The Apgar score was introduced in 1952 to quickly evaluate the clinical status of newborns and is still routinely used in delivery wards worldwide.1,2 Although the predictive value of adverse neonatal outcome after a low Apgar score has been questioned, an Apgar score less than 7 at 5 minutes after birth has been associated with an increased risk for cerebral paresis, epilepsy, and mental retardation.3–5 Recent studies also have shown an association between low 5-minute Apgar scores and reduced scoring on intelligence quotient tests in male draftees in Sweden and Denmark.6,7
Perinatal asphyxia is a major cause of neurologic sequelae in term newborns.8 Most follow-up studies have focused on short-term severe neurological deficits such as mental retardation and cerebral paresis.9,10 The theory of a “continuum of reproductive casualty” implies that severe hypoxic events can cause death or obvious neurological deficit, but mild events might cause subtle defects in cognitive function that are only detectable as the child grows older.11 Mild hypoxic events are more common than severe events and do not usually lead to obvious neurological dysfunction. However, these events could lead to more subtle cognitive impairment that could have a widespread effect on a larger population basis.12
No well-defined definition of perinatal asphyxia exists, and the definition has evolved from a single indicator definition to a multiple indicator definition.9 A low Apgar score has been considered a weak marker of asphyxia because of its low sensitivity and specificity,8,13 but few studies have addressed the long-term cognitive effects of a low Apgar score.
A good marker for neonatal outcome is essential for an efficient evaluation of obstetric care. In the absence of other good markers, the Apgar score is the most commonly used outcome to measure neonatal health after birth. However, the clinical relevance of a low Apgar score is questioned, and investigations with attempts to measure the effect of low Apgar scores on the long-term neurological outcome are needed. The aim of our study was to elucidate the sole effect of an Apgar score less than 7 at 5 minutes after birth and cognitive performance as adolescents, as measured by school graduation grades from compulsory school at 16 years of age.
MATERIALS AND METHODS
A linkage was performed between the Swedish Medical Birth Registry and the Swedish School Grade Registry. All singletons born between 1973 and 1986 after 36 6/7 weeks of gestation to Swedish-born women were included. All fathers were Swedish citizens, but country of birth was not obtainable. Fetuses that were stillborn and newborns who had congenital malformations or were small for gestational age were excluded in the analysis. All children who died or emigrated before 16 years of age were also excluded from the analysis. Emigration dates were available from public authority records. The study was approved by the Ethical Committee, University of Lund, Lund, Sweden.
The Medical Birth Registry contains medical information on 99% of all deliveries in Sweden.14,15 All women are offered free antenatal care. The Apgar scores were assigned by the midwife who is in charge of delivery or by a pediatrician when present. Standardized record forms are used at all antenatal clinics, all delivery units, and of all pediatric examinations of newborns. Information regarding maternal smoking is collected from interview forms from the first visit at the antenatal care clinic. Maternal smoking habits are available from 1983. Copies of these forms are sent to the National Board of Health, where they are computerized. The Medical Birth Registry was linked with Statistics Sweden to, among other things, obtain information on maternal parity, maternal country of birth, parental citizenship, and date of death.
Information regarding diagnosis was also extracted from the Medical Birth Registry. The following International Classification of Diseases codes, version 8, were collected: premature separation of placenta (770.1), prolapse of cord without mention of compression (771.1), delivery with rupture of uterus (659.9), respiratory distress syndrome (776.20), asphyxia of newborn, unspecified (776.9; 776.30; 776.40), central nervous system depression of the newborn (772.49), central nervous system irritability of the newborn (772.59), and infantile spasms (345.11). The School Grade Registry16 kept by Statistics Sweden contains information on all school grades at the final examinations and graduation grades (1988–2003) for each pupil leaving the Swedish compulsory school (usually at the age of 16 years). The Swedish Educational Register, also kept by Statistics Sweden, contains information on the highest final education of each citizen. By linking the Educational Register with the School Grade Registry, it is possible to adjust for maternal educational level when studying the children's achievements in compulsory school.
