- Email: [email protected]
Caffeine-cigarette interaction on fetal growth
Caffeine-cigarette interaction on fetal growth Louise Beaulac-Baillargeon, Ph.D., and Carole Desrosiers, B.Sc.(Pharm.), M.Sc. Ste-Foy, Quebec, Canada The influence of caffeine and cigarette consumption on fetal growth during pregnancy was studied retrospectively in 913 newborn infants. Analysis of variance was adjusted for length of gestation and sex. A significant caffeine-cigarette interaction was found on birth weight (F = 85.4, p < 0.01). Among women smoking 15 cigarettes or more per day, birth weight was 206 gm (±57.7 gm) lighter for babies whose mothers consumed 300 mg or more of caffeine per day. Head circumference and length of the newborns were affected only by cigarette smoking. Another caffeine-cigarette interaction was observed on placental weight (F = 15.0, p < 0.01). Among women who consumed less than 300 mg of caffeine daily, placental weight increased with cigarette consumption. However, it diminished (p < 0.05) among women smoking 15 cigarettes or more per day if they consumed 300 mg or more of caffeine daily. This is the first time that such interactions were found to influence birth and placental weights, indicating that the concomitant consumption of caffeine and cigarette constitutes a higher risk for the developing fetus. (AM J OssrET GYNECOL 1987;157:1236-40.)
Key words: Caffeine, cigarette smoking, fetus, pregnancy, fetal growth The caffeine, cigarette, and alcohol consumption of mothers during pregnancy has been the subject of many studies of its effects on fetal growth. Because these substances can easily reach the fetus by placental transfer, it is of prime importance to study the effect of interaction of these substances on fetal growth. The influence of smoking on birth weight is well known. In fact, birth weight is reported to be inversely proportional to the number of cigarettes smoked by the mother, and this effect is independent on other factors related to the variation in birth weight. 1 The length of the newborn is also reduced when the mother smokes, and some authors have observed a decrease in head circumference! Placental weight is not affected by smoking, but the number of cigarettes smoked has been correlated with hypovascular and atrophic microscopic changes in placental villi3 and deterioration of placental perfusion type! The reduction in uteroplacental circulation caused by the vasoconstricting effect of nicotine4 and the relative fetal hypoxia caused by the increase of carbon monoxide concentrations in the fetus 5 are two factors responsible for the decrease in fetal growth observed among babies born to smokers. On the other hand, maternal alcoholism has been related to a series of physical and neurologic abnormalities of newborns that in 1973 Jones and Smith6 termed "fetal alcohol syndrome." A significant clinical reduction of newborn weight is observed only at a con-
From the Ecole de Pharmacie, Universite Laval. Received for publication August 12, 1986; revised june 5, 1987; accepted june 24, 1987. Reprint requests: Louise Beaulac-Baillargeon, Ph.D., Ecole de Pharmacie, Universite Laval, Ste-Foy, Quebec, G1K 7P4, Canada
1236
sumption level higher than 30 ml of absolute alcohol per day. 7 During pregnancy, the elimination kinetic profile of caffeine was reported to be modified. In fact, Aldridge et al. 8 reported an increase in the half-life of caffeine from 5.3 hours before pregnancy to 18.1 hours during the last two trimesters of pregnancy. These values return to normal soon during the postpartum period. This increase in the half-life of caffeine in the mother coincides with the period during which fetal growth is exponential so that an accumulation of this substance represents a potential risk for the fetus and the placenta, which is responsible for the transfer of nutrients. To date, no well-controlled study has found that caffeine influences human birth weight. However, it has been demonstrated that a unique dose of 200 mg of caffeine, which is equivalent to two to three cups of coffee, administered to the mother significantly reduces blood flow in placental villi and increases maternal serum epinephrine levels. 9 This effect of caffeine on placental circulation is quite similar to that observed after cigarette consumption. Smoking is often associated with high caffeine intake, and this has also been confirmed for pregnant women. 10 However, the influence of that concomitant intake on fetal growth was not fully investigated. The association of these two consumption habits could represent a greater risk of disturbing the full development of the fetus. However, to date research has not fully investigated this hypothesis because the effects of caffeine and cigarette consumption on fetal growth have been studied separately. The effects of caffeine and cigarettes on placental circulation raise the assumption that these substances
Volume 157 Number 5
can act in an additive or synergistic way on fetal growth. The main objective of this study was to evaluate the independent effect and the interaction of caffeine and cigarette on weight, head circumference, length of the newborn, placental weight, and placental ratio (placental weight/birth weight x 100). Methodology
