Abstract

Objective: To describe the placental characteristics and neonatal outcomes of Central Australian Aboriginal women based on maternal self-report of tobacco use.

Methods: Placental and neonatal variables were collected from a prospective maternal cohort of 19 smokeless tobacco chewers, 23 smokers and 31 no-tobacco users.

Results: Chewers had the lowest placental weight (460 g) while the no-tobacco group had the heaviest placental weight (565 g). Chewers and the no-tobacco group had placental areas of similar size (285 cm2 and 288 cm2, respectively) while the placentas of smokers were at least 13 cm2 smaller (272 cm2). There were two stillbirths in the study and more than one-third (36%) of neonates (newborns) were admitted to the Special Care Nursery, with the chewers’ neonates having a higher admission rate compared with smokers’ neonates (44% vs. 23%). The cohort mean birthweight (3348 g) was not significantly different between the groups. When stratified for elevated maternal glucose, the chewers’ neonates had the lowest mean birthweight (2906 g) compared to the neonates of the no-tobacco group (3242 g) and smokers (3398 g).

Conclusions: This research is the first to demonstrate that the maternal use of Australian Nicotiana spp. (pituri) as smokeless tobacco may negatively impact placental and neonatal outcomes.

Implications for public health: Maternal smokeless tobacco use is a potential source of placental and foetal nicotine exposure. Maternal antenatal screening should be expanded to capture a broader range of tobacco and nicotine products, and appropriate cessation support is required.

Maternal tobacco smoking and maternal exposure to combusted tobacco vapour are associated with adverse maternal, placental, foetal and neonatal outcomes, with sequelae through childhood and adolescence1-10 and into adulthood.11-16 Reducing both the maternal use of smoked tobacco and exposure to the vapour from combusted tobacco are recognised as the most significant foetal and neonatal risk-reducing behaviours that can occur during pregnancy.17, 18

Tobacco smoking is commonplace in Westernised populations, however, in low- and middle-income countries and in Indigenous populations, the use of smokeless tobacco (ST) is often more common.19 ST is a term that describes tobacco products that are not combusted and are instead chewed, sucked or applied to the gums or nasal lining as powders and pastes. Products include chewing tobacco, snuff, chimo and khaini, and there are an estimated 356 million adult users in 140 countries (one in 10 males and 1 in 20 females).19 This prevalence is likely an underestimate due to non-reporting of ST by more than 58 World Health Organization member states (including Australia and New Zealand), nevertheless, an estimated 90 million women >15 years of age use ST globally.20

Note to readers: In this research, the Central Australian Aboriginal women chose the term ‘Aboriginal’ to refer to themselves, and ‘Indigenous’ to refer to the broader group of Australian First Peoples. That choice has been maintained in the reporting of the research findings.

In Australia, Central Australian Aboriginal populations in the Northern Territory (NT) and adjacent state areas utilise wild tobacco plants (Nicotiana spp.) called pituri, as ST.21 Portions of the dried plant are chewed to a pulp and placed in the lip, cheek and buccal space for extensive periods. If the quid of tobacco is removed from the oral space, it is stored on skin sites, for example under a headband, armband or behind the ear, possibly providing transdermal nicotine administration.22 Female pituri use commences in early life and does not cease during pregnancy and lactation.23 There has been no general health research related to pituri use, or any specific research examining the effects of pituri use on placental or neonatal outcomes.

Tobacco and nicotine: placental and foetal impacts

The question as to whether the maternal use of ST could affect placental and neonatal outcomes is premised on the pharmacological knowledge of nicotine. Nicotine is a potent vaso-constrictor and is the primary active and dose-dependent lethal component of tobacco.22 Nicotine binds with and activates nicotinic acetylcholine receptors (nAChRs) in central and peripheral neuronal and non-neuronal tissue.24-27 Receptor type and individual variability, including genetics and pregnancy, result in receptor up-regulation or desensitization in a biphasic manner.22, 28