All Swedish women are offered a free routine ultrasound examination at 17–18 postmenstrual weeks to confirm fetal life, estimate the expected date of delivery, identify twin pregnancies, and to detect severe fetal malformations. Thus, information regarding gestational age used in the current study is mostly based on ultrasound estimates. Newborns with birth weights of more than 2 standard deviations below the expected weight for gestational age were classified as small for gestational age.17 The Swedish school system during 1988 to 1997, henceforth referred to as period one, included school grades that were numerical on a five-level scale, with 1 as the lowest and 5 as the highest grade. The numerical school grades were comparative, meaning that for each year in each subject, the grades in the entire nation were expected to have a normal distribution, with 3 as the grand mean. The numerical grades were standardized using national standardized achievement tests in three subjects: mathematics, English, and Swedish.
From 1998 onward, henceforth referred to as period two, the grades were alphabetic, with four levels: not passed, passed, passed with distinction, and passed with excellence. The alphabetic grades are criterion-referenced, and the criterion for passed in each subject is thought to be possible for each pupil in the compulsory school (with special training if needed) to fulfill.
The children born between 1973 and 1981 received numerical grades, and the children born between 1982 and 1986 received alphabetical grades when they graduated from compulsory school.
The grades in chemistry, biology, and physics were recorded in each separate subject in some schools, and in some schools they were combined into one grade called natural sciences. If no composite grade in natural sciences was obtainable, then we used the mean grade of the separate subjects. The same procedure was used when estimating the grade in the composite subject social sciences, including geography, social science, history, and religion.
During period one, it was possible for the students to choose between a common or an advanced course in mathematics. During period two, all students completed the same mathematics course. A possible confounder influencing the results could be that adolescents who had low Apgar scores at birth would be more likely to choose the common course in mathematics rather than the advanced course. Therefore, results of low Apgar scores and school grades in mathematics are only from period two; during period one, we present only the odds of choosing common or advanced mathematics.
Children with mental retardation diagnosed do not enter the Swedish compulsory school. They are educated in special schools and their possible grades are not reported to the School Grade Registry. Children who are educated in special schools because of severe hearing or vision disabilities receive school grades, but these grades were not reported to the School Grade Registry during the study period. After excluding children who died or emigrated before 16 years of age, the assumption was made that children without graduation grades who were not enrolled in compulsory school went to a special school because of cognitive impairment. In Sweden it is possible to graduate with incomplete grades. If students “drop out” of 9th grade, then they will still graduate, but without graduation grades. They cannot continue to high school. Adolescents who do not have prerequisites (because of intellectual disabilities) to attend high school are offered specially designed education. If students are admitted to a special school, then they will be discharged from compulsory school and are not found in the School Grade Registry.
The association between Apgar scores less than 7 at 5 minutes after birth and school achievement was evaluated using the Mantel-Haenszel (1959) procedure to produce a stratified odds ratio (OR). Approximate 95% confidence intervals (CIs) were calculated using the method proposed by Miettinen.18
When specified, stratification was made for the year of birth (1-year interval), maternal age at delivery (classes: younger than 20, 20–24, 25–29, 30–34, 35–39, and 40 years or older), parity (previous children +1), maternal smoking (not known, none, smoking less than10 cigarettes per day, smoking 10 or more cigarettes per day), and maternal final educational level at January 1, 2005 (not known, graduating from 9 years of compulsory school only, graduating from high school 2-year program, graduating from high school 3-year program, graduating from university 2-year program, graduating from university 3-year program, graduating from university 3- to 4-year program, graduating from university 5-year program or more). Information regarding maternal smoking is only available from 1983; therefore, it is not possible to adjust for smoking fully. Group one comprised children who went to special schools and therefore never received graduation grades. Group two comprised children who entered compulsory school but did not obtain any graduation grades, presumably because they dropped out of school or were not capable of achieving grades. Group three comprised children who entered compulsory school and obtained graduation grades (although not necessarily in all school subjects).
First, the association between Apgar score less than 7 at 5 minutes after birth and need of education in special school or leaving school without any graduation grades, respectively, were evaluated. Second, the level of the grades within group three was investigated. For that investigation, the grades were divided into class variables (for period one: lacking, 1, 2, 3, 4, or 5; for period two: lacking, not passed, passed, passed with distinction, or passed with excellence). The mentioned class variables were compared with most common level (3 or passed, respectively). The effect of maternal risk factors on the offspring's need for special school was evaluated by the Manzel-Haenszel test and adjusted for year of child's birth, maternal age, parity, smoking, and educational level.
When comparing two adjusted ORs, two-tailed z tests were performed under the assumption of normal distribution of the log ORs. To detect a putative linear trend in stratified ORs along i one-step Apgar score strata, weighted linear regression analyses of the log (ORi) were performed, in which each log (ORi) was weighted according to precision (weighti=1/variance of log(ORi); Appendix).