All women delivering at either Centre Hospitalier Chauveau or Hopital St-Fran~ois d'Assise in Quebec City, Canada, during a 5-month period from February to July 1985 were requested to complete a questionnaire on their caffeine, cigarette, and alcohol consumption habits during pregnancy. Of these 35 women refused to participate while 1014 women entered the study. This low refusal rate of 3% is not thought to bias the results. Standardized verbal and written information was given along with the questionnaires. Data collection was completed by a review of the medical and obstetric records of the mother and baby. The consumption pattern of caffeine, cigarettes, and alcohol of these women during pregnancy has been the subject of a previous publication.'' The questionnaire gathered information about the socioeconomic status, participation at prenatal classes, parity, obstetric history, weight before and at the end of pregnancy, height, health problems, and pattern of consumption of caffeine, cigarettes, and alcohol during pregnancy. The reported intake of these substances represents an average across pregnancy. The caffeine content of coffee, tea, cola, and chocolate used is that reported by the quantitative analysis of Gilbert et al., 12 and the publication of Graham." The amount of caffeine varied depending on the substance consumed: coffee: expresso = 170 mg/cup of 180 ml, filter or percolator = 110 mg/cup, instant = 75 mg/cup, and decaffeinated = 1 mg/cup; tea: bag = 50 mg/cup and leaves or instant == 30 mg/cup; chocolate = 40 mg/bar of 40 gm; cola: diet or regular == 45 mg/180 ml. These values were used to calculate total caffeine consumption by adding the intake from coffee, tea, cola, and chocolate. Cigarette consumption represents the number of cigarettes smoked per day. Alcohol consumption was calculated in milliliters of absolute alcohol consumed per day. Beer (350 ml), wine (120 ml), appetizer (60 ml), and spirits/digestive (30 ml) contained 15, 13, 13, and 15 ml, respectively, of absolute alcohol per serving. To study the influence of caffeine, cigarette, and alcohol consumption on fetal growth and exclude other major factors affecting fetal growth, it was necessary to exclude 101 subjects from the results analysis: multiple pregnancies (n = 16), mothers suffering from preeclampsia and chronic hypertension (n = 51), diabetes (n = 8), organic heart disease (n = 5), congenital
Caffeine-cigarette interaction on fetal growth
1237
Table I. Physical characteristics of mothers and newborns Characteristics
Mothers Weight before pregnancy (kg) Weight at the end of pregnancy (kg) Weight gain (kg) Height (em) Length of gestation (wk) Placental weight Newborns Birth weight (gm) Head circumference (em) Length (em)
Mean± SD
56.3 69.6 13.3 160.7 39.4 526.1
± 9.7 ± 10.5 ± 4.4 ± 6.0 ± 1.6 ± 121.0
3352.1 ± 496.3 34.5 ± 1.5 51.0 ± 2.3
spherocytosis (n = 1), Val Kecklinhausen (hyperparathyroidia) (n = 1), hypoglycemia (n = 3), babies with major malformations (n == 11), and stillbirths (n == 5). Statistical analysis was done by means of the statistical analysis system. Level of significance was fixed at 0.05. Analysis of variance for unbalanced groups was done by the introduction of independent variables: caffeine, cigarette, and alcohol consumption and combinations of these variables adjusted for length of gestation and the sex of the babies. The consumption of these substances was divided as follow: alcohol: <2, 2 to 5.9, ;;.6 ml of absolute alcohol per day; cigarettes: 0, 1 to 14, ;;.IS/day; and caffeine <300 and ;;.300 mg/day. Dependent variables included birth weight, head circumference, length of newborns, placental weight, and placental ratio. Tukey-Kramer's test for unbalanced groups was used to define the influence of different levels of consumption on fetal growth. Reported means for birth weight, head circumference, and length were adjusted for length of gestation and the sex of the babies, both of which were unevenly distributed in the subgroups studied. Placental weight was adjusted for length of gestation only since it did not vary with the sex of the babies. Results
After data collection, 101 subjects were excluded since they did not meet the inclusion criteria mentioned. Results are presented for the remaining 913 subjects. Physical characteristics of babies and mothers are presented in Table I. Alcohol at doses consumed by women in the present study had no significant effect on the studied parameters. Mean absolute alcohol consumption of this population was 1.5 ml (SD = ± 3.2 ml) per day, which represented about three servings of alcoholic beverages per month. Subjects were then grouped according to caffeine and cigarette consumption. Caffeine consumption was
1238
Beaulac-Baillargeon and Desrosiers
November 1987 Am J Obstet Gynecol
Table II. Groups according to the caffeine and cigarette consumption of the mothers Consumption level Caffeine (mglday)
<300 <300 <300 :;;.300 :;;.300 :;;.300
I
Mean consumption
Cigarettes (No.I day)
No. of subjects
0 1-14 >15 0 1-14 >15
536 128 114 42 24 69
divided into two categories: (1) subjects whose consumption was less than 300 mg/day and (2) those whose daily consumption was greater than 300 mg/day. Among these two categories, women were divided according to their cigarette consumption, either 0, one to 14, or 15 or more per day. For each of these six groups, Table II shows the number of subjects and the mean daily consumption of caffeine and cigarettes. At the end of pregnancy, smokers represented 36.7% of the study population. Whereas 15.8% of the women who smoked one to 14 cigarettes/day and 37.7% of those who smoked 15 or more cigarettes daily consumed 300 mg or more of caffeine daily, only 7.3% of nonsmokers belonged to the latter category. Among the 135 women (14.8% of the population) whose daily consumption of caffeine rose to 300 mg or more, 68.9% smoked. Birth weight. Analysis of variance revealed an influence of cigarette (F = 32.3, p < 0.01) and caffeinecigarette interaction (F = 5.8, p < 0.01) on birth weight. For nonsmokers, mean birth weight of babies adjusted for length of gestation and sex of babies was 3441.9 gm (SEM = 16.8 gm). This weight was significantly reduced (p < 0.05) for babies born to mothers who smoked; it was 3204.4 gm (SEM = ± 39.0 gm) and 3141.7 gm (SEM = ±28.8 gm) for a consumption of one to 14 and 15 or more cigarettes per day, respectively. Fig. 1 presents mean birth weight of babies divided according to maternal caffeine and cigarette consumption. All four smoking subgroups had mean birth weights significantly lower than both nonsmoking subgroups. For nonsmokers and women who smoked one to 14 cigarettes daily, the birth weight was not statistically different with a caffeine consumption less than or greater than 300 mg/day. However, among women smoking 15 cigarettes or more per day, the birth weight of babies of those women who consumed 300 mg or more caffeine daily was significantly (p < 0.05) lighter than that of babies whose mothers consumed less caffeine. The difference in birth weight of babies between these two groups was 206 gm (SEM = ±57.7 gm). When we consider both caffeine and cigarette consumption, the birth weight of babies born to women
Caffeine (mglday) mean± SD
100.8 119.3 135.2 407.8 423.9 442.2
± 83.2 ± 90.6
± 86.5
± 222.7 ± 116.9 ± 142.7
I
Cigarettes (No./day) mean± SD
0 7.6 ± 3.8 20.3 ± 5.4 0 9.3 ± 3.4 22.8 ± 6.8
who smoked 15 cigarettes or more per day was significantly reduced if the mother consumed 300 mg of caffeine or more daily. Head circumference. Head circumference was influenced by cigarette consumption (F = 8.2, p < 0.01) but not by caffeine consumption. No caffeine-cigarette interaction was found. Head circumference of babies born to nonsmokers was 34.6 em (SEM = ± 0.06 em), and it diminished significantly (p < 0.05) for babies of mothers who smoked, being 34.3 em (SEM = ± 0.11 em) and 34.1 em (SEM = ± 0.11 em) in newborns whose mothers smoked one to 14 and 15 or more cigarettes per day, respectively. Length. Length of babies was also influenced by cigarette consumption (F = 17.8, p < 0.01), but again there was no interaction between cigarette smoking and caffeine consumption. Babies born to nonsmokers had a mean length of 51.3 em (SEM = ± 0.09 em), whereas those born to mothers who smoked one to 14 and 15 or more cigarettes per day were 50.7 em (SEM = ±0.16 em) and 50.3 em (SEM = ±0.15 em) long, respectively. Placental weight. Placental and birth weights underwent a caffeine-cigarette interaction (F = 8.9, p < 0.01). There was no significant difference between placental weight of nonsmokers and placental weight of the two other groups of smokers. However, when we compared this weight between smokers groups according to their caffeine consumption, a dose-related interaction was seen. Fig. 2 shows mean placental weight adjusted for length of gestation for each consumption group. Among women who consumed less than 300 mg of caffeine daily, placental weight was significantly higher for those who smoked 15 cigarettes or more per day (mean ± SEM = 566.6 ± 12.5 gm) as opposed to nonsmokers (mean ± SEM = 517.5 ± 4.8 gm) (p < 0.05) and smokers of one to 14 cigarettes per day (mean ± SEM = 521.2 ± 10.4 gm). Among women who smoked 15 cigarettes or more per day, placental weight was significantly reduced for those who consumed 300 mg or more of caffeine daily compared with those whose caffeine consumption was
Caffeine-cigarette interaction on fetal growth 1239
Volume 157 Number 5
3700 575
3600
--
3500 ........
~ 1:I: C!J
w
3:
3400
550
01
3300
1:I:
525
w 3:
500
C!J
3100
475
3000 2900
450
2800
0
1-14
15+
CIGARETTE CONSUMPTION (/day) Fig. 1. Effect of caffeine-cigarette interaction on birth weight. Means ( ±: SEM) are adjusted for the length of gestation and sex of newborns. (0 = Caffeine consumption <300 mg/day; fZl = caffeine consumption "'300 mg/day.)
less than 300 mg/day (p < 0.05); placental weight was 502.3 gm (SEM = ± 13.8 gm) and 566.6 gm (SEM = ± 12.5 gm), respectively. Placental ratio. The ratio of placental weight to birth weight was affected by cigarette consumption (F = 12.8, p < 0.01) and underwent a caffeinecigarette interaction (F = 3.3, p < 0.05) (Fig. 3). Placental ratio increased with cigarette consumption, going from 15.1 (SEM = ±0.12) for nonsmokers to 16.5 (SEM = ± 0.26) and 17.3 (SEM = ± 0.26) for women who smoked one to 14 and 15 or more cigarettes per day, respectively. This reflects the increase in placental weight and the decrease in birth weight observed with an increase in the number of cigarettes smoked. Considering both caffeine and cigarette consumption, the placental ratio increased significantly (p < 0.05) for every consumption level of cigarettes for women who consumed less than 300 mg of caffeine/day. However, for women who consumed 300 mg or more of caffeine daily, there was no significant difference in placental ratio according to cigarette consumption. Among women who smoked 15 or more cigarettes a day, the placental ratio was not statistically different with a caffeine consumption of less than or greater than 300 mg/day, since both birth and placental weights diminished when caffeine consumption was 300 mg or higher. Comment
This study confirms that cigarette consumption during pregnancy reduces weight, head circumference, and length of the newborn infant. Placental weight is not affected by smoking. These results are consistent
CIGARETTE CONSUMPTION (/day) Fig. 2. Effect of caffeine-cigarette interaction on placental weight. Means ( ±: SEM) are adjusted for the length of gestation. (0 = Caffeine consumption <300 mg/day; ~ = caffeine consumption "'300 mg/day.)