In the placenta, nicotine effects are thought to be due to mediation of the placental nAChRs, adversely affecting placental vessels and inhibiting trophoblast interstitial migration, invasion and differentiation, which impacts the establishment and development of the placenta.29, 30 An observational study in Pakistan compared 40 placentas of ST users to those of 40 non-ST users and demonstrated no difference in placental weights and the absence of gross morphological differences. However, at the micro-morphology level, there were significant abnormalities in the placental structures of ST users that would impact perfusion at the placental-uterine interface including: doubling of villi with excessive collagen (p<0.001), doubling of sub-trophoblastic basement membrane (p<0.001), doubling of syncytial buds (p<0.001), and tripling of apoptotic cells (i.e. programmed cell deaths; p<0.001).31

The placenta is critical to the survival and growth of the neonate (newborn), consequently, nicotine-induced changes to the placenta and its functionality flow on to changes in neonatal outcomes including lower birthweight, intrauterine growth retardation (which results in a small for gestational age [SGA] neonate), prematurity and admission to Special Care Nursery (SCN).32 Birthweight is one of the sentinel predictors for neonatal mortality and morbidity,33 and maternal cigarette exposure research shows a dose-dependent neonatal birthweight reduction of 320–435 grams (g),34 while ST research shows birthweight reduction of between 100 g and 395 g.35-39 In addition, there is a dose-dependent increased risk for preterm birth (<37 weeks gestation) and decreased length of gestation associated with both smoking and ST exposure.36, 38-44

Foetal nicotine exposure also results in a dampened response to intrauterine and post-birth hypoxic episodes.45-47 Research in low- and middle-income countries shows a stillbirth rate of 27–89/1000 births in ST chewers compared to 6–31/1000 in non-ST chewers.36, 48, 49 Systematic reviews of studies in high-income countries show a 36–46% increase in the likelihood of stillbirth in the presence of maternal smoking.50, 51

Nicotine: glucose homeostasis impacts

In addition to the foetal impact of nicotine exposure at CNS nAChRs, nicotine increases maternal blood glucose levels via the stimulation of the pancreas.52 Insulin resistance is a normal aspect of pregnancy;53 however, smoking in pregnancy increases the risk (AOR 1.9) for gestational diabetes.54 At the foetal level, rat studies13, 55 show that the pancreas is vulnerable to nicotine exposure both during foetal life and during lactation, resulting in a loss of beta-cell mass and impaired glucose homeostasis. This permanent beta-cell damage triggers a cascade of physiological responses including glucose and insulin intolerance, hyperinsulinaemia and increased body weight.56, 57

For Aboriginal neonates, exposure to elevated maternal glucose levels increases birthweight (mean increase 280 g), while for Caucasian neonates, exposure decreases birthweight (mean decrease 43 g).58 Both Aboriginal and Caucasian neonates exposed to elevated glucose have lower APGAR scores at 5 minutes (8.78 and 9.08, respectively), compared to non-exposed neonates (9.09 and 9.15, respectively).58 Gestational age (both preterm and post-term), birthweight (both low birthweight <2500 g and higher birthweight > 4000 g59) and elevated glucose independently increase the likelihood of caesarean section (CS) birth,60, 61 which in turn increase the likelihood of admission to SCN.62 A well neonate is expected to have an APGAR score at 5 minutes ≥7 and is not expected to be admitted to SCN. A neonatal APGAR score at 5 minutes <7 indicates the neonate is taking longer to transition to extra-uterine life and signals the likelihood of admission to SCN.62 Admission to SCN physically distances the mother from her neonate and extends the neonate's length of stay in hospital with resultant social, cultural, clinical and economic implications.

Objective

In Central Australia, the maternal use of Nicotiana spp. exposes the placenta and neonate to nicotine63 and this exposure has not previously been examined. This paper presents the placental and neonatal findings from a larger descriptive research study, the complete protocol of which is published elsewhere.64 The purpose of this paper is to address the primary research question: What are the placental characteristics and neonatal outcomes of Aboriginal women who self-report no-tobacco use, chewing pituri or smoking during pregnancy? Pregnancy, labour and birth outcomes65 suggest an association between chewing pituri and the development of elevated maternal glucose levels. Elevated maternal glucose adversely affects gestational age and birthweight and increases the likelihood of admission to SCN and of a CS birth.62, 66 Consequently, a secondary research question was added: Is there an association between elevated maternal glucose and neonatal outcomes?

