The anesthesiologist Virginia Apgar introduced her scoring system in 1953, intended as classification or “grading” of the condition of newborn infants when evaluating the effects of resuscitation.1 Later, a low Apgar score became widely used as a proxy for asphyxia. A low Apgar score at 1 minute is often caused by a temporary depression, whereas low 5‐minute and 10‐minute scores usually imply complications of clinical importance, indicating that the newborn has not responded optimally to resuscitation. Regardless of the cause of low Apgar score, it indicates compromise of the infant and is as such an unwanted outcome. The definition of birth asphyxia also incorporates hypoxia and acidosis, which is beyond the aim of this study.
The aims of this study were to determine the rate of low (under 7), and very low (under 4) 5‐minute Apgar scores in term infants born in Sweden during 10 years (1988–1997), look for time trends, evaluate possible obstetric risk factors, and relate low Apgar score to infant mortality, neonatal neurologic morbidity, and long‐term outcome. Our hypothesis was that each of the examined obstetric factors would influence the risk of low Apgar score, and that a low Apgar score would be associated with adverse outcome.
MATERIALS AND METHODS
The study was based on data from the Swedish Medical Birth Registry, compiled of copies of medical records from antenatal care, delivery, and the pediatric examination of the newborn. It covers whole Sweden and contains data on 98–99% of all deliveries. A quality analysis of the register has been published.2
We analyzed data for 10 years (1988–1997) when 1,121,992 children were born according to official birth statistics, 98.9% (1,109,826) of whom were registered. Infants born before 37 completed weeks, or with unknown gestational age combined with birth weight below 2500 g, were excluded (n = 71,909; 6.5%), leaving 1,037,917 infants. We also excluded 1416 fetal deaths and 1773 infants with severe malformations likely to affect Apgar score (central nervous system malformations, diaphragmatic hernias, pulmonary hypoplasia, severe kidney malformations, chromosomal anomalies, and multiple malformations or syndromes). Information on Apgar score was missing in 4327 cases (0.38%) and on maternal age in 341. If the 5‐minute Apgar score was missing and 1‐minute Apgar score was 9 or 10, the value of the 5‐minute score was considered to at least 7. Left for analysis were 1,028,705 infants, of whom 13,141 were multiple births.
During the study period, the expected date of delivery was estimated by ultrasonography, or if not available, the last menstrual period. Birth weight deviation from the expected weight was calculated using national growth curves, standardized for gestational age and gender.3 In the study, the following variables were considered: birth year, maternal age, parity, smoking, multiple birth, gender, birth weight, gestational age, presentation, mode of delivery, and the hour and month of birth. We also evaluated the associations between low Apgar score and the neonatal diagnoses: seizures and cerebral hemorrhage. Infant mortality was investigated using the Statistics Sweden Death Registry. Stillborns were excluded.
In addition, we linked the register with the National Hospital Discharge Registry, including all diagnoses of in‐patients in Swedish hospitals, and identified the diagnoses epilepsy, “seizures” (including febrile convulsions), cerebral palsy, and mental retardation, using codes from the International Classification of Diseases, 9th and 10th Revisions (ICD‐9 and ICD‐10).4
Statistical analyses were made by Mantel‐Haenszel procedure with stratifications as stated, and risks were expressed as odds ratio (OR) or risk ratio (RR). We used a test‐based method for the estimation of the 95% confidence intervals (CI).5
The study included 1,028,705 term infants of whom 7787 (0.76%) had 5‐minute Apgar scores below 7 and 2155 (0.21%) below 4. The rate of infants with 5‐minute Apgar score below 7 decreased from 0.77% in 1988 to 0.63% in 1992, then increased continuously to 0.82% in 1997 (Figure 1). An increased rate of low Apgar score during 1997 (OR 1.16, CI 1.07, 1.26) remained after excluding multiple births (OR 1.15, CI 1.06, 1.26), deliveries with epidural analgesia (OR 1.13, CI 1.02, 1.26), births of non‐Swedish mothers (OR 1.15, CI 1.06, 1.25), which all increased during this period. Excluding all three subgroups gave OR 1.11 (CI 0.98, 1.26).
The OR for Apgar score under 7 at 5 minutes increased with maternal age and in neonates of primiparous, compared with multiparous women. The age‐related risk was higher in secundiparous than in primiparous women. Smoking was a risk factor for moderately depressed Apgar score (4–6). The analyses were stratified for year of birth, and respectively maternal age, parity, and smoking (Table 1).
