Despite the hundreds of exposure–disease associations studied by perinatal epidemiologists, few have become accepted as causal. Maternal smoking and its effects on preterm delivery and fetal growth restriction are two such causal associations, whereas direct maternal smoking and spontaneous abortion is not.1 The causal role of environmental tobacco smoke (ETS) on perinatal disease has proved particularly difficult to document despite the plausibility of the hypothesis. The current evidence for an association of ETS with spontaneous abortion remains mixed.
In this issue of Epidemiology, George et al2 report an association between ETS, as assessed by urinary cotinine level, and spontaneous abortion. Cotinine is a widely used biomarker for cigarette smoking, and this is the first study to use it to assess ETS exposure as a risk factor for spontaneous abortion.
George et al report that the increased risks are similar for normal and abnormal karyotype. This may not be an implausible result given that ETS might induce the spontaneous abortion of abnormal karyotypes that survive very early pregnancy as well as the abortion of normal pregnancies. Perhaps less plausible is the finding that spontaneous abortion risk with exposure to ETS is similar, in these data, to the risk with direct smoking, an association with an order-of-magnitude greater level of cotinine exposure. In contrast, the associations of ETS on fetal growth restriction are weaker than the associations found for direct smoking,3 which seems more plausible. So what might account for the surprisingly strong risk found for ETS in the current report?
The authors used a case–control study with many strong design features: it is well powered, measured several potentially important covariates, validated the spontaneous abortions, used an objective biomarker for exposure, accounted for possible bias in women seeking induced abortion, and stratified on fetal karyotype. Given that the authors have dealt well with the usual methodologic problems, what other possibilities might explain their findings?
Nicotine is metabolized to cotinine by CYP (including CYP1A2) activity,4 which declines markedly in later pregnancy.5 This decline leads to slower nicotine metabolism and lower rates of cotinine for any given exposure as pregnancy progresses. Systematic bias in cases being interviewed earlier in pregnancy than controls would bias the cotinine in the observed direction. However, in this study, gestational age was closely matched,6 (Table 1) so this does not seem a likely confounding factor.
Residual confounding can be responsible for artifactually inflated risks if the confounders are not modeled precisely.7 In this article, several potentially important confounders are considered: older maternal age, history of spontaneous abortion, and caffeine consumption, all of which are associated with both an increased risk of spontaneous abortion and higher cotinine levels. The article is silent as to how these factors are parameterized in the modeling; simple categorization (as shown in the tables) may have been insufficient to eliminate their effects completely. All of these would tend to bias in favor of showing risk of spontaneous abortion with higher cotinine.
Biomarkers incur their own measurement problems, some of which are exemplified in this article. Cotinine is a measure of recent nicotine exposure and has a half-life of some 17 hours; this pattern leads to gradual accumulation of cotinine during the day.8 In the study by George et al, the case urines were systematically collected later in the day, which would lead to overestimation of risk. This is particularly true for ETS exposure, which comes from commuting, work and social environments. Moreover, the true etiologic event (fetal death) can precede the spontaneous abortion by many days. In this study, cotinine was measured several weeks after the abortion was diagnosed and even longer after fetal death. Is it possible that the case women, knowing they were no longer pregnant, may have been less inclined to avoid ETS than the still-pregnant controls?
The role of caffeine as a residual confounder in this study remains possible. The same study group has previously reported from the same case–control study6 an increased risk of spontaneous abortion among high caffeine consumers (odds ratio = 2.2; 95% confidence interval = 1.3–3.8), which is stronger than the ETS association reported here.
Also in the same database, high CYP1A2 activity has been reported to independently increase risk for spontaneous abortion.9 Because nicotine is more quickly metabolized to cotinine with increased CYP1A2 activity,10 the previously identified associations in this data between high CPY1A2 activity and spontaneous abortion would predict an association between high cotinine and spontaneous abortion for any given level of actual ETS exposure. Unfortunately, measures of ETS exposure other than cotinine are not reported in this article.
Another study has shown poor agreement of urinary cotinine with self-report or with personal ETS nicotine monitors,11 perhaps reflecting confounding due to differential drug metabolism. Additional evidence now also suggests that high CYP1A2 activity may be independently associated with perinatal outcomes.12 As mentioned, the same research group has focused on caffeine exposure in this same study to suggest that high CPY1A2 activity “may increase the risk of spontaneous abortion, independently or by modifying the effect of caffeine.” They similarly concluded regarding another important metabolic process (NAT2) that “slow acetylators may be at elevated risk for spontaneous abortion.”9 These observations raise the question of whether CYP1A2 metabolism, known to be increased by nicotine exposure,13 may be responsible for the observed effects on the same spontaneous abortion cases associated with CYP1A2 variants and caffeine exposure.9 Indeed, caffeine consumption and metabolism, the induction of CYP1A2 activity by nicotine, and the common concurrent exposure to caffeine and nicotine are so inextricably linked that it is difficult in this article (or in most other related research) to demonstrate the independent effects of each.