Methods

The research questions were addressed through a cohort study of prospectively enrolled pregnant Central Australian Aboriginal women who planned to birth at the Alice Springs Hospital, NT, Australia. The design, methodology and protocols were informed, directed and approved by a regional Aboriginal Women's Council. Ethical approval was obtained from the Central Australian (#2010.06.04) and The University of Queensland Human Research Ethics Committees (#2010000548 and # 2015001429). All participants provided written informed consent for themselves and their neonate prior to enrolling in the study, and all methods were conducted in accordance with the guidelines and regulations of the National Statement for the Ethical Conduct in Human Research.67

Sampling frame, sample size and exclusion criteria: The extent of pituri chewing by pregnant Central Australian Aboriginal women had not been previously established. Observation suggested approximately 33% did not smoke or use pituri, 33% of women used pituri, and 33% of women smoked (R. Carroll, personal communication, 16 July 2009). This smoking estimate was supported by the NT Mothers and Babies Report,68 which indicates a self-reported rate of 30% for Indigenous mothers in the Alice Springs District compared with an NT-wide rate of 52% in the first 20 weeks of pregnancy. The lower smoking rates in Central Australia are surmised to be a result of the ‘local practice of chewing tobacco (pituri) in that region’.69, 70

Detecting statistically significant adverse placental and neonatal outcomes in association with maternal tobacco and nicotine exposure requires an extensive sample size, which was not feasible in this study due to the relatively small Central Australian Aboriginal maternal population. Based on the observed rates of no-tobacco use, pituri use, and smoking, a sample of 20 participants in each group was considered sufficient to provide preliminary descriptive placental and neonatal data to inform further studies. Known maternal predictors of adverse placental and neonatal outcomes were not exclusion criteria for this study. These predictors are endemic in the target population and include: limited access to health and antenatal care, unemployment and poverty, poor nutrition, cardiac and renal disease, anaemia and hypertension.70-72 Exclusion due to the presence of these predictors would have significantly reduced the eligible population sample. The only maternal exclusion criterion was self-reported dual pituri and cigarette use as the study was not resourced to distinguish between placental and neonatal effects arising from tobacco and nicotine absorbed through the maternal respiratory tract, and those arising from the maternal oral and transdermal routes.73

Population and recruitment: This study population is the singleton neonates ≥28 weeks gestation and their placentas born from a maternal sample of conveniently recruited Central Australian Aboriginal women who planned to birth at the Alice Springs Hospital, who were ≥18 years of age at the time of enrolment, and who consented to enrol themselves and their neonate in the research.

Data collection: Informed by the neonatal and placental literature74-80 and by the pregnancy, labour and birth outcomes,65 maternal, neonatal and placental variables were included in the data collection. The maternal variables were: age at enrolment (years and months); parity (number of births after 20 weeks or of at least 400 g); elevated glucose – any pre-pregnancy or gestational report of diabetes (yes/no); CS (yes/no); and tobacco use (yes/no and type if yes). The placental variables were: weight (g), and area (cm2), and the neonatal outcomes were: viability (liveborn/stillborn); gender (male/female); gestational age (weeks and days); birthweight (g); head circumference in centimetres (cm); neonatal length (cm); APGAR score (</≥ 7); and admission to SCN (yes/no).

Accordingly, three data collection strategies were used. The first was a maternal interview on enrolment conducted by Aboriginal Health Workers (AHWs), Aboriginal Liaison Officers (ALOs) and midwives to capture self-report of tobacco use and variables such as maternal education, which are not routinely collected. Secondly, routinely collected data were drawn from the maternal and neonatal demographic and birth record contained in the NT Perinatal Data Report housed in CARESYS® (the NT electronic medical record system). Thirdly, placental weight and area were measured by midwives.

