Combined Effects of Cigarette Smoking and Alcohol Consumption on Perinatal Outcome

Abstract

Background:

An increase in various congenital abnormalities associated with cigarette smoking and the use of alcohol during pregnancy has been reported in many studies. These exposures also increase the risk of pregnancy complications such as abruptio placentae, unexplained stillbirth, preterm labor and intrauterine growth restriction. However, very few studies have addressed the combined effect of smoking and drinking on pregnancy outcomes.

Methods:

In this review, the adverse effects of smoking or drinking on pregnancy were obtained from publications in which both substances were addressed in the same study population. A special effort was made to find studies in which the combined effect of these substances was investigated.

Results:

Preterm labor occurred more frequently in women who drank and smoked during pregnancy. This increased odds ratio was more than the sum of the effects of either smoking or drinking, indicating that the use of both substances by the same woman has a synergistic effect that increases the risk of preterm labor. This synergistic effect was also found for low birth weight and growth restriction.

Conclusions:

As most of the women who drink during pregnancy also smoke cigarettes, attention should be given to the prevention or reduced use of both substances during pregnancy.

The prevalence of cigarette smoking during pregnancy varies between 11 and 56.9% [1–5]. The prevalence of drinking during pregnancy is also high. Studies from different countries reported a prevalence between 23 and 60% [6–10]. It is therefore not surprising to find that the prevalence of fetal alcohol syndrome ranges from 3.7 to 74.2 per 1,000 children [11, 12]. Furthermore, it has recently been found that exposure to alcohol during pregnancy, in particular binge drinking during the first trimester, is associated with an increased risk of sudden infant death syndrome (SIDS) [13].

The rate of perinatal mortality in many countries and communities has not improved much in recent years in spite of the great increase in medical knowledge. For example, the stillbirth and early neonatal mortality rates at 7 centers participating in the WHO calcium study were 12.5 and 9 per 1,000, respectively. Spontaneous preterm labor (28.7%) was the main obstetric event leading to perinatal death and preterm delivery (62%), the most common cause of early neonatal death [14].

Although the more common adverse effects from alcohol consumption and cigarette smoking on pregnancy are well described [15], little is known about the combined effect of both substances on the outcome of pregnancy and how combined exposure might affect common causes of perinatal mortality such as intrauterine death or preterm delivery. Therefore, we tried to find more information on the combined effect of smoking and prenatal alcohol exposure.

Methods

We completed a Medline literature search for articles on the effects of prenatal exposure to alcohol and smoking on the risk of adverse outcomes during pregnancy (miscarriage, congenital abnormalities, preterm labor, prematurity, low birth weight, growth restriction, placental abruption and stillbirths) and in the first year of life (neonatal deaths and SIDS). We then searched the identified articles for studies on effects of smoking and drinking on any of the mentioned adverse pregnancy outcomes in the same study population. Lastly, we identified the studies which report data on the combined effects on adverse pregnancy outcome. In table 1 and figure 1 we present a summary of the outcomes.

Table 1.

Fetal and perinatal risks of maternal smoking or drinking

	Smoking			Drinking
	OR or RR1	95% CI	comment	OR or RR1	95% Cl	comment
Spontaneous miscarriage
Harlap and Shiono [16]	RR 1.01	0.53–2.11	1st trimester	RR 1.98	1.04–3.77	1–2 drinks/day
	RR 1.21	0.67–2.19	2nd trimester	RR 3.53	1.77–7.01	≥3 drinks/day
Dominquez-Rojas et al. [17]	3.35	1.65–6.92	≥11 cigarettes/day	-	-
Rasch [18]	0.95	0.40–2.20	≥20 cigarettes/day	4.84	2.87–8.16	≥5 units/week
Armstrong et al. [19]	1.22	1.13–1.32	10–19 cigarettes/day	1.11	1.05–1.18	1–2 drinks/week
Orofacial clefts
Lorente et al. [20]	1.79	1.07–3.04	Both lip and palate	2.28	1.02–5.09	Palate only
Anal atresia
Yuan et al. [21]	1.4	0.5–3.6	20+ cigarettes/day	4.8	1.2–19.1	-
Preterm labor
Berkowitz et al. [29]	1.7	1.0–2.9	≥10 cigarettes/day	3.8	1.1–13.0	≥14 drinks/week
Shiono et al. [30]	1.2	1.1–1.4	≥1 pack/day	1.3	1.0–1.7	≥1 drink/day
McDonald et al. [31]	1.22	1.05–1.41	<10 cigarettes/day	1.21	0.68–2.17	21+ drinks/week
Bada et al. [32]	1.2	1–1.44	≥0.5 pack/day	1.11	0.86–1.43	≥1 drink/week
Low birth weight
Bada et al. [32]	2.0	1.56–2.57	≥0.5 pack/day	1.57	1.12–2.22	≥1 drink/week
Growth restriction
Spinillo et al. [37]	2.61	1.38–4.93	Throughout pregnancy	3.55	1.03–12.21	≥2 drinks/day
Abruption
Tikkanen et al. [44]	1.8	1.1–2.9	-	2.2	1.1–4.4	-
SIDS
Iyasu et al. [13]	2.2	0.8–5.8	Any smoking during pregnancy	8.2	1.9–35.3	Binge first trimester