18 17
0
1
a:
16 15 14 0
1-14
15+
CIGARETTE CONSUMPTION (/day) Fig. 3. Effect of caffeine-cigarette interaction on placental ratio. Means ( ±: SEM) are adjusted for the length of gestation and sex of newborns. (0 = Caffeine consumption <300 mg/day; IZ;I = caffeine consumption "'300 mg/day.)
with those of other studies.' Since placental weight is stable and birth weight diminishes with cigarette consumption, an increase in the placental ratio results. This effect was also observed by Wingerd et a!. 14 Wilson 15 suggested that smoking causes chronic fetal hypoxia, which results in a reduction of fetal growth, and that the relative placental hypertrophy is compensatory. Even if cigarette consumption does not seem to affect placental growth negatively in terms of weight, it has been shown that it produces a reduction of placental blood fl.ow 3 • 4 and that it increases morphologic anomalies of the placenta. 16 Many changes in the ultrastructure of the placenta of smokers have been observed 17 and might induce a decrease in metabolism and diffusion through the placental barrier.
1240
Beaulac-Baillargeon and Desrosiers
Linn et al. 1" concluded that independent of cigarette consumption, the incidence of low birth weight was not affected by maternal consumption of coffee or tea. Van den Berg 10 observed that the percentage of babies weighing 2500 gm or less was 9% in smokers who consumed more than six cups of coffee per day. Because 4.4% of the nonsmokers consumed the same quantity of coffee daily and because the latter percentage was equal to that observed in the total cohort, he attributed the difference entirely to cigarette consumption without considering that this effect might be caused by the association between caffeine and cigarettes. These authors have calculated caffeine consumption in the number of cups of coffee per day without taking into account the eventual consumption of decaffeinated coffee. In addition, tea consumption was transformed into cups of coffee without correcting for its smaller caffeine content. Other sources of caffeine, such as cola and chocolate, were not considered. The results of our study show that caffeine consumption during pregnancy has a negative effect on the birth weight of babies born to smokers of 15 or more cigarettes per day. Doses of 300 mg or more of caffeine per day in these women accentuate the decrease caused by smoking. It is the first time that such an effect has been demonstrated. Moreover, the interaction between caffeine consumption and smoking affects placental weight in a particular manner. For women who consume less than 300 mg of caffeine/day, the placenta could compensate for the effect of cigarettes by real hypertrophy. Nevertheless, for women who smoke 15 or more cigarettes a day and consume 300 or more mg of caffeine daily, placental weight is reduced compared with women who consume less caffeine. To explain the underlying mechanisms of the caffeine-cigarette interaction on birth and placental weights, further microscopic and ultrastructural studies are required since morphometric measures alone do not permit complete definition of the functional capacity of the placenta. We thank the direction and the staff from the Departments of Gynecology and Obstetrics of Centre Hospitalier Chauveau and Hopital St-Franc;:ois d'Assise for their collaboration in this study.
November 1987 Am J Obstet Gynecol
REFERENCES l. Butler NR, Goldstein H, Ross EM. Cigarette smoking in pregnancy: its influence on birth weight and perinatal mortality. Br Med J 1972;2: 127-30. 2. Miller HC, Hassanein K. Maternal smoking and fetal growth of full term infants. Pediatr Res 1974;8:960-3. 3. Mochizuki M, Maruo T, Masuko K, Ohtsu T. Effects of smoking on fetoplacental-maternal system during pregnancy. AM J OBSTET GvNECOL 1984; 149:413-20. 4. Philipp K, Pateisky N, Endler M. Effects of smoking on uteroplacental blood flow. Gynecol Obstet Invest 1984; 17:179-82. 5. Socol ML, Manning FA, Murata Y, Druzin ML. Maternal smoking causes fetal hypoxia: experimental evidence. AM J 0BSTET GYNECOL 1982;142:214-8. 6. Jones KL, Smith DW. Recognition of the fetal alcohol syndrome in early infancy. Lancet 1973;7836:999-1001. 7. Little RE. Moderate alcohol use during pregnancy and decreased infant birth weight. Am J Public Health 1977;67: 1154-6. 8. Aldridge A, Bailey J, Neims AH. The disposition of caffeine during pregnancy and after pregnancy. Semin Perinatoll981;5:310-3. 9. Kirkinen P, Jouppila P, Koivula A, Vuori J, Puukka M. The effect of caffeine on placental and fetal blood flow in human pregnancy. AM j 0BSTET GYNECOL 1983;147: 939-42. 10. Van den Berg BJ. Epidemiologic observations of prematurity: effects of tobacco, coffee and alcohol. In: Reed DM, Stanley FJ, eds. The epidemiology of prematurity. Baltimore: Urban & Schwarzenberg, 1977:157-76. II. Beaulac-Baillargeon L, Desrosiers C. Profil de Ia consommation de cafeine, de cigarettes et d'alcool par les femmes quebecoises pendant Ia grossesse. L'Union Med 1986; Ill: 813-7. 12. Gilbert RM, Marshman JA, Schweider M, Berg R. Caffeine content of beverages as consumed. Can Med Assoc J 1976; 114:205-8. 13. Graham DM. Caffeine-its identity, dietary sources, intake and biological effects. Nutr Rev 1978;36:97-102. 14. Wingerd J, Christianson R, Lovitt WV, Schoen E.J. Placental ratios in white and black women: relation to smoking and anemia. AMj 0BSTET GYNECOL 1976;124:671-5. 15. Wilson EW. The effect of smoking in pregnancy on the placental co-efficient. N Z Med.J 1971;74:384-5. 16. Miller HC, Hassanein K, Hensleigh PA. Fetal growth retardation in relation to maternal smoking and weight gain in pregnancy. AM j 0BSTET GYNECOL 1976; 125:55-60. 17. Asmussen I. Ultrastructure of the villi and fetal capillaries in placentas from smoking and nonsmoking mothers. Br J Obstet Gynaecol 1980;87:239-45. 18. Linn S, Schoenbaum SC, Monson RR, Rosner B, Stubblefield PG, Ryan KJ. No association between coffee consumption and adverse outcomes of pregnancy. N Eng! J Med 1982;306:141-5.