CYP1A2 metabolism is increased by many other dietary, pharmacologic, and environmental factors, including Brassica (Cruciferae) vegetables. However, CYP1A2 activity is also influenced by various recently identified polymorphisms, including CYP1A2*1B, CYP1A2*1C, and CYP1A2*1F. 5 Some of these polymorphisms have been associated with the activation of carcinogens present in tobacco smoke.14 If those carcinogens are also teratogenic, this association suggests possible mechanisms for how high metabolism might increase the risk for spontaneous abortion with ETS exposure.
Further research is needed to explore whether high-risk polymorphisms and increased CYP1A2 activity are independently associated with spontaneous abortion, in which case high cotinine levels may be a marker of increased CYP1A2 activity. CYP1A2 is one of the human hepatic cytochrome P450 enzymes associated with activation of a large number of potentially teratogenic environmental agents, including many therapeutic drugs, aromatic or heterocyclic amines, caffeine, hormones, organochlorines, and polybrominated biphenyls.
George et al have already genotyped NAT2, and many of the relevant polymorphisms in CPP1A2 would appear to be available for genotyping.5 Analysis of polymorphisms in CYP genes in combination with smoking data is providing important insights into the etiology of cancer.15 A full exploration of the relevant polymorphisms with ETS may offer equally useful information on teratogenic risk.
This commentary has benefited from many hours of conversation over several years with my colleague, Laura Grosso.
ABOUT THE AUTHOR
MICHAEL B. BRACKEN is the Susan Dwight Bliss Professor of Epidemiology at Yale University. He is currently President of the Society for Epidemiologic Research and is former President of the American College of Epidemiology. He holds a Research Fellowship at Green College, Oxford University.
1. US Department of Health and Human Services. The Health Consequences of Smoking: A Report of the Surgeon General
. Atlanta: Centers for Disease Control and Prevention, Office on Smoking and Health; 2004.
2. George L, Granath F, Johansson ALV, et al. Environmental tobacco smoke and risk of spontaneous abortion. Epidemiology
3. Sadler L, Belanger K, Saftlas A, et al. Environmental tobacco smoke exposure and small-for-gestational age birth. Am J Epidemiol
4. Cashman JR, Park SB, Yang Z-C, et al. Metabolism of nicotine by human liver microsomes: stereoselective formation of trans-nicotine N′-oxide. Chem Res Toxicol
5. Grosso LM, Bracken MB. Caffeine metabolism, genetics, and perinatal outcomes: a review of exposure assessment considerations during pregnancy. Ann Epidemiol
6. Cnattingius S, Signorello LB, Anneren G, et al. Caffeine intake and the risk of first-trimester spontaneous abortion. N Engl J Med
7. Benedetti A, Abrahamowicz M. Using generalized additive models to reduce residual confounding. Stat Med
8. Benowitz NL. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol Rev
9. Signorello LB, Nordmark A, Granath F, et al. Caffeine metabolism and the risk of spontaneous abortion of normal karyotype fetuses. Obstet Gynecol
10. Nakajima M, Iwata K, Yamamoto T, et al. Nicotine metabolism in liver microsomes from rats with acute hepatitis or cirrhosis. Drug Metab Dispos
11. O'Connor TZ, Holford TR, Leaderer BP, et al. Measurement of exposure to environmental tobacco smoke in pregnant women. Am J Epidemiol
12. Grosso LM, Triche EW, Belanger K, et al. Caffeine metabolites in umbilical cord blood, cytochrome P-450 1A2 activity, and intrauterine growth restriction. Am J Epidemiol
13. Pavanello S, Simioli P, Lupi S, et al. Exposure levels and cytochrome P450 1A2 activity, but not N
-acetyltransferase, glutathione S
-transferase (GST) M1
, influence urinary mutagen excretion in smokers. Cancer Epidemiol Biomarkers Prev
14. Pavanello S, Pulliero A, Lupi S, et al. Influence of the genetic polymorphism in the 5′-noncoding region of the CYP1A2
gene on CYP1A2 phenotype and urinary mutagenicity in smokers. Mutat Res
15. Bartsch H, Nair U, Risch A, et al. Genetic polymorphism of CYP genes, alone or in combination, as a risk modifier of tobacco-related cancers. Cancer Epidemiol Biomarkers Prev