Gynecology & Obstetrics

Gynecology & Obstetrics
Open Access

ISSN: 2161-0932

Research Article - (2013) Volume 3, Issue 1

Infant Mortality Attributable to Birth Defects in Singletons and Twins in Japan, 1995-2008

Yoko Imaizumi* and Kazuo Hayakawa
Department of Health Sciences, Graduate School of Medicine, Osaka University, Suita City, Osaka, Japan
*Corresponding Author: Yoko Imaizumi, A-1302, Kokubo 120-55, Akashi City, Hyogo Prefecture 673-0005, Japan, Tel: +81-78-928-6027 Email:

Abstract

Objective: To evaluate the Infant Mortality Rate (IMR) associated with birth defects in singletons and twins and to identify risk factors for IMR.

Study design: IMR was estimated using Japanese vital statistics from 1995 to 2008.

Results: All IMRs associated with birth defects significantly decreased from 1995 to 2008 for both singletons and twins. IMR was approximately 3-fold higher in the latter than in the former during the period 1995–2008. In contrast, IMR with chromosomal abnormalities was the same in singletons and twins. The proportion of infant deaths associated with all birth defects among the total number of infant deaths was nearly constant (40%) during the examination period in singletons. In contrast, the value in twins was 20% in 1995 and increased to 25% in 2008. During the period, IMR significantly decreased in six categories of birth defects in singletons and two categories in twins. The Relative Risk (RR) in twins vs. singletons was 52-fold for anencephaly and 14-fold for patent ductus arteriosus. RR in twins vs. singletons was 2.9-fold for IMR and 3.8-fold for the neonatal mortality rate. With regard to the circulatory system (Q20-Q28), RR of IMR in twins vs. singletons decreased with maternal age (MA) (from 7.6-fold to 2-fold). For chromosomal abnormalities (Q90-Q99), IMR in singletons was 15-fold higher for the oldest MA group than in the youngest group. With regard to the Gestational Age (GA) and birth weight, IMR associated with all birth defects decreased with an increase in GA from the shortest week of GA to the longest and from the lowest birth weight to the highest.

Conclusion: IMR associated with all birth defects decreased significantly from 1995 to 2008 for both singletons and twins. RR in twins vs. singletons was approximately 3-fold during the period 1995-2008. IMR for Q90-Q99 was similar in singletons and twins. IMR of anencephaly in twins increased following in vitro fertilization and stimulation of ovulation. To reduce different IMRs between twins and singletons, it is important to provide intensive care for twin babies during the neonatal period.

Keywords: Infant mortality; Neonatal mortality; Birth defects; Twins; Singletons; Risk factors

Introduction

According to Little and Bryan [1], Congenital Malformations (CMs) occurred more frequently in twins than in singletons in most studies. In Japan, the prevalence of CMs was higher in twins (1.2%) than in singletons (1.07%) [2]. Several studies have reported that the prevalence of birth defects in multiple births increased following Assisted Reproductive Technology (ART) [3-6]. In Japan, the twinning rate has increased since 1987 and is attributed to ovulation stimulation treatments and ART [7]. It is of interest to know the yearly change in the Infant Mortality Rate (IMR) associated with birth defects for both singletons and twins.

According to Broussard et al. [8], birth defects were a major contributor to Infant Mortality (IM) in the US. In Japan, the proportion of CMs among all infant deaths was 1.9% in 1947 and increased to 10.7% in 1965 and 26.9% in 1981 [9] and to 35.3% in 1995 [10].

The purpose of this study was to compare all-cause IMRs associated with birth defects in singletons and twins during the period 1995-2008 and to identify risk factors for IMR associated with birth defects.

Materials and Methods

Data on Live Births (LBs) and infant deaths were obtained from the vital statistics of Japan for the years 1995-2008 (Health and Welfare Statistics and Information Department, Ministry of Health, Labour and Welfare, Japan). These data cover the entire population of Japan. Death certificates provide information regarding nationality, sex, dates, Birth Weight (BW), Gestational Age (GA), parental age, single or multiple births, birth order of multiple births, cause of death, and other details. LB certificate records contain the same information, except for data related to cause of death. In Japan, Neonatal Mortality (NM) refers to the death of a live-born infant occurring within 28 days from the time of birth. The NM Rate (NMR) is defined as the number of NM cases per 10000 LBs. IM refers to the death of a live-born infant occurring within 1 year from the time of birth. IMR is defined as the number of IMs per 10000 LBs.

