Click on the links below to access all the Data Supplements for this article.
Please note that Data Supplement files may launch a viewer application outside of your web browser.
In Canada, the rate of multiple births (twins, triplets, etc) continues to accelerate. Between the years 1993 and 2002 the incidence of multiple births increased by 18%, whereas the overall Canadian birth rate dropped 19%.1–2 Consequently, much attention has been paid recently to the increasing rate of multiple births and the potential risk of adverse perinatal outcomes that accompany such births, because the majority of multiple-birth infants are born either preterm, with low birth weight, or both. Some studies have reported that multiple births are associated with adverse perinatal outcomes and increased risk of infant mortality and morbidity,1,3–8 whereas others noted that multiple-birth infants have similar or in certain cases even better outcomes than singletons.9–13 Although the association of multiple births with infant mortality and morbidity has been previously described,1,3–13 it remains unclear whether the perinatal outcomes of multiple-birth infants are worse than that of singletons of similar gestational age. There is little information about very preterm multiple-birth infants born at or before 32 weeks of gestation, or about the relationship between perinatal risk factors and neonatal illness severity at birth and multiple birth outcomes. Moreover, few studies have taken into account the nonindependence of observations for multiple births and the hierarchical structure of the risk factors. The purpose of this study, therefore, was to compare mortality, major morbidity and use of resources among singleton and multiple-birth very preterm infants born at or before 32 weeks of gestational age while controlling the risk factors at the child, mother, and neonatal intensive care unit (NICU) levels simultaneously.
MATERIALS AND METHODS
The study population was comprised of 3,242 infants born at or before 32 weeks of gestation who were admitted to any of 24 tertiary level NICUs in the Canadian Neonatal Network between January 1, 2005, and December 31, 2005. Within this group, there were 958 multiple-birth infants and 2,284 singletons. In Canada, perinatal care is highly regionalized and infants born at or before 32 weeks of gestation are routinely admitted to a tertiary level NICU. The 24 NICUs in this study provided care for a geographic area that includes more than 90% of all births in Canada. Since 1996, the Canadian Neonatal Network has maintained a national standardized database of sociodemographic, neonatal illness severity, health care practice, and health outcome information on infants admitted to tertiary NICUs across Canada. Data are collected by trained abstractors from mothers' and infants' charts, using a standard manual of operations and definitions, and is electronically transmitted to the Canadian Neonatal Network Coordinating Centre at the Integrated Centre for Care Advancement through Research in Edmonton, Alberta. The data collection system has been described previously.14 This study was approved by institutional review boards at all 24 hospital sites.
Study variables were defined according to the Canadian Neonatal Network Data Abstractor's Manual.14 Gestational age was defined as the best obstetric estimate based on early prenatal ultrasound, obstetric examination, and obstetric history, unless the postnatal pediatric estimate of gestation differed from the obstetric estimate by more than 2 weeks. In that case, the best pediatric estimate of gestational age was used instead. An infant was defined as small for gestational age (SGA) if the birth weight was less than the third percentile for gestational age according to the growth charts established by Kramer et al15 for the Canadian population. Score for Neonatal Acute Physiology Version II16 is a neonatal illness severity score calculated from six empirically weighted physiologic measurements made during the first 12 hours of admission to the NICU. Chronic lung disease was defined as oxygen dependency at 36 weeks corrected gestational age for an infant who was born at or before 32 weeks of gestation.17 Intraventricular hemorrhage was defined according to the Canadian Pediatric Society criteria18 from a head ultrasonogram performed before 14 days of life. Necrotizing enterocolitis was defined according to Bell's criteria (stage 2 or higher)19 and was classified as medical (clinical symptoms and signs plus evidence of pneumatosis on abdominal X-ray) or surgical (histologic evidence of necrotizing enterocolitis on surgical specimen of intestine). Retinopathy of prematurity was defined according to the International Classification for Retinopathy of Prematurity20 and the Reese Classification of cicatricial disease.21 Nosocomial infection was defined using blood and cerebrospinal fluid culture results according to Freeman's criteria.22 Patent ductus arteriosus was defined as clinical diagnosis plus treatment with indomethacin or surgical ligation or both.