Data analysis: All analyses were completed using SPSS® and Stata 15 (Statacorp, Texas). The maternal self-report of tobacco use at interview was used to categorise the neonates into maternal tobacco exposure groups: a) no-tobacco exposure; b) pituri exposure; or c) cigarette smoking exposure. Missing data were examined for patterns and outliers identified through standard processes,81 and both were reported but excluded from the analysis. Descriptive statistics with means and 95% confidence intervals (CI) if normally distributed, or medians with ranges for continuous data, or frequency and proportions for categorical data were reported. Maternal self-reported tobacco use was used as the independent variable for comparative analyses with statistical significance reported at the 0.05 alpha level.

Results Maternal and birthing characteristics

The maternal and birthing characteristics are reported in full elsewhere,65 in summary, the data in Table 1 show that of the 73 maternal participants, 31 mothers reported no-tobacco use, 19 reported pituri use and 23 reported cigarette use. The median maternal age of the cohort on enrolment was 24 years with no significant differences between the groups. The pituri chewing group experienced less than half the rate of CS (21%) compared with the smoking (48%) and slightly less than half of the no-tobacco group (39%). The rate of elevated glucose was higher in the pituri users (n = 9, 47%) compared with the smoking group (n = 5, 22%) and no-tobacco group (n = 5, 16%). The likelihood of the pituri chewing group having elevated glucose levels was four times that of no-tobacco use (OR: 4.68, 95%CI 1.26–17.42).

Table 1. Maternal and birth characteristics by maternal self-reported tobacco use (N=73).

Variable

Maternal self-reported tobacco use

Total

N=73

No-tobacco use

n=31

Chewer

n=19

Smoker

n=23

Maternal age in years

Median (range)

24 (18–38)

22 (18–37)

26 (18–34)

23 (18–38)

Maternal age categorised, n (col %, 95%CI)

< 20

15 (21, 13–32)

11 (35, 20–54)

1 (5, 1–31)

3 (13, 4–35)

20–29

41 (56, 44–67)

15 (48, 31–66)

10 (53, 30–74)

16 (70, 48–85)

30–39

17 (23, 15–35)

5 (16, 7–34)

8 (42, 22–65)

4 (17, 6–39)

Elective caesarean section

10 (14, 7–24)

2 (6, 2–23)

3 (16, 5–40)

5 (22, 9–44)

Emergency caesarean section

17 (23, 15–35)

10 (32, 18–51)

1 (5, 1–31)

6 (26, 12–48)

Elevated glucosea

Yes

19 (26, 17–38)

5 (16, 7–34)

9 (47, 26–70)

5 (22, 9–44)

No

54 (74, 62–83)

26 (84, 66–93)

10 (53, 30–74)

18 (78, 56–91)

Odds ratio (95%CI) of elevated glucose

Reference group

4.68 (1.26–17.42)

1.44 (0.36–5.73)

Notes: a: CARESYS® data did not specify if condition was pre-existing or gestational. Placental characteristics

Placental weight and size data were not available for 30 (41%) births. There were no reports of antepartum haemorrhage, and one chewer was reported to have a retained placenta. Table 2 shows that the mean placental weight was 598 g (95%CI 502–694). The smoker group had two very large placentas (1740 g and 1960 g) and one very small placenta (90 g) that met the outlier criteria and were removed from the analysis. The median placental weight for pituri chewers was 460 g, which was 40 g below that of the smokers and 105 g below that of the no-tobacco use group. Comparison of the placental area data shows the cohort median placental area was 288 cm2, with the placentas of no-tobacco use group slightly larger in area (288 cm2) than those of pituri users (285 cm2), and the placentas of smokers (272 cm2) were approximately 16 cm2 smaller than those of the no-tobacco use group.

Table 2. Placental characteristics by maternal self–reported tobacco use (N=43).