All birth defects were classified into 11 categories according to the ICD 10th revision [11]: Q00-Q07 (CMs of the nervous system), Q10-Q18 (CMs of the eye, ear, face, and neck), Q20-Q28 (CMs of the circulatory system), Q30-Q34 (CMs of the respiratory system), Q35-Q37 (the cleft lip and cleft palate), Q38-Q45 (other CMs of the digestive system), Q50-Q56 (CMs of genital organs), Q60-Q64 (CMs of the urinary system), Q65-Q79 (CMs and deformations of the musculoskeletal system), Q80-Q89 (other CMs), and Q90-Q99 (chromosomal abnormalities, not elsewhere classified). Of the 19 specific causes of deaths in Table 1, 18 causes of deaths were chosen in accordance with the large number of infant deaths resulting from each cause of deaths in singletons and twins as well as conjoined twins. In Table 2, the linear regression coefficients of IMR on the year was used to test the change in IMR from 1995-1996 to 2007-2008. The odds ratio (OR) and 95% Confidence Intervals (CIs) were used to test IMRs between the two categories. The correlation coefficient was also used to test the relationship between IMR and MA in Table 3. The average of MA in each group was computed in each CM category for both singletons and twins. The following numbers of infant deaths did not differ between singletons and multiple births: 366 (7.2%) in 1995, 313 (6.9%) in 1996, 337 (7.7%) in 1997, 330 (7.5%) in 1998, 271 (6.8%) in 1999, 268 (7.0%) in 2000, 259 (7.2%) in 2001, 220 (6.3%) in 2002, 206 (6.1%) in 2003, 199 (6.4%) in 2004, 209 (7.1%) in 2005, 185 (6.5%) in 2006, 169 (6.0%) in 2007, and 192 (6.9%) in 2008.

Cause of death (ICD-10th code)     Number of deathsIMR1 Odds ratio [95% CI]
Singletons Twins Singletons Twins  
Q00-Q07 CMs of the nervous system 714 (4.1%)  151 (15.1%) 0.45 4.82 10.63 [8.92-12.67]*
Q00.0  Anencephaly 109 112 0.07 3.58 51.64 [39.67-67.22]*
Q04.2Holoprosencephaly 200 18 0.13 0.57 4.52 [2.79-7.32]*
Q10-Q18 CMs of eye, ear, face and neck 3 1 0 0.03 -
Q20-Q28 CMs of the circulatory system 8289 (47.6%)  451 (45.1%) 5.27 14.41 2.74 [2.49-3.01]*
Q20.1Double outlet right ventricle 401 12 0.26 0.38 1.5 [0.85-2.67]
Q20.3  Discordantventriculoarterial connection 523 23 0.73 0.33 2.21 [1.46-3.34]*
Q20.4  Double inlet ventricle 435 11 0.28 0.35 1.27 [0.70- 2.31]
Q21.0  Ventricularseptal defect 662 25 0.42 0.8 1.90 [1.27-2.83]*
Q21.2  Atrioventricularseptal defect 636 11 0.4 0.35 0.87 [0.48 1.58]
Q23.4 Hypoplastic left heart syndrome 1086 29 0.69 0.93 1.34 [0.93-1.94]
Q24.9  CM of heart, unspecified 599 23 0.38 0.73 1.93 [1.27-2.93]*
Q25.0  Patentductusarteriosus 542 147 0.34 4.7 13.68 [11.36-16.36]*
Q25.1 Coarctation of aorta 469 23 0.3 0.73 2.46 [1.62-3.74]*
Q25.5  Atresia of pulmonary artery 309 28 0.2 0.89 4.55 [3.09-6.73]*
Q26.2  Total anomalous pulmonary venous connection 576 10 0.37 0.32 0.87 [0.47-4.24]
Q30-Q34    CMs of the respiratory system 1947 129 1.24 4.12 3.33 [2.79- 3.98]*
Q33.6      Hypoplasia and dysplasia of lung 1567 109 1 3.48 3.50 [2.88-4.24]*
Q35-Q37    Cleft lip and cleft palate 15 2 0.01 0.06 -
Q38-Q45    Other CMs of the digestive system 526 39 0.33 1.25 3.73 [2.69-5.16]*
Q50-Q56    CMs of genital organs 3 2 0 0.06 -
Q60-Q64    CMs of the urinary system 574 45 0.36 1.44 3.94 [2.91- 5.34]*
Q60.6  Potter’s syndrome 431 23 0.27 0.73 2.68 [1.76-4.08]*
Q65-Q79  CMs anddeformations of themusculoskeletal system 1150 45 0.73 1.44 1.97 [1.46 -2.65]*
Q79.0  Congenital diaphragmatic hernia  644 16 0.41 0.51 1.25 [0.76-2.05]
Q80-Q89    Other CMs 969 80 0.62 2.56 4.15 [3.30-5.21]*
Q89.4  Conjoined twins - 20 - 0.64 -
Q90-Q99    Chromosomal abnormalities, not elsewhere classified    3241 (18.6%) 56 (5.6%) 2.06 1.79 0.87 [0.67-1.13]
Q91.3Edwards syndrome, unspecified 2076 29 1.32 0.93   0.70 [0.49 1.01]
Q91.7Patau syndrome 455 11 0.29 0.35   1.21 [0.67-2.21]
Q00-Q99CMs, deformations and chromosomal abnormalities  17431 (70.2%) 1001 (65.7%) 11.08 31.98   2.89 [2.71-3.08]*

Note: 1 per 10000 live births (n=313,045 for twins and n=15,727,022 for singleton); CI: Confidence interval; *Significant at the 5% level; , CMs: Congenital malformations

Table 1: Number of Infant Deaths due to 11 Categories of All CMs and 19Specific Cause of Deaths and Infant MortalityRatein Singletons and Twins, 1995-2008.