Key words: Caffeine, cigarette smoking, fetus, pregnancy, fetal growth The caffeine, cigarette, and alcohol consumption of mothers during pregnancy has been the subject of many studies of its effects on fetal growth. Because these substances can easily reach the fetus by placental transfer, it is of prime importance to study the effect of interaction of these substances on fetal growth. The influence of smoking on birth weight is well known. In fact, birth weight is reported to be inversely proportional to the number of cigarettes smoked by the mother, and this effect is independent on other factors related to the variation in birth weight. 1 The length of the newborn is also reduced when the mother smokes, and some authors have observed a decrease in head circumference! Placental weight is not affected by smoking, but the number of cigarettes smoked has been correlated with hypovascular and atrophic microscopic changes in placental villi3 and deterioration of placental perfusion type! The reduction in uteroplacental circulation caused by the vasoconstricting effect of nicotine4 and the relative fetal hypoxia caused by the increase of carbon monoxide concentrations in the fetus 5 are two factors responsible for the decrease in fetal growth observed among babies born to smokers. On the other hand, maternal alcoholism has been related to a series of physical and neurologic abnormalities of newborns that in 1973 Jones and Smith6 termed "fetal alcohol syndrome." A significant clinical reduction of newborn weight is observed only at a con-
From the Ecole de Pharmacie, Universite Laval. Received for publication August 12, 1986; revised june 5, 1987; accepted june 24, 1987. Reprint requests: Louise Beaulac-Baillargeon, Ph.D., Ecole de Pharmacie, Universite Laval, Ste-Foy, Quebec, G1K 7P4, Canada
1236
sumption level higher than 30 ml of absolute alcohol per day. 7 During pregnancy, the elimination kinetic profile of caffeine was reported to be modified. In fact, Aldridge et al. 8 reported an increase in the half-life of caffeine from 5.3 hours before pregnancy to 18.1 hours during the last two trimesters of pregnancy. These values return to normal soon during the postpartum period. This increase in the half-life of caffeine in the mother coincides with the period during which fetal growth is exponential so that an accumulation of this substance represents a potential risk for the fetus and the placenta, which is responsible for the transfer of nutrients. To date, no well-controlled study has found that caffeine influences human birth weight. However, it has been demonstrated that a unique dose of 200 mg of caffeine, which is equivalent to two to three cups of coffee, administered to the mother significantly reduces blood flow in placental villi and increases maternal serum epinephrine levels. 9 This effect of caffeine on placental circulation is quite similar to that observed after cigarette consumption. Smoking is often associated with high caffeine intake, and this has also been confirmed for pregnant women. 10 However, the influence of that concomitant intake on fetal growth was not fully investigated. The association of these two consumption habits could represent a greater risk of disturbing the full development of the fetus. However, to date research has not fully investigated this hypothesis because the effects of caffeine and cigarette consumption on fetal growth have been studied separately. The effects of caffeine and cigarettes on placental circulation raise the assumption that these substances
Volume 157 Number 5
can act in an additive or synergistic way on fetal growth. The main objective of this study was to evaluate the independent effect and the interaction of caffeine and cigarette on weight, head circumference, length of the newborn, placental weight, and placental ratio (placental weight/birth weight x 100). Methodology
All women delivering at either Centre Hospitalier Chauveau or Hopital St-Fran~ois d'Assise in Quebec City, Canada, during a 5-month period from February to July 1985 were requested to complete a questionnaire on their caffeine, cigarette, and alcohol consumption habits during pregnancy. Of these 35 women refused to participate while 1014 women entered the study. This low refusal rate of 3% is not thought to bias the results. Standardized verbal and written information was given along with the questionnaires. Data collection was completed by a review of the medical and obstetric records of the mother and baby. The consumption pattern of caffeine, cigarettes, and alcohol of these women during pregnancy has been the subject of a previous publication.'' The questionnaire gathered information about the socioeconomic status, participation at prenatal classes, parity, obstetric history, weight before and at the end of pregnancy, height, health problems, and pattern of consumption of caffeine, cigarettes, and alcohol during pregnancy. The reported intake of these substances represents an average across pregnancy. The caffeine content of coffee, tea, cola, and chocolate used is that reported by the quantitative analysis of Gilbert et al., 12 and the publication of Graham." The amount of caffeine varied depending on the substance consumed: coffee: expresso = 170 mg/cup of 180 ml, filter or percolator = 110 mg/cup, instant = 75 mg/cup, and decaffeinated = 1 mg/cup; tea: bag = 50 mg/cup and leaves or instant == 30 mg/cup; chocolate = 40 mg/bar of 40 gm; cola: diet or regular == 45 mg/180 ml. These values were used to calculate total caffeine consumption by adding the intake from coffee, tea, cola, and chocolate. Cigarette consumption represents the number of cigarettes smoked per day. Alcohol consumption was calculated in milliliters of absolute alcohol consumed per day. Beer (350 ml), wine (120 ml), appetizer (60 ml), and spirits/digestive (30 ml) contained 15, 13, 13, and 15 ml, respectively, of absolute alcohol per serving. To study the influence of caffeine, cigarette, and alcohol consumption on fetal growth and exclude other major factors affecting fetal growth, it was necessary to exclude 101 subjects from the results analysis: multiple pregnancies (n = 16), mothers suffering from preeclampsia and chronic hypertension (n = 51), diabetes (n = 8), organic heart disease (n = 5), congenital
