EFFECT OF MATERNAL SMOKING ON PLACENTAL WEIGHT AND PLACENTAL WEIGHT/BIRTHWEIGHT RATIO AMONG FULL-TERM SINGLETON: A BIRTH REGISTRY STUDY

Full Text

INTRODUCTION

Maternal smoking during pregnancy is one of the most common modifiable risk factors of adverse short- and long-term pregnancy outcomes such as miscarriages [1], preterm deliveries [2, 3], low birthweight (BW) [3], some birth defects  [2-4]. Some of these complications are related to placenta development and can be partially explained by disfunction of trophoblasts and hormonal disbalance in early pregnancy caused by smoking [1] and vascular effects of nicotine [5].  All women are advised to quit smoking before conception or during first trimester of pregnancy. It was found that the greatest benefits can be observed in case of cessation before week 15 of pregnancy [6].  It can improve the prognosis and reduce risk of the adverse outcomes. In spite that fact the prevalence of smoking among pregnant women in many countries including Russia remains relatively high: our previous study in the North-West of Russia reveals up to 25% of smoking mothers before and 18% during pregnancy respectively [7].

 Research on association between smoking and placenta’s morphology is scarce. Smoking cessation effects on placenta also remain unclear. Some studies showed significant difference in the placenta’s structure and weight between smoking and non-smoking mothers [8], higher placental weight/birthweight ratio (PW/BW) among smoking mothers [9, 10], but others consider that in spite of lower first-trimester placental vascularization flow indices, a negative impact of smoking on PW is not evident yet [11]. Another important issue is how smoking cessation during pregnancy affects placenta’s development. Some studies revealed positive impact of smoking cessation on anthropometric measurements in newborns in comparison with offspring of smoking mothers due to an improvement of placental transfer [3, 12].

The associations of smoking before and during pregnancy with high PW and PW/BW ratio have, to our knowledge, not been studied. Two population-based birth registries established in Murmansk County (MC) in the North-West of Russia in two decades ago contain information on maternal smoking behavior before and during pregnancy and data of placenta morphological examinations for all deliveries in these regions. It allows to examine potential associations between maternal smoking status during pregnancy (including effect of smoking cessation and daily number of smoked cigarettes) and PW as well as PW/BW. Thus, our specific aims are to: (i) establish PW and BW for gestational age and sex and (ii) examine a potential association between smoking status during pregnancy (including daily number of smoked cigarettes) and z-scores of PW and PW/BW.

MATERIAL AND METHODS

Study setting, Design and Sample size

The Murmansk County Birth Registry (MCBR) has registered all births from 22 weeks of gestation in MC from 1 January 2006 to 31 December 2011. Detailed information about its design and implementation and data collection has been provided previously [13].

For the study, we included cases from the registry with singleton pregnancy delivered at > 37 completed weeks of gestation.  Our study focused on two main tobacco-smoking issues related to pregnancy: (i) smoking status, and (ii) number of cigarettes smoked daily. Sampling details are summarized in Figure 1.

 

Figure 1. Study population selection procedure

Data collection

Based on medical records and personal interviews with pregnant women, the MCBR contains information on maternal characteristics, such as age, parity, weight and height measured at the first antenatal visit, marital status, residence, ethnicity, education, and self-reported smoking (number of cigarettes per day before and during pregnancy). Information in the MCBR on mode of delivery, gestational diabetes, preeclampsia/eclampsia, gestational age, year of delivery, and anemia was derived from individual obstetric records. Based on newborn delivery records, the MCBR also contains data about sex of baby, PW, and BW.

Dependent variables

PW and BW were measured in grams. PW/BW was calculated by dividing PW by BW in grams. We calculated z-scores in the sample by using means and standard deviation of PW and PW/BW for each gestational age. Moreover, z-scores were calculated separately for boys and girls. PW and PW/BW z-scores were classified as < -1, -1 to +1, and > +1.

Independent variables

Smoking status during pregnancy was defined as “smokers” (women who smoked before and during pregnancy), “quitter” (women who smoked before but not during pregnancy), and “non-smoker” (women neither smoked before or during pregnancy). Number of cigarettes smoked per day during pregnancy was treated as a categorical variable, specifically as 0, 1–5, 6–10, and ≥11.