Results

Yearly change in IMRs associated with birth defects

Table 2 shows IMRs in singletons and twins according to 11 categories of birth defects from 1995-1996 to 2007-2008. The percentage of infant deaths associated with birth defects among the total infant deaths was nearly constant (39%-42%) in singletons during the period, whereas it was 21% in 1995-1996 and gradually increased to 27% in 2007–2008 in twins. Excluding the small number of infant deaths in three CM categories (Q10-Q18, Q35-Q37, and Q50-Q56) among 11 categories, IMR decreased significantly for Q00-Q07, Q20-Q28, Q30-Q34, Q38-Q34, Q65-Q79, and Q80-Q89 in singletons from 1995-1996 to 2007-2008 and decreased significantly for Q20-Q28 and Q30-Q34 in twins. For Q00-Q07 in twins, IMR increased during the period; however, the increase was not statistically significant. IMR associated with all birth defects decreased significantly in both singletons and twins. The Relative Risk (RR) in twins vs. singletons was nearly constant between 1995-1996 and 2007-2008 (2.7-3.4-fold) for all birth defects. In contrast, the ratio for Q00-Q07 was 5.5-fold in 1997- 1998, increased to 19-fold in 2005-2006, and decreased to 13.2-fold in 2007–2008. Table 2 also shows IMR associated with all birth defects in both sexes. With one exception, IMRs between males and females were similar in each period in both singletons and twins. The exception was IMR for the period 2005–2006 in twins where the rate was significantly higher in males than in females.

ICDICDICD-10th Number of infant deaths Infant mortality rate (IMR1) per 10000 live births 1995-2008 NMR1) (%) Regretion coefficient
1995     1997 1999 2001  2003 2005 2007 1995             1997 1999 2001  2003 2005 2007
Singletons                                
Q00-Q07   146 133 104 98 79 68 82 0.62 0.57 0.45 0.43 0.36 0.32 0.38 0.27(61) -0.02*
Q10-Q18 2 0 0 1 0 0 0 0.01 0 0 0 0 0 0 0 -
Q20-Q28 1614 1419 1265 1204 1082 845 838 6.86 6.04 5.45 5.29 4.95 4.01 3.93 2.49(47) -0.24*
Q30-Q34   394 322 299 268 254 246 156 1.68 1.37 1.29 1.18 1.16 1.17 0.73 1.04(84) -0.06*
Q35-Q37 0 4 3 2 3 2 1 0 0.02 0.01 0.01 0.01 0.01 0 0.01 -
Q38-Q45    128 91 72 62 66 46 60 0.54 0.39 0.31 0.27 0.3 0.22 0.28 0.12(36) -0.02*
Q50-Q56 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0  
Q60-Q64 91 80 86 87 93 64 71 0.39 0.34 0.37 0.38 0.43 0.3 0.33 0.33(92) -0.003
Q65-Q79    212 206 169 157 155 119 129 0.9 0.88 0.73 0.69 0.71 0.56 0.6 0.60(82) -0.03*
Q80-Q89 162 179 165 155 111 107 90 0.69 0.76 0.71 0.68 0.51 0.51 0.42 0.38(61) -0.03*
Q90-Q99 488 494 439 508 413 396 501 2.08 2.1 1.89 2.23 1.89 1.88 2.35 1.01(49) 0.01
Total 3237 2929 2602 2542 2256 1894 1929 13.76 12.46 11.21 11.17 10.33 8.99 9.04 6.26(57) -0.39*
Males 1697 1489 1354 1298 1172 995 967 14.06 12.34 11.34 11.1 10.46 9.21 8.82 6.56(59)  
Females 1540 1440 1248 1244 1084 899 962 13.46 12.58 11.06 11.24 10.19 8.77 9.26 5.93(54)  
% of infant deaths due to CM among total infant deaths 39.8 39.6 39.6 42.1 41.3 38.6 39.9    
Twins                                
Q00-Q07 15 13 20 20 30 29 23 3.74 3.11 4.52 4.34 6.21 6.13 5.08 4.31(90) 0.21
Q10-Q18 1 0 0 0 0 0 0 0.25 0 0 0 0 0 0 0.03 -
Q20-Q28 78 77 74 68 61 43 50 19.47 18.43 16.73 14.77 12.63 9.09 11.04 9.39(65) -0.86*
Q30-Q34 26 25 24 15 18 16 5 6.49 5.98 5.43 3.26 3.73 3.38 1.1 3.77(92) -0.41*
Q35-Q37 1 0 1 0 0 0 0 0.25 0 0.23 0 0 0 0 0.1 -
Q38-Q45     5 5 10 5 4 5 5 1.25 1.2 2.26 1.09 0.83 1.06 1.1 0.45(37) -0.04
Q50-Q56 0 0 0 0 0 2 0 0 0 0 0 0 0.42 0 0.06 -
Q60-Q64 6 7 6 7 6 7 6 1.5 1.68 1.36 1.52 1.24 1.48 1.32 1.44 (100) -0.02
Q65-Q79 5 6 4 10 6 7 7 1.25 1.44 0.9 2.17 1.24 1.48 1.55  1.31 (91) 0.03
Q80-Q89 10 10 21 16 8 8 7 2.5 2.39 4.75 3.47 1.66 1.69 1.55 1.95(76) -0.130
Q90-Q99 4 5 10 5 15 10 6 1 1.2 2.26 1.09 3.11 2.11 1.32 0.99(56) 0.07
Total 151 148 170 146 148 127 109 37.7 35.41 38.43 31.7 30.65 26.84 24.06 23.80(75) -1.18*
Males 75 79 78 74 70 79 55 37.39 37.34 34.95 31.89 28.63 32.98* 24.26 24.21(75)  
Females 76 69 92 72 78 48 54 38.01 33.44 41.98 31.51 32.73 20.54 23.86 23.38(74)  
% of infant deaths due to CM among total infant deaths 21.2 22.7 25.7 26.6 25.9 26.3 26.8    
Ratio of twins to singletons (Q00-Q07) 6 5.5 10.1 10.1 17.2 19 13.2 16  
    (Q00-Q99)           2.7 2.8 3.4 2.8 3 3 2.7 3.8  
    (Q10-Q99)2.6             2.7 3.2 2.5 2.5 2.4 2.2 3.3  

