Air pollution, one of the most prevalent environmental hazards, which affects up to 100 % of the population living in urban areas, has gained considerable interest because of the multiple adverse effects reported on human health [19]. It is a known risk factor for cardiovascular [20–22] and respiratory disease [23–25], and the International Agency for Research on Cancer (IARC), the division of the World Health Organization (WHO) that coordinates cancer research, has classified outdoor air pollution as carcinogenic to humans [26].

The associations between air pollution and adverse reproductive outcomes have also been described, including a restricted fetal growth leading to low birth weight newborns small for gestational age, and preterm birth [27–31].

Air pollutants that have mainly been studied in relation to adverse birth outcomes are particulate matter with aerodynamic diameter of less than 2.5 μm (PM 2.5), NO₂, CO, SO₂, PAHs (polycyclic aromatic hydrocarbons), and ozone (O3).

The associations between air pollution and miscarriages also have been described by several authors. Faiz et al. examined the risk of fetal loss associated with ambient air pollution during pregnancy. Using live birth and fetal death data, the authors assigned daily concentrations of air pollution to each birth or fetal death. The relative odds of fetal loss in the first trimester were significantly increased with each 10-ppb increase in mean nitrogen dioxide concentration (OR = 1.16, 95 % CI 1.03–1.31) and each 3-ppb increase in mean sulphur dioxide concentration (OR = 1.13, 95 % CI 1.01–1.28) [32].

Mohorovic et al. conducted a retrospective epidemiological study to evaluate the role of environmental factors in miscarriages. Methaemoglobin in the bloodstream was used as the biomarker. The frequencies of miscarriages were significantly lower in the control than in the exposure period (p < 0.05) [33].

In a cohort study, living within 50 m of a road with maximum annual average daily traffic was significantly associated with spontaneous miscarriage among African Americans (OR = 3.11; 95 % CI 1.26–7.66) and non-smokers (OR = 1.47; 95 % CI 1.07–2.04) [34].

With regard to studies conducted in women undergoing IVF/ET, a significant increase in miscarriage rate among women in the quartile with higher exposure to PM10 (OR 5.05, 95 % CI 1.04–24.51) was observed by Perin et al. [35].

In a recent retrospective case–control study, 959 fetal losses and 959 normal intrauterine pregnancies (controls) were selected. The association between ambient air pollutants and fetal loss was examined. Logistic regression suggested that fetal loss within 14 weeks was associated with higher exposure to SO₂ (OR = 19.76, 95 % CI 2.34–166.71) in the first month of pregnancy [36].

Several mechanisms were reported in the literature with regard to the effect of air pollutants on pregnancy outcomes including miscarriages.

Animal data on the toxicity of caffeine have demonstrated teratogenicity from caffeine only at very high doses. The teratogenic effect of caffeine has been clearly demonstrated, for example in rodents. Consumption of caffeine by oral-gastric intubation or intraperitoneal injection with doses ranging from 6 to 250 mg/kg indicate that at a dose of 250 mg/kg, 50 % of the mothers die, and at 200 mg/kg survivors frequently develop seizures within minutes of dosing. Evidence of teratogenesis was seen when dosing rose above 75–80 mg/kg. This dose also resulted in a doubling of the number of fetal deaths. The incidence of congenital malformations was not significantly different from that in controls until a dose of 125 mg/kg was exceeded [47].

Caffeine can cross the placental and blood brain barriers and the human fetus may not have developed enzymes for detoxification of caffeine via demethylation [48]. It has been hypothesised that caffeine inhibits cyclic nucleotide phosphodiesterases with a consequent increase in cellular cyclic adenosine monophosphate (cAMP), [49] and the rise in cAMP may interfere with fetal cell growth and development [50].

Several studies have shown the association between caffeine exposure and miscarriage. In 1998, a meta-analysis of 12 studies, which compared a caffeine-exposed group (>150 mg/d) and controls (0–150 mg/d), concluded that the Mantel-Haenszel odds ratio (95 % CI) for spontaneous miscarriage in 42,988 pregnancies was 1.36 (1.29–1.45) [51].

