Arch Gynecol Obstet DOI 10.1007/s00404-015-3773-3
MATERNAL-FETAL MEDICINE
Macrosomic and low birth weight neonates in Pacific Islanders from Samoa: a case–control study Michael Tsitas1 • Bernd C. Schmid2 • Martin K. Oehler2 • Clemens B. Tempfer1
Received: 29 March 2015 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Background To assess the perinatal morbidity and mortality of macrosomic ([4500 g) and low birth weight (LBW) (\2500 g) neonates in a Pacific Islander population (PIP) from Samoa compared to a Caucasian population (CP). Methods Case–control study. Clinical data were extracted by chart review. Results In 3166 (PIP) and 2101 (CP) deliveries, macrosomia was more prevalent and LBW less prevalent in the PIP [76/3166 (2.4 %) vs. 21/2101 (0.9 %); p \ 0.0001 and 149/3166 (4.7 %) vs. 163/2101 (7.7 %); p \ 0.0001, respectively]. Among macrosomic neonates, perinatal mortality and composite severe neonatal morbidity (CNM) were higher in the PIP compared to the CP [2/76 (3 %) vs. 0/21 (0 %) and 6/76 (7 %) vs. 1/21 (4 %), respectively]. Among LBW neonates, mortality, but not CNM, was significantly higher in the PIP [16/149 (7 %) vs. 2/163 (1 %), p \ 0.0001 and 10/149 (6 %) vs. 5/163 (3 %), p = 0.2, respectively]. The proportion of macrosomic neonates transferred to the Neonatal Intensive Care Unit was significantly higher in the PIP [50/76 (65 %) vs. 0/21 (0 %), p \ 0.0001]. Age, body mass index, and delivery mode did not independently predict CNM.
& Clemens B. Tempfer [email protected] 1
Department of Obstetrics and Gynecology, Ruhr University Bochum-Marienhospital Herne, Hoelkeskampring 40, 44625 Herne, Germany
2
Gynecology Department, University of Adelaide, Adelaide, Australia
Conclusion Samoan women have higher rates of macrosomia and lower rates of LBW compared to Caucasians, suggesting an anthropomorphic basis of this phenomenon. Keywords Pacific Islanders Macrosomia Low birth weight Perinatal mortality
Introduction Labor and delivery outcomes are influenced by individual, cultural, and ethnic factors [1]. Specifically, ethnic background has been described as an independent predictor of neonatal morbidity and mortality in a variety of populations [2, 3]. In the literature, Asian and Pacific Islander women are often regarded as a homogenous group and are usually counseled regarding their risk of labor and delivery complications as an aggregate group [4, 5]. However, within Asian populations, marked differences in obstetric characteristics exist. For example, Wong et al. reported significant differences in preterm labor, pregnancy-associated hypertension, eclampsia, gestational diabetes, small for gestation age (LBW), macrosomia, and cephalopelvic disproportion among Asian subgroups [6]. A distinct ethnic group with characteristic features different from other Asian populations is Pacific Islanders. Obstetric outcomes in indigenous Pacific Islander populations have been rarely described. In one report, Rao et al. described perinatal outcomes in a retrospective cohort of expatriate Pacific Islanders who had moved to the US [4]. In this series of 545 deliveries, Pacific Islanders from Tonga, Samoa, Guam, and Polynesia were compared to other ethnic Asian Americans such as Filipina and Indian/Pakistani women. Of note, maternal and neonatal morbidity were markedly different among these groups of women. Whereas Filipina
123
Arch Gynecol Obstet
women had the highest risk of gestational hypertension and preeclampsia, Indian/Pakistani women had the highest risk of preterm delivery, gestational diabetes mellitus, and LBW at term. Pacific Islanders, on the other hand, were characterized by markedly higher rates of macrosomic neonates. These data demonstrate that significant differences in perinatal outcomes exist among Asian women with different ethnic backgrounds. Ekerome et al. examined 20-year trends in preterm birth, LBW, and late fetal death in Pacific women in New Zealand. In this study, Pacific women had the lowest rates of preterm birth and LBW when compared to Maori and European women. Late fetal death rates declined by 49 % but still remained higher for Pacific women than for Maori and European/other women [7]. In a recent population-based study of 8508 mothers, Hayes et al. investigated the rates of exclusive breastfeeding at 8 weeks postpartum among White, native Hawaiian, and expatriate Pacific Islander residents in Hawaii [8]. In this study, native Hawaiian and Pacific Islander mothers had significantly lower rates of breastfeeding compared to Whites. These data further underscore that Pacific Islanders have distinct characteristics regarding their obstetric outcomes and behavior suggesting both anthropomorphic and cultural influences. Thus, ethnic Asian women should be counseled regarding their perinatal risk according to their specific ethnic backgrounds. In view of the scarce data on differences within Pacific Islanders in the obstetric literature, counseling of these women is difficult. Apart from the studies of Rao and Hayes and collaborators, perinatal outcome data of specific Pacific Islander populations have not been reported so far (PUBMED search; search terms: Pacific Islander, Pacific, Samoa, Tonga, Guam, Polynesia, delivery, perinatal, neonatal, morbidity, mortality; access date: 06-19-2014). Therefore, we attempted to characterize rates and outcomes of neonates with macrosomia and LBW in a defined Pacific Islander population. We collected data in a cohort of indigenous women delivering neonates with macrosomia ([4500 g) and low birth weight (LBW; \2500 g) in a single institution on the island of Samoa and compared these data to a set of Caucasian controls from Germany. This study was designed to provide clinical data in order to better counsel native and expatriate women with a Pacific Islander ethnic background.
