OBJECTIVE. Our goal was to investigate whether outcome in extremely low birth weight infants changes over time in Finland.
PATIENTS AND METHODS. All infants with a birth weight <1000 g born in Finland in 1996–1997 and 1999–2000 were included in the study. Perinatal and follow-up data were collected in a national extremely low birth weight infant research register. Data concerning cerebral palsy and visual impairment were obtained from hospitals, the national discharge, and visual impairment registers.
RESULTS. A total of 529 and 511 extremely low birth weight infants were born during 1996–1997 and 1999–2000. No changes were detected in prenatal, perinatal, neonatal, and postneonatal mortality rates between the periods. The survival rates including stillborn infants were 40% and 44%. The incidence of respiratory distress syndrome and septicemia increased from 1996–1997 to 1999–2000 (75% vs 83% and 23% vs 31%). The overall incidence of intraventricular hemorrhage increased (29% vs 37%), but the incidence of intraventricular hemorrhage grades 3 through 4 did not (16% vs 17%). The rates of oxygen dependency at the age corresponding with 36 gestational weeks, retinopathy of prematurity stages 3 to 5, cerebral palsy, and severe visual impairment did not change. Mortality remained higher in 1 university hospital area during both periods compared with the other 4 areas, but no regional differences in morbidity were detected during the later period.
CONCLUSIONS. No significant changes were detected in birth or mortality rate in extremely low birth weight infants born in Finland during the late 1990s, but some neonatal morbidities seemed to increase. Regional differences in mortality were detected in both cohorts. Repeated long-term follow-up studies on geographically defined very preterm infant cohorts are needed for establishing reliable outcome data of current perinatal care. Regional differences warrant thorough audits to assess causalities.
The more active approach to antenatal and perinatal treatment of most immature infants has recently been shown to decrease stillbirth rates and increase the survival at gestational weeks (GWs) from 22 to 25.1 Surfactant and antenatal steroids have obviously decreased the mortality as a consequence of respiratory problems in live-born infants2–5 and in addition, antenatal steroids have been protective in the prevention of intraventricular hemorrhage (IVH)6,7 and retinopathy of prematurity (ROP),8,9 the diseases which, in addition to bronchopulmonary dysplasia (BPD), have been shown to predict the overall neurosensory outcome during later childhood in very premature infants.10–12 On the other hand, in relation to more aggressive treatment practices and improved survival, an alarming finding was that the postnatal time to death has increased, resulting in elongated suffering and augmented treatment costs in infants who eventually die.13,14
The study by Evans et al15 showed how survival rates might be exaggerated even by 100% and 56% at 23 and 24 GWs, respectively, by including only those admitted to the NICU. Because the development in antenatal and neonatal care affect the stillbirth and survival rates directly, as well as the opinion concerning how active treatment should be for the very immature infants,1 the definition of extremely low birth weight (ELBW) infant populations are crucial for evaluating outcome. Thus, reliable and generalizable outcome rates are only achieved from geographically defined ELBW infant populations, including stillbirths and delivery room deaths.
Since 1996, data for all live-born and stillborn ELBW infants born in Finland have been registered in a national ELBW infant research register. In our previous reports on the outcome of ELBW infants born in 1996 and 1997, we found high incidence of neonatal morbidity and long-term neurologic impairment and clear regional and hospital level differences in mortality and morbidity rates between university hospital areas.16,17 By using the register information with the data from the national birth, hospital discharge, and visual impairment registers, our study aimed to assess whether mortality and short-term and long-term outcomes had improved in ELBW infants born in Finland in 1999–2000 compared with an earlier cohort in 1996–1997, and to find out whether regional and hospital level differences in mortality and morbidity still exist during the later period.
SUBJECTS AND METHODS
The study population consisted of all stillborn and live-born ELBW infants (with a birth weight <1000 g and a gestational age at least 22 full weeks) born in Finland during two 2-year study periods, between January 1, 1996, and December 31, 1997, and between January 1, 1999, and December 31, 2000. All maternity hospitals participated in the study, and no infants fulfilling the criteria were excluded.