Caffeine-cigarette interaction on fetal growth
1237
Table I. Physical characteristics of mothers and newborns Characteristics
Mothers Weight before pregnancy (kg) Weight at the end of pregnancy (kg) Weight gain (kg) Height (em) Length of gestation (wk) Placental weight Newborns Birth weight (gm) Head circumference (em) Length (em)
Mean± SD
56.3 69.6 13.3 160.7 39.4 526.1
± 9.7 ± 10.5 ± 4.4 ± 6.0 ± 1.6 ± 121.0
3352.1 ± 496.3 34.5 ± 1.5 51.0 ± 2.3
spherocytosis (n = 1), Val Kecklinhausen (hyperparathyroidia) (n = 1), hypoglycemia (n = 3), babies with major malformations (n == 11), and stillbirths (n == 5). Statistical analysis was done by means of the statistical analysis system. Level of significance was fixed at 0.05. Analysis of variance for unbalanced groups was done by the introduction of independent variables: caffeine, cigarette, and alcohol consumption and combinations of these variables adjusted for length of gestation and the sex of the babies. The consumption of these substances was divided as follow: alcohol: <2, 2 to 5.9, ;;.6 ml of absolute alcohol per day; cigarettes: 0, 1 to 14, ;;.IS/day; and caffeine <300 and ;;.300 mg/day. Dependent variables included birth weight, head circumference, length of newborns, placental weight, and placental ratio. Tukey-Kramer's test for unbalanced groups was used to define the influence of different levels of consumption on fetal growth. Reported means for birth weight, head circumference, and length were adjusted for length of gestation and the sex of the babies, both of which were unevenly distributed in the subgroups studied. Placental weight was adjusted for length of gestation only since it did not vary with the sex of the babies. Results
After data collection, 101 subjects were excluded since they did not meet the inclusion criteria mentioned. Results are presented for the remaining 913 subjects. Physical characteristics of babies and mothers are presented in Table I. Alcohol at doses consumed by women in the present study had no significant effect on the studied parameters. Mean absolute alcohol consumption of this population was 1.5 ml (SD = ± 3.2 ml) per day, which represented about three servings of alcoholic beverages per month. Subjects were then grouped according to caffeine and cigarette consumption. Caffeine consumption was
1238
Beaulac-Baillargeon and Desrosiers
November 1987 Am J Obstet Gynecol
Table II. Groups according to the caffeine and cigarette consumption of the mothers Consumption level Caffeine (mglday)
<300 <300 <300 :;;.300 :;;.300 :;;.300
I
Mean consumption
Cigarettes (No.I day)
No. of subjects
0 1-14 >15 0 1-14 >15
536 128 114 42 24 69
divided into two categories: (1) subjects whose consumption was less than 300 mg/day and (2) those whose daily consumption was greater than 300 mg/day. Among these two categories, women were divided according to their cigarette consumption, either 0, one to 14, or 15 or more per day. For each of these six groups, Table II shows the number of subjects and the mean daily consumption of caffeine and cigarettes. At the end of pregnancy, smokers represented 36.7% of the study population. Whereas 15.8% of the women who smoked one to 14 cigarettes/day and 37.7% of those who smoked 15 or more cigarettes daily consumed 300 mg or more of caffeine daily, only 7.3% of nonsmokers belonged to the latter category. Among the 135 women (14.8% of the population) whose daily consumption of caffeine rose to 300 mg or more, 68.9% smoked. Birth weight. Analysis of variance revealed an influence of cigarette (F = 32.3, p < 0.01) and caffeinecigarette interaction (F = 5.8, p < 0.01) on birth weight. For nonsmokers, mean birth weight of babies adjusted for length of gestation and sex of babies was 3441.9 gm (SEM = 16.8 gm). This weight was significantly reduced (p < 0.05) for babies born to mothers who smoked; it was 3204.4 gm (SEM = ± 39.0 gm) and 3141.7 gm (SEM = ±28.8 gm) for a consumption of one to 14 and 15 or more cigarettes per day, respectively. Fig. 1 presents mean birth weight of babies divided according to maternal caffeine and cigarette consumption. All four smoking subgroups had mean birth weights significantly lower than both nonsmoking subgroups. For nonsmokers and women who smoked one to 14 cigarettes daily, the birth weight was not statistically different with a caffeine consumption less than or greater than 300 mg/day. However, among women smoking 15 cigarettes or more per day, the birth weight of babies of those women who consumed 300 mg or more caffeine daily was significantly (p < 0.05) lighter than that of babies whose mothers consumed less caffeine. The difference in birth weight of babies between these two groups was 206 gm (SEM = ±57.7 gm). When we consider both caffeine and cigarette consumption, the birth weight of babies born to women