Odds ratios (ORs) refer to adjusted odds ratios, whenever available. Only the lowest amount of smoking or drinking that has caused a significant change in outcome in a particular study was listed. RR = Relative risk; CI = confidence interval; SIDS = sudden infant death syndrome.

¹OR is referred to unless RR is mentioned.

Fig. 1.

Combined effects of alcohol and drinking in pregnancy. Dew et al. [33] refer to preterm birth, Okah et al. [38] to term low birth weight and Jackson et al. [36] to low birth weight. OR = Odds ratio; CI = confidence interval.

Results

Spontaneous Miscarriage

Few articles could be found where the effects of smoking and alcohol were examined in the same study population. Harlap and Shiono [16] analyzed questionnaires completed by 34,344 women. Regular drinkers (one or more drinks per day) had more spontaneous losses, mainly during the second trimester. Occasional drinkers had a relative risk (RR) of 1.03 (95% CI 0.57–1.86). Regular drinkers taking 1–2 or more drinks, or more than 3 drinks daily had RRs of 1.98 and 3.53, respectively (table 1). Smokers had RRs of 1.01 (95% CI 0.53–2.11) and 1.21 (95% CI 0.67–2.19) in the first and second trimesters, respectively, when compared with non-smokers. The increased risk of second-trimester miscarriage in drinkers was not explained by age, parity, race, marital status, smoking, or the number of previous spontaneous or induced abortions. The combined effect was variable, but it seems that smoking did not increase the existing risk from heavy drinking.

Dominquez-Rojas et al. [17] studied the possible relationships between spontaneous abortion and caffeine, tobacco and alcohol intake in a well-controlled group of hospital workers using a retrospective cohort study design in 711 women, 20–41 years old in Madrid. All data regarding the purpose of this study were extracted from clinical histories registered. The dependent variable was spontaneous abortion and the independent variables were tobacco, coffee, and alcohol intake. Age, previous spontaneous abortion, menarcheal age and marital status were considered as potential confounders. Using multiple logistic regression the authors found that caffeine consumption and the smoking of more than 10 cigarettes/day increased the risk of spontaneous miscarriage (table 1). As the number of alcohol consumers was low, the effect of alcohol could not be assessed and therefore also not the combined effect with smoking.

In a case-control study, Rasch [18] examined 330 women who had spontaneous miscarriages and 1,168 women who received antenatal care. They examined the association between caffeine consumption, alcohol and cigarettes and used logistic regression to control for confounding variables. It was found that consumption of 5 or more units of alcohol/week during pregnancy may increase the risk of spontaneous abortion (table 1). The combined effect of alcohol and smoking was not addressed.

From 1982 to 1984 Armstrong et al. [19] examined questionnaires completed by 56,000 women after miscarriage or spontaneous abortion. A statistically significant association was found between spontaneous abortion and smoking or alcohol use. It was estimated that cigarette smoking contributed to 11% of spontaneous miscarriages and alcohol to 5% (table 1). However, the combined effect of both substances was not addressed.

Orofacial Clefts

Lorente et al. [20] examined the data derived from a European multicenter case-control study including 161 infants with oral clefts and 1,134 control infants. Multivariate analyses showed an increased risk of cleft lip with or without cleft palate associated with smoking odds ratio ((OR) 1.79; table 1) and an increased risk of cleft palate associated with alcohol consumption (OR 2.28, 95% CI 1.02–5.09). The former risk increased with the number of cigarettes smoked. This study provides further evidence of the possible role of prevalent environmental exposures such as tobacco and alcohol in the etiology of oral clefts. The combined effect of smoking and drinking was not addressed.

