|Home | About | Journals | Submit | Contact Us | Français|
To examine prevalence, characteristics, interventions and mortality of VLBW infants with trisomy 21 (T21), trisomy 18 (T18), trisomy 13 (T13) or triploidy.
Infants with birth weight 401–1500 g admitted to centers of the Vermont Oxford Network during 1994–2009 were studied. A majority of the analyses are presented as descriptive data. Median survival times and their 95% CIs were estimated using the Kaplan-Meier approach.
Of 539509 VLBW infants, 1681 (0.31%) were diagnosed with T21, 1416 (0.26%) with T18, 435 (0.08%) with T13, and 116 (0.02%) with triploidy. Infants with T18 were the most likely to be growth restricted (79.7%). Major surgery was reported for 30.4% of infants with T21, 9.2% with T18, 6.4% with T13, and 4.8% with triploidy. Hospital mortality occurred among 33.1% of infants with T21, 89.0% with T18, 92.4% with T13, and 90.5% with triploidy. Median survival time was 4 days (95% CI, 3–4) among infants with T18 and 3 days (95% CI, 2–4) among both infants with T13 and infants with triploidy.
In this cohort of VLBW infants, survival among infants with T18, T13 or triploidy was very poor. This information can be used to counsel families.
Trisomy 21 (T21), trisomy 18 (T18), and trisomy 13 (T13) represent the most commonly diagnosed autosomal trisomies in live-born infants.1 Previous literature described extensively the physical features, associated anomalies, management, and survival of these infants.2–8 However, a majority of the published articles addressed survival and interventions among term or near-term infants or were limited by small numbers of patients. Very-low-birth-weight (VLBW) infants with chromosomal anomalies have different challenges, as VLBW newborns are known to be at a higher risk of mortality and several neonatal morbidities. We used data from the Vermont Oxford Network (VON) to examine the frequency, associated anomalies, interventions, mortality, and neonatal morbidities of VLBW infants with T21, T18, T13, or triploidy.
Data were collected prospectively by US and international Neonatal Intensive Care Units (NICUs) participating in the VON. VON is a nonprofit voluntary collaboration of health care professionals dedicated to improving the outcomes of high-risk newborn infants. The use of the VON database for research was approved by the Committee for Human Research at the University of Vermont. Eligibility criteria for the centers participating in the VON database included from 1994–1995, infants with birth weight 501–1500 g; and from 1996–2009, infants with birth weight 401–1500 g. Accordingly, depending on the birth year, infants with birth weight 401–1500 or 501–1500 g, born between January 1, 1994 and December 31, 2009, at one of the VON participating centers or transferred to one of the study centers within the first 28 days after birth were studied. Participating centers followed a consistent set of rules for identifying and collecting data for eligible infants as outlined in the VON’s Manual of Operations.9
Neonatal information, including demographic measures and major birth defects, was collected for all eligible infants. Data on neonatal morbidities diagnosed during the hospital stay were collected for infants admitted to the NICU and included respiratory distress syndrome, pneumothorax, patent ductus arteriosus (PDA), early bacterial sepsis (positive blood and/or cerebrospinal fluid culture before day 3 of life), coagulase-negative staphylococcus sepsis (after day 3 of life), late bacterial sepsis (after day 3 of life), nosocomial infection (after day 3 of life), fungal infection (after day 3 of life), necrotizing enterocolitis (NEC), gastrointestinal perforation, severe intraventricular hemorrhage (grades 3–4), periventricular leukomalacia, retinopathy of prematurity (ROP), severe ROP (stages 3–5), and chronic lung disease at 36 weeks’ corrected gestational age (GA). Small for gestational age (SGA) was defined by birth weight below the 10th percentile.10
Major birth defects were entered according to a predefined list in the Manual of Operations or as a text field for defects considered lethal or life threatening by the reporting unit. Chromosomal anomalies including T21, T18, and T13 had predefined codes. Prior to 2008, triploidy was recorded in text fields in response to a general question about other major chromosomal anomalies. In 2008, triploidy was also assigned a predefined code. Specific surgery codes were added in 2006. Before 2006, a general question asked if any major surgical procedure was conducted in the operating room. This excluded PDA ligation, NEC surgery, and ROP surgery, as they were collected as individual questions. Other changes in the collection of variables are noted in table footnotes as appropriate. Worth noting is that chromosomal microarray analysis was not considered in the current study.
To identify all infants with T21, T18, T13 or triploidy, relevant text fields were reviewed. Additionally, among infants with T21, T18, T13, or triploidy, surgery codes specific for certain types of birth defects were reviewed to identify infants with associated lesions that had not been recorded elsewhere. Final discharge status and length of hospital stay (LOHS), defined as the sum of stay at all hospitals before the first discharge to home, death, or first birthday, whichever occurred first, were assessed for all infants.