Caffeine (mglday) mean± SD
100.8 119.3 135.2 407.8 423.9 442.2
± 83.2 ± 90.6
± 86.5
± 222.7 ± 116.9 ± 142.7
I
Cigarettes (No./day) mean± SD
0 7.6 ± 3.8 20.3 ± 5.4 0 9.3 ± 3.4 22.8 ± 6.8
who smoked 15 cigarettes or more per day was significantly reduced if the mother consumed 300 mg of caffeine or more daily. Head circumference. Head circumference was influenced by cigarette consumption (F = 8.2, p < 0.01) but not by caffeine consumption. No caffeine-cigarette interaction was found. Head circumference of babies born to nonsmokers was 34.6 em (SEM = ± 0.06 em), and it diminished significantly (p < 0.05) for babies of mothers who smoked, being 34.3 em (SEM = ± 0.11 em) and 34.1 em (SEM = ± 0.11 em) in newborns whose mothers smoked one to 14 and 15 or more cigarettes per day, respectively. Length. Length of babies was also influenced by cigarette consumption (F = 17.8, p < 0.01), but again there was no interaction between cigarette smoking and caffeine consumption. Babies born to nonsmokers had a mean length of 51.3 em (SEM = ± 0.09 em), whereas those born to mothers who smoked one to 14 and 15 or more cigarettes per day were 50.7 em (SEM = ±0.16 em) and 50.3 em (SEM = ±0.15 em) long, respectively. Placental weight. Placental and birth weights underwent a caffeine-cigarette interaction (F = 8.9, p < 0.01). There was no significant difference between placental weight of nonsmokers and placental weight of the two other groups of smokers. However, when we compared this weight between smokers groups according to their caffeine consumption, a dose-related interaction was seen. Fig. 2 shows mean placental weight adjusted for length of gestation for each consumption group. Among women who consumed less than 300 mg of caffeine daily, placental weight was significantly higher for those who smoked 15 cigarettes or more per day (mean ± SEM = 566.6 ± 12.5 gm) as opposed to nonsmokers (mean ± SEM = 517.5 ± 4.8 gm) (p < 0.05) and smokers of one to 14 cigarettes per day (mean ± SEM = 521.2 ± 10.4 gm). Among women who smoked 15 cigarettes or more per day, placental weight was significantly reduced for those who consumed 300 mg or more of caffeine daily compared with those whose caffeine consumption was
Caffeine-cigarette interaction on fetal growth 1239
Volume 157 Number 5
3700 575
3600
--
3500 ........
~ 1:I: C!J
w
3:
3400
550
01
3300
1:I:
525
w 3:
500
C!J
3100
475
3000 2900
450
2800
0
1-14
15+
CIGARETTE CONSUMPTION (/day) Fig. 1. Effect of caffeine-cigarette interaction on birth weight. Means ( ±: SEM) are adjusted for the length of gestation and sex of newborns. (0 = Caffeine consumption <300 mg/day; fZl = caffeine consumption "'300 mg/day.)
less than 300 mg/day (p < 0.05); placental weight was 502.3 gm (SEM = ± 13.8 gm) and 566.6 gm (SEM = ± 12.5 gm), respectively. Placental ratio. The ratio of placental weight to birth weight was affected by cigarette consumption (F = 12.8, p < 0.01) and underwent a caffeinecigarette interaction (F = 3.3, p < 0.05) (Fig. 3). Placental ratio increased with cigarette consumption, going from 15.1 (SEM = ±0.12) for nonsmokers to 16.5 (SEM = ± 0.26) and 17.3 (SEM = ± 0.26) for women who smoked one to 14 and 15 or more cigarettes per day, respectively. This reflects the increase in placental weight and the decrease in birth weight observed with an increase in the number of cigarettes smoked. Considering both caffeine and cigarette consumption, the placental ratio increased significantly (p < 0.05) for every consumption level of cigarettes for women who consumed less than 300 mg of caffeine/day. However, for women who consumed 300 mg or more of caffeine daily, there was no significant difference in placental ratio according to cigarette consumption. Among women who smoked 15 or more cigarettes a day, the placental ratio was not statistically different with a caffeine consumption of less than or greater than 300 mg/day, since both birth and placental weights diminished when caffeine consumption was 300 mg or higher. Comment
This study confirms that cigarette consumption during pregnancy reduces weight, head circumference, and length of the newborn infant. Placental weight is not affected by smoking. These results are consistent
CIGARETTE CONSUMPTION (/day) Fig. 2. Effect of caffeine-cigarette interaction on placental weight. Means ( ±: SEM) are adjusted for the length of gestation. (0 = Caffeine consumption <300 mg/day; ~ = caffeine consumption "'300 mg/day.)
18 17
0
1
a:
16 15 14 0
1-14
15+
CIGARETTE CONSUMPTION (/day) Fig. 3. Effect of caffeine-cigarette interaction on placental ratio. Means ( ±: SEM) are adjusted for the length of gestation and sex of newborns. (0 = Caffeine consumption <300 mg/day; IZ;I = caffeine consumption "'300 mg/day.)
with those of other studies.' Since placental weight is stable and birth weight diminishes with cigarette consumption, an increase in the placental ratio results. This effect was also observed by Wingerd et a!. 14 Wilson 15 suggested that smoking causes chronic fetal hypoxia, which results in a reduction of fetal growth, and that the relative placental hypertrophy is compensatory. Even if cigarette consumption does not seem to affect placental growth negatively in terms of weight, it has been shown that it produces a reduction of placental blood fl.ow 3 • 4 and that it increases morphologic anomalies of the placenta. 16 Many changes in the ultrastructure of the placenta of smokers have been observed 17 and might induce a decrease in metabolism and diffusion through the placental barrier.