Materials and methods We performed a case–control study of women who delivered a macrosomic neonate [4500 g as defined by Boulet et al. [9] or a LBW neonate \2500 g as defined by Alexander et al. [10] at the Department of Obstetrics, TTM National Hospital, Apia, Samoa and Department of
123
Obstetrics, Marienhospital Herne, Germany. Data in Samoa were collected by one of the authors (BS) during a research visit at the TTM National Hospital from 1/2007 to 12/2007. Perinatal morbidity and mortality data were collected at the Departments of Obstetrics and Gynecology and Neonatology, TTM National Hospital, Apia, Samoa. In a case–control approach, data from this Pacific Islander population (PIP) were compared to a German Caucasian population (CP) with neonates [4500 g or \2500 g delivered in the Marienhospital Herne, Germany between 1/2005 and 12/2010. We applied no matching criteria. Patient characteristics were extracted from patient charts. Approval for this study was obtained by the Ethics Committee of the Ruhr University Bochum, Bochum, Germany (Ethics Committee of the Ruhr University Bochum, Bochum, Germany, number 4540-12, issue date December 18, 2012). Values are given as median or mean values where appropriate. Comparisons between groups were made by chi-square test in case of categorical variables or by Mann– Whitney U test in case of continuous variables. p values \0.05 were considered statistically significant. We tested the influence of maternal age, parity, gravidity, twin delivery, maternal body mass index, and neonatal sex on composite severe neonatal morbidity (CNM) [11], i.e., respiratory distress syndrome, intracranial hemorrhage, sepsis, necrotizing enterocolitis, Erb’s palsy (brachial plexus injury), and severe hypoglycemia (defined as blood glucose \30 mg/dL [12]) in a multiple linear regression model. Some mothers prematurely took their babies out of the NICU and left the hospital for reasons not accessible to the authors (n = 10). In these cases, the neonatal status at the time of discharge was used for analysis. We used the statistical software SPSS 22.0 (SPSS Inc., Chicago, IL) for statistical analyses.
Results During the study period, 3166 and 2101 deliveries [24 weeks gestation and/or with a birth weight [500 g were identified in the PIP and CP, respectively. Rates of macrosomia were significantly higher and rates of LBW were significantly lower in the PIP compared to the CP [76/ 3166 (2.4 %) vs. 21/2101 (0.9 %); p \ 0.0001 and 149/3166 (4.7 %) vs. 163/2101 (7.7 %); p \ 0.0001, respectively]. Patient characteristics of the study populations are shown in Tables 1 and 2. The proportion of neonates transferred to the Neonatal Intensive Care Unit was significantly higher in the PIP for macrosomic [50/76 (65 %) vs. 0/21 (0 %), p \ 0.0001], but not for LBW neonates [106/149 (71 %) vs. 110/163 (67 %), p = 0.5]. Neonatal morbidity is described in detail in Table 3.
Arch Gynecol Obstet Table 1 Characteristics of women with neonates with macrosomia [4500 g from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany PIP
CP
p value
Total
76
21
–
Birth weighta (g)
4752 ± 314
4674 ± 110
0.07
Week of gestationa Agea (years)
40.1 ± 0.6 32.4 ± 5.9
40.4 ± 0.9 30.3 ± 4.8
0.1 0.09 0.001
Body Mass Indexa
39.5 ± 5.2
33.7 ± 5.9
Parity
3.3 ± 1.9
2.3 ± 1.1
0.008
Graviditya
4.2 ± 1.7
2.3 ± 1.1
\0.0001
Vaginal delivery
50/76
11/21
0.3
Cesarean section
26/76
10/21
Female
30/76
4/21
Male
46/76
17/21
7/76
1/21
Delivery mode
Fetal sex
GDM 0
0.1 1
5 APGAR \7
2/76
0/21
1
Transfer to NICU
50/76
0/21
\0.0001
Days in NICUa
2.4 ± 2.9
0
\0.0001
BSLa
114.0 ± 29.2
94.4 ± 16.1
0.01
PIP Pacific Islander population, CP Caucasian population, NICU neonatal intensive care unit, BSL blood sugar level a
Mean (±standard deviation)
Among macrosomic neonates, perinatal mortality and CNM, i.e., respiratory distress syndrome, intracranial hemorrhage, sepsis, necrotizing enterocolitis, Erb’s palsy Table 2 Characteristics of women with neonates with low birth weight \2500 g from a Pacific Islander population in Apia, Samoa, and a German Caucasian Population in Herne, Germany
