Article Text

PDF

Survival of children born with congenital anomalies
  1. S Dastgiri,
  2. W H Gilmour,
  3. D H Stone
  1. University of Glasgow, UK
  1. Correspondence to:
    Professor D Stone, Paediatric Epidemiology and Community Health (PEACH) Unit, Department of Child Health, University of Glasgow, Yorkhill Hospital, Glasgow G3 8SJ, UK;
    D.h.stone{at}clinmed.gla.ac.uk

Abstract

Aim: To describe the survival to age 5 years of children born with congenital anomalies.

Methods: Between 1980 and 1997, 6153 live born cases of congenital anomaly were diagnosed and registered by the population based Glasgow Register of Congenital Anomalies. They were retrospectively followed to assess their survival status from birth up to the age of 5 years.

Results: The proportions of all live born infants with congenital anomalies surviving to the end of the first week, and first and fifth year were 94%, 89%, and 88%, respectively. Survival to age 5, the end point of follow up, was significantly poorer for infants with chromosomal anomalies (48%) compared to neural tube defects (72%), respiratory system anomalies (74%), congenital heart disease (75%), nervous system anomalies (77%), and Down’s syndrome (84%).

Conclusion: Although almost 90% of all live born infants with congenital anomalies survive to 5 years, there are notable variations in survival between anomaly types. Our findings should be useful for both clinicians and geneticists to assess the prognosis of congenital anomalies. This information is also important for affected families and for the planning of health care needs for this high risk population.

  • congenital anomaly
  • survival
  • mortality

Statistics from Altmetric.com

Congenital anomalies make an important contribution to infant mortality. They remain a leading cause of death among infants in many countries in the world.1–9 Epidemiologists have reported numerous investigations of the prevalence and aetiology of congenital anomalies, but analyses of mortality have tended to focus on the contribution of these disorders to perinatal and infant death rates rather than to the survival of affected infants.

Accurate quantification of the survival pattern of congenital anomalies is required for genetic counselling, clinical decision making in the antenatal and neonatal periods, and public health policy making. There are, however, relatively few published data on this topic. Leek10 observed that mortality varies with the type of anomaly, being highest among those with respiratory, cardiovascular, central nervous system, and genetic disorders. The survival of infants with chromosomal anomalies has been examined repeatedly. An Italian study,11 for example, found that deaths in Down’s syndrome babies were mainly due to cardiac and respiratory causes, and that poor survival was associated with high parity and advanced maternal age.

The development of ultrasound and other techniques for prenatal diagnosis will inevitably lead to a steadily increasing demand for evidence based estimates of the prognosis for infants born with a variety of congenital anomalies. As good quality epidemiological data are necessary to generate such information, we decided to utilise a long established British congenital anomaly register to describe the survival pattern of children born with congenital anomalies in the latter part of the twentieth century.

METHODS

Congenital anomalies refer to structural defects (congenital malformations, deformations, disruptions, and dysplasias), chromosomal abnormalities, inborn errors of metabolism, and hereditary disease present at birth.12

The source of data was the Glasgow Register of Congenital Anomalies (GRCA), a population based registry operated by the Greater Glasgow NHS Board. The GRCA was established in the early 1970s as an epidemiological surveillance system. It records standard information on all prenatally and postnatally diagnosed congenital anomalies in the offspring of mothers resident within the boundaries of the Board. The register uses multiple sources of ascertainment and has no age limit for registration of cases. All notified anomalies are subjected to careful diagnostic validation by scrutiny of clinical and laboratory records. Anomalies are classified according to the ICD based coding of the British Paediatric Association Classification of Disease.13 Mortality data on children born with congenital anomalies are updated weekly by linking registry data with routine reports of immediate and underlying cause of death information that is sent to GRCA by the Registrar General for Scotland. Further information on the GRCA is available elsewhere.14

All cases (n = 6351) of live born congenital anomalies notified to the GRCA for the birth years 1980 to 1997 were included in the analysis. We retrospectively followed 6153 (97%) cases of live born anomaly on whom we had complete survival data from birth up to 5 years. Of these, 740 (12%) died, and 5413 (88%) were still alive at 5 years of age. No information was available on the remaining 198 (3%) cases, including 133 (2%) who had left Glasgow for elsewhere within the UK and 65 (1%) cases who moved abroad. They were then excluded from the study. We carried out a statistical comparison between cases lost to follow up and the study population to determine whether any particular type of anomaly was disproportionately represented in the cases lost to follow up.

Survival rates with 95% confidence intervals (CI) were calculated15 using the Kaplan-Meier method. Anomalies were analysed within selected groups as used by the EUROCAT network, of which the GRCA is part.12 The log rank test was used to compare survival between groups of anomalies.

χ2 tests were performed to try to assess changes in survival rates at one week, four weeks, and one year between three successive cohorts born in the time periods 1980–85, 1986–91, and 1992–97 respectively. This analysis was necessarily confined to all anomalies, as small numbers did not permit subdivision of the data.