Data analysis

Distribution of continuous variables were checked for normality with Shapiro-Wilk test and presented as mean and standard deviation. Simple linear regression was used to estimate the relationship between PW, BW and gestational age for boys and girls separately. ANCOVA was conducted with gestational age as a covariate and examined difference in PW between maternal smoking categories. By multinomial logistic regression, we examined the associations between z-scores (low, medium, and high) of PW and PW/BW ratio and smoking status during pregnancy, including daily number of smoked cigarettes. Medium z-scores were chosen as the base outcome; therefore, it was used as the reference category in constructing the binary logistic regression. Crude and adjusted relative risk ratios (RRR) were calculated with 95% confidence intervals (CI). Adjustment was done for maternal age (≤19 years, 20–24 years, 25–29 years, 30–34 years, and ≥35 years), parity (0, 1, and ≥ 2 deliveries), marital status (married, cohabitation or single (includes divorced or widowed)), residence (urban and rural), ethnicity (Russian or other), education (university and other), year of delivery, body mass index (≤18.4, 18.5-24.9, 25.0-29.9, ≥30 and unknown), mode of delivery (vaginal and caesarean section), pregnancy diabetes, preeclampsia/eclampsia, anemia, and birthweight All statistical analyses were conducted using STATA version 14 (StataCorp LLC).

RESULTS

Placenta weight and birthweight for gestational age and sex

The mean PW for boys was 534.1 g (SD 117.9 g) and for girls was 523.7 g (SD 116.6 g). The mean BW for boys was 3500 g (SD 457.1 g) and for girls was 3366 g (SD 439.0 g) (Table 1).

 

Table 1. Mean values of placental weight and birthweight for gestational age and sex

 

Gestational age	Boys, Mean (SD)			Girls, Mean (SD)
	N	Placental weight	Birthweight	N	Placental weight	Birthweight
37 weeks	1555	508.9 (123.2)	3148 (467)	1275	498.9 (117.8)	3023 (423)
38 weeks	3788	524.2 (119.1)	3352 (433)	3236	511.8 (117.9)	3205 (430)
39 weeks	6765	534.9 (118.2)	3493 (432)	6133	523.1 (115.8)	3354 (418)
40 weeks	6854	540.6 (117.2)	3594 (428)	6458	528.8 (115.4)	3440 (408)
41 weeks	2794	541.8 (112.3)	3633 (448)	3024	532.3 (116.2)	3500 (425)
≥ 42 weeks	1244	539.2 (116.6)	3612 (474)	1285	536.0 (116.7)	3483 (455)
Total	23,000	534.1 (117.9)	3500 (457)	21,411	523.7 (116.6)	3366 (439)

 

Moreover, mean PW and BW increased with gestational age. Results of linear regression were as follows - PW_boys: B=6.25 (95%CI 5.0-7.5) and PW_girls: B=7.05 (95%CI 5.8-8.3); BW_boys: B=97.5 (95%CI 92.9-102.1) and BW_girls: B=94.3 (95%CI 89.8-98.9). Difference in PW in groups according smoking status stratified with stratification by gestational age and sex is presented in the table 2.

 

Table 2. Mean values of placental weight for different smoking status with stratification by age and sex

 

Gestational age	Placental weight (Boys), Mean (SD)			Placental weight (Girls), Mean (SD)
	Non-smoker	Quitter	Smoker	Non-smoker	Quitter	Smoker
37 w	507 (122)	520 (134)	513 (122)	501 (121)	505 (122)	491 (106)
38 w	524 (120)	531 (110)	522 (118)	511 (115)	516 (128)	514 (125)
39 w	532 (117)	546 (117)	542 (124)	521 (116)	529 (115)	528 (116)
40 w	538 (115)	544 (124)	551 (122)	526 (114)	535 (111)	540 (121)
41 w	540 (112)	543 (112)	547 (113)	531 (114)	536 (125)	536 (120)
≥ 42 w	540 (116)	511 (110)	542 (119)	537 (120)	553 (107)	527 (106)
Total	532 (117)	539 (118)	539 (120)	522 (116)	530 (117)	527 (119)
p	<0,001			0,002