(brachial plexus injury), and severe hypoglycemia, were higher in the PIP compared to the CP [2/76 (3 %) vs. 0/21 (0 %) and 6/76 (7 %) vs. 1/21 (4 %), respectively], but these differences were not statistically significant (p = 1.0 and p = 1.0, respectively). Among LBW neonates, mortality, but not CNM was significantly higher in the PIP compared to the CP [16/149 (7 %) vs. 2/163 (1 %), p \ 0.0001 and 10/149 (6 %) vs. 5/163 (3 %), p = 0.2, respectively]. In a multivariable logistic regression model, maternal age, body mass index, delivery mode, and gravidity did not independently predict CNM in the PIP and CP (Table 4). Screening for gestational diabetes mellitus (GDM) was performed in all 21 CP women and in 32/76 PIP women with a macrosomic fetus. A diagnosis of GDM was established in 1/21 and 7/76 women, respectively. Of note, this difference was notably higher despite the fact that fewer women in the PIP were screened for GDM. Mean serum blood sugar levels at the day of delivery were 114.0 ± 29.2 and 94.4 ± 16.1 mg/dL (p = 0.01) in PIP and CP women, respectively (Table 1). Mode of delivery was not significantly different between the PIP and CP. Specifically, PIP women with macrosomic neonates delivered vaginally after spontaneous labor in 50/76 (65 %) cases. Primary and secondary cesarean sections were performed in 17 and 9 cases, respectively. In comparison, mode of delivery among the 21 CP women with macrosomic neonates was vaginal delivery after spontaneous labor and labor induction in 7 and 4 cases, respectively, [11/21, (52 %)]. Primary cesarean section and secondary cesarean section
PIP
CP
p value
Total (women/neonates)a
135/149
150/163
–
Birth weight (g)b
2117.0 ± 432.5
2151.4 ± 350.6
0.4
Week of gestationb
37.6 ± 4.1
34.9 ± 2.3
\0.0001
Age (years)b
26.5 ± 7.1
27.9 ± 5.9
0.07
Body Mass Indexb
30.7 ± 4.7
27.8 ± 5.6
\0.0001
Parityb
1.6 ± 2.1
1.7 ± 1.0
0.5
2.6 ± 2.1
2.1 ± 1.3
0.01
Vaginal spontaneous
104
41
\0.0001
Cesarean section
45
122
Female
77
81
Male
72
82
50 APGAR \7 Transfer to NICU
4/149 106/149
3/163 110/163
0.3 0.1
Days in the NICUb
8.3 ± 13.1
11.8 ± 23.3
0.2
Gravidity
a
Delivery mode
Fetal sex 0.7
PIP Pacific Islander population, CP Caucasian population, NICU neonatal intensive care unit a
Number of women and neonates differ due to twin deliveries
b
Median (±standard deviation)
123
Arch Gynecol Obstet Table 3 Composite severe neonatal morbidity in macrosomic and low birth weight neonates from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany
Macrosomic neonates PIP (n = 76)
LBW neonates PIP (n = 149)
Macrosomic neonates CP (n = 21)
LBW neonates CP (n = 163)
Total CNM
7
19
1
6
RDS
0
4
0
2
ICH
0
0
0
1
Sepsis
3
4
0
0
NEC
0
0
0
0
Erb’s palsy
1
0
0
0
SH
3
2
1
3
Other
0
9
0
0
PIP Pacific Islander population, CP Caucasian population, LBW low birth weight, CNM composite neonatal morbidity, RDS respiratory distress syndrome, ICH intracranial hemorrhage, NEC necrotizing enterocolitis, SH severe hypoglycemia; Neonates may have [1 CNM diagnoses
Table 4 Multivariable logistic regression analysis of composite severe neonatal morbidity in women from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany Parametera
PIP multivariable logistic regression OR (95 % CI)
p value
CP multivariable logistic regression OR (95 % CI)
p value
Age (B35 vs. [35 years)
1.2 (0.1–13.2)
0.9
0.5 (0.1–4.2)
0.5
Delivery mode (vaginal vs. CS)
0.1 (0.6–81.0)
0.1
1.1 (0.3–4.1)
0.9
Maternal BMI (\32 vs. C32) Gravidity (0 vs. [1)
0.3 (0.0–4.7) 6.9 (0.2–55.5)
0.3 0.4
2.4 (0.7–8.6) 0.9 (0.3–3.1)
0.2 0.9
PIP Pacific Islander population, CP Caucasian population, OR odds ratio, CI confidence interval, CS Cesarean section, BMI Body Mass Index a
Parameters in the multivariable analysis
were performed in 5 and 5 cases, respectively, in CP women. Delivery complications, especially severe vaginal/perineal lacerations, were noted less often in PIP women [10/76 (13 %)] (intrauterine asphyxia: n = 2, postpartum hemorrhage: n = 3, shoulder dystocia: n = 1; severe vaginal/perineal laceration: n = 4) compared to 10/21 (62 %) women in the CP (intrauterine asphyxia: n = 0, postpartum hemorrhage: n = 1, shoulder dystocia: n = 2; severe vaginal/perineal laceration: n = 7). This difference was statistically significant (p = 0.001). PIP women with LBW neonates had a vaginal delivery more often than CP women [104/149 (69 %) vs. 41/163 (25 %), p \ 0.0001]. Specifically, mode of delivery among the 149 PIP women with LBW neonates was vaginal delivery after spontaneous labor in 103 cases and vaginal delivery after labor induction in 1 case [104/149 (69 %)]. Primary cesarean section and secondary cesarean section were performed in 30 and 15 cases, respectively. In contrast, 122 women with LBW neonates in the CP were delivered by cesarean section [122/163 (75 %); primary cesarean section: n = 80; secondary cesarean section: n = 42]. Vaginal delivery after spontaneous labor occurred in 25, vaginal delivery after labor induction in 16 cases. Delivery complications were noted in 13/149 (9 %)
123
deliveries in the PIP (intrauterine asphyxia: n = 5, postpartum hemorrhage: n = 7, shoulder dystocia: n = 0; severe vaginal/perineal laceration: n = 1) and in 41/150 (27 %) women in the CP (intrauterine asphyxia: n = 2, postpartum hemorrhage: n = 6, shoulder dystocia: n = 0; severe vaginal/perineal laceration: n = 33).