Birth weight strongly influenced the risk of low Apgar score, with a five‐fold risk for the smallest and a six‐fold risk for the heaviest infants (Figure 2A, Table 2). The high risk for large infants remained after exclusion of mothers with pre‐existing and gestational diabetes mellitus. Gestational age at birth also influenced the risk of low Apgar score (Figure 2B). From a minimum at 38–40 gestational weeks, the risk increased continuously. The risk of low Apgar score increased with increasing deviation from the expected weight, both with positive and negative weight deviations (Figure 2C). In girls, low 5‐minute Apgar score occurred less frequently than in boys (OR 0.78, CI 0.74, 0.81). The difference remained after stratification for birth weight.
There were 12,886 twins in the study. The risk of low Apgar score was higher in twins than in singletons, and more pronounced for the second twin (OR 4.14, CI 3.52, 4.87) than for the first (OR 2.40, CI 1.97, 2.93), after standardization for birth weight, 2.41 (2.01, 2.89) and 1.62 (1.31, 2.00), respectively. Singleton infants with breech presentation had OR 2.57 (CI 2.22, 2.98) for low Apgar score at 5 minutes, compared with infants in vertex presentation. For vaginally born infants, the OR for low Apgar score in breech deliveries compared with vertex deliveries was 6.67 (CI 5.87, 7.58). Epidural analgesia was associated with an increased risk of low Apgar score, with an OR of 2.10 (1.94, 2.29) for vaginally delivered singleton infants in vertex presentation.
In June, July, August, and December, the risk of low Apgar score was higher than in the rest of the year, but only in August the increased risk was significant (OR 1.10, CI 1.02, 1.18). The risk of low Apgar score was higher in infants born during “non‐office hours” (5.00 PM–6.59 AM) as compared with infants born during office hours (8.30 AM–3.59 PM) (OR 1.21, CI 1.14, 1.29). The risk during intermediate hours did not differ significantly from the other two groups (OR 1.05, CI 0.95, 1.16).
The infant mortality was 48 per 1000 (371 of 7787) in infants with 5‐minute Apgar score under 7, and 107 per 1000 (230 of 2155) with Apgar score under 4, compared with 1.8 per 1000 in those with 5‐minute Apgar score 7 or more (OR 14.4, CI 12.5, 16.5). The neonatal mortality was 40, 95, and 0.6 per 1000, respectively, and the first week of life 34, 86, and 0.3 per 1000, respectively. Infants who had an Apgar under 7 at 5 minutes and also under 7 at 10 minutes had an infant mortality rate of 129 per 1000 (239 of 1849).
Of infants who had Apgar scores below 7 at 5 minutes, 64 per 1000 (498 of 7787) had seizures as compared with 0.9 per 1000 (904 of 1,020,825) in infants with Apgar 7 or more (RR 71.5, CI 64.2, 74.5). Among infants with Apgar score below 7 at 5 minutes who also had Apgar score below 7 at 10 minutes, the rate of seizures was 149 per 1000 (283 of 1895). Of infants with an Apgar score under 7 at 5 minutes, 11.4 per 1000 (89 of 7787) had intracranial hemorrhage, compared with 0.27 per 1000 (273 of 1,020,825) in infants with an Apgar 7 or more at 5 minutes (RR 42.4). In infants with Apgar score under 7 at 5 minutes and also under 7 at 10 minutes, 31 per 1000 (50 of 1612) had intracranial hemorrhage during the neonatal period.
By linking the Medical Birth Registry to the Registry of Hospital Discharges, we investigated the risk of long‐term sequels for infants with a 5‐minute Apgar score under 7 (Table 3). The analyses were stratified for birth year, maternal age, parity, and smoking. The OR for developing cerebral palsy was 31.4 (CI 27.3, 36.1), for epilepsy 7.9 (CI 6.6, 9.4), for “seizures” (including febrile convulsions) 1.6 (CI 1.4, 1.9), and for mental retardation 9.5 (CI 7.2, 12.5).
The Swedish Medical Birth Registry, providing data on 98–99% of births in Sweden, is one of the most complete birth registers in the world. We used the register to perform a population‐based study on term infants with low Apgar score at 5 minutes after birth.