1Neonatal mortality rate per 10000 live births; *: Significant at the 5% level between IMRs for males and females

Table 2: Number of Congenital Malformations, Deformations and Chromosomal Abnormalities (CM) of Infant Deaths and Death Rates in Singleton and Twins, 1995-2008.

Table 2 also shows the proportion of neonatal deaths among infant deaths in each CM category. Excluding the small number of infant deaths in three categories of CMs, the proportions ranged between 36% and 92% in singletons and 37% and 100% in twins. For all birth defects, these values were 57% in singletons and 75% in twins. The highest difference between percentages of singletons and twins was 29% (Q00-Q07), followed by 18% (Q20-Q28 and the total birth defects). NMR for all birth defects was significantly higher in twins (23.80 per 10000 LBs) than in singletons (6.26) (OR, 3.81; 95% CI, 3.54-4.11).

Comparison of IM between singletons and twins

Table 1 shows IMRs for 11 CM categories and 19 specific causes of deaths during the period 1995-2008. Excluding the small number of deaths in three categories (Q10-Q18, Q35-Q37, and Q50-Q56), IMRs were significantly higher in twins than in singletons for seven CM categories, where the highest RR in twins vs. singletons was 10.6- fold for Q00-Q07 and the lowest was 2-fold for Q65-Q79. In contrast, RR in twins vs. singletons was 0.87-fold for Q90–Q99. Of IMRs for 18 specific causes of deaths, IMRs for 10 specific causes of deaths were higher in twins than in singletons. In singletons, the highest IMR was 1.32 for Edward’s syndrome, unspecified (Q91.3), followed by 1.00 for hypoplasia and dysplasia of the lung (Q33.6) and 0.69 for hypoplastic left heart syndrome (Q23.4). In twins, the highest IMR was 4.70 for patent ducts arteriosus (Q25.0), followed by 3.58 for anencephaly (Q00.0) and 3.48 for Q33.6. RR in twins vs. singletons was 51.6-fold for Q00.0, followed by 13.7-fold for Q25.0. IMR in conjoined twins (Q89.4) was 0.64; in this case, the cause of death is peculiar to twins.

IMRs by maternal age (MA)

Table 3 shows IMRs for Q00-Q07, Q20-Q28, Q90-Q99, and the total birth defects according to MA during 1999-2008. IMRs for Q00-Q07 in singletons were 0.3-0.4 for MA <25 to 35-39 groups and 1.0 for the oldest MA group, where the differences between the lowest IMR and those for the oldest MA group were statistically significant at the 5% level. In contrast, IMRs in twins were also similar (4.5-5.6) for MA <25 to 35-39 years and the lowest (1.8) for MA ≥ 40 years, where the number of infant deaths was only one. IMRs in twins were independent of MA. Excluding the small number of infant deaths, RRs in twins vs. singletons were between 11- and 17-fold. For Q20– Q28, IMR in singletons was 4.2 for MA <25 years to 25-29 years and increased thereafter (11.1) up to MA ≥ 40 years. IMR was significantly lower for MA <25 years than for MA 35-39 years and ≥ 40 years. In contrast, IMR in twins decreased from 17.4 for MA <25 years to 11 at 30–34 years and increased thereafter to 22 up to MA ≥ 40 years. IMR for MA 30–34 years was significantly lower than that for MA <25 years or ≥ 40 years. IMR was significantly higher in twins than in singletons for each MA group. RR in twins vs. singletons was the highest for the youngest MA group (4.1-fold) and decreased (2-fold) for MA ≥ 40 years. For Q90–Q99, IMR in singletons was the lowest (1.1) for MA <25 years and increased (13.5) with an increase in MA up to ≥ 40 years. IMR was significantly lower for MA <25 years than for MA of 30-34, 35-39, and ≥ 40 years. In contrast, IMR in twins was 0.5 for MA <25 years and increased to 9.2 with an increase in MA up to ≥ 40 years. IMR for MA 25-29 years was significantly lower than those for MA of 35-39 years and ≥ 40 years. IMR was similar in both singletons and twins in each MA group. With regard to the birth defects, IMR in singletons was the lowest (8.5) for MA of 25-29 years, which was significantly lower than those for MA of 30-34 (9.3), 35-39 (13.6), and ≥ 40 years (30.8). Similarly, IMR in twins was the lowest (28) for MA 25-29 years, whereas IMRs in the oldest MA groups were the highest (46). The former IMR was significantly lower than the latter. IMR was significantly higher in twins than in singletons for each MA group. RR in twins vs. singletons was the highest in the youngest MA group (3.7- fold) and decreased (1.5-fold) with MA up to ≥ 40 years.