Discussion In this case–control study of Pacific Islander women from Samoa and Caucasian women from Germany with macrosomic and LBW neonates, Pacific Islander women had higher rates of macrosomic neonates but lower rates of LBW neonates, suggesting an anthropomorphic basis of this phenomenon. PIP neonates with both macrosomia and LBW had comparatively poor perinatal outcomes with increased morbidity and mortality, potentially attributable to the poor local resource setting. Epidemiologic studies in Pacific Islanders have shown high rates of obesity and diabetes. For example, in a survey of 15 countries in the Western Pacific and South-east Asia including data from 330,374 individuals, Martiniuk and colleagues found the highest population attributable fraction of elevated BMI for diabetes mortality in Samoa with
Arch Gynecol Obstet
98 % in males and 95 % in females [13]. In line with these data, a cohort study of five ethnic groups (South-east Asian, South Asian, Middle Eastern, Anglo-European and Pacific Islander) found that women from the Pacific Islands had the highest parity, BMI, fasting glucose levels, HbA(1c), and the greatest need for insulin therapy [14]. Elevated BMI and preexisting diabetes are well-known risk factors of fetal macrosomia [15]. In accordance with these epidemiologic data, it has been previously demonstrated that expatriate women from Pacific Islands have higher rates of macrosomia compared to other Asian women [4]. We confirm these data demonstrating that both high rates of macrosomia and low rates of LBW infants are obstetric phenomena, which have to be expected both in indigenous and expatriate women from this population. Also, PIP women had higher rates of blood sugar levels and GDM. To our best knowledge, this is the first study assessing the frequency and perinatal outcomes associated with macrosomia and LBW in an indigenous PIP. Our data suggest that anthropomorphic factors, cultural factors such as nutritional habits, or both, have important clinical consequences for obstetricians caring for these women during pregnancy and delivery. On a political level, nutritional education programs, GDM screening, and shoulder dystocia trainings for midwives and obstetricians may be important and efficacious interventions in Samoa and other Pacific Islands. For obstetricians caring for expatriate Pacific Islander women, it may be helpful to note that this specific ethnic subgroup has a substantially elevated risk of macrosomia and associated delivery complications. In addition, macrosomic and LBW neonates from PIP mothers may be more vulnerable to subpartum stress compared to Caucasian controls, although the poor outcomes observed in our study may be attributable to both anthropomorphic effects as well as the level of obstetric and NICU sophistication present in Samoa. Our study has limitations. For example, we did not record sociodemographic variables such as level of education, income, socioeconomic status, availability to health care ressources, and marital status. Therefore, we cannot quantify the effect of between these factors on neonatal morbidity and mortality as described in this study. In a developing country such as Samoa, however, such information is difficult to acquire. Due to this methodological limitation, we were not able to assess the influence of socioeconomic factors on obstetric outcome parameters. In addition, data on long-term neonatal outcomes are missing in our dataset due to the lack of systematic pediatric care in Samoa. Thus, the differences regarding neonatal morbidity and mortality may even be higher than those observed in our analysis. Also, the absolute numbers of the primary outcome, i.e., CNM, are low. Therefore, it is possible that
some of the specific CNM outcomes may be over- or underrepresented in our analysis. In summary, Samoan indigenous women have higher rates of macrosomia, but lower rates of LBW, compared to Caucasian controls, suggesting an anthropomorphic basis of this phenomenon. Neonates with macrosomia and LBW have poor perinatal outcomes in this population, potentially attributable to limited local health resources. These data may be helpful for health officials and obstetricians in Samoa and other PIP as well as for obstetricians caring for expatriate Samoan women. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Edmonds JK, Yehezkel R, Liao X et al (2013) Racial and ethnic differences in primary, unscheduled cesarean deliveries among low-risk primiparous women at an academic medical center: a retrospective cohort study. BMC Pregnancy Childbirth 13:168 2. Racape J, De Spiegelaere M, Dramaix M et al (2013) Effect of adopting host-country nationality on perinatal mortality rates and causes among immigrants in Brussels. Eur J Obstet Gynecol Reprod Biol 168(2):145–150 3. Cabacungan ET, Ngui EM, McGinley EL et al (2012) Racial/ ethnic disparities in maternal morbidities: a statewide study of labor and delivery hospitalizations in Wisconsin. Matern Child Health J 16(7):1455–1467 4. Rao AK, Daniels K, El-Sayed YY et al (2006) Perinatal outcomes among Asian American and Pacific Islander women. Am J Obstet Gynecol 195(3):834–838 5. Yu SM, Alexander GR, Schwalberg R et al (2001) Prenatal care use among selected Asian American groups. Am J Public Health 91:1865–1868 6. Wong LF, Caughey AB, Nakagawa S et al (2008) Perinatal outcomes among different Asian-American subgroups. Am J Obstet Gynecol 199(4):382 7. Ekeroma AJ, Craig ED, Stewart AW, Mantell CD, Mitchell EA (2004) Ethnicity and birth outcome: New Zealand trends 1980–2001: Part 3. Pregnancy outcomes for Pacific women. Aust N Z J Obstet Gynecol 44(6):541–544 8. Hayes DK, Mitchell KM, Donohoe-Mather C et al (2014) Predictors of exclusive breastfeeding at least 8 weeks among Asian and Native Hawaiian or other Pacific Islander race subgroups in Hawaii, 2004–2008. Matern Child Health J 18(5):1215–1223 9. Boulet SL, Alexander GR, Salihu HM, Pass M (2003) Macrosomic births in the united states: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol 188(5):1372–1378 10. Alexander GR, Kogan M, Bader D, Carlo W, Allen M, Mor J (2003) US birth weight/gestational age-specific neonatal mortality: 1995–1997 rates for whites, hispanics, and blacks. Pediatrics 111(1):61–66 11. Witt A, Do¨ner M, Petricevic L, Berger A, Germann P, Heinze G, Tempfer C (2011) Antibiotic prophylaxis before surgery vs after cord clamping in elective cesarean delivery: a double-blind, prospective, randomized, placebo-controlled trial. Arch Surg 146(12):1404–1409 12. Rozance PJ, Hay WW Jr (2010) Describing hypoglycemia—definition or operational threshold? Early Hum Dev 86(5):275–280
123
Arch Gynecol Obstet 13. Martiniuk AL, Lee CM, Colagiuri S et al (2011) Higher-thanoptimal body mass index and diabetes mortality in the Asia Pacific region. Diabetes Res Clin Pract 94(3):471–476 14. Wong VW (2012) Gestational diabetes mellitus in five ethnic groups: a comparison of their clinical characteristics. Diabet Med 29(3):366–371
123
15. Alberico S, Montico M, Barresi V et al (2014) The role of gestational diabetes, pre-pregnancy body mass index and gestational weight gain on the risk of newborn macrosomia: results from a prospective multicentre study. BMC Pregnancy Childbirth 14:23
MATERNAL-FETAL MEDICINE
Macrosomic and low birth weight neonates in Pacific Islanders from Samoa: a case–control study Michael Tsitas1 • Bernd C. Schmid2 • Martin K. Oehler2 • Clemens B. Tempfer1
Received: 29 March 2015 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Background To assess the perinatal morbidity and mortality of macrosomic ([4500 g) and low birth weight (LBW) (\2500 g) neonates in a Pacific Islander population (PIP) from Samoa compared to a Caucasian population (CP). Methods Case–control study. Clinical data were extracted by chart review. Results In 3166 (PIP) and 2101 (CP) deliveries, macrosomia was more prevalent and LBW less prevalent in the PIP [76/3166 (2.4 %) vs. 21/2101 (0.9 %); p \ 0.0001 and 149/3166 (4.7 %) vs. 163/2101 (7.7 %); p \ 0.0001, respectively]. Among macrosomic neonates, perinatal mortality and composite severe neonatal morbidity (CNM) were higher in the PIP compared to the CP [2/76 (3 %) vs. 0/21 (0 %) and 6/76 (7 %) vs. 1/21 (4 %), respectively]. Among LBW neonates, mortality, but not CNM, was significantly higher in the PIP [16/149 (7 %) vs. 2/163 (1 %), p \ 0.0001 and 10/149 (6 %) vs. 5/163 (3 %), p = 0.2, respectively]. The proportion of macrosomic neonates transferred to the Neonatal Intensive Care Unit was significantly higher in the PIP [50/76 (65 %) vs. 0/21 (0 %), p \ 0.0001]. Age, body mass index, and delivery mode did not independently predict CNM.