Anal Atresia

Anal atresia also seems to be associated with smoking and maternal alcohol use during pregnancy [21]. Using the data compiled from 216,707 births from the population-based Kanagawa Birth Defects Monitoring Program, a case-control study was done to evaluate the effect of maternal smoking and/or alcohol exposure during pregnancy on the risk of infants’ anal atresia in 1989–1994. The frequency of maternal smoking (including passive smoking) and/or maternal alcohol use during pregnancy among 84 infants with anal atresia was compared with 174 matched controls. Maternal alcohol use during early pregnancy was associated with an increased risk of isolated anal atresia (table 1). A slightly increased trend was also observed in the association of maternal smoking during pregnancy with syndromal and isolated anal atresia. The combined effect of smoking and drinking was not reported.

Preterm Labor

There are several reports on the association between the consumption of alcohol during pregnancy and preterm labor [22–26]. Smoking is also a well-known risk factor for preterm labor. Using meta-analysis, Castles et al. [27] examined the effects of smoking on 5 common complications of pregnancy. Smoking was found to be strongly associated with an elevated risk of placenta praevia, abruptio placentae, ectopic pregnancy, and preterm prelabor rupture of membranes, and a decreased risk of preeclampsia. In a more recent meta-analysis from 20 prospective studies, Shah and Bracken [28] found that any maternal smoking increased the risk of preterm delivery by 27%. Heavy drinking has a 3-fold increase in the risk of preterm labor [29]. This confirmed earlier findings that smoking or alcohol use were associated with preterm deliveries (table 1) [30]. The combined effect of both substances was not addressed. However, two other studies failed to identify an increase in risk for alcohol (table 1) [31, 32]. The combined effects were not addressed.

In a recent retrospective study, Dew et al. [33] looked at the combined effect of alcohol and smoking on the rates of preterm labor. They examined data on 83,683 births in Kansas City from 1990 to 2002. Over 13% of infants born to mothers who smoked were preterm in comparison to 9.6% for non-smokers (adjusted OR (aOR) 1.22). For alcohol alone the aOR was 1.08 (fig. 1). The sum of the aOR for smoking and alcohol use indicated a 30% (22 and 8%) risk of preterm birth. However, the aOR for the combined effect was 1.46. As this indicates an increase in risk of 46%, cigarette smoking and alcohol use appear to have a synergistic effect on the likelihood of preterm birth. The combination of the two behaviors was associated with an 18% risk of a preterm birth. The additional use of illicit drugs increased the risk of preterm labor by 31.4%.

Birth Weight

Both alcohol use and cigarette smoking were found to affect birth weight adversely (table 1) [32]. However, the combined effect of both substances was not examined in this study. In a prospective study of 7,301 births, associations were sought between maternal alcohol, tobacco and cannabis use and several adverse outcomes of pregnancy. Infants born to smokers had lower birth weights. However, the same effect was not observed after exposure to alcohol [34].

Verkerk et al. [35] analyzed the relationship between moderate maternal alcohol consumption during pregnancy and both birth weight corrected for gestational age and preterm delivery in 3,447 women. After adjustment for possible confounders they found that for most women alcohol consumption was unrelated to birth weight when corrected for gestational age and preterm delivery. However, in a subgroup of women who smoked 20 cigarettes or more a day and who drank more than 120 g of alcohol a week in early pregnancy, birth weight was decreased by 7.2% (95% CI 0.2–14.2). However, they examined the effect on birth weight and not the proportion of low-birth-weight newborns.

In a recent case-control study, Jackson et al. [36] compared 200 infants with a birth weight of <2,500 g with 200 unmatched infants of normal birth weight. Smoking had an aOR of 2.67 and drinking an aOR of 1.32 (95% CI 0.8–2.2). However, the use of both substances increased the aOR for low birth weight to 4.24, suggesting an interaction between the use of the two substances (fig. 1).

Growth Restriction

Spinillo et al. [37] estimated the proportion of ultrasonically diagnosed fetal growth restriction that may be attributable to cigarette smoking, alcohol use, and illicit drug use. They computed multivariate odds ratios for fetal growth restriction associated with cigarette smoking and alcohol and illicit drug consumption in 350 singleton pregnancies complicated by fetal growth restriction and 700 controls. The OR for fetal growth restriction among women who smoked throughout pregnancy was 2.61 (95% CI 1.38–4.93; table 1) compared to women who claimed to have stopped smoking by the 18th week of gestation. For alcohol use the OR was 3.55 (95% CI 1.03–12.21). The OR for the combined smoking of cigarettes and use of alcohol was not given but the authors stated that the attributable risk for fetal growth restriction associated with behavioral variables measured during the first trimester and thereafter ranged from 18 to 21%. The combined effect of smoking and drinking was not examined.

