International Perspectives

Since 1984, Tokyo Women’s Medical University has played a pioneering role in improving the treatment results of ultrapreterm infants in Japan. The results were cited in the newest edition of Avery’s Neonatology: Pathophysiology & Management of the Newborn. (1) Like many western countries, the survival rate of extremely low-birthweight (ELBW) and very low-birthweight (VLBW) infants improved in Japan in the 1990s, with the spread of new treatment methods such as surfactant therapy, new methods of mechanical ventilation such as high-frequency oscillatory ventilation (HFOV), and prenatal steroid therapy. (2) The Committee on Newborn Infants, Japan Society of Pediatrics, has performed nationwide surveys on the outcomes of ELBW and VLBW infants for reports on the current status of neonatal intensive care units and overall neonatal mortality in Japan at 5-year intervals since 1980. The mortality rate of infants weighing less than 1,000 g at birth significantly declined in the previous 2 decades from 55.3% in 1980 and 28.2% in 1990 to 17.8% in 2000 (TableT1). In view of this, the Ministry of Health, Labor and Welfare of the Japanese government changed the definition of the threshold of viability in the Maternal Protection Law from 24 weeks of gestation to 22 weeks of gestation in January 1991. After this change, neonatologists in Japan have increasingly intervened in deliveries at 22 to 23 weeks of gestation.Although Sumida and associates (3) and Tsuchida and colleagues (4) reported in 1999 and 2001, respectively, survival rates at 22 weeks’ gestation of approximately 30%, their results were published in Japanese, and the number of patients was small. At the Hot Topics in Neonatology conference (Washington, DC) in 2002, Nishida reported the outcomes of infants born at 22 to 23 weeks of gestation in 1995 to 2001 in Japan and documented relatively active interventions by neonatologists for such infants. (5) However, no report has been published in English about the impact of this change on the outcomes of resuscitated infants after delivery at 22 to 23 weeks’ gestation in Japan.In 1997, Oishi and associates (6) reported the outcomes of “micropremies” born weighing less than 600 g in Japan from 1983 to 1993. Although the Committee on Newborn Infants, Japan Society of Pediatrics, mentioned the short-term outcomes of VLBW and ELBW infants according to birthweight in the reports at 5-year intervals, no reports have been published in English on the outcomes of infants whose birthweights were less than 500 g in Japan since Oishi’s report.In view of this background, we performed a nationwide survey of infants born at 22 to 23 weeks’ gestation and infants born weighing less than 500 g in August 2005 by a method similar to that used by Nishida in his report at the Hot Topics in Neonatology conference in 2002. (7) The purpose of this review is to introduce the survey results and to show changes in treatment outcomes in infants born at 22 to 23 weeks’ of gestation in Japan from 1995 to 2004. (5) The treatment outcomes of infants born weighing less than 500 g, irrespective of the weeks of gestation, also are considered. Limitations of the study also are discussed.To assess the impact of the change in the definition of the threshold of viability in the Maternal Protection Law from 24 to 22 weeks of gestation, Nishida and colleagues (5) performed a nationwide survey in September 2002 of infants born at 22 to 23 weeks’ gestation between 1995 and 2001. A similar survey was conducted in August 2005 to evaluate the outcomes of extremely preterm infants born at 22 to 23 weeks’ gestation in Japan between 2002 and 2004. Questionnaires were sent to institutions belonging to the Japan Neonatologist Association, which comprises 205 institutions having tertiary and secondary neonatal intensive care units (NICUs). Of 205 NICUs, 114 (56%) responded to the questionnaire, and 580 infants born at 22 to 23 weeks’ gestation during 2002 to 2004 were included. The manager of the NICU (neonatologist in charge) at each institution was asked to answer questions concerning survival/death, birth year, weeks and days of gestation at birth, birthweight, and sex. Among survivors, severe cerebral palsy (eg, spastic quadriplegia), mental retardation (developmental quotient <70), visual defect (including stage 4 retinopathy of prematurity), and bronchopulmonary dysplasia requiring home oxygen therapy were investigated to evaluate major adverse short-term outcomes. The patients were recognized as having sequelae if any of these severe complications