& Clemens B. Tempfer [email protected] 1
Department of Obstetrics and Gynecology, Ruhr University Bochum-Marienhospital Herne, Hoelkeskampring 40, 44625 Herne, Germany
2
Gynecology Department, University of Adelaide, Adelaide, Australia
Conclusion Samoan women have higher rates of macrosomia and lower rates of LBW compared to Caucasians, suggesting an anthropomorphic basis of this phenomenon. Keywords Pacific Islanders Macrosomia Low birth weight Perinatal mortality
Introduction Labor and delivery outcomes are influenced by individual, cultural, and ethnic factors [1]. Specifically, ethnic background has been described as an independent predictor of neonatal morbidity and mortality in a variety of populations [2, 3]. In the literature, Asian and Pacific Islander women are often regarded as a homogenous group and are usually counseled regarding their risk of labor and delivery complications as an aggregate group [4, 5]. However, within Asian populations, marked differences in obstetric characteristics exist. For example, Wong et al. reported significant differences in preterm labor, pregnancy-associated hypertension, eclampsia, gestational diabetes, small for gestation age (LBW), macrosomia, and cephalopelvic disproportion among Asian subgroups [6]. A distinct ethnic group with characteristic features different from other Asian populations is Pacific Islanders. Obstetric outcomes in indigenous Pacific Islander populations have been rarely described. In one report, Rao et al. described perinatal outcomes in a retrospective cohort of expatriate Pacific Islanders who had moved to the US [4]. In this series of 545 deliveries, Pacific Islanders from Tonga, Samoa, Guam, and Polynesia were compared to other ethnic Asian Americans such as Filipina and Indian/Pakistani women. Of note, maternal and neonatal morbidity were markedly different among these groups of women. Whereas Filipina
123
Arch Gynecol Obstet
women had the highest risk of gestational hypertension and preeclampsia, Indian/Pakistani women had the highest risk of preterm delivery, gestational diabetes mellitus, and LBW at term. Pacific Islanders, on the other hand, were characterized by markedly higher rates of macrosomic neonates. These data demonstrate that significant differences in perinatal outcomes exist among Asian women with different ethnic backgrounds. Ekerome et al. examined 20-year trends in preterm birth, LBW, and late fetal death in Pacific women in New Zealand. In this study, Pacific women had the lowest rates of preterm birth and LBW when compared to Maori and European women. Late fetal death rates declined by 49 % but still remained higher for Pacific women than for Maori and European/other women [7]. In a recent population-based study of 8508 mothers, Hayes et al. investigated the rates of exclusive breastfeeding at 8 weeks postpartum among White, native Hawaiian, and expatriate Pacific Islander residents in Hawaii [8]. In this study, native Hawaiian and Pacific Islander mothers had significantly lower rates of breastfeeding compared to Whites. These data further underscore that Pacific Islanders have distinct characteristics regarding their obstetric outcomes and behavior suggesting both anthropomorphic and cultural influences. Thus, ethnic Asian women should be counseled regarding their perinatal risk according to their specific ethnic backgrounds. In view of the scarce data on differences within Pacific Islanders in the obstetric literature, counseling of these women is difficult. Apart from the studies of Rao and Hayes and collaborators, perinatal outcome data of specific Pacific Islander populations have not been reported so far (PUBMED search; search terms: Pacific Islander, Pacific, Samoa, Tonga, Guam, Polynesia, delivery, perinatal, neonatal, morbidity, mortality; access date: 06-19-2014). Therefore, we attempted to characterize rates and outcomes of neonates with macrosomia and LBW in a defined Pacific Islander population. We collected data in a cohort of indigenous women delivering neonates with macrosomia ([4500 g) and low birth weight (LBW; \2500 g) in a single institution on the island of Samoa and compared these data to a set of Caucasian controls from Germany. This study was designed to provide clinical data in order to better counsel native and expatriate women with a Pacific Islander ethnic background.