In 2003, A case–control study of 474 women indicated that high caffeine consumption during pregnancy (>300 mg/day), in particular coffee consumption, is an independent risk factor for increased risk of miscarriage. Adjusted odds ratios were 1.94 [95 % CI 1.04–3.63] for 301–500 mg/day and 2.18 [95 % CI 1.08–4.40] for >500 mg/day [52].

In a prospective cohort study involving 3135 women, the relative risk for spontaneous miscarriage for women consuming over 150 mg of caffeine daily was 1.73 (p = .03) [53]. In a case–control study, Kline et al. [54] karyotyped 900 pregnancy losses prior to 28 weeks of gestation and employed 1423 controls. In women who consumed more than 225 mg of caffeine daily during pregnancy, the adjusted odds ratio for chromosomally normal losses versus controls was 1.9 (1.3–2.6), which was statistically significant.

A case–control study of 331 women with spontaneous miscarriage showed that caffeine intake before and during pregnancy was associated with an increased risk of fetal loss. After controlling for confounding factors, there was a strong association of caffeine intake during pregnancy and fetal loss, compatible with a linear trend on the logistic scale in which ORs increased by a factor of 1.22 (1.10–1.34) for each 100 mg of caffeine ingested daily during pregnancy. Consumption of less than 162 mg/day were not associated with an increased risk of fetal loss [55].

Another case–control study of 330 women with spontaneous miscarriages found that consumption of 375 mg or more caffeine per day during pregnancy may increase the risk of spontaneous miscarriage OR 2.21 (1.53–3.18) [56].

In a recent prospective cohort study that included 5132 women planning pregnancy, women reported their daily caffeine and caffeinated beverage consumption on questionnaires before conception and during early pregnancy; 732 women (14.3 %) had spontaneous miscarriages. In the preconception period, caffeine consumption was not materially associated with spontaneous miscarriage risk (Hazard Ratio comparing ≥300 with <100 mg/day: 1.09; 95 % CI 0.89–1.33). In early pregnancy, the Hazard Ratios for 100–199, 200–299, and ≥300 mg/day of caffeine consumption were 1.62 (95 % CI 1.19–2.22), 1.48 (95 % CI 1.03–2.13), and 1.23 (95 % CI 0.61–2.46), respectively, compared with <100 mg/day. They concluded that preconception caffeine consumption was not materially associated with an increased risk of spontaneous miscarriage, but consumption during early pregnancy was associated with a small increased risk, although the relation was not linear [57].

It is important to emphasise that in regard to caffeine consumption there may be several confounders that may be relevant: In a cohort study, an increased mean daily caffeine intake in women with spontaneous abortions was reported. But, heavier caffeine consumers were also significantly older and more likely to smoke cigarettes, which could have confounded the results of this study [58]. Successful pregnancies may also be more often associated with food aversion, nausea, and vomiting than pregnancies destined to result in miscarriage, and because coffee is one of the foods most commonly found unappealing under these circumstances, women with successful pregnancies may therefore decrease their coffee intake, whereas women destined to early pregnancy loss may not [59]. Therefore, even though several studies have reported an association between higher caffeine intake and spontaneous miscarriage, the relationship may not necessarily be causal.

Maternal cigarette smoking is a well-known cause associated with adverse reproductive outcomes including increases incidence of abruptio placentae, placenta previa, bleeding during pregnancy, premature rupture of membranes, and reduced fertility [60]. Exposed infants are more likely to be of low birth weight (<2500 g) and have twice the risk of infant mortality from all causes, specifically from sudden infant death syndrome [61].

Cigarette smoking contains plenty of toxic components . Nicotine, the main addictive compound, is a strong vasoconstrictor that reduces uterine and placental blood flow. Other toxic components include carbon monoxide, which binds to haemoglobin and decreases the availability of oxygen to the fetus and cyanide, which depletes vitamin B 12, a necessary cofactor for fetal growth and development [62]. Cigarette smoking is measured in the studies by self-reports or by urine analysis (urine cotinine concentrations).

Several studies have shown the association between cigarette smoking and miscarriage, but not all of them. In 1996, in a systematic review of the literature, including seven studies evaluating spontaneous miscarriage, suggested a small increased risk among female smokers (OR 0.83–1.8). The dose–response effect was consistent [63]. A retrospective study of 12,914 pregnancies found a significant increase in risk for spontaneous miscarriage with maternal cigarette smoking. The risk of spontaneous miscarriage for smokers was as much as 1.7 times that of the nonsmoker group [64].

