt Lake City, UT 84112.

0 1994 Wiley-Liss, Inc.

KEY WORDS: trisomy 18, trisomy 13, chro- mosome 18, chromosome 13, natural history

INTRODUCTION Trisomy 18 and trisomy 13 are two well known chro-

mosomal disorders which have been reviewed exten- sively [Finley et al., 1963; Hodes et al., 1978; Lewis, 1964; Rohde et al., 1964; Smith, 1964; Smith, 1969; Taylor, 1967, 1968; Warkany et al., 1966; Zellweger et al., 19641. In spite of the hundreds of medical articles describing trisomy 18 and trisomy 13, there is still a lack of precise data about their natural history. There also tends to be a skewing of information toward the extreme negative, with families often being told that their child has no chance for survival, and, more impor- tantly, little chance for meaningful interaction with their families. Although it is clear that these conditions are severely handicapping and the rate of early death is high, it is crucial that accurate information about the range of the natural history of these conditions be avail- able to health care personnel. Accurate information is especially important to pediatricians and obstetricians who may help families make important decisions involv- ing care and interruption of pregnancy, help the family adjust to the birth ofa child with trisomy 18 or 13, and in some cases provide health care for many years.

The objectives of this study were to help delineate the natural history of trisomy 18 and trisomy 13, and to obtain better information to answer parents’ questions in a clinical setting. Some of the important goals were (1) to gather data for construction ofgrowth curves, (2) to assess whether immunizations are as safe for children with trisomy 18 and 13 as the rest of the pediatric popu- lation, (3) to develop a profile of psychomotor develop- ment, and (4) to learn more about what congenital de- fectdphysical handicaps and morbidity were occurring. The developmental information is presented in a sepa- rate publication [Baty et al., 19941.

MATERIALS AND METHODS Questionnaires were sent to 483 families nationwide

who were members of S.O.F.T. (Support Organization for Trisomy 18,13, and Related Disorders) and who were

Page 2: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

176 Baty et al.

listed as having a child with trisomy 18 or trisomy 13. The response rate was 1141334 (34%) for trisomy 18 and 481149 (32%) for trisomy 13. Questionnaires reporting individuals with mosaicism, translocations, or other chromosome abnormalities were not included in the analysis (30 questionnaires). One individual with both trisomy 18 and Klinefelter syndrome (48,XXY, + 18) and another individual with both trisomy 18 and Turner syndrome (46,X, + 18) were also excluded. Data were analyzed from 98 individuals with trisomy 18 and 32 individuals with trisomy 13. Cytogenetic confirmation was achieved for 97% of trisomy 18 individuals and 97% of trisomy 13 individuals. Cytogenetic confirmation was defined as either a copy of the chromosome report or a statement in the medical records that the child had full trisomy 18 or trisomy 13. The remaining families stated that their child had full trisomy 18 or 13, and the ques- tionnaires and medical records were consistent with that diagnosis. A subset of 85 trisomy 18 individuals and 27 trisomy 13 individuals had documentation of how many cells were counted.

All but 4 families agreed to be recontacted to discuss release of medical records. Since some families did not follow through on sending their medical releases, and other families could not be contacted, records were ob- tained on a subset of 89 (91%) trisomy 18 individuals and 27 (84%) trisomy 13 individuals. The medical re- cords were analyzed for the same information collected on the parent questionnaires, as well as some additional information.

The data were analyzed using Ingres or DataEase for database management. Statistics were generated using SAS (Statistical Analysis System) and BMDP Statisti- cal Software. Information from the medical records was stored in a completely separate data base from the ques- tionnaire data and analyzed separately. In a few cases, information from both sources was added together, with duplicate information deleted. In the results section, the source of data on each result is identified.

Growth curves were constructed for combined male and female weight, height, head circumference (OFC), and weight vs height, using the method described in Stevens et al. [19901. There were 76 trisomy 18 individ- uals and 17 trisomy 13 individuals with at least one measurement available. Of these individuals, 83’Z of trisomy 18 individuals and 88% of trisomy 13 individ- uals had multiple measurements available. Measure- ments on each child were limited to one per month in the first year, one per 3 months in the second and third years, and one per year after the age of 3 years. The mean and 95% confidence limits of the measurements within each interval were calculated and the curves were smoothed using least-squares regression. Eighty- four percent of the trisomy 18 measurements and 69% of the trisomy 13 measurements were from females. We used combined male and female data, due to the sparsity of data for males. Normal curves (averaged for mailes and females, and indicated by the dotted lines) were included for reference [Hammil et al., 1979; Nellhaus, 19681. The curves were not significantly different for parental and medical data, so the data were added to- gether and overlapping data removed. Data with less

than 4 separate measurements at a given age were excluded.

Immunizations were counted only if the administra- tion date was recorded. Immunizations given in combi- nation and half doses are listed separately. Tine tests given alone were not included. The parental and medical data were combined with duplicate data removed.

Many of the results are compared t o equivalent values in the normal population. We looked at the birth dates of all study subjects and used normative data from the literature which best corresponded to the birth years of most of the study subjects. Close to 90% of the cases were born in the 1980s, with most of the rest born in the 1970s, and a few trisomy 18 children born in the 1960s. When normative data were taken from U.S. census data, the years 1970-1988 were used. Some normative data were available for only a subset of these years, and this is indicated in the references. Many tables contain values for n (number of individuals). This indicates the number of positive findings in those cases which specifically stated their presence or absence. In some instances, the denominator will refer to all study patients, and this is clearly stated. This is because for some items using the whole patient population for the denominator is more accurate than using the subset that filled in that partic- ular blank. For instance, most questionnaires leaving a blank for “omphalocele” probably did not have an omphalocele.

RESULTS Pregnancy

Pregnancy complications noted were 29 cases with intrauterine growth retardation (IUGR) and 17 cases with polyhydramnios for trisomy 18. The other compli- cations noted were extremely varied for both conditions. Reported average maternal weight gain was 13.3 kg (n = 48) for trisomy 18 and 14.1 kg (n = 14) for trisomy 13. The average weight gain in normal pregnancies is 10.8-12.6 kg [Gross, 19851.

Six trisomy 18 patients were noted to have declined amniocentesis, 3 cases were diagnosed by amniocen- tesis, and 22 cases had an abnormal prenatal ultrasound finding. The most common ultrasound abnormalities noted were IUGR (17 cases, almost all in the third tri- mester) and polyhydramnios (10 cases). Three cases had a choroid plexus cyst, 3 had oligohydramnios, 2 had congenital heart disease, 2 had hydrocephalus (both slight or resolving), and one each had an omphalocele, an umbilical hernia, a single umbilical artery, a bowel abnormality, decreased fetal movement, polycystic kid- neys, and mild hydronephrosis.

One trisomy 13 patient declined amniocentesis, and 3 had abnormal ultrasound findings of IUGR (3 cases), oligohydramnios (3 cases), hydrocephalus (1 case), and lack of fetal movement (1 case). All data on pregnancy came from the medical records.