After these exclusions, 877,618 individuals with available Apgar scores were included in the analysis (Fig. 1 shows exclusion criteria). Maternal and fetal characteristics in relation to school graduation grades are shown in Table 1; 848,828 (96.7%) individuals graduated from compulsory school with graduation grades, 6,002 (0.7%) graduated from compulsory school without grades, and the remaining 22,788 (2.6%) went to special schools. Young maternal age, low or high parity, maternal smoking, and low educational level were all significant (P<10−6) risk factors for need of education in special schools. Furthermore, the risk of going to a special school was dependent of the child's birth year, with the risk increasing in children being born after 1984 (OR 1.15, 95% CI 1.10–1.20). We interpret this as an effect of an increased inclination to admit students to special schools rather than substandard obstetric care.
Among the children in need of education in a special school, 910 (4%) had neonatal asphyxia diagnosed. The corresponding number for children educated in compulsory school and not receiving any graduation grades is 151 (2.5%). Among children who graduated with grades, 24,123 (2.8%) had neonatal asphyxia diagnosed. Table 2 shows neonatal and obstetric diagnosis in relation to school performance at 16 years of age. The association between neonatal asphyxia and need of education in a special school was highly statistically significant (OR 1.43, 95% CI 1.33–1.52). The association between the need for education in a special school and an Apgar score less than 7 at 5 minutes after birth was stronger than the association between the need of education in a special school after neonatal asphyxia was diagnosed (P<10−6).
Among children who had Apgar scores 0–3 at 5 minutes after birth who also had neonatal asphyxia diagnosed, 80 were in need of education in a special school, seven graduated from compulsory school without grades, and 810 graduated with grades from compulsory school. The corresponding numbers for Apgar scores 4–6 together with neonatal asphyxia diagnosed were 186, 26, and 3,546, respectively. Among children who had Apgar scores of 7–10 and neonatal asphyxia diagnosed, the corresponding numbers were 644, 118, and 19,776, respectively.
Newborns with Apgar scores less than 7 at 5 minutes after birth showed a significantly increased risk of never receiving graduation grades, presumably because they went to a special school because of cognitive impairment or other special educational needs (OR 1.93, 95% CI 1.75–2.14). The risk of needing education in a special school in relation to Apgar scores is shown in Table 3 and Table 4. A significant linear relationship (P<10−6) was found between increasing risk of needing education in a special school and each one-step decrease of the Apgar score. The strong association between Apgar score less than 7 at 5 minutes after birth and the need for special school is concordant with an estimate of numbers needed to harm of 44, meaning that 1 out of 44 newborns with an Apgar score less than 7 at 5 minutes will need education in a special school because of the antenatal or perinatal factors that caused the low Apgar score. The corresponding numbers needed to harm for Apgar scores 0–3 or 4–6 at 5 minutes after birth were 36 and 49, respectively.
Sixty-one individuals leaving compulsory school without any graduation grades in any subject had had Apgar scores less than 7 at 5 minutes after birth. Among newborns with Apgar scores less than 7 at 5 minutes after birth, the OR with 95% CI for graduating from normal compulsory school without receiving any graduation grades in any subject was 1.23 (95% CI 0.95–1.58). In Table 5, the distribution of school grades in six different school subjects (mathematics, Swedish, natural sciences, social sciences, handicraft, and physical education) when graduating from compulsory school are shown. Nearly all school children who had Apgar scores less than 7 at 5 minutes after birth showed an increased risk of graduating from compulsory school without graduation grades in that specific subject or receiving the lowest possible grades (not passed or grade 1 or 2). When graduating from compulsory school, the odds of not receiving a grade in the core subject Swedish language was 50% higher among those who had an Apgar score less than 7 at 5 minutes after birth. Furthermore, as shown in Table 5, in nearly all school subjects, those who had Apgar scores less than 7 at 5 minutes after birth showed a lower probability of receiving higher grades (passed with distinction, passed with excellence, or grade 4 or 5) than those who had Apgar score more than 7 at 5 minutes after birth. When graduating from compulsory school, the odds of having the highest grade (passed with excellence or grade 5) in the core subject Swedish language was 15% lower among those who had Apgar score less than 7 at 5 minutes after birth.