Maternal age  Singletons    Twins Odds ratio [95%CI] Odds ratio [95%CI] Twins vs. singletons Odds ratio [95%CI] Singletons Twins
Number IMR1 Number IMR1
Q00-Q07   -0.73   (-0.60)      
<25 64 0.4 9 4.5 11.09 [5.52-22.28]* 1.25 [0.92-1.69] 1.00 Reference
25-29 124 0.3 38 5.6 17.38 [12.08-24.99]* 1.00 Reference 1.26 [0.61-2.60]
30-34 149 0.4 51 5.4 14.28 [10.39-19.63]* 1.17 [0.92-1.48] 1.20 [0.59-2.44]
35-39 63 0.4 23 5.2 12.03 [7.46-19.39]* 1.34 [0.99-1.81] 1.16 [0.54-2.51]
≥40 18 1 1 1.8 - 2.95 [1.80-4.83]* -
Q20-Q28   -0.8   -0.31      
<25 670 4.2 35 17.4 4.12 [2.93-5.79]* 1.00 Reference 1.59 [1.08-2.33]*
25-29 1623 4.2 88 13 3.08 [2.48-3.81]* 1.00 [0.92-1.10] 1.19 [0.89-1.58]
30-34 1765 4.5 104 11 2.46 [2.02-3.00]* 1.06 [0.99-1.15] 1.00 Reference
35-39 795 5.5 55 12.4 2.28 [1.73-3.00]* 1.29 [1.17-1.43]* 1.13 [0.82-1.57]
≥40 209 11.1 12 22.2 2.00 [1.12-3.59]* 2.62 [2.24-3.06]* 2.02 [1.11-3.68]*
Q90-Q99   (0.84)   -0.82      
<25 180 1.1 1 0.5 - 1.00 Reference  
25-29 453 1.2 6 0.9 0.75 [0.34-1.68] 1.04 [0.88-1.24] 1.00 Reference
30-34 686 1.7 14 1.5 0.85 [0.50-1.44] 1.53 [1.30-1.80]* 1.66 [0.64-4.33]
35-39 634 4.4 20 4.5 1.04 [0.67-1.62] 3.84 [3.25-4.53]* 5.10 [2.05-12.71]*
≥40 254 13.5 5 9.2 0.69 [0.28-1.66] 11.87 [9.80-14.37]* 10.43 [3.18-34.17]*
Q00-Q99   -0.81   -0.67      
<25 1405 8.9 66 32.8 3.71 [2.90-4.75]* 1.04 [0.98-1.11] 1.16 [0.88-1.54]
25-29 3264 8.5 191 28.2 3.32 [2.87-3.85]* 1.00 Reference 1.00 Reference
30-34 3699 9.3 270 28.5 3.05 [2.69-3.45]* 1.10 [1.04-1.15]* 1.01 [0.84-1.21]
35-39 1975 13.6 146 33 2.44 [2.06-2.89]* 1.59 [1.51-1.69]* 1.17 [0.94-1.45]
≥40     cde581 30.8 25 46.2 1.50 [1.01-2.24]* 3.62 [3.32-4.00]* 1.64 [1.08-2.49]*

1 IMR per 10000 live births; CI: Confidence interval; *Significant at the 5% level. Figures in parentheses indicate correlation coefficients between IMR and MA.

Table 3: Infant Mortality Rate (IMR) according to Maternal Age (MA), and Correlation Coefficients between IMR and MA, 1999-2008.

Correlation coefficients between MA groups and IMRs are shown in Table 3 for Q00-Q07, Q20-Q28, Q90-Q99, and the total birth defects. For singletons, correlation coefficients indicated higher values (0.73- 0.84); however, these were not significant at the 5% level. In contrast, these coefficients in twins were -0.60 for Q00-Q07, 0.31 for Q20-Q28, 0.82 for Q90-Q99, and 0.67 for the total birth defects, which were not significant at the 5% level.

IMRs by GA

Table 4 shows IMR for three categories of CMs and the total birth defects according to GA in singletons and twins during the period 1999- 2008. With one exception, IMRs in singletons and twins decreased with an increase in GA from the shortest GA to the longest for three CM categories and the total birth defects, with the exception of Q00-Q07 at GA of 37-38 weeks in twins. IMR for Q00-Q07 was significantly higher in twins than in singletons for four GA groups ≥ 33-34 weeks. In contrast, IMRs for Q20-Q28 (33-34 and 35-36 weeks), Q90-Q99 (five GA groups <39 weeks), and the total birth defects (three GA groups ranging between 31-32 and 35-36 weeks) were significantly higher in singletons than in twins. IMRs for the total birth defects at GA of 37-38 and ≥ 39 weeks were significantly higher in twins than in singletons, where the corresponding RRs in twins vs. singletons were 1.4-fold and 2.9-fold, respectively.