RESULTS

Table 1 shows survival of the 6153 live born children with a diagnosis of congenital anomaly. The median survival time for all ascertained cases who died was 11 days (95% CI: 8 to 14). The shortest median survival time for those who died was in cases with urogenital tract and kidney anomalies (1 day, 95% CI: 0.5 to 1), musculoskeletal and connective tissue anomalies (1 day, 95% CI: 1 to 7), and respiratory system anomalies (1 day, 95% CI: 0.5 to 36). In contrast, children with Down’s syndrome who died had the longest median survival time (100 days, 95% CI: 72 to 205). The median survival time for those who died from digestive system anomalies, congenital heart disease, and nervous system anomalies ranged from 18 to 31 days.

Table 1

Survival of congenital anomalies among cases who died from birth to 5 years of age (ranked in order of median survival time)

Table 2 and fig 1 present the cumulative proportion of live born children with congenital anomalies surviving from the first week to five years. The proportions of all live born infants with congenital anomalies surviving up to the first week, and to 1 and 5 years of age, were 94%, 89%, and 88%, respectively.

Table 2

Cumulative percentages of live born children with congenital anomalies surviving to 5 years (ranked in order of cumulative percentages surviving to age 5)

Figure 1

Survival of children with selected congenital anomalies (with 95% CI). x and y axes refer to “age at death” and “cumulative proportion surviving”, respectively. Note that in some graphs, the large number of cases for total congenital anomalies results in extremely small confidence intervals, which are not properly visible.

Infants with chromosomal anomalies were found to have the poorest prognosis compared to other groups of anomalies. Survival at the end point of follow up was poorer for children with (all) chromosomal anomalies (48%) compared to neural tube defects (72%, p = 0.0002), respiratory system anomalies (74%, p = 0.0005), congenital heart disease (75%, p < 0.0001), nervous system anomalies (77%, p < 0.0001), and Down’s syndrome (84%, p < 0.0001).

To assess whether survival changed over time, three successive six-year time periods (1980–85, 1986–91, 1992–97) were compared in terms of proportions alive at the endpoint of three short follow up age intervals with complete data—that is, one week, four weeks, and one year. There was no statistically significant difference in survival rates over time. In the first time period (1980–85), the proportions alive at the end of one week, four weeks, and one year were 94.2%, 92.1%, and 88%, respectively. These proportions slightly increased to 94.9% (p = 0.37), 93.1% (p = 0.25), and 89.9% (p = 0.35), respectively in the last time period (1992–97).

A further statistical analysis was performed to compare between cases lost to follow up (n = 198) and the study population (n = 6351) to determine whether any particular type of anomaly was disproportionately represented in the cases lost to follow up. A statistical comparison between this group and the study population showed no significant difference in terms of type of anomaly (p = 0.27).

DISCUSSION

Our study used a population based and systematically validated registry of congenital anomalies so that possible sources of error in ascertainment were minimised. We were able to follow up 97% of the study cohort between birth and 5 years.

The study had, however, some limitations. First, in some groups of anomalies (cleft lip with/without palate, anomalies of eye, limb, and integument), few subjects died so that the results are based on very small numbers in these groups. Second, we lost 198 (3%) subjects as they had left the study area and we had no information about their current status. A statistical comparison between this group and the study population showed no difference in terms of type of anomaly, suggesting that there was unlikely to be selection bias. Finally, all registry data are subject to a degree of inaccuracy and underascertainment, although the GRCA is widely considered one of the best of its kind.

We found that almost 90% of all live born infants survived to at least 5 years, with notable variations in survival between anomaly types. Infants with chromosomal anomalies had the poorest prognosis.

In our study, the proportion of live born infants with Down’s syndrome surviving to the first year was high (87%), much better than chromosomal anomalies as a whole, and comparable with the proportions surviving in South America (74%),16 Ireland (88%),17 and Taiwan (94%).18

The survival of children with neural tube defects fell from 83% at the end of the first week to 72% at age 5 years. A Danish study reported 57% surviving to age 7 years.19 Adams et al reported that 57% of their cohort of infants with spina bifida in Atlanta, USA, survived one year or more.20

The survival of infants with congenital heart disease to 1 year of age was similar in Glasgow (78%) to that reported from northern England (82%)21 and the Czech Republic (80%).22

Although an improvement in the survival of congenital anomalies might have been expected over time given the likely improvements in medical care generally, and especially the ever more sophisticated techniques of surgical repair, we found no significant change in the survival of successive cohorts over the study period.

Because survival analysis of congenital anomaly births has rarely been performed, our findings should provide geneticists, obstetricians, and neonatalogists a valuable source of data on the prognosis of anomalies in various age groups. Promoting the appropriate use of termination of pregnancy following prenatal diagnosis depends on professionals and parents having access to information to aid decision making. This is particularly important for those anomalies that are known to be potentially lethal—including chromosomal anomalies, neural tube defects, respiratory system anomalies, congenital heart disease, and nervous system anomalies.

As we have described survival only up to age 5, we suggest that further outcome studies with longer follow up be carried out. Moreover, the paucity of published literature in this field suggests that further research is required to assess the quality of life and health care needs as well as the survival of such children.

Acknowledgments

We thank Mrs H Jordan and colleagues at the Greater Glasgow NHS Board for their assistance. Saeed Dastgiri was sponsored by Tabriz University, Iran. The GRCA is funded by the Greater Glasgow NHS Board.

REFERENCES

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.