Materials and methods We performed a case–control study of women who delivered a macrosomic neonate [4500 g as defined by Boulet et al. [9] or a LBW neonate \2500 g as defined by Alexander et al. [10] at the Department of Obstetrics, TTM National Hospital, Apia, Samoa and Department of
123
Obstetrics, Marienhospital Herne, Germany. Data in Samoa were collected by one of the authors (BS) during a research visit at the TTM National Hospital from 1/2007 to 12/2007. Perinatal morbidity and mortality data were collected at the Departments of Obstetrics and Gynecology and Neonatology, TTM National Hospital, Apia, Samoa. In a case–control approach, data from this Pacific Islander population (PIP) were compared to a German Caucasian population (CP) with neonates [4500 g or \2500 g delivered in the Marienhospital Herne, Germany between 1/2005 and 12/2010. We applied no matching criteria. Patient characteristics were extracted from patient charts. Approval for this study was obtained by the Ethics Committee of the Ruhr University Bochum, Bochum, Germany (Ethics Committee of the Ruhr University Bochum, Bochum, Germany, number 4540-12, issue date December 18, 2012). Values are given as median or mean values where appropriate. Comparisons between groups were made by chi-square test in case of categorical variables or by Mann– Whitney U test in case of continuous variables. p values \0.05 were considered statistically significant. We tested the influence of maternal age, parity, gravidity, twin delivery, maternal body mass index, and neonatal sex on composite severe neonatal morbidity (CNM) [11], i.e., respiratory distress syndrome, intracranial hemorrhage, sepsis, necrotizing enterocolitis, Erb’s palsy (brachial plexus injury), and severe hypoglycemia (defined as blood glucose \30 mg/dL [12]) in a multiple linear regression model. Some mothers prematurely took their babies out of the NICU and left the hospital for reasons not accessible to the authors (n = 10). In these cases, the neonatal status at the time of discharge was used for analysis. We used the statistical software SPSS 22.0 (SPSS Inc., Chicago, IL) for statistical analyses.
Results During the study period, 3166 and 2101 deliveries [24 weeks gestation and/or with a birth weight [500 g were identified in the PIP and CP, respectively. Rates of macrosomia were significantly higher and rates of LBW were significantly lower in the PIP compared to the CP [76/ 3166 (2.4 %) vs. 21/2101 (0.9 %); p \ 0.0001 and 149/3166 (4.7 %) vs. 163/2101 (7.7 %); p \ 0.0001, respectively]. Patient characteristics of the study populations are shown in Tables 1 and 2. The proportion of neonates transferred to the Neonatal Intensive Care Unit was significantly higher in the PIP for macrosomic [50/76 (65 %) vs. 0/21 (0 %), p \ 0.0001], but not for LBW neonates [106/149 (71 %) vs. 110/163 (67 %), p = 0.5]. Neonatal morbidity is described in detail in Table 3.
Arch Gynecol Obstet Table 1 Characteristics of women with neonates with macrosomia [4500 g from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany PIP
CP
p value
Total
76
21
–
Birth weighta (g)
4752 ± 314
4674 ± 110
0.07
Week of gestationa Agea (years)
40.1 ± 0.6 32.4 ± 5.9
40.4 ± 0.9 30.3 ± 4.8
0.1 0.09 0.001
Body Mass Indexa
39.5 ± 5.2
33.7 ± 5.9
Parity
3.3 ± 1.9
2.3 ± 1.1
0.008
Graviditya
4.2 ± 1.7
2.3 ± 1.1
\0.0001
Vaginal delivery
50/76
11/21
0.3
Cesarean section
26/76
10/21
Female
30/76
4/21
Male
46/76
17/21
7/76
1/21
Delivery mode
Fetal sex
GDM 0
0.1 1
5 APGAR \7
2/76
0/21
1
Transfer to NICU
50/76
0/21
\0.0001
Days in NICUa
2.4 ± 2.9
0
\0.0001
BSLa
114.0 ± 29.2
94.4 ± 16.1
0.01
PIP Pacific Islander population, CP Caucasian population, NICU neonatal intensive care unit, BSL blood sugar level a
Mean (±standard deviation)
Among macrosomic neonates, perinatal mortality and CNM, i.e., respiratory distress syndrome, intracranial hemorrhage, sepsis, necrotizing enterocolitis, Erb’s palsy Table 2 Characteristics of women with neonates with low birth weight \2500 g from a Pacific Islander population in Apia, Samoa, and a German Caucasian Population in Herne, Germany