Delivery The mean maternal and paternal ages at birth (Table

I) are from parental data, which are more complete. The mean maternal and paternal ages were significantly higher (P < 0.001) than the population mean [Riccardi,

Page 3: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

Natural History of Trisomy 18 and 13: I 177

TABLE I. Delivery Data for Trisomy 18 and Trisomy 13 "risomy 18 Trisomy 13 Normal

Mean maternal age, years

Mean paternal age, years

Percent delivered at

Range

Range

<38 weeks 38-42 weeks 242 weeks

Percent born C-section Percent with breech pres-

Mean birthweight, lbs Percent SGA by

entation

Weight Height OFC

Percent low birth weight

31.2 (n = 98)

33.2 ( n = 98) 18.1-44.9

21.2-51.6

16 43 41

n 98 54 (n = 95) 31 (n = 44)

4.84 (n = 96)

87 (IE = 70) 68 (n = 50) 62 (IE = 53) 79 (n = 98)

31.3 (n = 31)

33.7 (n = 31) 21.7-40.8

25.3-42.5

19 77 3

n = 30 25 (n = 32) 13 (n = 16)

5.90 (n = 31)

26 (n = 19) 20 (n = 15) 40 ( n = 15) 35 (n = 31)

26.0

28.2

8 85

7

15 3

7.34

4-7

7.3 Apgar scores 1 min 5 min 1 min 5 min l m i n 5min

7-10 (%) 4-6 (%) 0-3 (%)

28 70 38 67 90 98 37 21 24 14 7 1 35 9 38 19 2 1

n = 79 n = 21

1988; National Center for Health Statistics: Vital Sta- tistics, 19861 for both trisomy 18 and trisomy 13, using either parental or medical data.

The maternal and paternal age distributions have no obvious bimodality. The maternal age distribution in our data (Fig. 1) was adjusted for the maternal age distribution for all births in the US population for simi- lar birth years [National Center for Health Statistics: Vital Statistics of the United States, 1970-19881. This was done by dividing our cases for each year of maternal age by the total number of U.S. livebirths for the same

year's interval. Since we do not have total ascertain- ment of trisomy 18 and trisomy 13, these figures do not reflect incidence.

The distribution of gestational ages at birth is differ- ent from population norms [Creasy, 1989a; Resnick, 1989; National Center for Health Statistics: Vital Sta- tistics, 19861 for trisomy 18 (P < O.OOOl) , but not for trisomy 13 (P > O.l), in both parental and medical data. This is accounted for mainly by an excess of postterm births for trisomy 18 (Table I).

Mean birth weights were similar in parental and med-

15 25 35 4 5

Maternal age Maternal age

Fig. 1. Maternal age distribution for (A) trisomy 18 and (B) trisomy 13. The Y axis represents the number of cases born to women at a given age divided by the total number of U.S. births a t the same age,

Page 4: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

178 Baty et al.

ical data. We used parental data, which were more com- plete (Table I). In order to determine the percentage of cases that was small for gestational age (SGA), birth weights, heights, and head circumferences (OFC) from the medical data were compared to normal curves [Lubchenco et al., 1963, 1966; National Center for Health Statistics: Vital Statistics of the United States, 1970-19871 by gestational age (Table I). SGA was de- fined as below the 10th centile for gestational age. Among all infants born in developed countries, about 4-7% are classified as growth retarded [Creasy and Resnick, 1989bl. We also looked at low birth weight (LBW, defined as 52500 g), regardless ofgestational age (Table I), compared to the US. population norm for equivalent years [National Center for Health Statistics: Vital Statistics of the United States, 1982-19871.

Data from the medical records showed that for trisomy 18,23 cases (66%) had spontaneous labor and 12 (34%) were induced. Kettell [1968] reported 3.8% elective in- ductions of labor in a normal population. For trisomy 13, 10 (83%) were spontaneous and 2 (17%) were induced. A total of 10 patients (7 with trisomy 18 and 3 with tri- somy 13) had documented sterilization procedures after the birth of their trisomy child.

Caesarian-section (C-section) deliveries (P < 0.0001 for trisomy 18; P < 0.03 for trisomy 13) and breech deliveries (P < 0.0001 for trisomy 18; P < 0.005 for trisomy 13) were increased for both conditions compared to the general population [Benacerraf et al., 1988; Eowes, 1989; US. Bureau of the Census, Statistical Ab- stract of the United States, 19881 (Table I). The frequen- cies of breech deliveries are probably underestimated, since presentation was often not reported for C-section deliveries, which usually have a higher incidence of breech presentations. There was not a significant cor- relation between babies who were delivered by C-section and babies who were postterm, or between babies who were breech and babies who were postterm. However, there was a positive correlation between babies who were breech and babies delivered by C-section (P < 0.0001). The same conclusions were reached using parental or medical data and Table I uses parental data, which are more complete. Of the 51 trisomy 18 cases with a C-section, 41 cases had a reason recorded in the medical records. Fetal distress was a factor in 21 cases (51%) and was the only reason in 15 cases (37%). Breech position was the indication in 29% and failure to pro- gress in 12%.

Apgar scores from the medical data (Table I) were assessed in two different ways, using one cutpoint of 56) and using two cutpoints of 26 and 5 4 [normative data from the National Center for Health Statistics: Vital Statistics of the United States, 1978-19871. There was no association between Apgar score and gestational age at delivery (n = 79 in both medical and parental data), using chi square 2 x 2 contingency tables. There was a significant association in trisomy 18 cases between Apgar score at 1 min and C-section delivery, using ei- ther one (P < 0.02) or two cutpoints (P < 0.05) for Apgar data (n = 75 in medical data). There was not a signifi- cant association using the 5 min Apgar data. There was also a significant association in trisomy 18 cases be-

tween Apgar score at 1 min and breech delivery, using one cutpoint (P < 0.02, N = 52 in medical data and P < 0.05, n = 37 in parental data). There was not a signifi- cant association using two cutpoints, or using 5 min Apgar data. In both cases, C-section and breech delivery were associated with lower Apgar scores.

The incidence of single umbilical artery (SUA) re- ported in the medical records was 23/60 (38%) for tri- somy 18 and 5/18 (28%) for trisomy 13. Placental find- ings mentioned in trisomy 18 cases were small placenta (18 cases), small, short or thin cord (10 cases), eccentric cord insertion (2 cases), cyst of the cord (3 cases), cal- cifications of the placenta (2 cases), and abruption (2 cases). For trisomy 13, 2 cases had a small cord.

Survival For both trisomy 18 and trisomy 13, survival varied by

sex (Fig. 2). There were more girls than boys at all ages for both conditions, and the sex ratio decreased with time (Fig. 3). The preponderance of girls is more striking for trisomy 18 than trisomy 13 and increases with longer survival. At age 1 year, 41 (42%) of trisomy 18 and 12 (38%) of trisomy 13 individuals were living. At age 5 years, 11 (11%) of trisomy 18 and 4 (13%) of trisomy 13 individuals were living. At age 10 years, 5 (5%) of tri- somy 18 and 1 (3%) of trisomy 13 individuals were liv- ing. The oldest trisomy 18 individual was 22 years old at the time of the questionnaire, and was last known to be surviving at age 24. The oldest trisomy 13 individual was 9 years old and was last known to be surviving at age 11. The average length of survival (not including those still living) was 6.6 months (n = 64) for trisomy 18 and 12.8 months (n = 22) for trisomy 13. For tri- somy 18 males, it was 1.4 months (n = 23) and for trisomy 18 females, it was 9.6 months (n = 47). For trisomy 13 males, it was 3.6 months (n = 10) and for trisomy 13 females, it was 20.4 months (n = 12). Al- though we looked at survival statistics obtained through the medical records, it is clear that a record of the child's death was missing on too many records for the survival data to be meaningful, so the survival data are taken entirely from the parent questionnaires.