Harlap et al. in a prospective study of 32,019 women, found after adjustment for alcohol use, that the only subgroup in which smoking had a significant adverse effect was in those women smoking more than two packs of cigarettes per day. In this group, the odds ratio for second-trimester pregnancy loss was 2.02 (95 % CI 1.01–4.01) [65].

In another prospective study of 970 women, the presence of cotinine in urine was independently associated with an increased risk of spontaneous miscarriage (OR 1.8; 95 % CI 1.3–2.6) [66].

Armstrong et al. analysed data of occupational factors and pregnancy outcome from 47,146 women, to examine the effects of cigarette smoking on pregnancy outcome. Clear and statistically significant associations were found between cigarettes and spontaneous miscarriage. If the associations were causal, 11 % of the spontaneous miscarriages could be attributed to smoking [67].

On the other hand, Wisborg et al. in a prospective study found no association between smoking and first- and second-trimester miscarriages. Adjustment for alcohol, coffee, maternal age, marital status, occupation, education, prepregnancy body mass index, and parity did not change the result substantially [68].

What about environmental tobacco smoke and passive smoking? The chemical exposure from passive smoking is qualitatively similar but quantitatively different from that of the smoker. The undiluted sidestream smoke contains many harmful chemicals in greater amounts than the inhaled cigarette smoke. For example, the amount of nicotine in the undiluted sidestream is seven times more [69]. But when it is diluted, the concentration in the air is low. In meta-analyses, the scientific evidence on the effects of preconception and prenatal exposure to environmental tobacco smoke on reproductive health has been described. The associations noted between passive exposure and spontaneous miscarriage are of similar magnitude as the associations found between spontaneous miscarriage and active smoking, although the effect of passive exposure might be expected to be much lower than that of active smoking [70].

In summary, the data evaluating smoking and miscarriage suggest an increased risk for early pregnancy loss that is dose dependent. The evidence on the effects of environmental tobacco smoke on spontaneous miscarriage is weak but a potential relationship between exposure and spontaneous miscarriage cannot be excluded.

Maternal alcohol consumption is known to be teratogenic and associated with fetal alcohol syndrome. Fetal alcohol syndrome includes growth restriction , a pattern of craniofacial anomalies, neurological effects , and behavioural effects [71]. In 1983, Kaufman described that ethanol consumption at the time of conception may be the cause of certain types of chromosomal defects commonly observed in human spontaneous miscarriages [72]. Animal data on the toxicity of ethanol have demonstrated the risk for pregnancy failure . Animals were given a 1.8 g/kg dose of ethanol once per week for the first 3, 6, or 24 weeks (full gestation) of pregnancy. Peak plasma ethanol levels ranged from 175 to 250 mg/dl. Weekly maternal exposure to this intoxicating dose of ethanol, starting early in pregnancy, did not influence risk of pregnancy failure during the first 30 days of gestation but appeared to be associated with an increased risk of miscarriage occurring between gestational days 30 and 160 [73]. Studies on the association between alcohol consumption and miscarriage have yielded inconsistent results:

Harlap et al. in a prospective study of 32,019 women, found that the adjusted relative risks of second-trimester losses (15–27 weeks) were 1.03 (not significant) for women taking less than 1 drinks daily, 1.98 (p < 0.01) for women taking 1–2 drinks daily, and 3.53 (p < 0.01) for women taking more than 3 drinks daily, compared with non-drinkers. 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 miscarriages or induced abortions [65].

In a large case–control study of spontaneous miscarriages (626 cases, 1,300 controls), the odds ratio for alcohol consumption of seven or more drinks per week was 1.9 (95 % CI 1.1–3.4) when adjusted for maternal smoking, passive smoking, and maternal age [74].

Armstrong et al. analysed data of occupational factors and pregnancy outcome from 47,146 women to examine the effects of alcohol on pregnancy outcome. Clear and statistically significant associations were found between alcohol consumption and spontaneous miscarriage (odds ratios increased on average by a factor of 1.26 (1.19–1.33) for each drink per day). If the associations were causal, 5 % of the spontaneous miscarriages could be attributed to alcohol consumption [67].