In this study the statistical software packages SPSS 15.0 (SPSS Inc., Chicago, IL) and HLM 6.04 (SSI Inc., Lincolnwood, IL) were used for data analyses. Using the χ2 analysis (for categorical variables) and the unpaired t test (for continuous variables), the incidence of mortality and morbidity, and the use of resources in neonates were compared for very preterm singletons and multiples. Three-level hierarchical generalized linear models (for binary outcomes) and hierarchical linear models (for continuous outcomes) analyses were employed to take into account the nonindependence of observations for multiple births and the hierarchical structure of our data (eg, infants came from singleton or multiple birth mothers within NICUs). Odds ratios (ORs) or coefficients and 95% confidence intervals (CIs) for singleton and multiple birth very preterm infants were calculated, controlling for child characteristics (admission status, gestational age, gender, neonatal illness severity score, SGA, and Apgar score at 5 minutes) at the child level (level 1), and controlling maternal demographic characteristics (delivery type, maternal hypertension, antenatal corticosteroid, maternal age, gravida, number of abortions, and tobacco use) at the maternal level (level 2). At the NICU level (level 3), specification of the level 2 intercept coefficient was set as random. The level of significance was set at P<.05.
Table 1 shows child characteristics of singleton and multiple birth very preterm infants in the study cohort. Multiple birth very preterm infants were more likely to be of higher gestational age, have lower neonatal illness severity, and have higher Apgar scores at 5 minutes.
Table 2 displays the maternal characteristics for the risk of mortality and morbidity. Very preterm multiple birth mothers were more likely to be older, have maternal hypertension and caesarean delivery, and have received antenatal corticosteroid treatment and prenatal care. These mothers were less likely to have used drugs, tobacco, or alcohol during their pregnancy, and they had fewer abortions and previous pregnancies than singleton birth mothers.
Table 3 shows that multiple birth very preterm infants have lower overall incidences of chronic lung disease, severe intraventricular hemorrhage, stage 3 or higher retinopathy of prematurity, and nosocomial infection than singleton very preterm infants. However, the gestational age–specific analyses showed no significant (P>.05) differences in mortality, major morbidity, or survival free of major morbidity between singleton and multiple-birth very preterm infants, except for higher mortality (30.2% compared with 20.0%) among multiple-birth infants born at 26 weeks or less, and a higher incidence of patent ductus arteriosus (51.2% compared with 37.7%) among multiple-birth infants born at 27–28 weeks of gestational age. Table 4 presents the results from the unpaired t test analysis for use of resources among singleton and multiple birth very preterm infants. Multiple-birth infants had significantly (P<.05) shorter mean duration of ventilation than singleton birth infants (6.9 days compared with 8.7 days), but there were no significant differences in the mean duration of oxygen use, continuous positive airway pressure use, or NICU stay.
Hierarchical generalized linear models analyses (Table 5) showed that, after adjusting for perinatal risks factors at both child and mother levels, there were no significant differences in outcomes between singleton and multiple birth very preterm infants, with the exception of multiple-birth infants who had a higher incidence of respiratory distress syndrome (adjusted odds ratio [OR] 1.3, 95% confidence interval [CI] 1.0–1.6) but a lower incidence of stage 3 or higher retinopathy of prematurity (adjusted OR 0.5, 95% CI 0.3–0.9). There were no significant differences between multiple and singleton very preterm infants in the duration of oxygen treatment, assisted ventilation use, continuous positive airway pressure use, or the length of NICU stay (Table 6). Multivariate and multilevel analyses in this study also show significant variations across NICUs for both binary and continuous outcomes (estimations of level 3 variances were P<.001).
In contrast to many previous reports of higher mortality and morbidity among multiple-birth infants,1,3–8 we found that very preterm singleton and multiple-birth infants born at or before 32 weeks of gestation had similar outcomes, with the exception of a higher incidence of respiratory distress syndrome. Our findings also show that multiple-birth infants were less likely to develop severe retinopathy of prematurity than singleton birth infants (adjusted OR 0.5, 95% CI 0.3–0.9). This result is similar to that of the recent study by Friling et al.12
Our findings are important for several reasons. First, prevailing concerns by many caregivers that very preterm multiple-birth infants have poorer outcomes than singleton infants of similar gestational age may be unfounded. This has implications for decisions concerning delivery of very preterm multiple pregnancies, infant care, and counseling of parents. Second, because individual infant risks such as being small for gestational age and high neonatal illness severity are better predictors of adverse outcomes than multiple birth status, these risk factors should be used to target infants for neurodevelopmental follow-up and early childhood intervention. Third, the finding of consistent significant variations across NICUs reveals the important association between NICU factors and the outcomes of very preterm infants. It suggests the need for further research to examine the effect that different practices at individual NICUs may have on the outcomes of very preterm infants of singleton and multiple births. Finally, our results are reassuring for the prognosis of multiple birth very preterm infants and suggest that the quality of care for these infants in Canada is at least as good as that for singleton very preterm infants.