(brachial plexus injury), and severe hypoglycemia, were higher in the PIP compared to the CP [2/76 (3 %) vs. 0/21 (0 %) and 6/76 (7 %) vs. 1/21 (4 %), respectively], but these differences were not statistically significant (p = 1.0 and p = 1.0, respectively). Among LBW neonates, mortality, but not CNM was significantly higher in the PIP compared to the CP [16/149 (7 %) vs. 2/163 (1 %), p \ 0.0001 and 10/149 (6 %) vs. 5/163 (3 %), p = 0.2, respectively]. In a multivariable logistic regression model, maternal age, body mass index, delivery mode, and gravidity did not independently predict CNM in the PIP and CP (Table 4). Screening for gestational diabetes mellitus (GDM) was performed in all 21 CP women and in 32/76 PIP women with a macrosomic fetus. A diagnosis of GDM was established in 1/21 and 7/76 women, respectively. Of note, this difference was notably higher despite the fact that fewer women in the PIP were screened for GDM. Mean serum blood sugar levels at the day of delivery were 114.0 ± 29.2 and 94.4 ± 16.1 mg/dL (p = 0.01) in PIP and CP women, respectively (Table 1). Mode of delivery was not significantly different between the PIP and CP. Specifically, PIP women with macrosomic neonates delivered vaginally after spontaneous labor in 50/76 (65 %) cases. Primary and secondary cesarean sections were performed in 17 and 9 cases, respectively. In comparison, mode of delivery among the 21 CP women with macrosomic neonates was vaginal delivery after spontaneous labor and labor induction in 7 and 4 cases, respectively, [11/21, (52 %)]. Primary cesarean section and secondary cesarean section
PIP
CP
p value
Total (women/neonates)a
135/149
150/163
–
Birth weight (g)b
2117.0 ± 432.5
2151.4 ± 350.6
0.4
Week of gestationb
37.6 ± 4.1
34.9 ± 2.3
\0.0001
Age (years)b
26.5 ± 7.1
27.9 ± 5.9
0.07
Body Mass Indexb
30.7 ± 4.7
27.8 ± 5.6
\0.0001
Parityb
1.6 ± 2.1
1.7 ± 1.0
0.5
2.6 ± 2.1
2.1 ± 1.3
0.01
Vaginal spontaneous
104
41
\0.0001
Cesarean section
45
122
Female
77
81
Male
72
82
50 APGAR \7 Transfer to NICU
4/149 106/149
3/163 110/163
0.3 0.1
Days in the NICUb
8.3 ± 13.1
11.8 ± 23.3
0.2
Gravidity
a
Delivery mode
Fetal sex 0.7
PIP Pacific Islander population, CP Caucasian population, NICU neonatal intensive care unit a
Number of women and neonates differ due to twin deliveries
b
Median (±standard deviation)
123
Arch Gynecol Obstet Table 3 Composite severe neonatal morbidity in macrosomic and low birth weight neonates from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany
Macrosomic neonates PIP (n = 76)
LBW neonates PIP (n = 149)
Macrosomic neonates CP (n = 21)
LBW neonates CP (n = 163)
Total CNM
7
19
1
6
RDS
0
4
0
2
ICH
0
0
0
1
Sepsis
3
4
0
0
NEC
0
0
0
0
Erb’s palsy
1
0
0
0
SH
3
2
1
3
Other
0
9
0
0
PIP Pacific Islander population, CP Caucasian population, LBW low birth weight, CNM composite neonatal morbidity, RDS respiratory distress syndrome, ICH intracranial hemorrhage, NEC necrotizing enterocolitis, SH severe hypoglycemia; Neonates may have [1 CNM diagnoses
Table 4 Multivariable logistic regression analysis of composite severe neonatal morbidity in women from a Pacific Islander population in Apia, Samoa, and a German Caucasian population in Herne, Germany Parametera
PIP multivariable logistic regression OR (95 % CI)
p value
CP multivariable logistic regression OR (95 % CI)
p value
Age (B35 vs. [35 years)
1.2 (0.1–13.2)
0.9
0.5 (0.1–4.2)
0.5
Delivery mode (vaginal vs. CS)
0.1 (0.6–81.0)
0.1
1.1 (0.3–4.1)
0.9
Maternal BMI (\32 vs. C32) Gravidity (0 vs. [1)
0.3 (0.0–4.7) 6.9 (0.2–55.5)
0.3 0.4
2.4 (0.7–8.6) 0.9 (0.3–3.1)
0.2 0.9
PIP Pacific Islander population, CP Caucasian population, OR odds ratio, CI confidence interval, CS Cesarean section, BMI Body Mass Index a
Parameters in the multivariable analysis
were performed in 5 and 5 cases, respectively, in CP women. Delivery complications, especially severe vaginal/perineal lacerations, were noted less often in PIP women [10/76 (13 %)] (intrauterine asphyxia: n = 2, postpartum hemorrhage: n = 3, shoulder dystocia: n = 1; severe vaginal/perineal laceration: n = 4) compared to 10/21 (62 %) women in the CP (intrauterine asphyxia: n = 0, postpartum hemorrhage: n = 1, shoulder dystocia: n = 2; severe vaginal/perineal laceration: n = 7). This difference was statistically significant (p = 0.001). PIP women with LBW neonates had a vaginal delivery more often than CP women [104/149 (69 %) vs. 41/163 (25 %), p \ 0.0001]. Specifically, mode of delivery among the 149 PIP women with LBW neonates was vaginal delivery after spontaneous labor in 103 cases and vaginal delivery after labor induction in 1 case [104/149 (69 %)]. Primary cesarean section and secondary cesarean section were performed in 30 and 15 cases, respectively. In contrast, 122 women with LBW neonates in the CP were delivered by cesarean section [122/163 (75 %); primary cesarean section: n = 80; secondary cesarean section: n = 42]. Vaginal delivery after spontaneous labor occurred in 25, vaginal delivery after labor induction in 16 cases. Delivery complications were noted in 13/149 (9 %)
123
deliveries in the PIP (intrauterine asphyxia: n = 5, postpartum hemorrhage: n = 7, shoulder dystocia: n = 0; severe vaginal/perineal laceration: n = 1) and in 41/150 (27 %) women in the CP (intrauterine asphyxia: n = 2, postpartum hemorrhage: n = 6, shoulder dystocia: n = 0; severe vaginal/perineal laceration: n = 33).