Structural Defects and Medical Complications For parental data, the rate of structural defects and

medical complications was calculated by the number reported divided by the total study cases. For medical data, the rate was calculated by the number of struc- tural defects and medical complications documented in the medical records divided by the study cases for which we had at least some records. We included only informa- tion about the same defects and complications assessed in the parent questionnaire. The rates reported by par- ents and in the medical records were somewhat differ- ent, with no consistent pattern. Table I1 shows the rate obtained by both methods. It is likely that there is un- derascertainment in the medical data in some cases because records are incomplete. Polydactyly in trisomy 13 probably is an example of underascertainment in the medical records. However, in some cases there is under- ascertainment in the parental data, probably because of uncertainty about the medical terms used. For example,

Page 5: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

A Percent Surviving

1 week 1 mo. 3 mos. 6 rnos. 1 yr.

B% Trisomy 18 Surviving 100,

5 yrs. 10 yrs.

1 mo. 3 mos. 6 mos. 1 yr. 5 yrs. 10 yrs. 1 wk.

% Trisomy 13 Surviving C loo 1

Male

El Female

1 wk. 1 mo. 3 mos. 6 mos. 1 yr. 5 yrs. 10 yrs.

Age

Fig. 2. Survival of (A) trisomy 18 and trisorny 13 individuals, (B) trisomy 18 individuals by sex, and (C) trisomy 13 individuals by sex.

Page 6: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

180 Batyet al.

Sex Ratio (MIF)

"O 1 I Trisomy18 I 0.8

0.6

0.4

0.2

0.0

At birth Living to 1 mo. Living !o 1 yr.

Fig. 3. Sex ratios from birth to 1 year for trisomy 18 and trisomy 13.

eye defects and kidney problems (for trisomy 13) are deficient in the parental data.

In addition to the structural defects and medical com- plications explicitly ascertained on the questionnaire, a list of selected structural defects and medical complica- tions was gathered (Table 111). These came from the medical records, or from an "other" category on the parental questionnaire. The most common findings were contractures, abnormal genitalia, dislocations, scoliosis, and syndactyly.

The average neonatal hospital stay was 19.6 days for trisomy 18 (n = 93) and 11.0 days (n = 32) for trisomy 13. The average number of days in the newborn inten- sive care unit (NBICU) was 16.3 days (n = 70) for tri- somy 18 and 10.8 days ( n = 23) for trisomy 13. The average number of days on a ventilator was 10.1 days

(n = 23) for trisomy 18 and 3.3 days (n = 7) for trisomy 13. In each case, n equals the number of families with a recorded hospital stay, NBICU stay, or number of days on a ventilator. Table IV shows the percentages and types of operations in the newborn period. One can see that surgeries were either performed for life-threaten- ing conditions or in a few cases, such as polydactyly, for medically trivial conditions. The average number of op- erations or hospitalizations in all cases living through each age group is shown in Table IV.

Table V shows information regarding the period after going home. These numbers can be useful in counseling parents of a newborn infant with trisomy 18 or 13. These data on the neonatal period were all taken from parental data. The medical data in this area were extremely deficient.

Immunizations Most immunizations were received in the first 6

months. Trisomy 18 children received 146 immuniza- tions and trisomy 13 children received 47, for a total of 193 immunizations (Table VI). The only significant ill- nesses after immunizations occurred in 2 trisomy 18 children, one with a possible seizure 2 weeks after a DPT/polio immunization and one with septic shock and convulsions 9 days after an MMR immunization. These symptoms were not typical of complications from DPT, polio, or MMR immunizations.

Growth We generated growth curves for weight, height, head

circumference, and weight vs. height (Figs. 4 and 5). The weight and height curves for trisomy 18 are consistently below the normal curve, except for an overlap at birth. The weight and height curves for trisomy 13 have more overlap with the normal curve. The OFC curves for both conditions show overlap with the normal curve, al- though the medians are lower. The graphs of weight vs.

TABLE 11. Structural DefectsiMedical Comulications in Trisomy 18 and Trisomv 13* Parental Medical

Trisomy Trisomy Trisomy Trisomy 18 (76) 13 (%) 18 (%) 13 (%)

Congenital heart disease 80 64 87 59 Cleft lip 6 39 6 41 Cleft palate 7 42 7 48 Club foot 14 9 15 15 Omphalocele 7 6 9 11 Polydactyly 3 67 2 52 Scalp defect 3 42 0 44 Radial aplasia 5 0 6 0 Eye defect 20 64 27 74 Vision loss 6 46 2 37 Seizures 36 42 28 52 Kidney problems 18 6 17 37 Urinary tract infection 17 21 11 19 Hydrocephalus 5 0 3 7 Spina bifida 2 3 1 4 Brain defect 16 18 16 11 Totals n = 98 n = 89 n = 27 n = 32 * For parental data, n = total study cases. For medical data, n = total cases for which we had medical records.

Page 7: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

Natural History of Trisomy 18 and 13: I 181

TABLE IV. Operations in Trisomy 18 and Trisomy 13 TABLE 111. Other Selected Congenital Defects and Medical Complications in Trisomy 18 and Trisomy 13

Trisomv 18 Trisomv 13

Bicornuate uterus Choanal atresiaistenosis Clavicle, absent Contractures Craniosynostosis Cryptorchidism Diaphragmatic hernia/

eventration Dislocated hips Dislocated knees Ears, structurally abnormal Ear canals, abnormal Eartags Esophageal atresia Eye bands Female genitalia, abnormal Hernia Larynx, abnormal Macrostomia Male genitalia, abnormal Malrotation GI tract Meckel’s diverticulum Natal teeth Premature puberty Pyloric stenosis Reflux Scoliosis Skin adhesion behind ears Skull defect Spleen, abnormal Split hand variant Syndact yly TE fistula Torticollis Vertebrae, abnormal

1 2 1

27 0

14 9

8 3 3 7 5 4 4 8 8 3 1 7 8 6 1 0 2 2

20 1 0 3 0

24 8 2 4

1 1 0 1 2

10 0

3 0 0 0 0 0 0 0 5 0 0 8 2 3 1 1 0 0 4 0 1 1 1 2 0 1 0

height are generally at the 5th percentile except in newborn trisomy 18 and trisomy 13 children and older trisomy 18 children, who are average.

Cause of Death The most common causes of death listed were cardio-

pulmonary arrest (68% oftrisomy 18 and 69% of trisomy 13), congenital heart disease (11% of trisomy 18 and 13% of trisomy 131, and pneumonia (11% of trisomy 18 and 4% of trisomy 13). Medical and parental data were similar, but parental data were more complete.