Although our study does not provide definitive reasons why multiple birth and singleton very preterm infants have similar outcomes, we speculate that there may be several reasons. Most previous reports included infants of all gestational ages, whereas this study focused only on infants who were born at or before 32 weeks of gestation. Because the discrepancy between gestational age and fetal size and the fact that some complications of pregnancy such as gestational diabetes and hypertension tend to worsen with increasing gestational age, outcome differences between multiple birth and singleton infants may be less pronounced at the lower gestational ages. In a study of changes in stillbirths and infant mortality of preterm births among twins, Joseph et al23 reported that infant mortality rates among twins declined substantially in all categories of gestational age above 24 weeks except for live births at 32 to 33 and 34 to 36 weeks of gestation. Their report is consistent with this hypothesis and the findings in our study.
The reasons for preterm delivery in singleton and multiple birth pregnancies may also be different, potentially with singleton infants being more ill. In our study, singleton infants had higher mean neonatal illness severity scores and lower mean Apgar scores at 5 minutes, which is consistent with this hypothesis. Improvements in perinatal care, including increased access to obstetric and neonatal intensive care, availability of laboratories, improved medical technology, and more adequate staffing, may have contributed to better outcomes of preterm multiple-birth infants than in the past. For example, in this study, 81.6% of singleton and 86.9% of multiple-birth infants received antenatal corticosteroid treatment, which significantly reduces neonatal mortality and morbidity. Although it has been reported by Sen et al24 and Elimian et al,25 that a single dose of antenatal corticosteroids or incomplete course of antenatal corticosteroids is associated with a reduction in neonatal mortality and morbidity, Costa et al26 found the effects of incomplete antenatal corticosteroids are variable: they give some benefits to infants of 25 to 27 weeks of gestational age, but fail to show any difference in outcomes for the 32–34 weeks of gestational age group. Chien et al,27 in 2002 reported that increased use of antenatal corticosteroids for preterm deliveries can reduce neonatal mortality by up to 10% in Canada. Consequently, multiple-birth infants may have a better prognosis today than previously.
This study has several limitations. Because the study population comprises admissions to the NICU, it does not include deaths that occur in the delivery room and may underestimate neonatal mortality. Although perinatal care in Canada is highly regionalized, it is possible that in the rare instances when there is a shortage of hospital beds, infants may be transferred out of the geographic region for care. This may affect the incidences of outcomes reported. Because NICU admissions only represent a fraction of pregnancies with maternal complications before 33 weeks of gestation, the incidence of maternal hypertension may be underestimated. Finally, although perinatal outcomes from our study are generally favorable for multiple-birth infants, we should not downplay the overall risks resulting from multiple births. For instance, the need for specialized prenatal care to reduce complications and adverse outcomes in multiple pregnancies, and the need for ongoing social and medical care beyond the prenatal and perinatal periods, particularly for mothers of multiple infants, should not be overlooked.
2. Joseph KS, Allen AC, Dodds L, Vincer MJ, Armson BA. Causes and consequences of recent increases in preterm birth among twins. Obstet Gynecol 2001;98:56–64.
3. Blondel B, Kaminski M. Trends in the occurrence, determinants, and consequences of multiple births. Semin Perinatol 2002;26:239–49.
4. Misra DP, Ananth CV. Infant mortality among singletons and twins in the United States during 2 decades: effects of maternal age. Pediatrics 2002;110:1163–8.
6. Russell RB, Petrini JR, Damus K, Mattison DR, Schwarz RH. The changing epidemiology of multiple births in the United States. Obstet Gynecol 2003;101:129–35.