The gestational age was mainly based on ultrasound examination before the end of 20 GWs (73% and 76% in 1996–1997 and in 1999–2000, respectively) or on the last menstrual period. Infants with a birth weight below −2 SD according to Finnish intrauterine growth curves18 were classified as small for their gestational age (SGA).
Finnish maternity hospitals were classified into the following 3 levels on the basis of size, equipment, and availability of staffing: level III: university hospitals with a NICU and a neonatologist available (n = 5); level II: all central hospitals and 2 local hospitals with an obstetrician available 24 hours a day and special wards for newborn infants with or without possibility to provide basic neonatal intensive care (n = 18); level I: local hospitals with equipment and staff primarily for healthy newborns. Obstetricians and pediatricians are usually not at the hospital outside office hours (n = 20). Finland is geographically divided into 5 university hospital catchment areas, each served by 1 level III hospital. The 5 university hospitals with their catchment areas were labeled from A to E for the outcome analysis.
In every hospital, the same international criteria were used for defining respiratory distress syndrome (RDS),19 necrotizing enterocolitis (NEC),20 IVH,21 and ROP.22 Septicemia was defined as a blood culture-positive disease. The use of supplementary oxygen was recorded at the age corresponding with 36 GWs. Short-term survival was defined as the infant being alive at discharge or at the postconceptional age of 40 weeks.
Cerebral palsy (CP) was defined as a nonprogressive motor impairment with delayed motor development, spastic or dystonic muscle tone, accelerated peripheral reflexes, positive Babinski's sign, and persistent primitive reflexes. Severe visual impairment was defined as a visual acuity of <20 of 200.
Data-Collection Procedures and Statistical Methods
Data were obtained from all maternity hospitals by a prospectively designed questionnaire and were cross-linked with the Finnish National Birth Register and the death certificates from the Central Statistical Office of Finland.
CP and severe visual impairment should have been diagnosed before the end of 2000 in ELBW infants born in 1996–1997 and before the end of 2003 in those born in 1999–2000. Data were obtained from hospitals, the national discharge register (containing information of all hospital care periods in Finland), and the national visual impairment register. According to prevailing practice in Finland, infants with neurosensory handicaps or developmental disabilities are followed in hospitals.
All statistical analyses were performed with SPSS (SPSS Inc, Chicago, IL). Mantel-Haenzel and Pearson's χ2 test, Fisher's exact test, and Student's t test were used as appropriate to distinguish the outcome differences between the gestational age and birth weight groups and between the university hospital catchment areas. The outcome of ELBW infants born in each university hospital area was compared with the combined results of infants born in the other areas. In all analyses, P values of <.05 were considered significant.
The study was approved by the ethics committees of the Hospital for Children and Adolescents and the Departments of Obstetrics and Gynecology, Helsinki University Hospital, by the Ministry of Social Affairs and Health, and by the national Data Protection Ombudsman.
A total of 529 and 511 ELBW infants were born during the years 1996–1997 and 1999–2000, respectively. ELBW infants accounted for 0.4% of all infants born during the both periods. The study cohorts included 97% and 95%, respectively, of all infants born before 27 GW. In both cohorts, 64% of ELBW infants born at 27 GWs or more were SGA.
Background characteristics of ELBW infants born in 1996–1997 and 1999–2000 are presented in Table 1.
Of the live-born ELBW infants born at 22 to 26 GWs, 79% in 1996–1997 and 86% in 1999–2000 were born in tertiary care hospitals (P = .07), and 82% and 83%, respectively, were admitted to neonatal intensive care (P = .74). The rates of cesarean section and use of antenatal steroid treatment in pregnancies resulting in a birth of live-born ELBW infants at 22 to 26 GWs were 33% vs 43% (0.11) and 58% vs 70% (P = .035) in 1996–1997 vs 1999–2000, respectively.
No significant differences were detected in prenatal (34% in 1996–1997 vs 34% in 1999–2000), perinatal (55% vs 52%, respectively), neonatal (38% vs 34%, respectively), or postneonatal mortality (2% vs 1%, respectively). The overall survival rate up to the age corresponding with 40 GWs including both stillborn and live-born ELBW infants in the nominator were 40% and 44% (P = .22), and including only live-born ELBW infants were 60% and 65% (P = .12), respectively. In 1996–1997, a malformation was a primary reason for antenatal death in 13% of stillborn ELBW infants and in 1999–2000 in 10% (P = .37).
