How Far Can Prenatal Screening Go in Preventing Birth Defects?
Article Outline
The ideal screening program would have perfect sensitivity and specificity in detecting the condition of interest, the health condition being detected would be of substantial clinical and public health importance, once detected there would be effective and acceptable action to prevent the adverse outcome, and the health care community and public would fully embrace the strategy. Although screening for serious birth defects that result from single gene defects may not attain that standard of perfection, they come closer than perhaps any other health condition.
See related articles, p 15, p 20, and p 25
In this issue of The Journal, the studies of Down syndrome in Victoria, Australia,1 and cystic fibrosis in Brittany, France, and Trentino Alto-Adige, Italy,2 present carefully documented descriptive epidemiologic data on geographically defined populations over extended periods of time to examine the risk factors for occurrence and the effect of prenatal diagnosis and selective terminations of pregnancy on birth prevalence. The very fact that the health care system can have such a substantial influence on birth prevalence and resulting morbidity from these disorders is notable. Our understanding of the genetics of these conditions, the availability of relatively noninvasive, accurate prenatal screening tests, societal investment in screening programs, and reasonably widespread acceptance of selective termination of pregnancy have made it possible to prevent these conditions that carry lifelong morbidity for the affected child and their families. Even though these achievements warrant recognition, they also raise the question of whether further reduction in birth prevalence is possible and what the ultimate goal of such programs should be, including eradication of the disorder.
The birth prevalence of such disorders can be conceptualized as the cumulative result of: the following: (1) incidence among conceptuses, driven by endogenous factors such as maternal age, genetic predisposition to the disorder (often reflected by family history), and exogenous factors; (2) prenatal identification, determined by the many factors that result in receipt of a screening test including guidelines on the basis of maternal age or other risk factors, availability of resources, and individual motivation to participate; and (3) parental decisions regarding termination of affected pregnancies. These reports both provide data to enhance understanding of each component and thus inform discussion of how to further reduce the birth prevalence of Down syndrome and cystic fibrosis, respectively.
The pattern of Down syndrome in Victoria over the period 1986-20041 reflects some key changes that may well be occurring in other developed countries. Among mothers <35 years of age, the use of prenatal screening has increased markedly as the screening algorithms, and improved technology has resulted in reduced risk of pregnancy loss.3 This increased screening has resulted in a substantial increase in the proportion of Down syndrome pregnancies terminated before birth, rising from 3% to 60% among younger women. There is a competing influence, however, in the increased proportion of births to older mothers who are at intrinsically higher risk of Down syndrome conceptions, with older mothers also showing an increased proportion of terminations before birth (from around 50% to 80% over the study period). The net impact was declining birth prevalence through 1996, which was stable thereafter, counter to the perception that the disorder may be disappearing. There is every reason to believe that increasing the proportion of cases diagnosed prenatally (among younger or older mothers) will result in an increase in selective pregnancy terminations and reduced birth prevalence, a desirable and attainable goal.
The contrasting patterns for cystic fibrosis in selected regions of France versus Italy over the period 1990-20052 provide insight into the incidence of cystic fibrosis at conception, use of prenatal diagnosis and pregnancy termination, and the resulting impact on birth prevalence. Again, algorithms for efficient prenatal screening for cystic fibrosis have been developed and evaluated for public health impact.4 Both regions have sophisticated screening programs for cystic fibrosis, with the modest contrast in quality and cultural differences in the 2 regions limiting the inferences that can be drawn from a comparison. The birth prevalence was similar in both areas, yet the proportion of cases that were diagnosed prenatally and terminated was notably higher in Brittany, indicative of a greater risk at conception in that area. Over time, the birth prevalence of cystic fibrosis declined more markedly in Veneto/Trentino, a product of increasing use of prenatal diagnosis over the surveillance period. There is no suggestion that the prevalence among all conceptuses changed over the period of the study, leaving only the influences of changes in screening comprehensiveness and individual decision-making regarding pregnancy terminations to account for the shifts. Expanding the scope of the populations considered suitable for prenatal screening (as algorithms and technology become more efficient and safer) or more comprehensive use of screening in those populations defined as suitable will inevitably increase the proportion of cases that are detected prenatally, thereby increasing the proportion of all cases that are terminated, and reduce the birth prevalence.
The combination of refined surveillance methods and ambitious screening programs addressed in these studies sets current bounds on reduction in birth prevalence of these disorders. There are clear limitations inherent in these and the many other studies cited regarding screening for Down syndrome and cystic fibrosis. First, it is impossible to assess the true incidence of the disorder among conceptions because cases can only be identified prenatally by active surveillance or at birth, excluding conceptuses with the disorder that abort spontaneously (for example, an estimated one third of Down syndrome conceptions are lost spontaneously before birth5). Second, these studies are not addressing a universal, biologic phenomenon but rather the combination of the underlying biology of the disorders as it plays out in specific populations, each of which has a distinctive health care system and its own attitude toward screening and pregnancy termination. Generalization of these findings to other geographic settings or even the same communities in the future is not straightforward.
The goal is to offer safe and accurate screening methods to all mothers, allowing informed choices regarding pregnancy termination. What cannot be specified is the targeted birth prevalence because this depends on incidence at conception and use of screening and choice of pregnancy termination. Nevertheless, if one accepts the goal of further reducing birth prevalence, then such studies provide the opportunity to examine in a quantitative manner the changes that have the greatest potential to reduce birth prevalence. Guidelines regarding which women should be screened, completeness of screening uptake in the population, and availability of counseling and services pertaining to pregnancy termination each warrant examination and quantification for their impact on the birth prevalence of these disorders. There are few situations in which investment in health services has such a clear public health impact. Combined with financial estimates of costs and benefits, such research builds a sound empirical basis for societal decisions regarding investment in such screening programs.
References
- . Is Down syndrome a disappearing birth defect?. J Pediatr. 2008;152:20–24
- . Time trends in birth incidence of cystic fibrosis in two European regions: data from newborn screening programs. J Pediatr. 2008;152:25–33
- . Incorporating first-trimester Down syndrome studies into prenatal screening (Executive summary of the National Institute of Child Health and Human Development Workshop). Obstet Gynecol. 2006;107:167–173
- . Cystic fibrosis: selecting the prenatal screening strategy of choice. Prenat Diagn. 2003;23:474–483
- . Screening by test combination: a statistical overview. In: Grudzinskas JG, Chard T, Chapman M, Cuckle H editor. Screening for Down’s syndrome. Cambridge: Cambridge University Press; 1994;p. 47–72
PII: S0022-3476(07)00857-8
doi:10.1016/j.jpeds.2007.09.012
© 2008 Mosby, Inc. All rights reserved.
Refers to article:
- Prevalence, Neonatal Characteristics, and First-Year Mortality of Down Syndrome: A National Study , 19 November 2007
- Is Down Syndrome a Disappearing Birth Defect? , 22 October 2007
- Time Trends in Birth Incidence of Cystic Fibrosis in Two European Areas: Data from Newborn Screening Programs , 22 October 2007