Cytogenetics We also looked a t the number of cells counted for each

individual in the study for whom we could obtain data. The variation in the numbers of cells counted was quite wide. For trisomy 18, there were as few as 5 cells counted (2 individuals) and as many as 165. Seventy-eight per- cent (66185) had 15 or more cells counted. For trisomy 13, 74% (20127) had 15 or more cells counted. For both conditions, the modal count, as expected, was 20. All study individuals had full trisomy 18 or 13 in every cell with the exception of two. One trisomy 18 child had one cell with 46,XX in 104 cells counted in peripheral blood. The other individual had 47,XX, + 18 in every cell, but

Trisomy 18 Trisomy 13

Percent having surgery in

Number of operations for neonatal period

Omphalocele TE fistula Esophageal atresia Gastrostomy Repair scalp Myelomeningocele Colostomy Polydactyly Pyloric stenosis Tracheostomy Thoracotomy

13

4 3 2 2 0 1 0 0 1 1 1

23

1 0 0 1 2 0 1 2 0 0 0

Average number of operationsihospitalizations (for all cases living through each age group) Trisomy 13 Trisomy 18

First month 0.54 (n = 78) 0.55 (n = 20) First 3 months 1.08 (n = 13) First vear 2.20 (n = 411 2.08 (n = 12)

0.77 (n = 62)

also had two parentally derived balanced translocations [t(7;5) and t(7;16)1 in every cell.

Recurrence Risk The number of sibs born either before or after the

trisomy child was ascertained by phone at the time medical records were requested. Half sibs were not dif- ferentiated from full sibs. For trisomy 18, there were 168 sibs (6 were half sibs), with no occurrence of chromosome abnormality. For trisomy 13, there were 50 sibs (0 were half sibs) with no occurrence of chromosome abnormal- ity. The second method of data collection was to record all previous pregnancies from the birth records. There were a total of 98 liveborn previous sibs recorded for trisomy 18 and 31 for trisomy 13. None had any chromo- some abnormality. The number of abortions was also recorded. Any known therapeutic abortions were ex- cluded, but the number presumably includes some ther- apeutic abortions that we were unable to detect. There

TABLE V. Home Care for Trisomy 18 and Trisomy 13 Trisomy 18 Trisomy 13

Percent going home 80 66 Percent going home on oxygen 5 7

Percent on home monitor (of 12 10

Percent bottlebreast fed 44 53

Percent with gastrostomy 19 22

Percent with known hearing 17 22

Percent with hearing aid 7 0 Percent of those with hearing 41 0

(of those going home)

those going home)

Avg. length of time (weeks) 83.6 97.4

Avg. age at tube insertion 8.4 7.1 (months)

loss

loss with a hearing aid

Page 8: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

182 Baty et al.

TABLE VI. Immunizations in Trisomy 18 and Trisomy 13 Percent immunized (of those who survived to this age) Trisomy 18 Trisomy 13 First 6 months 100 (n = 12) First year 100 (n = 41) 100 (n = 12)

84 (n = 51)

Number Number with Immunizations received Side effect side effect

DPT DPT (1h dose) DT Polio PolioPTd DPTipolio

DPTipolio (1/2 dose) DTipolio Haemophilus Hepatitis B MMR

Flu Measles Measlesipolio Mumps Rubellaimumps Varicella

Total

DPT DPTipolio

DT DTipolio Flu Hemophilus B Hepatitis B MMR MMRipolio MMRipolim’DPT Pneumovax Polio

Total

19 5

12 19 1

30

5 10 10 1

18

6 3 1 3 2 1

146

5 13

2 3 4 3 1 8 1 2 1 4

47

Trisomy 18 Fever Fever Fever Fever

Fever Possible seizure 12-14

days later

Fever Shock, hypoglycemia,

convulsions 7 days later

Fever

Fever

Trisomy 13 Fever Fever Fever, RSV pneumonia Hard, red leg

Hard, red leg, fever

7 1 1 2

6 1

2 1

2

1

1 1 1 1

1

were 81 cases of trisomy 18 and 25 cases of trisomy 13 with information on previous pregnancies. We ascer- tained a total of 42 pregnancy losses for trisomy 18 (30% of previous pregnancies), and 9 pregnancy losses (23% of previous pregnancies) for trisomy 13. In addition, there was 1 documented chromosome abnormality in these pregnancy losses, a therapeutically aborted fetus with trisomy 18 in a family who later had a trisomy 13 live- born. The mother was 39 years old at the delivery of the trisomy 18 fetus and 40 at the birth of the trisomy 13 child.

Among relatives, in two trisomy 18 families one of the parents reported a sib with Down syndrome. One family reported a maternal first cousin with trisomy 18. One other trisomy 18 family and two trisomy 13 families reported a cousin with Down syndrome.

DISCUSSION When we began this study, we were concerned that

data derived from parental questionnaires would be viewed as unreliable. Because of this, we obtained medi- cal records and kept the data derived from both sources separate. In our experience, there are some areas in which parental data are clearly superior to medical re- cords and other areas in which the opposite is true. In each area presented, we made a judgment about the validity of each data source, and occasionally thought that it was appropriate to combine both types of data.

Pregnancy The information about pregnancy derived from the

medical records is consistent with the many reports of prenatal diagnosis. Our data probably underestimate

Page 9: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

A200 A" MI

1Bo

160

140 50

a 40 A - 2 120

E

3 $30

El00

g 80

60 20

40

20 10

0 0 0 2 4 6 8 10 12 14 16 18 0 1 2 3 4 5 6 7

Age (years) Age (Years)

L o r , , , , ! , , , , ! , , , , ! , , , , ! , , , , ! , , , , I

Height(cm) D 40 50 60 70 80 90 100 110 120

Height (in)

Fig. 4. Growth curves for trisomy 18: (A) weight, (B) height, (C) head circumference, (D) weight vs. height. Regression and 95% confi- dence limits (trisomy 18 = solid lines; normal = dotted lines).

Age (years)

0 0.5 1 1.5 2 2.5 3 Age (years)

Height(cm) - . .

1 5 20 25 30 35 40 Height (in)

Growth curves for trisomy 13: (A) weight, (B) height, (C) head circumference, (D) weight vs. height. Regression and 95% confi- dence limits (trisomy 13 = solid lines; normal = dotted lines).

Fig. 5.

Page 10: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

184 Baty et al.

the prenatal diagnostic findings, since this information was not asked on parent questionnaires and we did not have complete records on all individuals in the study.

Delivery The increased mean maternal and paternal ages are

compatible with all previous reports based on sizable numbers of cases [Bergin et al., 1988; Butler et al., 1965; Conen and Erkman, 1966b; Hecht et al., 1963; Lejeune, 1964; Magenis et al., 1968; Rohde et al., 1964; Smith, 1969; Taylor, 1968; Young et al., 1986; Zellweger et al., 19641. Several studies [Hecht et al., 1963; Magenis et al., 1968; Taylor, 1967, 19681 have noted a bimodal mater- nal age distribution of trisomy 18 and trisomy 13. Our data do not have a bimodal maternal age distribution. When we adjusted our maternal age data using U.S. census data for maternal age by single year intervals, the resulting maternal age distribution is similar to previously reported maternal age distributions for tri- somy 18 and trisomy 13 [Ferguson-Smith and Yates, 1984; Goldstein and Nielsen, 1988; Hassold and Chiu, 1985; Hook et al., 1979; Hook, 1981; Hook et al., 19831.