were confirmed. Surviving infants were followed from 8 months to 3 years, using standardized neurodevelopmental assessments (the Enjoji neurodevelopmental score prior to 18 mo of age and the New Kyoto method after reaching 18 mo of age). Gestational age was determined by obstetric methods, including fetal size assessment by ultrasonography in the first trimester, and confirmed by neonatal assessment after birth.Overall, the survival rate in infants born at 22 to 23 weeks’ gestation was 49.1% (285/580). The survival rate increased with birthweight (Fig. 1). The survival rates at 22 weeks’ and 23 weeks’ gestation were 31.1% (51/164) and 56.3% (234/416), respectively (P<0.01). In the evaluation of outcomes according to days of pregnancy among infants born between 22 weeks and 0 days of gestation and 23 weeks and 6 days of gestation, outcomes improved with the prolongation of gestational days (Fig. 2).Overall, the complication rate of neurologic sequelae in the survivors born at 22 to 23 weeks’ gestation was 32.6% (86/264) (Fig. 1). No significant difference was observed between infants born at 22 weeks of gestation (36.2%) and those born at 23 weeks of gestation (31.8%).The survival rate of all live births at 22 to 23 weeks’ gestation was 35% in 1995 to 2001 and 49.1% in 2002 to 2004 (Fig. 3). Approximately 90% of the responding centers were the same in the two surveys. Significant improvement was observed between the two periods (P<0.01).During the survey period, questions also were asked about infants born weighing less than 500 g irrespective of the weeks of gestation. The contents of the survey were the same as those for infants born at 22 to 23 weeks of gestation. Results were compared with those reported by Oishi and associates (6) and the Committee on Newborn Infants, Japan Society of Pediatrics (Table 1). The survival rate in infants born weighing less than 500 g in 2002 to 2004 was significantly increased compared with those born before 2000. The survival rate increased with an increase in the weeks of gestation at birth (Figs. 4, 5). Rieger-Fackeldey and associates (8) reported a survival rate of 60.4% in infants born weighing less than 501 g. However, 42 of 48 infants in their study were small for gestational age, and only three infants born at 22 to 23 weeks of gestation were included. In our survey, the survival rate was 53.5% (99/185) at 24 or more weeks of gestation and 31.0% (49/158) at 22 to 23 weeks of gestation (P<0.05).The limit of viability has changed over time due to medical advances and may also differ slightly, even in the same period, among countries according to the standard of medical care and ethical and religious views. The 1990s witnessed great advances in neonatology. For example, prenatal steroids, surfactant therapy and HFOV aided in successful treatment and increased survival chances of extremely preterm babies. The result was a reduction of the mortality rate of ELBW infants and decrease in the gestational age limit allowing extrauterine growth.Several countries have published national guidelines concerning the limit of viability, suggesting when it is appropriate to treat actively and when to offer compassionate, but not intensive, care. (2)(9) These guidelines differ in specificity and clarity about considered limits of gestational age and weight.The World Health Organization (WHO) has stated that a fetus should be regarded as a child at a gestational age of 22 weeks’ or older or at a birthweight of more than 500 g. (2) On the other hand, the American Academy of Pediatrics and American Heart Association describes in the Neonatal Resuscitation Program (NRP) that noninitiation of resuscitation in the delivery room is appropriate for newborns who have a confirmed gestation of less than 23 weeks or a birthweight of less than 400 g. (10)(11) However, it is footnoted that “The limit of viability has undergone considerable changes over the past 30 years. It is anticipated that this change will continue.” (10)(11)In contrast to the WHO and NRP, guidelines from different countries almost exclusively deal only with gestational age and not birthweight. The following conclusions can be drawn from different national guidelines: (2) These descriptions suggest that 22 to 24 weeks’ gestation is a gray zone in the statements from national guidelines.In Japan, the definition of the limit of viability in the Maternal Protection Law was changed in 1991 from 24 weeks to 22 weeks of gestation. Since then, neonatologists have intervened increasingly in deliveries at 22 weeks of gestation. Therefore, in