Discussion In this case–control study of Pacific Islander women from Samoa and Caucasian women from Germany with macrosomic and LBW neonates, Pacific Islander women had higher rates of macrosomic neonates but lower rates of LBW neonates, suggesting an anthropomorphic basis of this phenomenon. PIP neonates with both macrosomia and LBW had comparatively poor perinatal outcomes with increased morbidity and mortality, potentially attributable to the poor local resource setting. Epidemiologic studies in Pacific Islanders have shown high rates of obesity and diabetes. For example, in a survey of 15 countries in the Western Pacific and South-east Asia including data from 330,374 individuals, Martiniuk and colleagues found the highest population attributable fraction of elevated BMI for diabetes mortality in Samoa with
Arch Gynecol Obstet
98 % in males and 95 % in females [13]. In line with these data, a cohort study of five ethnic groups (South-east Asian, South Asian, Middle Eastern, Anglo-European and Pacific Islander) found that women from the Pacific Islands had the highest parity, BMI, fasting glucose levels, HbA(1c), and the greatest need for insulin therapy [14]. Elevated BMI and preexisting diabetes are well-known risk factors of fetal macrosomia [15]. In accordance with these epidemiologic data, it has been previously demonstrated that expatriate women from Pacific Islands have higher rates of macrosomia compared to other Asian women [4]. We confirm these data demonstrating that both high rates of macrosomia and low rates of LBW infants are obstetric phenomena, which have to be expected both in indigenous and expatriate women from this population. Also, PIP women had higher rates of blood sugar levels and GDM. To our best knowledge, this is the first study assessing the frequency and perinatal outcomes associated with macrosomia and LBW in an indigenous PIP. Our data suggest that anthropomorphic factors, cultural factors such as nutritional habits, or both, have important clinical consequences for obstetricians caring for these women during pregnancy and delivery. On a political level, nutritional education programs, GDM screening, and shoulder dystocia trainings for midwives and obstetricians may be important and efficacious interventions in Samoa and other Pacific Islands. For obstetricians caring for expatriate Pacific Islander women, it may be helpful to note that this specific ethnic subgroup has a substantially elevated risk of macrosomia and associated delivery complications. In addition, macrosomic and LBW neonates from PIP mothers may be more vulnerable to subpartum stress compared to Caucasian controls, although the poor outcomes observed in our study may be attributable to both anthropomorphic effects as well as the level of obstetric and NICU sophistication present in Samoa. Our study has limitations. For example, we did not record sociodemographic variables such as level of education, income, socioeconomic status, availability to health care ressources, and marital status. Therefore, we cannot quantify the effect of between these factors on neonatal morbidity and mortality as described in this study. In a developing country such as Samoa, however, such information is difficult to acquire. Due to this methodological limitation, we were not able to assess the influence of socioeconomic factors on obstetric outcome parameters. In addition, data on long-term neonatal outcomes are missing in our dataset due to the lack of systematic pediatric care in Samoa. Thus, the differences regarding neonatal morbidity and mortality may even be higher than those observed in our analysis. Also, the absolute numbers of the primary outcome, i.e., CNM, are low. Therefore, it is possible that
some of the specific CNM outcomes may be over- or underrepresented in our analysis. In summary, Samoan indigenous women have higher rates of macrosomia, but lower rates of LBW, compared to Caucasian controls, suggesting an anthropomorphic basis of this phenomenon. Neonates with macrosomia and LBW have poor perinatal outcomes in this population, potentially attributable to limited local health resources. These data may be helpful for health officials and obstetricians in Samoa and other PIP as well as for obstetricians caring for expatriate Samoan women. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Edmonds JK, Yehezkel R, Liao X et al (2013) Racial and ethnic differences in primary, unscheduled cesarean deliveries among low-risk primiparous women at an academic medical center: a retrospective cohort study. BMC Pregnancy Childbirth 13:168 2. Racape J, De Spiegelaere M, Dramaix M et al (2013) Effect of adopting host-country nationality on perinatal mortality rates and causes among immigrants in Brussels. Eur J Obstet Gynecol Reprod Biol 168(2):145–150 3. Cabacungan ET, Ngui EM, McGinley EL et al (2012) Racial/ ethnic disparities in maternal morbidities: a statewide study of labor and delivery hospitalizations in Wisconsin. Matern Child Health J 16(7):1455–1467 4. Rao AK, Daniels K, El-Sayed YY et al (2006) Perinatal outcomes among Asian American and Pacific Islander women. Am J Obstet Gynecol 195(3):834–838 5. Yu SM, Alexander GR, Schwalberg R et al (2001) Prenatal care use among selected Asian American groups. Am J Public Health 91:1865–1868 6. Wong LF, Caughey AB, Nakagawa S et al (2008) Perinatal outcomes among different Asian-American subgroups. Am J Obstet Gynecol 199(4):382 7. Ekeroma AJ, Craig ED, Stewart AW, Mantell CD, Mitchell EA (2004) Ethnicity and birth outcome: New Zealand trends 1980–2001: Part 3. Pregnancy outcomes for Pacific women. Aust N Z J Obstet Gynecol 44(6):541–544 8. Hayes DK, Mitchell KM, Donohoe-Mather C et al (2014) Predictors of exclusive breastfeeding at least 8 weeks among Asian and Native Hawaiian or other Pacific Islander race subgroups in Hawaii, 2004–2008. Matern Child Health J 18(5):1215–1223 9. Boulet SL, Alexander GR, Salihu HM, Pass M (2003) Macrosomic births in the united states: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol 188(5):1372–1378 10. Alexander GR, Kogan M, Bader D, Carlo W, Allen M, Mor J (2003) US birth weight/gestational age-specific neonatal mortality: 1995–1997 rates for whites, hispanics, and blacks. Pediatrics 111(1):61–66 11. Witt A, Do¨ner M, Petricevic L, Berger A, Germann P, Heinze G, Tempfer C (2011) Antibiotic prophylaxis before surgery vs after cord clamping in elective cesarean delivery: a double-blind, prospective, randomized, placebo-controlled trial. Arch Surg 146(12):1404–1409 12. Rozance PJ, Hay WW Jr (2010) Describing hypoglycemia—definition or operational threshold? Early Hum Dev 86(5):275–280
123
Arch Gynecol Obstet 13. Martiniuk AL, Lee CM, Colagiuri S et al (2011) Higher-thanoptimal body mass index and diabetes mortality in the Asia Pacific region. Diabetes Res Clin Pract 94(3):471–476 14. Wong VW (2012) Gestational diabetes mellitus in five ethnic groups: a comparison of their clinical characteristics. Diabet Med 29(3):366–371
123
15. Alberico S, Montico M, Barresi V et al (2014) The role of gestational diabetes, pre-pregnancy body mass index and gestational weight gain on the risk of newborn macrosomia: results from a prospective multicentre study. BMC Pregnancy Childbirth 14:23