In a retrospective cohort study, Okah et al. [38] examined the effects of smoking, alcohol, and drug use in 78,397 term live births in Kansas City. Overall term low-birth-weight rate was 3.3%, and it continued to increase as the number of health-compromising behaviors increased. Unadjusted ORs for term low birth weight were: 2.3 for smoking; 0.9 for alcohol; 4.6 for smoking and alcohol, and 8.4 (95% CI 6.2–11.5) for smoking, alcohol and drugs (fig. 1).

Placental Abruption

Previous studies demonstrated an association between abruptio placentae or intrapartum asphyxia and the use of alcohol during pregnancy [39–41]. The association between cigarette smoking and abruptio placentae is well known [27, 42]. Another large meta-analysis on cigarette smoking and abruptio placentae was published in 1999 [43]. A total of 1,358,083 pregnancies was examined from seven case-control studies and six cohort studies. The overall incidence of abruptio placentae was 0.64%. Smoking was associated with an increased risk of abruptio placentae (OR 1.9, 95% CI 1.8–2.0). It was estimated that 15–25% of episodes of abruptio placentae were caused by smoking. None of these studies have addressed the combined effects of smoking and alcohol.

Tikkanen et al. [44] compared 198 women with placental abruption to 396 control women who were identified among 46,742 women who delivered at a tertiary referral university hospital between 1997 and 2001. Clinical variables were compared between the groups. Multivariate logistic regression analysis was applied to identify independent risk factors. The overall incidence of placental abruption was 0.42%. Among the independent risk factors were maternal smoking (aOR 1.8; table 1) and paternal smoking (aOR 2.2, 95% CI 1.3–3.6) and use of alcohol. Smoking by both partners multiplied the risk. However, the authors did not report on the combined effects of alcohol and smoking.

Stillbirths

Marbury et al. [45] were among the first investigators to find an association between alcohol use during pregnancy and adverse pregnancy outcome. In the crude data, alcohol intake of 14 or more drinks per week was associated with a variety of adverse pregnancy outcomes such as stillbirth. Most of the alcohol-related stillbirths were due to placental abruption. Kesmodel et al. [46] studied the outcome of 24,768 singleton pregnancies which included 116 stillbirths. For women who consumed >5 drinks/week during pregnancy the risk ratio was 2.96 (95% CI 1.37–6.41) when compared to women who drank less than 1 drink/week. Adjustments for various socio-economic and lifestyle factors did not change the effect of alcohol. The rate of stillbirth from fetoplacental dysfunction increased by 6.5-fold from 1.37/1,000 births for women who took <1 drink/week to 8.83/1,000 for women who consumed >5 drinks/week. No interaction with smoking or caffeine was observed.

Henderson et al. [47] recently systematically reviewed the available evidence on studies in humans, from 1970 to 2005, on the effects of alcohol exposure (up to 10.4 UK units or 83 g/week) compared with pregnancies with no alcohol exposure on pregnancy outcomes. The main outcome measures considered were miscarriage, stillbirth, intrauterine growth restriction, prematurity, birth weight, small for gestational age at birth, and birth defects including fetal alcohol syndrome. At low/moderate levels of alcohol exposure, there were no consistently significant effects of alcohol on any of the outcomes considered. However, the authors concluded weaknesses in the evidence preclude the conclusion that drinking at these levels during pregnancy is safe.

The association between cigarette smoking during pregnancy and stillbirths is well known [48, 49]. In addition, Wisborg et al. [50] examined the association between exposure to tobacco smoke in utero and the risk of stillbirth and infant death in a cohort of 25,102 singleton children of pregnant women. Exposure to tobacco smoke in utero was associated with an increased risk of stillbirth (OR 2.0, 95% CI 1.4–2.9). In a recent review article on the epidemiology of stillbirth, the authors confirmed the well-accepted association between cigarette smoking and stillbirth. The underlying mechanisms were most likely the effects of carbon monoxide and nicotine and placental insufficiency [51]. In addition, an association between cigarette smoking and sudden intrauterine unexplained deaths has been found in several studies [52–54].

However, not all studies agreed. Huang et al. [55] assessed fetal, maternal, and pregnancy-related determinants of unexplained antepartum fetal death in 84,294 births weighing 500 g or more from 1961–1974 and 1978–1996. Unexplained fetal deaths were defined as fetal deaths occurring before labor without evidence of significant fetal, maternal, or placental pathology. One hundred and ninety-six unexplained antepartum fetal deaths accounted for 27.2% of 721 total fetal deaths. Two thirds of the unexplained fetal deaths occurred after 35 weeks of gestation. Cigarette smoking and alcohol use were not significantly associated with unexplained fetal death.