Japan, resuscitation procedures seem to be performed more earnestly in neonates born with viability signs at 22 weeks of gestation than in other countries.Only a few reports on the prognosis of infants born at 22 weeks of gestation have been published, probably because only a few countries define the limit of viability as 22 weeks of gestation. Even in these reports, the number of subjects was small, and the survival rate differed markedly between infants born at 22 weeks and those born at 23 weeks of gestation. (12)(13)(14)(15)(16)(17)(18) The present survey showed a survival rate of 31.1% (51/164) in infants born at 22 weeks of gestation. To our knowledge, this survival rate is higher than any previously reported values. The survival rate of infants born at 23 weeks of gestation in this survey (56.3%, 234/416) also was higher than that in the other studies.Nishida’s previous study in 2002 and our recent study in 2005 involved the greatest numbers of subjects among studies on neonates born at 22 to 23 weeks of gestation performed in Japan after the change in the interpretation of the Maternal Protection Law in 1999. However, the response rate to the questionnaire was 55% (106/198) in the first study and 56% (114/205) in the second. Consequently, these two surveys do not necessarily represent the attitudes toward, and therapeutic results in, most ultrapremature infants in NICUs in Japan. Nevertheless, because about 90% of responding centers were the same in the two surveys, comparisons between responding centers are probably valid. Comparison of therapeutic results other than the survival rate with those in other reports is difficult because these surveys lacked careful investigation of the evaluation criteria for sequelae as performed by Wood and associates. (17)(18)The results of analysis according to the gestational age at birth from 22 weeks and 0 days to 23 weeks and 6 days are interesting. In our study, the gestational age was estimated from the fetal size determined by transvaginal ultrasonography examination at an early stage of pregnancy as well as from the date of the last menstruation. In infants born between 22 weeks and 0 days and 23 weeks and 6 days of gestation, the outcomes improved day by day with prolongation of the gestational period and did not improve suddenly between 22 weeks and 23 weeks of gestation.As mentioned previously, this survey included less than 60% of all ultrapremature infants, and the results may not represent the results at all facilities in Japan. However, these relatively good treatment outcomes in ultrapremature infants may be associated not only with medical advances such as the spread of surfactant therapy, new methods of mechanical ventilation represented by HFOV, and prenatal steroid therapy from the 1990s, but also with relatively active interventions by neonatologists after the amendment of the Maternal Protection Law. In most previous studies on ultrapremature infants, the subjects were infants born at 23 or more weeks of gestation. To our knowledge, the 164 infants born at 22 weeks of gestation included in this study is the highest number in the literature.A comparison of the two periods (1995 to 2001 and 2002 to 2005), according to birthweight, revealed a significant improvement (P<0.01). Thus, improvement in outcomes was observed even in infants born at 22 to 23 weeks’ gestation.According to our survey, the short-term outcome of ultrapremature infants born at 22 to 23 weeks of gestation and infants born with a birthweight of less than 500 g in Japan improved, but to date we know little about the long-term outcome of these children. This short-term prognostic survey had a follow-up period of 8 months to 3 years and 8 months. Interestingly, no significant difference was observed in the complication rate of severe neurologic sequelae between survivors born at 22 weeks and those born at 23 weeks of gestation, despite the same evaluation method.Not only was there an increase in the survival rate in infants born at 22 weeks’ gestation and those born weighing less than 500 g, but there was a steady increase in deliveries at 22 weeks’ gestation in which neonatologists intervene. These results, together with no significant difference in the short-term follow-up results between infants born at 22 weeks’ and those born at 23 weeks’ gestation, indicate that the efforts of neonatologists in Japan will continue to improve the prognosis of infants born at 22 weeks of gestation.We express deep gratitude to the members of the Japan Neonatologist Association for their cooperation in this survey.