The comparisons of mortality in different birth weight groups and in different GW groups in ELBW infants born at 22 to 26 GWs are presented in Table 2.
Of all ELBW infants, who died before the age corresponding with 40 weeks, 49% died during the first 12 hours and 82% died before the age of 7 days in 1996–1997. The respective proportions were 46% and 83% in 1999–2000. The average survival time in live-born nonsurviving infants was 5.8 days (median: 0.5 days) in 1996–1997 and 5.2 days (median: 1 day) in 1999–2000 (P = .77).
The overall survival was 47% in live-born infants born before 27 GWs in 1996–1997 and 54% in those born in 1999–2000 (P = .17). In the extremely premature infants born at 22 to 25 GW, the respective survival rates were 38% and 48% (P = .07) and of these survivors, 38% and 36%, respectively, (P = .81) had intact short-term outcome, which means that infants had neither IVH (grades 3–4), ROP (stages 3–5), or oxygen dependency at the age corresponding with 36 GW.
The incidence of blood culture-positive septicemia and RDS in infants admitted to the NICU were higher in the later cohort than in earlier one (23% in 1996–1996 vs 31% in 1999–2000; P = .022 and 75% vs 83%; P = .010, respectively). The overall incidence of IVH (all grades) increased remarkably (29% vs 37%; P = .024, respectively), but the incidence of severe bleeding (grades 3–4) did not (16% vs 17%; P = .61, respectively). NEC with bowel perforation seemed to decline (8% vs 4%, respectively), but the difference was not significant (P = .08). Figure 1 summarizes disease rates at GWs from 22 to 26. More live-born infants born in 1999–2000 were treated with surfactant when compared with those born during the earlier period (67% in 1996–1997 vs 81% in 1999–2000; P < .001). There was also a significant difference in infants born at 22 to 26 GWs (59% vs 74%; P = .003). In 1996–1997, 26 infants (8% of those admitted to NICU) needed surgical closure of persistent ductus arteriosus (PDA) and 5 (2% of those admitted to NICU) needed a shunt operation because of post hemorrhagic ventricular dilatation before the age corresponding with 40 GWs. Respective numbers in 1999–2000 were 36 (12% those admitted to NICU) and 3 (1% of those admitted to NICU). Differences were not significant. Of the surviving ELBW infants (n = 211 in 1996–1997 and n = 219 in 1999–2000) 39% vs 49% (P = .11), respectively, were oxygen-dependent at the age corresponding with 36 GWs and 9% vs 5% (P = .09), respectively, had ROP (stages 3–5). The incidences of IVH (grades 3–4), ROP (stages 3–5), and oxygen dependency in ELBW infants born from 22 to 26 GWs are summarized in Fig 2.
The rate of CP in surviving ELBW infants remained stable during the 5-year study period (11% in 1996–1997 vs 12% in 1999–2000; P = .93). In infants born before the age of 27 GWs, the respective rates were 14% vs12% (P = .55) and did not differ significantly from rates of ELBW infants born at or after 27 GWs.
Severe visual impairment was diagnosed in 1% (n = 2) and in 3% (n = 6) of infants in 1996–1997 and 1999–2000, (P = .29). The rates in infants born at the lowest GWs (from 22–26) were 2% and 4% (P = .45). Neonatal and long-term problems were common in infants with severe visual impairment: all 8 infants had had ROP stage 3 or worse, 4 had had septicemias, 5 had IVH (3 had grades 2 or more severe), 6 had the prolonged need for supplementary oxygen, 2 had an operation because of PDA, 2 because of perforated NEC, and 2 because of hydrocephalus.