A low Apgar score is not a specific indicator of birth asphyxia because there may be other causes of this depression at birth. In this study, we excluded the two most apparent, prematurity and severe malformations. Biochemical markers, measuring acidemia, provide more specific evidence of intrauterine hypoxia, but are not available in the register. On the other hand, the Apgar score provides information about compromise of the neonate, which is not obtainable by biochemical measurements. The diagnosis “birth asphyxia” in ICD‐9 and ICD‐10 is based on a low 1‐minute Apgar score, which often is caused by a temporary depression. A 10‐minute Apgar score is often lacking in the Medical Birth Registry when a 5‐minute Apgar score is 10, and does not always reflect intrapartum events. Thus, to investigate the 5‐minute Apgar score may be as close we can come to “asphyxia” in a register study.
The incidence of 5‐minute Apgar score under 7 in term infants increased from 1992 to 1997 (after a preceding decrease during 1988–1992). Multiple birth, immigration, and epidural analgesia, three potential risk factors for low Apgar score, which all had increased during 1992–97, may have contributed to the increase.
Many studies on asphyxia are based on data from 1‐minute Apgar score,6 and some include preterm infants,7 making comparisons difficult. A study from western Sweden showed a rate of 6.9 per 1000 for 5‐minute Apgar score under 7 in term infants during 1985–91,8 comparable with the lowest annual rate in our study. In England, the incidence of “clinically significant birth asphyxia” (post‐asphyxic encephalopathy) was 6 per 1000, with no changes in incidence over a 10‐year period.9 Our incidence of Apgar score below 4 at 5 minutes was comparable with an earlier report from Sweden.10 However, the perinatal mortality rate of 210 per 1000 during 1973–1979 in those infants was much higher than the mortality before 7 days in our study, 86 per 1000 (stillbirths excluded). In Sweden, like in other western countries, the perinatal mortality rate is still decreasing. The official perinatal mortality rate in Sweden decreased from 14 per 1000 in 1973 to 5 per 1000 in 1998.
It has been debated whether a low Apgar score and asphyxia are related to subsequent cerebral palsy.6,11,12 The OR for having cerebral palsy, for an infant with 5‐minute Apgar score under 7, was in our study as high as 31, when calculated from hospital diagnoses. We only estimated the OR because the Hospital Discharge Registry, which includes only in‐patients, might underestimate the incidence.
Hypoxic ischemic encephalopathy, a consequence of birth asphyxia, was included in the register only since 1997 (ICD‐10). However, neonatal seizures, a symtom related to moderate and severe hypoxic ischemic encephalopathy,13 and known to be a risk factor for later neurological sequele,14 occurred in 6% of infants with Apgar score under 7 at 5 minutes (RR 71.5, CI 64.2, 74.5), and in 15% (283 of 1895) of those with Apgar score below 7 also at 10 minutes.
We found an increased risk of being born with a low Apgar score during “non‐office hours” (OR 1.2), as well as during the periods of general holiday (OR 1.1). This was in accordance with a Welsh study, where the RR for infant death was 2.18 for infants delivered 9 PM–8.59 AM and 1.99 during July and August.15 This may suggest that the personnel resources (quantitatively or qualitatively) may be a factor of importance.
In the present study, as in a previous,16 smoking was found to increase the risk of a low 5‐minute Apgar score, particularly in older multiparas. Confounding factors for smoking are social class, alcohol intake, and drug abuse, which could not be investigated.
Our study supported the results of another large register study reporting higher rates of asphyxia among the neonates of mothers beyond age 40 and in primiparas women.17 In the present study, the age‐related risk was moderate. The risk increment is probably multifactorial. Several complications of pregnancy increase with maternal age.17,18 Some complications, such as preeclampsia, predominantly affect the first pregnancy. Women with severe complications during their first pregnancy are also more likely to receive optimal care during subsequent pregnancies and to have fewer children.
An important finding in this study was the strong influence on low Apgar score by birth weight and gestational age. A low birth weight is known to be a risk factor for fetal compromise, being a typical finding in cases of chronic placental insufficiency,19 whereas fetal macrosomia has not gained the same interest. We found that birth weight deviation in either direction was accompanied with a similar increased risk of a low 5‐minute Apgar score. The risk of fetal compromise in post‐term pregnancies is in agreement with previous reports.20,21 In our study, the risk was obvious in week 41 and very pronounced in week 43.