As specified, the distribution of grades in mathematics was only analyzed during period two, during which only one combined course in mathematics was offered. During period one, the adolescents could choose between a common or advanced course in mathematics. During period one, the adolescents with Apgar score less than 7 at 5 minutes after birth showed a higher risk of graduating without a grade in advanced math (OR 1.14, 95% CI 1.08–1.21), assumingly because they were more prone to choose common mathematics.
The results from our study, comprising 877,618 deliveries, show an association between an Apgar score less than 7 at 5 minutes after birth and long-term cognitive function. Individuals who had Apgar score less than 7 at 5 minutes after birth were approximately twice as likely to lack graduation grades from compulsory school, presumably reflecting placement in a special school because of cognitive impairment or other special educational needs. As many as 1 out of 44 individuals who had an Apgar score less than 7 at 5 minutes after birth will need education in a special school because of antenatal or perinatal factors that caused the low Apgar score. Also, among adolescents for whom school records were obtainable, in nearly all school subjects, individuals who had Apgar scores less than 7 at 5 minutes after birth showed an increased risk of graduating from compulsory school without graduation grades in that specific subject or receiving the lowest possible grade. Furthermore, for almost all school subjects, the odds of having the highest grade (passed with excellence or grade 5) was significantly lower among adolescents who had Apgar scores less than 7 at 5 minutes after birth.
As shown in Table 4, it is evident that newborns with Apgar scores 7 or 8 at 5 minutes after birth were more likely to need education in special schools as compared with newborns with Apgar score 9 or 10 at 5 minutes after birth. Thus, the results suggest that the negative association between Apgar scores and need of education in special school is not just valid for strongly asphyxiated newborns, and conditions that mildly influence the Apgar scores could have an effect on future cognitive capacities. Furthermore, our results revealed a stronger association between the need for education in a special school and an Apgar score less than 7 at 5 minutes after birth than the association between the need of education in a special school after receiving the diagnosis neonatal asphyxia. However, it must be kept in mind that our results are based on the diagnoses registered in the Medical Birth Registry, and some diagnosis could be missing.
We chose to investigate the association between a low Apgar score at birth and performance in theoretical subjects (mathematics, Swedish language, natural science, and social science), practical subjects (handicraft) and physical education, postulating that these subjects would mirror the individual's cognitive, fine motor, and gross motor skills, respectively. Interestingly, an association between a low 5-minute Apgar score and school grades at 16 years of age was found irrespective of school subject, suggesting a general cerebral engagement. The long-term predictive value of a low Apgar score has been a matter of discussion, and different authors have showed diverging results. Few studies, however, have a follow-up time longer than 10 years. Misra et al19 studied 64 asphyxiated term newborns (defined as Apgar score less than 6 at 5 minutes) and 90 control newborns. At 3, 7, and 11 months of follow-up, the neurodevelopmental outcomes of newborns whose 5-minute Apgar scores were less than 6 were similar to those of the control newborns.19 In a study of 111 newborns, a 5-minute Apgar score less than 7 has not been shown to predict developmental dysfunction at 5 years of age.20 However, a low Apgar score at 5 minutes has been associated with late speech development at 18 months of age21 and at the age of 6 to 7 years.22 A Norwegian study on 727 children with 5-minute Apgar scores of 0–3 and neonatal symptoms during the first week of life showed a sevenfold increased risk of need for extra resources in kindergarten, a threefold increased risk of intervention at school, nearly sixfold need for follow-up by a child psychologist, and nearly fourfold the need for follow-up by the special resource center for children with educational or behavioral problems.23 These studies all have been small. However, a somewhat larger study (N=1,942) found a poor correlation between intelligence test scoring at 17 years of age and a low 5-minute Apgar score.24
This study has a long follow-up time and addresses the long-term effects of a low 5-minute Apgar score. Previous studies might not have had enough statistical power to detect small differences in cognitive outcome. A limitation to our study was that maternal substance abuse, maternal medical conditions, and socioeconomic status cannot be adjusted for. However, we have adjusted for maternal smoking and maternal educational level. One should keep in mind that the school grades do not necessarily measure cognitive skills; they could also reflect behavioral problems. Even if the association between a low Apgar score at birth and poor school performance in adolescence to some extent could be attributable to behavioral problems, that would not change the fact that Apgar score at birth is a powerful marker of neonatal health and effect on future academic achievements. Our results show a strong association between low 5-minute Apgar scores and future cognitive function and are a useful tool in the assessment of obstetric care.
1. Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 1953;32:260–7.
2. Rüdiger M, Wauer RR, Schmidt K, Küster H. The Apgar score. Pediatrics 2006;118:1314–5.
3. Thorngren-Jerneck K, Herbst A. Low 5-minute Apgar score: a population-based register study of 1 million term births. Obstet Gynecol 2001;98:65–70.
4. Moster D, Lie RT, Irgens LM, Bjerkedal T, Markestad T. The association of Apgar score with subsequent death and cerebral palsy: A population-based study in term infants. J Pediatr 2001;138:798–803.
5. Lie KK, Grøholt EK, Eskild A. Association of cerebral palsy with Apgar score in low and normal birthweight infants: population based cohort study. BMJ 2010;341:c4990.
6. Ehrenstein V, Pedersen L, Grijota M, Nielsen GL, Rothman KJ, Sørensen HT. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14.
7. Odd DE, Rasmussen F, Gunnell D, Lewis G, Whitelaw A. A cohort study of low Apgar scores and cognitive outcomes. Arch Dis Child Fetal Neonatal Ed 2008;93:115–20.
8. Robertson CM, Finer NN. Long-term follow-up of term neonates with perinatal asphyxia. Clin Perinatol 1993;20:483–500.
9. Dilenge ME, Majnemer A, Shevell MI. Long-term developmental outcome of asphyxiated term neonates. J Child Neurol 2001;16:781–92.
10. Thornberg E, Thiringer K, Odeback A, Milsom I. Birth asphyxia: incidence, clinical course and outcome in a Swedish population. Acta Paediatr 1995;84:927–32.
11. Pasamanick B, Rogers ME, Lilienfeld AM. Pregnancy experience and the development of behavior disorders in children. Am J Psychiatry 1956;112:613–8.
12. Odd DE, Lewis G, Whitelaw A, Gunnell D. Resuscitation at birth and cognition at 8 years of age: a cohort study. Lancet 2009;373:1615–22.
13. Simon NP. Long-term neurodevelopmental outcome of asphyxiated newborns. Clin Perinatol 1999;26:767–78.
14. Cnattingius S, Ericson A, Gunnarskog J, Källén B. A quality study of a medical birth registry. Scand J Soc Med 1990;18:143–8.
15. The Swedish Medical Birth Register-A summary of content and quality. Research Report from EpC. Centre for Epidemiology, The National Board of Health and Welfare, 2003. Available at: www.sos.se
. Accessed June 2, 2011.
16. National Agency for Education. Education Results 2004–National level. Report 240, 2004. Available at: http://www.skolverket.se
. Retrieved June 2, 2011.
17. Marsál K, Persson PH, Larsen T, Lilja H, Selbing A, Sultan B. Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr 1996;85:843–8.
18. Miettinen OS. Simple interval estimation of risk ratio. Am J Epidemiol 1974;100:515–6.
19. Misra PK, Srivastava N, Malik GK, Kapoor RK, Srivastava KL, Rastogi S. Outcome in relation to Apgar score in term neonates. Indian Pediatr 1994;31:1215–8.
20. Blackman JA. The value of Apgar scores in predicting developmental outcome at age five. J Perinatol 1988;8:206–10.
21. Marschik PB, Einspieler C, Garzarolli B, Prechtl HF. Events at early development: are they associated with early word production and neurodevelopmental abilities at the preschool age? Early Hum Dev 2007;83:107–14.
22. Stanton-Chapman T, Chapman D, Bainbridge N, Scott K. Identification of early risk factors for language impairment. Research Develop Disab 2002;23:390–405.
23. Moster D, Lie RT, Markestad T. Joint association of Apgar scores and early neonatal symptoms with minor disabilities at school age. Arch Dis Child Fetal Neonatal Ed 2002;86:16–21.
24. Seidman DS, Paz I, Laor A, Gale R, Stevenson DK, Danon YL. Apgar scores and cognitive performance at 17 years of age. Obstet Gynecol 1991;77:875–8.
Let xm=Σwixi/Σixi be the weighted averages of the xi:s, and ym the corresponding weighted average of the yi=log(ORi*):s, where wi=1/V(log(ORi*)).
To estimate the slope b in the regression equation log(ORx)=ym+β*(x-xm), let Sxy=Σwi*(xi-xm)(yi-ym). Then β*=Sxy/Sxx has 95% CI: β*±1.96/√Sx.