Gestational age (weeks)     Singletons Twins Odds ratio[95%CI] Odds ratio[95%CI]       Twins vs. singletons Odds ratio[95%CI] Singletons Twins
Number IMR1 Number   IMR1
Q00-Q07              
<31 41 8.5 15 12.6 1.48 [0.82-2.68] 58.95 [41.12-84.51]* 4.74 [2.51-8.95]*
31-32 32 9.8 10 11 1.12 [0.55-2.28] 68.15 [45.91-101.16]* 4.14 [2.00-8.59]*
33-34 46 6.4 26 12.8 2.00 [1.24-3.24]* 44.27 [31.33-62.55]* 4.82 [2.80-8.30]*
35-36 78 2.3 39 4.9 2.14 [1.46-3.14]* 15.84 [11.83-21.21]* 1.84 [1.12-3.02]*
37-38 108 0.3 26 2.77 65 [4.99-11.74]*  2.41 [1.84-3.14]* 1.00 Reference
≥39 107 0.1 6 5 34.52 [15.17-78.57]* 1.00 Reference 1.87 [0.77-4.56
Q20-Q28              
<31 433 89.8 118 99 1.10 [0.90-1.35] 31.32 [28.23-34.74]* 17.20 [8.02-36.88]*
31-32 158 48.5 31 34.1 0.70 [0.48-1.03] 16.85 [14.33-19.82]* 5.88 [2.59-13.36]*
33-34 243 33.7 29 14.3 0.42 [0.29-0.62]* 11.70 [10.2413.36]* 2.46 [1.08-5.61]*
35-36 529 15.5 47 5.9 0.38 [0.51-0.58]* 5.37 [4.88-5.90]* 1.01 [0.46-2.24]
37-38 1589 5.1 64 6.5 1.28 [0.997-1.64] 1.77 [1.66-1.89]* 1.13 [0.52-2.46]
≥39 2144 2.9 7 5.8 2.01 [0.96-4.22] 1.00 Reference 1.00 Reference
Q90-Q99              
<31 125 25.9 11 9.2 0.36 [0.19-0.66]* 33.56 [27.65-40.73]* 11.13 [1.44-86.20]*
31-32 155 47.6 2 2.2 0.05 [0.01-0.19]* 61.76 [51.70-73.77]* 2.65 [0.24-29.22]
33-34 221 30.7 11 5.4 0.18 [0.10-0.32]* 39.74 [34.02-46.42]* 6.52 [0.84-50.52]
35-36 431 12.6 10 1.3 0.10 [0.05-0.19]* 16.33 [14.41-18.50]* 1.51 [0.19-11.79]
37-38 708 2.3 11 1.1 0.49 [0.27-0.90]* 2.94 [2.64-3.28]* 1.35 [0.17-10.49]
≥39 574 0.8 1 0.8 1.07 [0.15-7.62] 1.00 Reference 1.00 Reference
Q00-Q99              
<31 996 206.5 222 186.2 0.90 [0.78-1.04] 45.23 [42.12-48.56]* 14.27 [8.59-23.72]*
31-32 596 183 66 72.5 0.39[0.30-0.51]* 39.98 [36.63-43.64]* 5.50 [3.18-9.49]*
33-34 922 128 96 47.2 0.37 [0.30-0.45]* 27.81 [25.85-29.91]* 3.57 [2.10-6.06]*
35-36 1667 48.8 153 19.2 0.39 [0.33-0.46]* 10.53 [9.93-11.16]* 1.44 [0.86-2.42]
37-38 3327 10.7 147 15 1.40 [1.20-1.66]* 2.30 [2.19-2.41]* 1.13 [0.68-1.90]
≥39 3455 4.7 16 13.3 2.85 [1.74- 4.66]* 1.00 Reference 1.00 Reference

1IMR per 10000 live births; CI: Confidence Interval; *Significant at the 5% level

Table 4: Infant Mortality Rate according to Gestational Age, 1999-200.

IMRs by BW

Table 5 shows IMR for the three categories of CMs and the total birth defects according to BW in singletons and twins during 1995- 2008. With two exceptions, BW-specific IMRs in singletons and twins decreased with an increase in BW from the lightest to the heaviest weight in both singletons and twins. The exceptions were for the total birth defects at BW 1000-1499 g in singletons and ≥ 3000 g in twins. For Q00-Q07, IMRs were significantly higher in twins than in singletons for BW <1000g and 1000–1499 g. For Q20-Q28, IMR was significantly higher in twins than in singletons for BW <1000 g and ≥ 3000 g. In contrast, IMRs were higher in singletons than in twins from 1000-1499 g to 2000-2499 g. For Q90-Q99, IMRs were significantly higher in singletons than in twins from BW <1000 g to 2000-2499 g. With regard to the total birth defects, IMR was significantly higher in singletons than in twins from BW 1000-1499 g to 2000-2499 g, whereas the opposite result was obtained at the heaviest weight.