Several studies have noted a large percentage of tri- somy 18 babies that are small a t birth and/or small for gestational age [Bergin et al., 1988; Butler et al., 1965; Khoury et al., 1988; Zellweger et al., 19641. There is probably a systematic bias toward underreporting of SGA babies due to underascertainment of stillbirths and early deaths in many of the series, and miscalcula- tion of gestational age based on ultrasound growth mea- surements which often underestimate gestational age. Our data show a somewhat asymmetric pattern of growth retardation at birth, with weight being most affected in trisomy 18 and head size being most affected in trisomy 13. There are more trisomy 18 babies and fewer trisomy 13 babies that are SGA compared to LBW (Table I). In the general population, about 30% of LBW newborns are SGA [Villar and Belizan, 19821. In our data, the ratios of SGAiLBW were 110% for trisomy 18 and 74% for trisomy 13, showing that not only are there more SGA babies than the general population, but most ofthe LBW newborns are SGA, unlike the general popu- lation. Previously reported mean birth weights have varied for trisomy 18, but are well below the 10th centile [Bergin et al., 1988; Rohde et al., 1964; Smith, 1969; Taylor, 1968; Warkany et al., 1966; Young et al., 1986; Zellweger, 19641. Birth weights for trisomy 13 are just below the 10th centile [Scarbrough et al., 1982; Smith, 1969; Taylor, 1968; Warkany et al., 19661. Our average birth weights are similar to those previously reported in the literature.

Our series is the first large series of Apgar scores. Nationally, 5 rnin Apgar scores of <7 are seen in 1.5 to 2.5% of births, depending on maternal age [National Center for Health Statistics: Advance Report of Final Natality Statistics, 19861. About 30% of our cases had 5 minute Apgar scores <7. It is of interest that 1 min, but not 5 min, Apgar scores have a significant association with C-section and breech deliveries. It has been noted that 5 min Apgar scores have a better correlation with outcome that 1 min Apgar scores [MacDonald et al., 19801.

In our series, there was an excess of postterm births [Smith, 1964, 1969; Warkany et al., 19661 and C-sec- tions in trisomy 18, which has also been observed previ- ously. Prior to widespread prenatal diagnosis of trisomy 18, about half the cases delivered by C-section [Bergin et al., 1988; David and Glew, 1980; Inagaki et al., 1987; Young et al., 19861. Currently, it is a somewhat contro- versial recommendation not to do a C-section for fetal distress in a pregnancy with trisomy 18 [Benacerraf et al., 1988; Bergin et al., 1988; Bundy et al., 1986; Karp and Meis, 1977; MacDonald et al., 1980; Schneider et al., 19811. The justification is that the procedure increases maternal mortality and morbidity somewhat, and that the procedure does not benefit the baby. Unfortunately, no previous data exist to demonstrate whether or not doing a C-section in a pregnancy with trisomy 18 or 13 improves the outcome. In light of this, it is interesting that in our series, 1 min Apgar scores were lower in babies that were breech or delivered by C-section. At 5 min, the association was no longer significant. One might interpret this as evidence supporting the asser- tion that C-section delivery does not benefit babies with trisomy 18 and 13. However, there are so many factors involved, such as nonrandom assignment of C-sections, whether the cases were prenatally diagnosed or not, and the reasons for C-section, that it is not clear that one can draw this conclusion from our data. In our study, there was no significant correlation between babies who were breech and babies who were postterm, or babies who delivered by C-section and babies who were postterm. Babies who were breech were much more likely to de- liver by C-section, as expected.

Single umbilical artery (SUA) has a known associa- tion with chromosome abnormalities, as well as other congenital malformations [Bjoro, 19831. Its frequency in newborn infants in the general population is 1% or less [Bryan and Kohler, 1974; Byrne and Blanc, 1985; Heinonen et al., 1977:. Saller et al. [1990] reported 2 fetuses with SUA among 9 fetuses with trisomy 18 (22%) and 2 fetuses with SUA among 6 fetuses with trisomy 13 (33%), from a series of 53 cytogenetical:y abnormal pregnancies in which the number of umbilical cord ves- sels could be documented. Overall, 6/53 (11%) of chromo- somally abnormal pregnancies had a SUA, and none was seen in trisomy 21 fetuses. In Byrne and Blanc’s series of 879 consecutively collected, spontaneously aborted em- bryos and fetuses, 17% of their chromosomally abnor- mal fetuses had a single umbilical artery, including 3 of 6 fetuses with trisomy 18 (50%) [Bjoro, 19831. In a re- view, Smith [1969] listed SUA as occurring in 50% or more of trisomy 18 and 13 patients. The incidence in our study was 23/60 (38%) for trisomy 18 and 5/18 (28%) for trisomy 13. This is the largest series oftrisomy 18 and 13 cases with determination of the number of umbilical cord vessels.

Survival The statistics most commonly used for survival in

trisomy 18 come from a literature study by Weber et al. [1964]. The authors stated that the chance of surviving to 1 month, 1 year, and 10 years is 70, 10, and 1%, respectively. Magenis et al. [1968] reported lifetable

Page 11: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

Natural History of Qisomy 18 and 13: I 185

her CA and closure of the PDA. She was discharged 5 days after surgery, was doing well, and breast feeding. She underwent another cardiac catheterization at age 12 months 28 days because of increasing cyanosis and anticipation of surgical repair of her DORV. She had DORV with a membranous ventricular septal defect (VSD) and an intact CA repair. At age 14 months 3 days, surgery for repair of DORV was done, but the child could not be weaned from cardiopulmonary bypass and ex- pired. Other findings at surgery were an overriding aorta and pulmonary valve disease. Since she lived for 10 months after her CA repair, she did have significant survival after surgery. We are aware of one other report of significant survival after major cardiac surgery in a child with full trisomy 18. Van Dyke and Allen, [19901 reported a 5-year-old girl with full trisomy 18 and a VSD and PDA which were repaired without incident at age 2. We are also aware of a 6-month-old child with full tri- somy 13 who is doing well 2 months after surgery to repair an atrial septal defect and VSD [personal commu- nication of child’s family].

Immunizations After 198 immunizations, two children had serious

medical illnesses which were not characteristic of either the timing or type of complications of the immunization received. We conclude that there is no evidence for an increase in adverse reactions to immunizations in chil- dren with trisomy 18 and trisomy 13 compared to chro- mosomally normal children, although the numbers are small.

Growth The growth curves are the only ones available for

trisomy 18 and 13, with the exception of a study report- ing fetal growth [Droste et al., 19901. It has been a frustrating experience to plot these children on normal growth curves, especially for trisomy 18, because the children are so far below the curves that comparison has not aided in assessing what is normal for a child with trisomy 18. These growth curves provide a way to assess the growth of children with trisomy 18 and 13 compared with their peers.

Cause of Death When we looked at the causes of death, we were sur-

prised at the lack of variety of the diagnoses. We had hoped to discover useful information in this area, since parents usually have questions about what to expect, but neither the parental nor medical data were very enlightening. Most causes of death were a variation of cardiopulmonary arrest.