Neonatal characteristics, delivery room (DR) interventions, surgeries, in-hospital morbidity outcomes and mortality were examined for all infants with T21, T18, T13, or triploidy. In some tables, infants without chromosomal anomalies are also included for comparison purposes. A majority of the analyses are presented as descriptive data. Median survival times and their 95% CIs were estimated using the Kaplan-Meier approach. The Cochran-Armitage trend test was also used to examine trends in mortality rates across the study period. All analyses were conducted using SAS 9.2 (SAS Institute, Cary, NC).
A total of 539509 VLBW infants were born or cared for at one of the 915 institutions participating in VON between 1994 and 2009, of which 456279 (84.6%) infants were cared for at US centers. VLBW infants with major birth defects accounted for 25,634 (4.8%) with chromosomal anomalies reported for 5257 (0.97%) infants. Table 1 (online) shows the frequency of VLBW infants with T21 (1681, 0.31%), T18 (1416, 0.26%), T13 (435, 0.08%) or triploidy (116, 0.02%) in the VON cohort. A majority of these infants were cared for at US centers (range 83.0%–87.9%). No significant differences were noted in the percentage of triploidy cases before and after the triploidy code was added in 2008. Nine infants with suspected, but not proved T21, T18, or T13 and 24 infants with T21, T18, or T13 plus other associated chromosomal anomalies were excluded from subsequent analyses.
Infants with T18 were more likely to be growth restricted than infants in the other groups (Table 2). Worth noting are the lower percentages of multiple births among infants with the more severe chromosomal anomalies (multiple births: 17.0% of infants with T21, 0.86% of infants with triploidy). Among T18 or T13 multiples, a shift towards a higher distribution of females was also noted (T13 multiples: 57.6% females; T18 multiples: 67.6% females) (data not shown).
A total of 583 (34.7%), 574 (40.5%), 168 (38.6%), and 35 (30.2%) infants with T21, T18, T13, or triploidy, respectively, had one or more associated structural malformations (Table 3; online). Congenital heart defects (CHDs) were most commonly reported for infants with T21, and gastrointestinal defects were the most prevalent type of birth defect among infants with T18 or T13.
Among DR survivors, major surgery was reported for 30.4% of infants with T21, 9.2% of infants with T18, 6.4% of infants with T13, and 4.8% of infants with triploidy (Table 4). After examining the surgery-specific codes added in 2006, the most common procedures involved the abdomen. Only 3 (0.70%) infants with T18 and 1 (0.76%) infant with T13 had heart surgery. As expected, procedures among infants with triploidy were very rare; 4 (4.8%) infants were reported to have had major surgery. The types of surgeries were available for 2 infants; 1 had omphalocele repair and 1 had PDA ligation.
Cesarean-section delivery was performed for >50% of infants with chromosomal anomalies. Any type of DR intervention among infants with a chromosomal anomaly ranged between 66.6% for infants with T13 and 80.8% for infants with T21, and 90.9% of infants without a chromosomal anomaly had DR interventions. When NICU interventions among DR survivors were examined, the percentages of infants receiving any type of respiratory support were comparable among the groups. Infants with T18, T13, or triploidy were more likely to be discharged on oxygen and a monitor and were more likely to have received no enteral feeding before discharge or death compared to infants with T21 and infants without a chromosomal anomaly (Table 5).
In-hospital mortality was 33.1% for infants with T21, 89.0% for infants with T18, 92.4% for infants with T13, and 90.5% for infants with triploidy. Mortality was highest among infants with the lowest GA and was >95% among infants with T18, T13, or triploidy with a GA ≤29 weeks (Table 6). No significant trends in mortality rates were noted in any group across the study period (P value trend-test >0.05). Among the 153 infants with T18 (including 5 reported to be mosaic), 33 infants with T13 (including 1 reported to be mosaic), and 11 infants with triploidy (including 1 reported to be mosaic diploid/triploid) discharged home, the median LOHS was 33 days (25–75%, 13–58), 45 days (25–75%, 31–77) and 43 days (25–75%, 20–72) with a mean (SD) discharge weight of 1732 (673) g, 1992 (748) g, and 1838 (666) g, respectively. The median survival time was 4 days (95% CI, 3–4) among infants with T18 and 3 days (95% CI, 2–4) among both infants with T13 and infants with triploidy. By 1 week of life, 67.4% of infants with T18 and 73.1% of infants with T13 had died.
Table 7 (online) shows the outcomes of DR survivors among the 4 groups of infants with chromosomal anomalies and among infants without chromosomal anomalies. Approximately 50% of infants with T21 had a PDA with 19.8% treated with indomethacin and 9.4% having a surgical ligation. Infections and sepsis rates were lower among infants with T18, T13, or triploidy in comparison with infants with T21 and infants without chromosomal anomalies, reflecting the higher rates of early mortality among the former groups of infants. NEC surgery was performed for 6 (0.56%) infants with T18 and 4 (1.4%) infants with T13. Cranial imaging and ROP examination were more likely obtained for infants with T21 and infants without a chromosomal anomaly than among the other groups of infants. ROP surgery was performed among 10 (0.67%) infants with T21 and 2 (0.19%) infants with T18. Chronic lung disease was more common among infants with T18, T13, or triploidy than among infants with T21 and infants without a chromosomal anomaly, with steroid administration being more common among infants in the 2 latter groups.