7. Fanaroff AA, Stoll BJ, Wright LL, Carlo WA, Ehrenkranz RA, Stark AR, et al. Trends in neonatal morbidity and mortality for very low birthweight infants. Am J Obstet Gynecol 2007;196:147e1–8.
8. Ombelet W, Martens G, De Sutter P, Gerris J, Bosmans E, Ruyssinck G, et al. Perinatal outcome of 12,021 singleton and 3108 twin births after non-IVF-assisted reproduction: a cohort study. Hum Reprod 2006;21:1025–32.
9. Wolf EJ, Vintzileos AM, Rosenkrantz TS, Rodis JF, Lettieri L, Mallozzi A. A comparison of pre-discharge survival and morbidity in singleton and twin very low birth weight infants. Obstet Gynecol 1992;80:436–9.
10. Alexander GR, Slay Wingate M, Salihu H, Kirby RS. Fetal and neonatal mortality risks of multiple births. Obstet Gynecol Clin North Am 2005;32:1–16.
11. Garite TJ, Clark RH, Elliott JP, Thorp JA. Twins and triplets: the effect of plurality and growth on neonatal outcome compared with singleton infants. Am J Obstet Gynecol 2004;191:700–7.
12. Friling R, Axer-Siegel R, Hersocovici Z, Weinberger D, Sirota L, Snir M. Retinopathy of prematurity in assisted versus natural conception and singleton versus multiple births. Ophthalmology 2007;114:321–4.
13. Cheung YB, Yip P, Karlberg J. Mortality of twins and singletons by gestational age: a varying-coefficient approach. Am J Epidemiol 2000;152:1107–16.
14. Lee SK, McMillan DD, Ohlsson A, Pendray M, Synnes A, Whyte R, et al. Variations in practice and outcomes in the Canadian NICU Network: 1996-1997. Pediatrics 2000;106:1070–9.
15. Kramer MS, Platt RW, Wen SW, Joseph KS, Allen A, Abrahamowicz M, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 2001;108:E35.
16. Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr 2001;138:92–100.
17. Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in preterm infants: prediction from oxygen requirement in the neonatal period. Pediatrics 1988;82:527–32.
18. Fetus and Newborn Committee, Canadian Pediatric Society. Routine screening cranial ultrasound examinations for the prediction of long term neurodevelopmental outcomes in preterm infants. J Paediatr Child Health 2001;6:39–52.
19. Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1987;187:1–7.
20. An international classification of retinopathy of prematurity. Pediatrics 1984;74:127–33.
21. Reese AB, Ling MJ, Owens WC. A classification of retrolental fibroplasias. Am J Opthalmol 1956;36:1333.
22. Freeman J, Epstein MF, Smith NE, Platt R, Sidebottom DG, Goldmann DA. Extra hospital stay and antibiotic usage with nosocomial coagulase-negative staphylococcal bacteremia in two neonatal intensive care unit populations. Am J Dis Child 1990;144:324–9.
23. Joseph KS, Marcoux S, Ohlsson A, Liu S, Allen AC, Kramer MS, et al. Changes in stillbirth and infant mortality associated with increases in preterm birth among twins. Pediatrics 2001;108:1055–61.
24. Sen S, Reghu A, Ferguson SD. Efficacy of a single dose of antenatal steroid in surfactant-treated babies under 31 weeks' gestation. J Matern Fetal Neonatal Med 2002;12:298–303.
25. Elimian A, Figueroa R, Spitzer AR, Ogburn PL, Wiencek V, Quirk JG. Antenatal corticosteroids: are incomplete courses beneficial? Obstet Gynecol 2003;102:352–5.
26. Costa S, Zecca E, De Luca D, De Carolis MP, Romagnoli C. Efficacy of a single dose of antenatal corticosteroids on morbidity and mortality of preterm infants. Eur J Obstet Gynecol Reprod Biol 2007;131:154–7.
27. Chien L, Ohlsson A, Seshia MM, Boulton J, Sankaran K, Lee SK, et al. Variations in antenatal corticosteroid therapy: a persistent problem despite 30 years of evidence. Obstet Gynecol 2002;99:401–8.
Figure. No caption available.
Supplemental Digital Content
© 2008 The American College of Obstetricians and Gynecologists