Sudden Infant Death Syndrome

The association between periconceptional alcohol use (aOR 6.2, 95% CI 1.6–23.3) and in particular first trimester binge drinking (aOR 8.2; table 1) and SIDS was described for the first time in a recent population-based case-controlled study. Cigarette smoking was not identified as a risk factor [13]. This finding was confirmed in a later literature review on paternal and maternal alcohol exposure and SIDS [56]. On the other hand, little association between alcohol intake and infant death was found when Kesmodel et al. [46] studied the outcome of 24,768 singleton pregnancies which included 116 stillbirths and 119 infant deaths.

To investigate maternal and obstetric risk factors for SIDS, data from a cohort of 15,627,404 live births from 1995 to 1998 were analyzed [57]. A nested case-control analysis was done to examine the risk factors for SIDS. The overall incidence of SIDS was 81.7 per 100,000 births. Cigarette smoking significantly increased the risk of SIDS (OR 3.19, 95% CI 3.03–3.37). This large study also confirmed the association between SIDS and abruptio placentae as found previously when 239 mothers of SIDS infants were compared with 239 controls matched for sex, race, birth hospital and date of birth. In this study abruptio placentae was associated with a nearly sevenfold increase in SIDS (OR 7.94, 95% CI 1.34–47.12) [58]. In addition, in a recent case-control study, the problem of SIDS was addressed in 20 regions in Europe. Data for more than 60 variables were obtained from the records of 745 SIDS cases. Logistic regression was used to calculate odds ratios for every factor in isolation and to construct multivariate models. There were 2,411 live controls. If the mother smoked, significant risks were associated with bed-sharing at 2 weeks (OR 27, 95% CI 13.3–54.9). It was found that this increase in risk was partly attributable to the mother’s use of alcohol [59]. None of these studies investigated the prevalence of SIDS in mothers who smoked and drank alcohol.

Conclusions

There are known examples of similar biochemical effects of alcohol use and cigarette smoking. The best example of this shared mechanism is the effect on folic acid metabolism. It has been shown that smoking compromises the folate status in risk groups with an inadequate daily allowance of folate [60]. In addition, alcohol intake and low methylenetetrahydrofolate reductase activity has an adverse effect on the total homocysteine concentration [61]. In addition, in a recent meta-analysis, Verkleij-Hagoort et al. [62] confirmed that hyperhomocysteinemia is a risk factor for congenital heart disease. Another known example is the effect of these substances on antioxidants other than folate. For example, there is evidence for alcohol-induced oxidant injury where pretreatment of antioxidants such as vitamins C and E reduce alcohol-induced effects [63]. Cigarette smoke also induces oxidative stress as reflected by low vitamin C levels in current smokers. Smokers also have lower levels of α-carotene, β-carotene and cryptoxanthin [64]. In addition, accelerated placental maturation, measured as tissue calcification is associated with maternal cigarette smoking, but dietary antioxidants may reduce villus calcification [65]. Furthermore, there are indications that the use of antioxidants such as vitamins C and E may reduce the teratogenic effect of alcohol [66]. Lastly, a recent meta-analysis on prenatal vitamin supplementation of folic acid-containing multivitamins showed an association with a decreased risk of several congenital abnormalities, not only neural tube defects [67]. However, it is not certain whether these protective effects are stronger in women who consume alcohol and/or smoke cigarettes during pregnancy. It is most likely that alcohol and cigarette smoking have synergistic effects on some adverse outcomes of pregnancy. This combined effect has been demonstrated on term growth restriction [36]. In addition, the combined effects of smoking and drinking increase the risk of preterm birth by 46% in contrast to the 30% reflected by the sum of the adjusted odds ratios for either smoking or drinking only [33]. It is likely that there are more subtle effects that are not yet known.

In this review it was extremely difficult to assess the effects of alcohol on the outcome of pregnancy as it was often not examined in large studies. In addition, alcohol exposure was not reported in the same way and was not quantified accurately. Comparisons of risks or meta-analyses will therefore not be accurate. Large carefully designed studies need to be developed to examine the effects of smoking, alcohol use and the combined effects from both. In addition, the social background of participants should also be reported as it could influence the drinking pattern or the effects of alcohol on the fetus. The public health implications would be of considerable importance if a combined effect was identified that exceeded the additive effects of both risk factors. This would identify potential preventive interventions during pregnancy for alcohol use, smoking or both that would reduce risk for a variety of adverse outcomes. Many interventions with demonstrated efficacy are available. The benefits from intervention would also have a beneficial effect on future pregnancies.