VLBW infants represent 1.5% of annual births in the US.11 For 2009, data collected by the VON database represented approximately 70% of the VLBW population born in the United States. The prevalence of the major chromosomal anomalies per 10000 live births in this population was: 31.2 for T21, 26.2 for T18, 8.1 for T13, and 2.2 for triploidy. The rates for the trisomies are much higher than those based on all US births, with recent estimates for T21, T18, and T13 per 10000 live births being: 13.5, 2.5, and 1.2, respectively.1 As this study was based on VLBW newborns, the elevated rates reflect the high proportions of growth restriction among these infants. This was especially the case for infants with T18, of whom 80% had low birth weight for gestational age (ie, SGA) consistent with previous findings.2,12 The higher rates are also consistent with the general observation that infants with significant abnormalities are more likely to be born prematurely and to be SGA.13,14
In general, infants with a chromosomal anomaly were less likely to receive any type of DR intervention than infants without a chromosomal anomaly. For infants with T13 or T18, DR resuscitation rates were 65% and 74%, respectively. In contrast, a recent survey found that at the mother’s request, 44% of U.S. neonatologists would consider resuscitating a preterm infant with confirmed T18 and a known CHD,15 and the most recent American Academy of Pediatrics neonatal resuscitation guidelines excluded T18 but included T13 in the list of conditions for which resuscitation is not indicated.16 We did not, however, have information on when the diagnosis of the chromosomal anomaly was suspected and when it was confirmed. In particular, we had no information about whether the diagnosis had been made prenatally, which might have affected both parent and medical team views on the mode of delivery and resuscitation. Beyond the DR, the use of respiratory support in the NICU among infants with T18 or T13 was as common as among infants without chromosomal anomalies, and major surgeries were performed on 9.2% of infants with T18 and 6.4% of infants with T13.
Median survival times for VLBW infants with T18 or T13 in our cohort were similar to some previous reports3,7,17 but lower than others.4,5 By 1 week of age, 67.4% of infants with T18 and 73.1% of infants with T13 had died. This is higher than the previously reported estimates from more recent studies (T18: range 40–63%; T13: 50–58%).3,4,5,7,18 At 1 year of age, 3 (0.21%) infants with T18 were still hospitalized. Others have reported 1-year survival among these infants to range between 0–8%.3,4,5,7,18,19,20 Our mortality rates, however, might have been confounded by withholding medical treatment from these infants given the dual risks of major chromosomal anomaly and VLBW. Although offering intensive management for infants with T21 is standard treatment,21,22 the decision to implement or continue invasive or other life-sustaining procedures among infants with T18 or T13 is a complex decision dependent on input from the family, family supporters, involved healthcare workers and sometimes bioethicists and/or independent guardians of the infant. Although some have indicated that cardiac surgery is not appropriate for these infants as they die of conditions unrelated to their cardiac defects,5,20,23 others have argued that cardiac surgery and the related intensive treatment are ethically acceptable as they can alleviate their symptoms and prolong their survival.6,24,25,26 An extensive body of literature addresses the challenges created by the birth of infants with T18 or T13, whether treatment is in their best interest or whether it is “medically futile” extending their suffering and pain given their profound neurodevelopmental disability and reduced life span.27 These difficult choices and their timing cannot be derived from algorithms based on empiric outcome data alone. We offer the findings in this study as one facet of the information that can help families and the health care community with these agonizing decisions.
Our study had several limitations. The age at surgery for infants in the VON cohort was not available, and thus the duration of postoperative survival for infants with T18 or T13 could not be established. Data on long-term follow-up beyond hospital discharge were not available, and the duration of survival for 10.8% of infants with T18, 7.6% of infants with T13, and 9.5% of infants with triploidy could not be examined. Some of these infants, given their short LOHS and their very low weight at discharge, were likely discharged home to die with comfort care only. In addition to the previously mentioned mosaic cases, we cannot exclude the possibility of other infants with mosaicism influencing survival in the VON cohort. The associated malformations among infants with chromosomal anomalies were, in general, likely to have been under-reported, as several hospitals consider these chromosomal anomalies to be a complete report and do not list other congenital anomalies separately. For example, among VLBW infants with T21 in our study, the rate of CHD was 20%, whereas most studies of T21 report a 40–55% CHD rate.28 The large sample size in our study makes our findings generalizable to a diverse group of NICUs with varying care practices. Such information is useful to health care providers in counseling families of VLBW infants affected with these chromosomal anomalies.
CONFLICT OF INTEREST: Dr. Horbar is the Chief Executive and Scientific Officer of the Vermont Oxford Network. Mr. Carpenter is the Director of Operations and Statistics at the Vermont Oxford Network. Both receive salary from the Vermont Oxford Network. None of the authors has any disclosure to report.