1240
Beaulac-Baillargeon and Desrosiers
Linn et al. 1" concluded that independent of cigarette consumption, the incidence of low birth weight was not affected by maternal consumption of coffee or tea. Van den Berg 10 observed that the percentage of babies weighing 2500 gm or less was 9% in smokers who consumed more than six cups of coffee per day. Because 4.4% of the nonsmokers consumed the same quantity of coffee daily and because the latter percentage was equal to that observed in the total cohort, he attributed the difference entirely to cigarette consumption without considering that this effect might be caused by the association between caffeine and cigarettes. These authors have calculated caffeine consumption in the number of cups of coffee per day without taking into account the eventual consumption of decaffeinated coffee. In addition, tea consumption was transformed into cups of coffee without correcting for its smaller caffeine content. Other sources of caffeine, such as cola and chocolate, were not considered. The results of our study show that caffeine consumption during pregnancy has a negative effect on the birth weight of babies born to smokers of 15 or more cigarettes per day. Doses of 300 mg or more of caffeine per day in these women accentuate the decrease caused by smoking. It is the first time that such an effect has been demonstrated. Moreover, the interaction between caffeine consumption and smoking affects placental weight in a particular manner. For women who consume less than 300 mg of caffeine/day, the placenta could compensate for the effect of cigarettes by real hypertrophy. Nevertheless, for women who smoke 15 or more cigarettes a day and consume 300 or more mg of caffeine daily, placental weight is reduced compared with women who consume less caffeine. To explain the underlying mechanisms of the caffeine-cigarette interaction on birth and placental weights, further microscopic and ultrastructural studies are required since morphometric measures alone do not permit complete definition of the functional capacity of the placenta. We thank the direction and the staff from the Departments of Gynecology and Obstetrics of Centre Hospitalier Chauveau and Hopital St-Franc;:ois d'Assise for their collaboration in this study.
November 1987 Am J Obstet Gynecol
REFERENCES l. Butler NR, Goldstein H, Ross EM. Cigarette smoking in pregnancy: its influence on birth weight and perinatal mortality. Br Med J 1972;2: 127-30. 2. Miller HC, Hassanein K. Maternal smoking and fetal growth of full term infants. Pediatr Res 1974;8:960-3. 3. Mochizuki M, Maruo T, Masuko K, Ohtsu T. Effects of smoking on fetoplacental-maternal system during pregnancy. AM J OBSTET GvNECOL 1984; 149:413-20. 4. Philipp K, Pateisky N, Endler M. Effects of smoking on uteroplacental blood flow. Gynecol Obstet Invest 1984; 17:179-82. 5. Socol ML, Manning FA, Murata Y, Druzin ML. Maternal smoking causes fetal hypoxia: experimental evidence. AM J 0BSTET GYNECOL 1982;142:214-8. 6. Jones KL, Smith DW. Recognition of the fetal alcohol syndrome in early infancy. Lancet 1973;7836:999-1001. 7. Little RE. Moderate alcohol use during pregnancy and decreased infant birth weight. Am J Public Health 1977;67: 1154-6. 8. Aldridge A, Bailey J, Neims AH. The disposition of caffeine during pregnancy and after pregnancy. Semin Perinatoll981;5:310-3. 9. Kirkinen P, Jouppila P, Koivula A, Vuori J, Puukka M. The effect of caffeine on placental and fetal blood flow in human pregnancy. AM j 0BSTET GYNECOL 1983;147: 939-42. 10. Van den Berg BJ. Epidemiologic observations of prematurity: effects of tobacco, coffee and alcohol. In: Reed DM, Stanley FJ, eds. The epidemiology of prematurity. Baltimore: Urban & Schwarzenberg, 1977:157-76. II. Beaulac-Baillargeon L, Desrosiers C. Profil de Ia consommation de cafeine, de cigarettes et d'alcool par les femmes quebecoises pendant Ia grossesse. L'Union Med 1986; Ill: 813-7. 12. Gilbert RM, Marshman JA, Schweider M, Berg R. Caffeine content of beverages as consumed. Can Med Assoc J 1976; 114:205-8. 13. Graham DM. Caffeine-its identity, dietary sources, intake and biological effects. Nutr Rev 1978;36:97-102. 14. Wingerd J, Christianson R, Lovitt WV, Schoen E.J. Placental ratios in white and black women: relation to smoking and anemia. AMj 0BSTET GYNECOL 1976;124:671-5. 15. Wilson EW. The effect of smoking in pregnancy on the placental co-efficient. N Z Med.J 1971;74:384-5. 16. Miller HC, Hassanein K, Hensleigh PA. Fetal growth retardation in relation to maternal smoking and weight gain in pregnancy. AM j 0BSTET GYNECOL 1976; 125:55-60. 17. Asmussen I. Ultrastructure of the villi and fetal capillaries in placentas from smoking and nonsmoking mothers. Br J Obstet Gynaecol 1980;87:239-45. 18. Linn S, Schoenbaum SC, Monson RR, Rosner B, Stubblefield PG, Ryan KJ. No association between coffee consumption and adverse outcomes of pregnancy. N Eng! J Med 1982;306:141-5.