 

 

Potential association between smoking status during pregnancy and z-scores of placental weight and placental weight to birthweight ratio

 

The associations between smoking status during pregnancy and z-scores of PW and PW/BW are presented in Table 2. Smokers and quitters were less likely to have low PW z-score only after adjustment for potential confounders compared to non-smokers (adjusted RRR_smoker of 0.75 with 95% CI 0.70-0.81 and RRR_quitter of 0.86 with 95% CI 0.76-0.97). Moreover, smokers and quitters were less likely to have low PW/BW z-score both before and after adjustment for socio-demographic and medical characteristics of pregnant women compared to non-smokers.

Compared to non-smokers in the crude and adjusted analyses summarized in Table 3, high z-scores of PW and PW/BW were more likely among smokers and quitters.

 

Table 3. Association between smoking status before and during pregnancy and z-scores of placental weight and placental weight to birth weight ratio in Murmansk County (N = 44,411)

 

Z-score of:		Crude RRR (95% CI)			Adjusted RRR (95% CI)
		Smoking status during pregnancy			Smoking status during pregnancy
		Non-smoker	Quitter	Smoker	Non-smoker	Quitter	Smoker
placental weight	Low	1.00	0.91 (0.81-1.02)	0.98 (0.92-1.05)	1.001	0.86 (0.76-0.97)	0.75 (0.70-0.81)
	High	1.00	1.14 (1.03-1.27)	1.16 (1.08-1.24)	1.001	1.21 (1.09-1.36)	1.35 (1.25-1.45)
placental weight to birth weight ratio	Low	1.00	0.85 (0.76-0.96)	0.71 (0.65-0.77)	1.002	0.87 (0.77-0.97)	0.76 (0.70-0.83)
	High	1.00	1.21 (1.09-1.35)	1.73 (1.62-1.84)	1.002	1.18 (1.06-1.31)	1.52 (1.43-1.63)

 

¹– Z-score (-1;1) was used as a reference category

²– RRR adjusted for the variables maternal age, parity, marital status, residence, ethnicity, education, year of delivery, body mass index, mode of delivery, pregnancy diabetes, preeclampsia/eclampsia, anemia, and birthweight;

³ - RRR adjusted for all variables except birthweight

 

A negative dose-response relationship was evident between the number of cigarettes smoked per day during pregnancy and the odds of having low z-scores of both PW and PW/BW. Adjustment for potential confounders did not change these associations (Table 4).

However, positive dose-response relationship was observed between the high z-scores of PW and PW/BW and the number of daily smoked cigarettes during pregnancy (Table 4). Moreover, mothers who smoked ≥ 11 cigarettes per day while pregnant were 2.97 and 3.55 times more likely to have high z-score of PW and PW/BW, respectively compared to none cig/per day.

 

Table 4. Association between number of smoked cigarettes per day during pregnancy and z-scores of placental weight and placental weight to birth weight ratio in Murmansk County (N = 40,464)

 

Z-score of:	Crude RRR (95% CI) 1			Adjusted RRR (95% CI) 2
	Daily numbers of smoked cigarettes during pregnancy (0 – reference)			Daily numbers of smoked cigarettes during pregnancy (0 – reference)
	1-5	6-10	≥ 11	1-5	6-10	≥ 11
placental weight1: - Low       - High	0.62 (0.52-0.73)   1.55 (1.38-1.74)	0.56 (0.47-0.67)     1.81 (1.62-2.03)	0.52 (0.38-0.72)   2.01 (1.66-2.42)	0.47 (0.39-0.56)     1.87 (1.65-2.13)	0.36 (0.29-0.43)     2.48 (2.19-2.82)	0.30 (0.21-0.42)   2.97 (2.41-3.64)
placental weight to birth weight ratio2: - Low       - High	0.38 (0.31-0.46)   2.16 (1.93-2.41)	0.31 (0.24-0.39)     3.10 (2.80-3.44)	0.29 (0.19-0.46)   4.14 (3.49-4.91)	0.42 (0.34-0.52)     1.87 (1.67-2.09)	0.34 (0.27-0.43)     2.69 (2.42-3.00)	0.32 (0.21-0.51)   3.55 (2.98-4.23)