Birth weight Singletons Twins Odds ratio[95%CI] Odds ratio[95%CI]       Twins vs. singletons Odds ratio[95%CI] Singletons Twins
Number IMR1 Number   IMR1
Q00-Q07              
<1000 46 13.9 23 28.3 2.04 [1.24-3.37]* 145.15 [101.30-207.97]* 48.88 [18.58-128.60]*
1000-1499 75 14.9 41 27.3 1.84 [1.25-2.69]* 155.49 [113.87-212.32]*   47.10 [18.61-119.22]*
1500-1999 132 10.3 58 11.6  1.13 [0.83-1.54] 107.05 [81.42-140.74]* 19.98 [8.02-49.83]*
2000-2499 194 2 23 1.6 0.80 [0.52-1.24] 21.00 [16.25-27.12]* 2.78 [1.06-7.33]*
2500-2999 163 0.3 5 0.6 2.06 [0.85-5.02] 2.94 [2.26-3.82]* 1.00 Reference
≥3000 84 0.1 0 0 - 1.00 Reference -
Q20-Q28              
<1000 469 141.7 153 188.4    1.34[1.11-1.61]* 149.89 [133.81-167.91]* 100.89 [25.00-407.19]*
1000-1499 455 90.3 76 50.6    0.56 [0.44-0.71]*    95.04 [84.76-106.56]* 26.72 [6.56-108.82]*
1500-1999 802 62.3 92 18.4    0.29 [0.24-0.36]* 65.38 [59.34-72.03]* 9.68 [2.39-39.31]*
2000-2499 1575 16.3 100 7    0.43 [0.35-0.53]* 17.07 [15.70-18.56]* 3.70 [0.91-15.01]*
2500-2999 2724 4.7 28 3.3    0.69 [0.48-1.003] 4.91 [4.54-5.30]* 1.71 [0.41-7.18]
≥3000 840 1 2 1.9    1.98 [0.50-8.0] 1.00 Reference  1.00 Reference
Q90-99              
<1000 200 60.4 14 17.2    0.28 [0.17-0.49]* 719.70 [551.10-939.87]*   81.75 [23.49-284.54]*
1000-1499  691 137.2 26 17.3    0.12 [0.08-0.18]* 1646.35 [1295.20-2092.70]*  82.09 [24.84-271.24]*
1500-1999  1175 91.3 12 2.4    0.03 [0.01-0.05]* 1090.53 [862.16-1379.38]*  11.36[3.21-40.27]*
2000-2499  797 8.3 3 0.2    0.03 [0.01-0.08]*   97.97 [77.21-124.32]* 1.00 Reference
2500-2999 242 0.4 0 0 -    4.95 [3.81-6.42]* -
≥3000 74 0.1 0 0 -    1.00 Reference -
Q00-Q99              
<1000 1094 330.6 268 329.9 0.998 [0.87-1.14] 104.20 [97.11-111.81]* 54.35 [40.53-72.87]
1000-1499 1803 357.9 228 151.8 0.42 [0.36-0.48]* 113.14 [106.60-120.08]* 24.55 [18.24-33.04]*
1500-1999  3109 241.5 254 50.8 0.21 [0.18-0.23]* 75.44 [71.69-79.39]* 8.13 [6.06-10.91]*
2000-2499  3785 39.3 182 12.8 0.33 [0.28-0.38]* 12.02 [11.45-12.62]* 2.04 [1.51-2.77]*
2500-2999  4317 7.5 54 6.3 0.84 [0.64-1.10] 2.28 [2.17-2.38]* 1.00 Reference
≥3000 2873 3.3 13 12.4 3.77 [2.19-6.51]* 1.00 Reference 1.97 [1.08-3.61]*

1 IMR per 10000 live births; CI: Confidence interval; *Significant at the 5% level

Table 5: Infant Mortality Rate (IMR) due to Congenital Malformations according to Birth Weight, 1995-2008.

IMRs according to birth order of twins

Table 6 shows IMR for categories of CMs according to the birth order of twins during 1995-2008. With respect to the total birth defects, IMR in the second birth order of twins was significantly higher than that in the first order of twins (1.3-fold). A similar tendency was observed for Q00-Q07 (1.5-fold), Q38-Q45 (2.2-fold), Q65-Q79 (2.0- fold), and Q80-Q89 (1.6-fold).

  First-born Second-born Odds ratio  
  No. of deaths   IMR No. of deaths   IMR [95% CI]
Q00-Q07 60 3.8 90 5.8 1.52 [1.09-2.10]*
Q20-Q28 220 14 31 14.8 1.06 [0.88-1.28]
Q30-Q 34 55 3.5 74 4.8 1.36 [0.96-1.93]
Q38-Q45 12 0.8 26 1.7 2.19 [1.10-4.34]*
Q60-Q64 21 1.3 24 1.5 1.15 [0.64-2.07]
Q65-Q79 15 1 29 1.9 1.95 [1.05-3.64]*
Q80-Q89 31 2 49 3.1 1.60 [1.02-2.50]*
Q90-Q99 26 1.7 29 1.9 1.13 [0.66-1.91]
Q00-Q99 443 28.2 555 35.6 1.27 [1.12-1.44]*

Table 6: Infant Mortality Rate (IMR) due to Congenital Malformations according to Birth Order of Twins, 1995-2008.