Totala

N=43

Maternal self–reported tobacco use

No–tobacco

n=22

Chewer

n=11

Smoker

n=10

Placental weight (g)

n=43 mean (95% CI)

598 (502–694)

574 (528–620)

485 (385–585)

774 (349–1199)

n=40 mean (95% CI)b

548 (509–587)

574 (528–620)

485 (385–585)

564 (466–663)

n=40 median (range)b

555 (280–800)

565 (380–800)

460 (280–750)

500 (460 –690)

Placental area (cm2)b

n=43 mean (95% CI)

286 (264–308)

301 (273–328)

279 (220–339)

259 (209–310)

n=40 mean (95% CI)b

290 (269–312)

301 (273–328)

279 (220–339)

276 (229–323)

n=40 median (range)b

288 (156–480)

288 (208–480)

285 (156–450)

272 (180–340)

Notes: a: Missing placental data n = 30 (no tobacco, n = 9 (30%); chewer, n = 8 (27%); smoker, n = 10 (43%)) b: Placenta of 3 smoker group participants not included in weight analysis and area as extreme outliers (90 g, 1740 g and 1960 g) Neonatal characteristics

The neonatal findings are shown in Table 3 and are summarised as follows:

Viability and gender. Of the 73 neonates born in the study, 71 were liveborn. Two foetal deaths in utero (stillbirths) were reported, both at 40 weeks gestation; a neonate from a pituri chewing mother and a neonate from a no-tobacco using mother. There were 40 male neonates (55%) and 33 female neonates (45%) born in the study with no gender difference apparent across the cohort, although when considered by the tobacco exposure group, differences in the male:female ratios were evident. The male:female ratio in the no-tobacco use group was 19:12, 61% male; the ratio shifted to a closer equivalence in the pituri chewing group (11:8, 57% male), and was reversed in the smoking group with the birth of fewer males than females (10:13, 43% male).

Table 3. Neonatal outcomes and characteristics by maternal self-reported tobacco use (N = 73).

Total

N=73

Maternal self–reported tobacco use

No–tobacco

n=31

Chewer

n=19

Smoker

n=1230

n (col %, 95% CI)

Viability

Liveborn

71 (97,89–99)

30 (97, 79–99)

18 (95, 69–99)

23 (100,–)

Still born

2 (3,1–11)

1 (3, 1–21)

1 (5, 1–31)

0 (0)

Neonate gender

Male

40 (55, 43–66)

19 (61, 43–77)

11 (58, 35–78)

10 (43, 25–64)

Female

33 (45, 34–57)

12 (39, 23–57)

8 (42, 22–65)

13 (57, 36–75)

Gestational age, weeks

< 28 weeks

1 (1, 0–9)

0 (0)

0 (0)

1 (4, 100)

28–36 weeks

6 (8, 4–17)

3 (10, 50)

2 (11, 33)

1 (4, 17)

37+ weeks

66 (90, 81–95)

28 (90, 42)

17 (89, 26)

21 (91, 32)

Mean (95 % CI)a

38.8 (38.3–39.2)

38.8 (38.2–39.5)

38.4 (37.6–39.1)

39.0 (38.1–40.0)

Birthweight

<1,500 g

2 (3, 1–11)

0 (0)

0 (0)

2 (9, 2–30)

1,500–2,499 g

6 (8, 4–17)

3 (10, 3–27)

3 (16, 5–40)

0 (0)

2,500–2,999 g

12 (16, 9–27)

5 (16, 7–34)

4 (21, 8–46)

3 (13, 4–35)

3,000–3,499 g

23 (32, 22–43)

12 (39, 23–57)

4 (21, 8–46)

7 (30, 15–52)

3,500–3,999 g

21 (29, 19–40)

9 (29, 16–48)

5 (26,11–51)

7 (30, 15–52)

4,000–4,499 g

6 (8, 4–17)

1 (3, 1–21)

3 (16, 5–40)

2 (9, 2–30)

> 4,500 g

3 (4, 1–12)

1 (3,1–21)

0 (0)

2 (9, 2–30)

Mean (95% CI)a

3,348 (3,189–3,507)

3,278 (3,065–3,491)

3,286 (2950–3,622)

3,500 (3,153–3,848)

Head circumference, cm

Mean (95% CI)

34.3 (33.8–34.8)

34.1 (33.2–34.9)

33.8 (32.9–34.7)

35.0 (34.0–35.9)

Missing,b n

7

3

3

1

Body length, cm

Mean (95% CI)

49.8 (49.0–50.6)

49.9 (48.5–51.2)

49.2 (47.7–50.7)

50.2 (48.9–51.6)