Of the ELBW infants born alive in level III hospitals, 66% (189 of 288) survived in the earlier period compared with 68% (193 of 285) in the later period (P = .60). The respective survival rates in level II hospitals were 41% (22 of 54) vs 59% (30 of 51) (P = .06). No live-born infants born in level I hospitals (n = 9 in 1996–1997 and n = 3 in 1999–2000) survived. In 1996–1997, significantly more live-born ELBW infants born in level III hospitals survived than in level II hospitals (P = .001), but in 1999–2000 the difference was no longer significant (P = .22).
In neither study period were any significant differences found in the rates of ROP (stages 3–5) or oxygen dependency at the age corresponding with 36 GWs between surviving infants born in secondary level hospitals and those born in tertiary level hospitals.
When the mortality and morbidity rates in the 5 university hospital districts were compared, some differences, however, remained. In 1996–1997, the mortality rate in university hospital area C was higher than in the rest of the country (58% in area C vs 35% in other areas; P = .001). A similar difference was observed during the later study period (48% in area C vs 31% in other areas; P = .008). In 1996–1997, the ELBW infants born in university hospital area C did not differ from those born in other areas with regard to background characteristics mentioned in Table 1. On the contrary, in the later cohort more ELBW infants in area C had a gestational age ≤25 weeks (28% vs 17%; P = .017) and were born outside the level III hospital (29% vs 13%; P = .001). However, the detected significant differences in mortality rates between centers persisted even in multiple regression models controlling several prenatal and postnatal factors (n = 32), such as gestational age, birth weight, Apgar scores, etc. Mortality was lowest in 1996–1997 in area A (25% in area A vs 44% in other areas; P < .001), whereas in 1999–2000, the lowest mortality was detected in university hospital area D (17% in area D vs 38% in other areas; P = .003).
No other morbidity differences detected in the earlier period persisted during the later period, such as higher rates of ROP (stages 3–5) and oxygen dependency at the age corresponding with 36 GWs in area D or lower oxygen dependency rate in area B.
Population-based surveys are needed to reliably follow the overall outcome of the extremely premature infants. In our study, we followed 2 population-based ELBW infant cohorts with all births included and, therefore, we consider our results to provide a real situation of the care and outcome of ELBW infants in our country. Because practically all infants born before the 27th GW were included in both cohorts, comparisons between the gestational-age groups in infants born at 22 to 26 GWs were possible. The birth weight-based inclusion criterion can be criticized because it will result in an increasing proportion of SGA infants with increasing gestational age, as was the case. We considered it, however, a relevant criterion because it was the criterion of the previous cohort, and also the criterion for the Finnish national ELBW infant register.
The stillborn rates were relatively high in both cohorts compared with reported rates elsewhere.1,23,24 In Finland, all infants without any signs of life born at ≥22 full weeks or weighing at least 500 g at birth are registered as stillbirths. The criterion of stillbirth varies widely between countries and the gestational age limit might be as high as 28 weeks.25
We also detected higher neonatal mortality than in recent studies.13,24 Including infants who were not actively treated because of their extreme immaturity or other diseases, such as anomalies, and those who urgently and unintentionally were born in primary or secondary level hospitals without full neonatal intensive care capacity increases the mortality rate in our birth cohorts compared with NICU-based study populations. The detected decline in overall mortality in our study was not statistically significant, although in absolute numbers, more live-born infants survived in the later period. The reported, prolonged time before death in nonsurviving ELBW infants and the increased number of postneonatal deaths in some studies13,14 could not be confirmed in our study.
In the later period, despite the more common use of antenatal steroids and surfactant, which should have decreased the incidence of RDS2 and IVH,6 the rates of RDS, IVH, and blood culture-positive septicemia seemed to increase. Rather than one explaining factor, several factors, including those not obvious in our study, might have influenced these increases. The main increase in IVH rate was detected in minor hemorrhages, which may be because of developed ultrasound equipment with a better capacity to detect minor changes, such as IVH grade 1. Practical and methodologic aspects, such as improved diagnosing and reporting, might, therefore, explain to a certain extent the increased morbidity rates.