Cytogenetics The question has often been raised whether long-term

survivors with typical trisomy 18 or trisomy 13 have mosaicism for normal cells in some of their tissues. There have been reports of long-term survivors with no detected mosaicism [Boraz, 1987; Carey, 1990; Frohlich, 1990; Kupke and Muller, 1989; Redheendran et al., 1981; Saller et al., 1990; Singh, 19901, and reports of individuals with long-term survival who do have mosai-

data on 172 full trisomy 13 patients. They found that 56% survived to 1 month, and 14% survived to 1 year. More recent papers using population-based data from Denmark [Goldstein and Nielsen, 19881, Australia [Carter et al., 19851, England [Young et al., 19861, and Utah [Root and Carey, 19941 conclude that survival in trisomy 18 is poorer than previously reported, either due to previous underascertainment [Carey, 19921 or changes in medical management. These studies indicate that about 50% of children with trisomy 18 die in the first week of life and about 5-10% survive the first year of life. Weber [1967] also reported that the sex ratio de- creased as the age ofthe subjects increased. The average number of days of survival in nonautopsy series varies widely, from 17-239 days for trisomy 18 [Bergin et al., 1988; Carter et al., 1985; Conen and Erkman, 1966b; Hodes et al., 1978; Rohde et al., 1964; Taylor, 1968; Young et al., 19861 and 24-252 days for trisomy 13 [Conen and Erkman, 1966a; Hodes et al., 1978; Red- heendran et al., 1981; Taylor, 19681. The higher figures most likely also underascertain children with short sur- vival. It is not always clear whether survival figures include children still living.

The survival of children& our study is weighted by the inclusion of a large number of survivors. This is probably because parents of a living trisomy 18 or 13 child are somewhat more likely to belong to a parent support group and complete a questionnaire than par- ents of a deceased child. This bias affects some aspects of the database (e.g., survival), but has little effect on most of the data. For example, we think the differences in survival by sex are unlikely to be affected by this bias. Our data show more girls than boys at all ages for both conditions, and a sex ratio which drops with increas- ing age.

Structural Defects and Medical Complications The data relating to the neonatal hospital stay, later

operations and hospitalizations, and other medical com- plications of childhood give us a sense of the burden of these conditions, and are useful in primary care and genetic counseling. Although the figures give a sense of numbers and averages, there was considerable vari- ability among individual children. Some of the older children were constantly in the hospital, while others were rarely seriously ill. In our experience, hearing tests utilizing a BAER test in young trisomy 18 children do not seem to be accurate [Page, 19921. Many of these children have abnormal early tests, but seem to hear normally as older children. It is possible that these figures inflate the hearing loss in trisomy 18 individ- uals. It is unclear why trisomy 13 children are not utili- zing hearing aids.

One operation of particular relevance is a repair of a congenital heart defect in a girl with full trisomy 18. This patient was diagnosed at 3I/a months by cardiac catheterization to have a double outlet right ventricle (DORV) with a patent ductus arteriosis (PDA) and co- arctation of the aorta (CA). She also had a large lum- bosacral myelomeningocele and Arnold Chiari I1 mal- formation with operations at l and 5 days. A t age 4 months 11 days, she had a Waldhausen flap correction of

Page 12: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

186 Baty et al.

cism [Bass et al., 19821. Some of the children in the present study have had many cells counted (3 children ages 7, 9, and 10 had more than 100 cells) and some children also had multiple studies, including more than one tissue. Seventy-eight percent had 15 or more cells counted (that we could verify). The numbers of cells counted are underascertained in some cases. We found instances where a second chromosome study was re- ferred to, but we could not obtain information about the exact number of cells counted in the second study. It is difficult to rule out the possibility of mosaicism. It is our clinical impression that children with mosaicism have a different prognosis, at least as older children. Every child we have seen with trisomy 18 mosaicism has had a phenotype distinct from full trisomy 18, including the ability to walk and higher skills in many areas. For this reason, it is doubtful that this is the explanation for long-term survival, and more likely that it represents the range of the condition. It is important to count enough cells to adequately exclude mosaicism, since the prognosis is different if significant mosaicism is de- tected. We suggest counting a minimum of 20 cells. If the diagnosis was made prenatally and the child sur- vives, the diagnosis should always be confirmed post- natally in a t least 20 cells.

Recurrence Risk There has been a lack of data regarding recurrence

risk in trisomy 18 and trisomy 13. There have been isolated reports of trisomy 18 in twins [Mulder et al., 1989; Schlessel et al., 1990; Shah et al., 19891 and sibs [Atkins et al., 1974; Pauli et al., 1978; Shih et al., 1974; Stene and Stene, 19841. There have been many addi- tional reports of trisomy 18 and another trisomy in sibs [Crandall and Ebbin, 1973; FitzPatrick and Boyd, 1989; Fryns et al., 1986; Girardet et al., 1972; Scarbrough et al., 1982; Turner et al., 19641. Hecht et al. [19641 re- ported on 44 index families with trisomy 18 with 104 liveborn sibs. Three of the trisomy 18 index cases had a sib with Down syndrome. He also reported 16 index cases with trisomy 13 with 37 liveborn sibs and no sibs with a trisomy. In addition, parental mosaicism for tri- somy 18 has been seen [Beratis et al., 1972; Fitzgerald et al., 19861, and there have been many reports of double aneuploidy involving trisomy 18 or trisomy 13 and an additional aneuploid finding. The most common recur- rence risk used is 1%, which is derived mostly by anal- ogy with data from trisomy 21.

There are many ways to combine our data to generate a “recurrence risk.” One critical factor is whether to include the case with both a trisomy 18 fetus and a liveborn with trisomy 13 as a (‘recurrence” (or precur- rence). We favor combining data from both trisomies, because the existence of families with, two different types of trisomies (including our case) indicates that there is some lack of specificity to nondisjunction. We favor including the affected fetus because this preg- nancy could have resulted in a liveborn child, and we think that occurrence in a fetus is still a significant burden for families. Combining data for trisomy 18 and trisomy 13, there was 1 in 181 pregnancies with an affected sib, with a recurrence risk figure of 0.55% (CL

= 0-1.63%). Considering the small numbers in this study, these data are consistent with earlier estimates.

SUMMARY Although some of our conclusions are consistent with

observations from other sources, such as the skewed sex ratio, maternal and secondary paternal age effect, in- crease in postmature and abdominally delivered babies, increased frequency of single umbilical artery, and the percentage of various structural congenital defects, there are several areas of new information. These in- clude growth curves, developmental data [reported in Baty et al., 1991, 19931, immunization data, and other information about the frequency of medical complications in infancy and childhood. Although the ascertainment of cases is biased toward survivors, the data provide infor- mation useful in counseling families of newborn infants with trisomy 18 and trisomy 13, and valuable insights into the natural history of trisomy 18 and trisomy 13. When prenatal or postnatal decisions need to be made regarding the care of a fetus or child with trisomy 18 or trisomy 13, it is important that the full spectrum of the natural history of these conditions be presented to the parents to enable them to make the best possible deci- sions for their family.

ACKNOWLEDGMENTS We would like to thank the Support Organization for

Trisomy 18, 13, and Related Disorders and the families that filled out questionnaires and submitted medical records. It was obviously a labor of love, hope, and shar- ing, but for some it was a painful process. A special thanks to Kris Holladay, who produced the original questionnaire entitled “What About Me” that inspired this study. We are also grateful to Lynn Jorde for statis- tical consultation and critical evaluation of the manu- script, to Craig Forster for computer programming as- sistance, and to Joan Wetzel for help with data entry.