 

¹– Z-score (-1;1) was used as a reference category

²– RRR adjusted for the variables maternal age, parity, marital status, residence, ethnicity, education, year of delivery, body mass index, mode of delivery, pregnancy diabetes, preeclampsia/eclampsia, anemia, and birthweight;

³ - RRR adjusted for all variables except birthweight

DISCUSSION

In our singleton pregnancies study, mean PW at any week of gestation was lower compared to mean PW (SD) – 612 (138) g – at term pregnancies (37-42 weeks) reported by Nascente et al., (2020) [14]. We found that mean BW increased with gestational age for both girls and boys born at 37-41 week of pregnancy. Other studies also showed trend of increasing BW in infants from 37 to 41 weeks processing in line with our findings [14-16].  Unlike previous reports [15, 17], our study showed lower mean BW in male and female newborns born at a gestational age greater than 42 completed weeks compared to that at 41 weeks. McLean et al. (1991) [18] demonstrated higher BW and greater length and head circumference in postterm infants compared to term babies. Abovementioned study found no evidence of large postterm weight loss. Interestingly, Beltrand et al (2012) [19] found post-term boys were heavier and longer than term boys at birth, whereas BW was similar in post-term girls and term girls. Zhang et al. (2010) [17] reported that with obvious trend of declining mean BW among US non-Hispanic white singleton live births ≥37 weeks of gestation from 1992 through 2003 postterm infants continued to be heavier than term babies. The observed lower mean BW in postterm infants in our study might be explained by higher proportion of small for gestational age infants among postterm male and female newborns. To the moment, the origin of this finding is unclear and requires further investigation.

In our study, male singleton infants had higher BW than female infants at each gestational age (37–42 weeks). These findings are in line with the data of the Intergrowth-21^st project [15] and other studies [20, 21] Wallas et al (2013) [22] showed the placentas of males were significantly heavier than those of females but this effect of sex was less marked than the effect of parity. In recently published study on singleton births in Poland, boys also exhibited higher BW than girls in the entire cohort at each gestational week from 23 to 42 [23]. According to the hypothesis by Clifton (2010) [24], sex differences in growth and, moreover, survival of the fetus were likely to be mediated by the sex specific function of the human placenta. Fetal growth is generally regulated by placental function and inherent growth perspectives of the fetus. Observed in many population-based studies gender-specific difference may be explained by “gender-dependent differences in “sensitivity” to the placental and fetal tissue to fasting plasma glucose or weight gain–associated factors” [25].  Differences in gene expression as well as the effects of steroids or proteins on placental function may also influence on BW differences in newborn boys and girls [26]. It is of great importance that in unsatisfactory conditions boys are at increased risk of pregnancy complications, morbidity, and mortality because of different growth strategies. In mammals, the relationship between fetal sex and molecular signaling in both the placenta and endometria across gestation were suggested [27].

In our study, high z-scores of PW and PW/BW were more likely among smokers and quitters than among non-smokers. These findings are in line with those of previous studies confirmed that continued smoking during pregnancy caused higher PWs [9; 10; 50]. Thus, in a birth cohort study in Japan with 91,951 records included, both PW and PW/BW were higher among smokers compared with non-smokers [50]. Heidari et al. (2018) also demonstrated higher mean PW among smokers (610 g) compared to non-smokers (455 g) [8].  In the recent meta-analysis showed a 182 g heavier placenta in smoking pregnant women compared to quitters [10]. In quitters, PW was higher than in non-smokers [51]. 