We detected an increasing trend in the need for supplementary oxygen at the age corresponding with 36 GWs. Because prolonged need for supplementary oxygen has several long-term effects in neurologic, cognitive, and physical outcome of ELBW infants,10,12,26,27 the oxygen dependency rate in our country should continuously be followed in new ELBW infants cohorts and reasons for possibly increasing rates evaluated. The need for supplementary oxygen has commonly been used as the criterion for BPD, but because centers have dissimilar saturation target levels and different practices to use supplementary oxygen, to assess oxygen needs, and to use diuretics, the oxygen dependency rate might not reflect well the real incidence of BPD. Instead, in future studies the recently published physiologic method for assessing the need for supplementary oxygen would give a more uniform criteria for BPD.28
The incidence of severe ROP (stages 3–5) has declined during the 1980s and 1990s.29,30 Factors associated with a declining incidence of ROP, such as more common use of antenatal steroids, better nutrition, and careful monitoring in oxygen treatment might affect a declining trend noticed in our study. Unfortunately, the incidence of late visual impairments remained stable in our population, which may be explained by several other risks for unfavorable prognosis in these infants, such as complicated neonatal course (septicemia, IVH, NEC, or PDA), the long need for supplementary oxygen, and extreme prematurity.
In a Swedish study, Hakansson et al1 showed that so-called proactive perinatal treatment increased significantly the incidence of live births and survival of infants born at 22 to 25 GWs without increasing morbidity in survivors. In Finland, the perinatal care of the most immature infants seemed to be somewhat more active in 1999–2000 compared with treatment in 1996–1997 because the use of antenatal steroids and centralization of deliveries to tertiary care centers were more common during the later period. The number of survivors born before 26 GWs increased in the later cohort with a similar high proportion of survivors with IVH (grades 3–4), oxygen dependency, or ROP (stages 3–5) as in the earlier cohort. Treatment of the extremely small or premature infants remains a constant dilemma. Recently published follow-up data of the former cohort show that only a few surviving children born at 22 or 24 GWs had no functional abnormalities at the age of 5 years,31 and numerous other reports also confirm an unfavorable long-term outcome in the most premature infants.3,4,32–34 However, because published outcome results are always those of cohorts born in earlier periods and because the treatment practices in neonatal care rapidly change, the outcome of present cohorts might have changed, thus stressing the importance of continuous long-term follow-up of these infants.
The centralization of deliveries with additional risk factors, such as extreme prematurity (22–26 GWs), to tertiary care hospitals seemed to have improved during the late 1990s. This might partly explain that the difference found in mortality rates between level II and III hospitals diminished from 1996–1997 to 1999–2000. During both study periods, the mortality in 1 university hospital area remained constantly higher compared with other areas. In 1999–2000, more infants in area C compared with the other areas were born outside the tertiary care center and had a gestational age of ≤25 weeks, both of which might have increased the mortality in area C during the later period. However, as in a study by Vohr et al,35 significant differences in mortality between centers persisted even after adjustment for maternal and infant characteristics and antenatal interventions, suggesting that differences in clinical treatment practices influence the outcomes.
We found no significant changes in birth and mortality rate of ELBW infants in Finland in the late 1990s. The incidence of RDS, septicemia, and IVH seemed to have increased, and an increasing trend in the rate of oxygen dependency at the age corresponding with 36 GWs was detected in the later period. We also detected some constant differences in outcome of ELBW infants between the 5 tertiary care centers. Online register-based follow-up of neonatal mortality and morbidity rates is needed to focus developmental efforts on care practices or care organizations leading to suboptimal outcome.
This study was supported by grants from Signe and the Ane Gyllenberg Foundation.
We thank all Finnish maternity hospitals for providing data in a national research register.
- Accepted September 19, 2006.