REFERENCES Atkins L, Milunsky A, Shahood JM (1974): Prenatal diagnosis: De-

tailed chromosomal analysis in 500 cases. Clin Genet 6:317. Bass HN, Fox M, Wulfsberg E, Sparkes RS, Crandall BF (1982): Tri-

somy 18 mosaicism: Clues to the diagnosis. Clin Genet 22:327-330. Baty B, Blackburn B, Carey J (1991): Development in trisomy 18 and

trisomy 13. Am J Hum Genet (Suppl) 49:43. Baty BJ, Jorde LJ, Blackburn BL, Carey J C (1994): The natural his-

tory of trisomy 18 and trisomy 13. 11. Psychomotor development. Am J Med Genet 49:189-194.

BenacerrafBR, Miller WA, F‘rigoletto FD (1988): Sonographic detection of fetuses with trisomies 13 and 18: Accuracy and limitations. Am J Obstet Gynecol 58:404-409.

Beratis NG, Kardon NB, Hsu LYF, Grossman D, Hirschhorn K (1972): Parental mosaicism in trisomy 18. Pediatr 50:908-911.

Bergin A, McManus SP, Clarke TA, Moloney M (1988): ‘hisomy 18: A nine year review. Int J Med Sci 157:5-7.

Bjoro K (1983): Vascular anomalies of the umbilical cord. 11. Perinatal and pediatric implications. Early Human Dev 8:279-287.

Boraz RA (1987): Trisomy 13 syndrome: A rare case of long term survival. J Pedodont 11:288-294.

Bowes WA (1989): Clinical aspects of normal and abnormal labor. In Creasy RK, Resnik H (eds): “Maternal-Fetal Medicine. Principles and Practice,” 2nd ed. Philadelphia: W.B. Saunders, pp 510-546.

BryanEM, Kohler HG (1974): The missing umbilical artery. I. Prospec- tive study based on a maternity unit. Arch Dis Child 49:844-852.

Page 13: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

Natural History of Trisomy 18 and 13: I 187

Bundy AL, Saltzman DH, Poben B, Fine C, Emerson D, Doublet PM (1986): Antenatal sonographic findings in trisomy 18. J Ultrasound Med 5:361-364.

Butler LJ, Snodgrass GJAI, France NE, Sinclair L, Russell A (1965): E (16-18) trisomy syndrome: Analysis of 13 cases. Arch Dis Child 40:600-611.

Byrne J, Blanc W (1985): Malformations and chromosome anomalies in spontaneously aborted fetuses with single umbilical artery. Am J Obstet Gynecol 151:340-342.

Carey J (1990): What is known about adolescents and young adults with trisomy 18? “The Soft Touch. Support Organization for Tri- somy and Related Disorders” (newsletter). 12(2):39-40.

Carey J (1992): Health supervision and anticipatory guidance for chil- dren with genetic disorders (including specific recommendations for trisomy 21, trisomy 18, and neurofibromatosis I). Pediatr Clin North Am 39:25-53.

Carter PE, Pearn JH, Bell J, Martin N, Anderson NG (1985): Survival in trisomy 18. Life tables for use in genetic counselling and clinical paediatrics. Clin Genet 2759-61.

Conen PE, Erkman B (1966a): Frequency and occurrence of chromo- somal syndromes. I. D-trisomy. Am J Hum Genet 18:374-386.

Conen PE, Erkman B (1966b): Frequency and occurrence of chromo- somal syndromes. 11. E-trisomy. Am J Hum Genet 18:387-398.

Crandall BF, Ebbin AJ (1973): Trisomy 18 and 21 in two siblings. Clin Genet 4:517-519.

Creasy RK (1989a): Preterm labor and delivery. In Creasy RK, Resnik R (eds): “Maternal-Fetal Medicine. Principles and Practice,” 2nd ed. Philadelphia: W.B. Saunders, pp 477-504.

Creasy RK, Resnick R (1989b): Intrauterine growth retardation. In Creasy RK, Resnik R (eds): “Maternal-Fetal Medicine. Principles and Practice,” 2nd ed. Philadelphia: W.B. Saunders, pp 547-564.

David TJ, Glew S (1980): Morbidity of trisomy 18 includes delivery by caesarean section. Lancet 2:1295.

Droste S, FitzSimmons J , Pascoe-Mason J, Shepard TH (1990): Growth of linear parameters in trisomy 18 fetuses. Am J Obstet Gynecol 163:158-161.

Ferguson-Smith MA, Yates JRW (1984): Maternal age specific rates for chromosome aberrations and factors influencing them: Report of a collaborative European study on 52965 amniocenteses. Pren Diag 4:5-44.

Finley WH, Finley SC, Carte ET (1963): 17-18 trisomy syndrome. Review and report of a case. Am J Dis Child 106:591-596.

Fitzgerald PH, Archer SA, Morris CM (1986): Evidence for the repeated primary non-disjunction of chromosome 21 as a result ofpremature centromere division (PCD). Hum Genet 7258-62.

FitzPatrick DR, Boyd E (1989): Recurrences of trisomy 18 and trisomy 13 after trisomy 21. Hum Genet 82:301.

Frohlich GS (1990): Long term survival of trisomy 18. “The Soft Touch. Support Organization for Trisomy 18, 13 and Related Disorders” (newsletter). 12(2):38-39.

Fryns JP, Kleczkowska A, Moerman P, Van den Berghe H (1986): Reciprocal translocations and full trisomy (trisomy 18 and trisorny 21) in the offspring. Ann Genet 29:272-274.

Girardet P, Grosset L, Juillard E (1972): Trisomy 21 and trisomy 18 in siblings. Helv Paediatr Acta 27:583-589.

Goldstein H, Nielsen KG (1988): Rates and survival of individuals with trisomy 13 and 18. Clin Genet 34:366-372.

Gross TL (1985): Normal fetal growth. In Gleicher N (ed): “Principles of Medical Therapy in Pregnancy.” New York: Plenum, pp 320-332.

Hammill PVV, Drizd TA, Johnson CI, Reed RB, Roche AF, Moore WM (1979): Physical growth: National Center for Health Statistics per- centiles. Am J Clin Nutr 32:607-629.

Haasold T, Chiu D (1985): Maternal age-specific rates of numerical chromosome abnormalities with special reference to trisomy. Hum Genet 70:ll-17.

Hecht F, Bryant JS, Motulsky AG, Giblett ER (1963): The No. 17-18 (E) trisomy syndrome. J Pediatr 63:605-621.

Hecht F, Bryant JS, Gruber D, Townes PL (1964): The nonrandomness of chromosomal abnormalities. Association of trisomy 18 and Down’s syndrome. N Engl J Med 271:1081-1086.

Heinonen OP, Slone D, Shapiro S (1977): “Birth Defects and Drugs in Pregnancy.” Littleton, MA: Publishing Sciences Group, p 74.

Hodes ME, Cole J , Palmer CG, Reed T (1978): Clinical experience with trisomies 18 and 13. J Med Genet 15:48-60.

Hook EB (1981): Rates of chromosome abnormalities at different ma- ternal ages. Obstet Gynecol 58:282-285.

Hook EB, Woodbury EF, Albright SG (1979): Rates of trisomy 18 in livebirths, stillbirths, and a t amniocentesis. BD:OAS 15(5C): 81-93.

Hook EB, Cross PK, Schreinemachers DM (1983): Chromosomal abnor- mality rates at amniocentesis and in live-born infants. JAMA 249: 2034-2038.

Inagaki M, Ando Y, Mito T, Ieshima A, Ohtani K, Takashima S, Takeshita K (1987): Comparison ofbrain imaging and neuropathol- ogy in cases of trisomy 18 and 13. Neurorad 29:474-479.