Smoking during pregnancy changes development and function of placenta; however, the underlying mechanisms remain unclear. Compared to non-smokers, heavy smokers had 1.5-times lower total volume of placenta blood vessels and 2-times lower volume density of blood vessels. These differences were statistically significant. In smokers, total volume of intervillous space, syncytiotrophoblast, as well as fibrin were almost 1.5-times higher compared to control (non-smokers) group [8]. Previously, it has been suggested that abnormal vascularisation of the placenta and subsequent placental insufficiency in smokers were suggested the leading causes of adverse pregnancy outcomes [52].  Also, previously done studies observed differential expression of angiogenic factors in placenta in pregnancy with complications [9, 35, 53]. Pfarrer et al. (1999) explained the increase of PW in smokers by adaptive angiogenesis in placental villi [34]. Their findings suggested an adaptive response of the capillary bed in fetus within placental villi in smoking pregnant women. It increases the surface area used for gas and nutrients exchange reducing negative effect of hypoxia. Gloria-Bottini et al. (2015) [54] confirmed a discordant effect of smoking on BW and PW in the Haptoglobin 2 phenotype mothers but a concordant effect on BW and PW in mothers carrying Haptoglobin 1 allele.  These results provide further support for the hypothesis that there is the interaction between BW and PW and maternal haptoglobin phenotype. In contrast, previously done research did not confirm expected negative impact of smoking on PW.  This inconsistency may be due to methodological limitations of the study [11].

In the present study, both the PW and PW/BW z-scores were associated with daily numbers of smoked cigarettes during pregnancy. Those women who smoked eleven or more cigarettes per day had heavier placentas and higher PW/BW compared to light (1-5 cigarettes) smokers. In agreement with Mendelian study [10], our results showed that continued smoking during pregnancy causes higher PW. We also found greater PW/BW in smokers compared to quitters and in quitters compared to non-smokers. Compared with non-smokers high values of PW/BW z-scores in quitters and continuing smoking were in approximately 1.2 and 1.5 times often, respectively. 

We observed positive dose-response relationship between the high z-scores of PW and PW/BW and the number of daily smoked cigarettes. Our results are in agreement with earlier findings which showed that in smokers the PW/BW increased by number of cigarettes, independent of smoking status in the third trimester [51].

Our study is the first one in the Russian Federation aimed to explore association between smoking before and during pregnancy and placenta weight using large population-based sample. This can attenuate risk of selection biases, however relatively high number of missing among registered pregnancies  can affect the results. We had not used any imputation technique to treat the problem of missing data, that lead to exclusion around 17-18% of observation. This potentially can decrease the power of the study, but it is still more than 80%. The pregnancies with missed information probably can be systematically different from those without missed information. We did not perform sensitivity analysis in this study, but previous research based on the MCBR did not show any difference between those with and without missing data on core characteristics.

Another limitation of the study which can affects the results is possibility of information bias: questions on smoking status can be sensitive for pregnant women and information collected by doctors on this behavior factor can be a subject of differential misclassification. However, the consistently of results we got with previously published research allows us to treat self-assessment of smoking with confidence.

CONCLUSIONS

The mean placenta weight for boys was 534.1 g (SD 117.9 g) and for girls was 523.7 g (SD 116.6 g) and increased with gestational age.  The study found higher placenta weight and greater placental-to-birth weight ratio for current smokers and whose who quite before pregnancy and this association has dose-response patters for smoker. Our findings confirm, that nor only quitting during pregnancy but decrease in number cigarettes daily smoked can be beneficial for placentas development and consequently for a fetus. This can potentially be used as a motivation tool for promoting primary prevention strategies aimed at mitigating adverse pregnancy outcomes.    

 

About the authors

Olga Kharkova

Northern state medical university

Email: harkovaolga@yandex.ru
ORCID iD: 0000-0002-3130-2920
SPIN-code: 2167-7550
Scopus Author ID: 55912362200
ResearcherId: AAO-8495-2020

Russian Federation, Arkhangelsk

Vitaly A. Postoev

Northern state medical university

Author for correspondence.
Email: v.postoev@nsmu.ru
ORCID iD: 0000-0003-4982-4169
SPIN-code: 6070-2486

Russian Federation, Arkhangelsk

Anna Usynina

Northern state medical university

Email: perinat@mail.ru
ORCID iD: 0000-0002-5346-3047

Russian Federation, Arkhangelsk

References

Supplementary files