- Address correspondence to Viena Tommiska, MD, PhD, Hospital for Children and Infants, University of Helskinki, PL 610, 00029 HUS, Finland. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Hakansson S, Farooqi A, Holmgren P, Serenius F, Hogberg U. Proactive management promotes outcome in extremely preterm infants: a population-based comparison of two perinatal management strategies. Pediatrics.2004;114 :58– 64
- ↵Costeloe K, Hennessy E, Gibson AT, Marlow N, Wilkinson AR. The EPICure study: outcomes to discharge from hospital for infants born at the threshold of viability. Pediatrics.2000;106 :659– 671
- ↵Foix-L'Helias L, Baud O, Lenclen R, Kaminski M, Lacaze-Masmonteil T. Benefit of antenatal glucocorticoids according to the cause of very premature birth. Arch Dis Child Fetal Neonatal Ed.2005;90 :F46– F48
- ↵Linder N, Haskin O, Levit O, et al. Risk factors for intraventricular hemorrhage in very low birth weight premature infants: a retrospective case-control study. Pediatrics.2003;111 . Available at: www.pediatrics.org/cgi/content/full/111/5/e590
- Msall ME, Phelps DL, DiGaudio KM, et al. Severity of neonatal retinopathy of prematurity is predictive of neurodevelopmental functional outcome at age 5.5 years. Behalf of the Cryotherapy for Retinopathy of Prematurity Cooperative Group. Pediatrics.2000;106 :998– 1005
- ↵Böhm B, Katz-Salomon M. Cognitive development at 5.5 years of children with chronic lung disease of prematurity. Arch Dis Child Fetal Neonatal Ed.2003;88 :F101– F105
- ↵Meadow W, Lee G, Lin K, Lantos J. Changes in mortality for extremely low birth weight infants in the 1990s: implications for treatment decisions and resource use. Pediatrics.2004;113 :1223– 1229
- ↵Hack M, Friedman H, Fanaroff AA. Outcomes of extremely low birth weight infants. Pediatrics.1996;98 :931– 937
- ↵Evans DJ, Levene MI. Evidence of selection bias in preterm survival studies: a systematic review. Arch Dis Child Fetal Neonatal Ed.2001;84 :F79– F84
- ↵Tommiska V, Heinonen K, Ikonen S, et al. A national short-term follow-up study of extremely low birth weight infants born in Finland in 1996–1997. Pediatrics.2001;107(1) . Available at: www.pediatrics.org/cgi/content/full/107/1/e2
- ↵Tommiska V, Heinonen K, Kero P, et al. A national 2-year follow-up study of extremely low birth weight infants born in 1996–1997. Arch Dis Child Fetal Neonatal Ed.2003;88 :F29– F35
- ↵Markestad T, Kaaresen PI, Ronnestad A, et al. Early death, morbidity, and need of treatment among extremely premature infants. Pediatrics.2005;115 :1289– 1298
- ↵Nordic Medical Statistical Committee. Births and Infants Mortality in the Nordic Countries. Copenhagen, Denmark: Nordic Medical Statistical Committee; 1993
- ↵Walsh MC, Yao Q, Gettner P, et al. Impact of a physiologic definition on bronchopulmonary dysplasia rates. Pediatrics.2004;114 :1305– 1311
- ↵Chow LC, Wright KW, Sola A; CSMC Oxygen Administration Study Group. Can changes in clinical practice decrease the incidence of severe retinopathy of prematurity in very low birth weight infants. Pediatrics.2003;111 :339– 345
- ↵Mikkola K, Riihela N, Tommiska V, et al. Neurodevelopmental outcome at 5 years of age of a national cohort of extremely low birth weight infants who were born in 1996–1997. Pediatrics.2005;116 :1391– 1400
- ↵Hoekstra RE, Ferrara B, Couser RJ, Payne NR, Connet JE. Survival and long-term neurodevelopmental outcome of extremely premature infants born at 23–26 weeks' gestational age at a tertiary center. Pediatrics.2004;113(1) . Available at: www.pediatrics.org/cgi/content/full/113/1/e1
- ↵Hintz SR, Kemdreck DE, Vohr BR, Poole K, Higgins RD; for the National Institute of Child Health and Human Development Neonatal Research Network. Changes in neurodevelopmental outcomes at 18 to 22 months' corrected age among infants of less than 25 gestational age born in 1993–1999. Pediatrics.2005;115 :1645– 1651
- ↵Vohr BR, Wright LL, Dusick AM, et al. for the Neonatal Research Network. Center differences and outcomes of extremely low birth weight infants. Pediatrics.2004;113 :781– 789
- Copyright © 2007 by the American Academy of Pediatrics