Karp LE, Meis PJ (1977): Trisomy-18 and antepartal fetal distress. J Reprod Med 19:345-347.

Kettell WC (1968): Inducing labor by rupturing membranes. Postgrad Med 44:199-204.

Khoury MJ, Erickson JD, Corder0 JF, McCarthy BJ (1988): Congenital malformations and intrauterine growth retardation: a population study. Pediatr 82:83-90.

Kupke KG, Muller U (1989): Parental origin of the extra chromosome in trisomy 18. Am J Hum Genet 45599-605.

Lejeune J (1964): “Les Chromosomes Humains,”Vol 1. Paris: Gauthier- Villars .

Lewis AJ (1964): The pathology of 18 trisomy. J Pediatr 6592-101. Lubchenco LO, Hansmann C, Dressler N, Boyd E (1963): Intrauterine

growth as estimated from liveborn birthweight data at 24 to 42 weeks of gestation. Pediatrics 32:793-800.

Lubchenco LO, Hansmann C, Boyd E (1966): Intrauterine growth in length and head circumference as estimated from livebirth at gesta- tional ages from 26 to 42 weeks. Pediatrics 37:403-408.

MacDonald HM, Mulligan JC, Allen AC, Taylor PM (1980): Neonatal asphyxia. I: Relationship of obstetric and neonatal complications to neonatal mortality in 38,405 consecutive deliveries. J Pediatr 96: 898-902.

Magenis RE, Hecht F, Milham S (1968): Trisomy 13 (DJ syndrome: Studies on parental age, sex ratio, and survival. J Pediatr 73: 222-228.

Mulder AFP, van Eyck J, Groenendaal F, Wladimiroff JW (1989): Trisomy 18 in monozygotic twins. Hum Genet 83:300-301.

National Center for Health Statistics: Advance report of final natality statistics, 1986. Monthly Vital Statistics Report, Vol. 37, No. 3, Supp. DHHS Pub No. (PHS) 88-1120. Public Health Service, Hyattsville, MD, p 12.

National Center for Health Statistics: Vital Statistics of the United States 1970-1988, Vol. 1, Natality. Public Health Service, Wash- ington, DC: U.S. Government Printing Office.

Nellhaus G (1968): Composite international and interracial graphs. Pediatrics 41:106-114.

Page M (1992): Hearing loss in long term survivors oftrisomy 18. Third Annual Scientific Session. Trisomy 18, 13 and Related Disorders. S.O.F.T. 6th Annual National Conference, Baltimore, MD.

Pauli RM, Pagon RA, Hall JG (1978): Trisomy 18 in sibs and maternal chromosome 9 variant. BD:OAS 14(6C):297-301.

Redheendran R, Neu RL, Bannerman RM (1981): Long survival in trisomy-13 syndrome: 21 cases, including prolonged survival in two patients 11 and 19 years old. Am J Med Genet 8:167-172.

Resnick R (1989): Post-term pregnancy. In Creasy RK, Resnick R (eds): “Maternal-Fetal Medicine. Principles and Practice,” 2nd ed. Phila- delphia: W.B. Saunders, pp 505-509.

Riccardi V (1988): American paternal age data for selected years from 1876 to 1981. Neurofibromatosis 1:93-99.

Rohde RA, Hodgman JE , Cleland RS (1964): Multiple congenital anom- alies in the El-trisomy (group 16-18) syndrome. Pediatrics 33: 258-270.

Root S, Carey JC (1994): Survival in trisomy 18. Am J Med Genet 49: 170-174.

Saller DN, Keene CL, Sun CJ, Schwartz S (1990): The association of single umbilical artery with cytogenetically abnormal pregnancies. Am J Obstet Gynecol 163:922-925.

Page 14: Natural history of trisomy 18 and trisomy 13: I. Growth ...trisomy.org/.../08/Trisomy-18-Trisomy-13-I-Baty-1994.pdfAmerican Journal of Medical Genetics 49:175-188 (1994) Natural History

188 Baty et al.

Scarbrough PR, Finley WH, Finley SC (1982): A review oftrisomies 21, 18, and 13. Alabama J Med Sci 19:174-188.

Schlessel JS, Brown WT, Lysikiewicz A, Schff R, Zaslav AL (1990): Monozygotic twins with trisomy 18: A report of discordant phe- notype. J Med Genet 27:640-642.

Schneider AS, Mennuti MT, Zackai EH (1981): High cesarean section rate in trisomy 18 births: A potential indication for late prenatal diagnosis. Am J Obstet Gynecol 140:367-370.

Shah DM, Jeanty P, Dev VG, Ulm JE, Phillips J (1989): Diagnosis of trisomy 18 in monozygotic twins by cordocentesis. Am J Obstet Gynecol 160:214-215.

Shih L, Hsu LY, Sujansky E, Kushnick T (1974): Trisomy 18 mosaicism in two siblings. Clin Genet 5420-427.

Singh KST (1990): Trisomy 13 (Patau’s syndrome): A rare case of survival into adulthood. J Ment Def Res 34:91-93.

Smith DW (1964): Autosomal abnormalities. Am J Obstet Gynecol

Smith DW (1969): The 18 trisomy and the 13 trisomy syndromes. 901055-1077.

BD:OAS 5(5):67-71. Stene J, Stene E (1984): Risk for chromosome abnormality at amnio-

centesis following a child with a non-inherited chromosome aberra- tion. Pren Diagn 4:81-95.

Stevens CA, Hennekam RCM, Blackburn BL (1990): Growth in the Rubinstein-Taybi syndrome. Am J Med Genet Suppl 6 5 - 5 5 ,

Taylor A1 (1967): Patau’s, Edwards’ and Cri du Chat syndromes: A

tabulated summary of current findings. Dev Med Child Neurol 9:78-86.

Taylor A1 (1968): Autosomal trisomy syndromes: A detailed study of 27 cases of Edwards’ syndrome and 27 cases of Patau’s syndrome. J Med Genet 5227-252.

Turner BT, den Dulk GM, Watkins G (1964): The 17-18 trisomy and 21 trisomy syndromes in siblings. J Pediatr 64:601-604.

U.S. Bureau of the Census, Statistical Abstract of the United States: 1988 (108th edition). Washington, DC, 1987, p 64.

Van Dyke DC, Allen M (1990): Clinical management considerations in long-term survivors with trisomy 18. Pediatrics 85:753-759.

Villar J, Belizan J M (1982): The relative contribution of prematurity and fetal growth retardation to low birth weight in developing and developed societies. Am J Obstet Gynecol 143:793-798.

Warkany J, Passarge E, Smith LB (1966): Congenital malformations in autosomal trisomy syndromes. Am J Dis Child 112502-517.

Weber WW (1967): Survival and the sex ratio in trisomy 17-18. Am J Hum Genet 19:369-377.

Weber WW, Mamunes P, Day R, Miller P (1964): Trisomy 17-18(E): Studies in long-term survival with report of two autopsied cases. Pediatrics 34533-541.

Young ID, Cook JP, Mehta L (1986): Changing demography of trisomy 18. Arch Dis Child 61:1035-1036.

Zellweger H, Beck K, Hawtrey CE (1964): Trisomy 18. Report of a case and discussion of the syndrome. Arch Int Med 113598-605.