A recent study showed an increased risk of perinatal morbidity and mortality in planned vaginal breech delivery compared with planned cesarean delivery in breech presentation.22 A Danish study showed a 15‐fold risk of low Apgar score in vaginal breech delivery.23 In a study from our department, vaginal breech delivery was the strongest among examined risk factors for acidemia at birth.24 The present study supported these results, showing an increased risk for a low 5‐minute Apgar score in breech presentation, mainly in vaginal delivery (OR 6.67), probably reflecting increased risks of both asphyxia and trauma. Twin birth was also a risk factor for low Apgar score, twice as high in the second born twin, also after adjusting for birth weight. Problems during delivery for the second born twin are likely, but also the possibility of a delay in delivery of a twin already compromised in utero.
We conclude that 5‐minute Apgar score under 7 in term infants is associated with an increased risk of neonatal morbidity, infant mortality, and neurologic impairment. In this large population‐based register study of over 1 million term births, it is not possible to investigate birth asphyxia. However, the outcome indicates that continued efforts to reduce the rate of low Apgar score are strongly indicated.
1. Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 1953;32:260–7.
2. Cnattingius S, Ericson A, Gunnarskog J, Källén B. A quality study of a medical birth registry. Scand J Soc Med 1990;18:143–8.
3. Källén B. A birth weight for gestational age standard based on data in the Swedish Medical Birth Registry, 1985–1989. Eur J Epidemiol 1995;11:601–6.
4. WHO. International Classification of Diseases, 9th Revision. Geneva: World Health Organization, 1997.
5. Miettinen OS. Simple interval estimation of risk ratio. Am J Epidemiol 1974;100:515–6.
6. Nelson KB, Ellenberg JH. Apgar scores as predictors of chronic neurologic disability. Pediatrics 1981;68:36–44.
7. Palme-Killander C. Methods of resuscitation in low Apgar score newborn infants — a national survey. Acta Paediatr Scand 1992;81:739–44.
8. 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.
9. Levene ML, Kornberg J, Williams TH. The incidence and severity of post-asphyxial encephalopathy in full-term infants. Early Hum Dev 1985;11:21–6.
10. Ergander U, Eriksson M, Zetterström R. Severe neonatal asphyxia. Incidence and prediction of outcome in the Stockholm area. Acta Paediatr Scand 1983;72:321–5.
11. Stanley FJ, Watson L. Trends in perinatal mortality and cerebral palsy in Western Australia, 1967 to 1985. BMJ 1992;304:1658–63.
12. Badawi N, Kurinczuk JJ, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, et al. Intrapartum risk factors for newborn encephalopathy: The Western Australian case-control study. BMJ 1998;317:1554–8.
13. Ronen GM, Penney S, Andrews W. The epidemiology of clinical neonatal seizures in Newfoundland: A population-based study. J Pediatr 1999;134:71–5.
14. Volpe JJ. Neonatal seizures. Clin Perinatol 1977;4:43–63.
15. Stewart JH, Andrews J, Cartlidge PH. Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993–5. BMJ 1998;316:657–60.
16. Garn SM, Johnston M, Ridella SA, Petzold AS. Effect of maternal cigarette smoking on Apgar scores. Am J Dis Child 1981;135:503–6.
17. Gilbert WM, Nesbitt TS, Danielsen B. Childbearing beyond age 40: Pregnancy outcome in 24,032 cases. Obstet Gynecol 1999;93:9–14.
18. Cnattingius S. Maternal age modifies the effect of maternal smoking on intrauterine growth retardation but not on late fetal death and placental abruption. Am J Epidemiol 1997; 145:319–23.
19. Golan A, Lin G, Evron S, Arieli S, Niv D, David MP. Oligohydramnios: Maternal complications and fetal outcome in 145 cases. Gynecol Obstet Invest 1994;37:91–5.
20. Gunn AJ, Gunn TR. Changes in risk factors for hypoxicischaemic seizures in term infants. Aust N Z J Obstet Gynaecol 1997;37:36–9.
21. Ingemarsson I, Källen K. Stillbirths and rate of neonatal deaths in 76,761 postterm pregnancies in Sweden, 1982–1991: A register study. Acta Obstet Gynecol Scand 1997;76:658–62.
22. Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, Willan AR. Planned caesarean section versus planned vaginal birth for breech presentation at term: A randomised multicentre trial. Term Breech Trial Collaborative Group. Lancet 2000;356:1375–83.
23. Krebs L, Langhoff-Roos J. Breech delivery at term in Denmark, 1982–92: A population-based case-control study. Paediatr Perinat Epidemiol 1999;13:431–41.
24. Herbst A, Wölner-Hanssen P, Ingemarsson I. Risk factors for acidemia at birth. Obstet Gynecol 1997;90:125–30.