Discussion

In the present study, IMRs for CMs of the nervous system in twins increased from 1995–1996 to 2005–2006. In contrast, the rate in singletons decreased during this period. The corresponding RRs in twins vs. singletons were 6-fold and 19-fold, respectively. Similarly, RR in twins vs. singletons was 51.6-fold for anencephaly during 1995– 2008. Lancaster [3] reported that the prevalence of neural tube defects increased following in vitro fertilization and stimulation of ovulation. A significant correlation was observed only between anencephaly and twin births [12]. In Japan, the twinning rate was 8.9 per 1000 births in 1995–1998 and increased to 10.2 in 1999–2003 and 11.1 in 2004–2008 [13]. According to Hansen et al. [4], children born following ART were at a high risk of birth defects. Ooki [5] reported that the early NMR was slightly higher in the ART group (5.09 per 10000 LBs) than in the general population (3.86) in Japan during the period 2004–2008. In Japan, the higher RR of anencephaly is related to the effect of in vitro fertilization and stimulation of ovulation.

RR in twins vs. singletons was 2.9–fold for IMR and 3.8–fold for NMR during the period 1995–2008. Therefore, intensive care of twins in the neonatal period is important to reduce the difference in IMRs between singletons and twins. IMR for patent ductus arteriosus (PDA; Q25.0) was 13.6–fold higher in twins than in singletons. Layde et al. [14] reported that incidence rates of PDA were 17.43 per 10000 LBs in white singletons and 86.07 in white twins in the US during 1969–1976, where RR in twins vs. singletons was 4.9–fold. The corresponding values in the black population were 24.38–, 184.44–, and 7.6–fold, respectively. They also reported that a high concordance rate was observed in samesex twin pairs, which suggested a strong genetic component to the etiology of PDA. In the present study, excluding infant deaths due to anencephaly, PDA, and conjoined twins from the total infant deaths associated with birth defects, IMR was recomputed using Table 1 and IMR of 10.67 was obtained in singletons and 23.06 in twins. RR in twins vs. singletons was 2.2–fold. The difference between IMRs in twins and singletons can be reduced by 25% compared with IMRs associated with all birth defects. It should be possible to reduce IMR associated with birth defects in twins through management for the reduction of premature and low BW and care for the second order of twins.

Acknowledgements

We are grateful to the staff of Statistics and Information Department, Ministry of Health, Labour and Welfare in Japan.

References

  1. Little J, Bryan EM (1988) Congenital malformations. In: Twinning and Twins. MacGillivray I, Campbell D M, Thompson B (Edr.), John Wiley & Sons, New York, USA.
  2. Imaizumi Y, Yamamura H, Nishikawa M, Matsuoka M, Moriyama I (1991) The prevalence at birth of congenital malformations at a maternity hospital in Osaka City, 1948-1990. Jpn Hum Genet36: 275-287.
  3. Lancaster PAL (1987) Congenital malformations after in vitro fertilization. Lancet2: 1392-1393.
  4. Hansen M, Bower C, Milne E, de Klerk N, Kurinczuk JJ (2005) Assisted reproductive technologies and the risk of birth defects-a systematic review. Human Reproduction 20: 328-338.
  5. Ooki S (2011) Birth defects in singleton versus multiple ART births in Japan (2004–2008). J Pregnancy.
  6. Boyle B, McConkey R, Garne E, Loane M, Addor MC, et al. (2013) Trends in the prevalence, risk and pregnancy outcome of multiple births with congenital anomaly: a registry-based study in 14 European countries 1984-2007. Inter J Obstet Gynaecol 120: 707-716.
  7. Imaizumi Y (2004) Demographic Trends in Japan and Asia. In: Blickstein I, Keith L (Eds.), Multiple Pregnancy. Parthenon Publishing Group, London, UK.
  8. Broussard CS, Gilboa SM, Lee KA, Oster M, Petrini J R, et al. (2012) Racial/ethnic differences in infant mortality attributable to birth defects by gestational age. Pediatrics 130: e518-e527.
  9. Imaizumi Y (1986) Genetic and biological consequences of mortality changes. In: United Nations. Consequenc of Mortality Trends and Differentials, Population Studies.
  10. Ministry of Health, Labour and Welfare (1998) Vital Statistics of Japan, Statistics and Information Department, Ministry of Health, Labour and Welfare, Tokyo, Japan.
  11. Ben-Ami I, Vaknin Z, Reish O, Sherman D, Herman A, et al. (2005) Is there an increased rate of anencephaly in twins? Prenat Diagn 25: 1007-1010.
  12. Imaizumi Y, Hayakawa K (2012) Deaths from twin-twin transfusion syndrome in Japan, 1995-2008. Gynecol Obsteric 2: 116.
  13. Layde PM, Erickson JD, Falek A, McCarthy BJ (1980) Congenital malformation in twins. Am J Hum Genet 32: 69-78.
Citation: Imaizumi Y, Hayakawa K (2013) Infant Mortality Attributable to Birth Defects in Singletons and Twins in Japan, 1995-2008. Gynecol Obstet 3:150.

Copyright: © 2013 Imaizumi Y, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Top