- Email: [email protected]
Effect of cigarette smoking upon reproductive hormones in women of reproductive age: a retrospective analysis
Reproductive BioMedicine Online (2010) 20, 861– 865
www.sciencedirect.com www.rbmonline.com
ARTICLE
Effect of cigarette smoking upon reproductive hormones in women of reproductive age: a retrospective analysis AL Waylen
a,*
, GL Jones b, WL Ledger
c
a University of Sheffield School of Medicine and Biomedical Sciences, Beech Hill Road, Sheffield S10 2RX, UK; b Health Services Research Section, ScHARR, Regent Court, 30 Regent Street, Sheffield S1 4DA, UK; c Academic Unit of Reproductive and Developmental Medicine, Jessop Wing Hospital, Tree Root Walk, Sheffield S10 2TJ, UK
* Corresponding author. E-mail addresses: [email protected], [email protected] (AL Waylen). Anna Waylon attended the University of Sheffield School of Medicine from September 2003 to July 2009 during which time she completed a Bachelor of Medical Science degree in the Academic Unit of Reproductive and Developmental Medicine. Her research involved investigating the effects of cigarette smoking on female reproductive hormones and on the clinical outcomes of assisted reproduction, and a qualitative assessment of women’s knowledge of factors that could affect their fertility.
Abstract There is continuing debate concerning the relationship between cigarette smoking and premature ovarian failure. The aim
of this retrospective data analysis was to investigate whether smoking has a measurable effect on early follicular serum concentrations of inhibin B hormone, FSH and anti-Mu ¨llerian hormone (AMH) in women of reproductive age. A database containing data on age, smoking status and serum concentrations of inhibin B, FSH and AMH was analysed. Pearson’s correlation coefficient was calculated to determine the correlation between hormone concentrations and age. One-way analysis of variance was used to determine any significant difference in age between smoking categories and a univariate general linear model was used to compare geometric means and geometric mean ratios of hormone concentrations in relation to smoking status. Serum concentrations of inhibin B were significantly lower in women who had ever smoked cigarettes: F(2,332) = 3.371, P = 0.036. There was no statistically significant difference in FSH or AMH concentrations although a trend towards lower AMH concentrations in smokers was observed. This analysis provides evidence of an advancement of ovarian ageing in women who smoke cigarettes and is relevant to women of childbearing age who wish to avoid premature decline in fertility. RBMOnline ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: cigarette, infertility, inhibin B, ovarian reserve, smoking
Introduction There is a trend in the UK towards deferral of childbearing until older maternal age, with the average age of women
giving birth for the first time inside marriage increasing by almost 6 years since 1971 (Office for National Statistics, 2007). Postponement of childbearing and societal changes in family planning have led to a significant increase in the
1472-6483/$ - see front matter ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.rbmo.2010.02.021
862 incidence of unwanted infertility due to the diminished ovarian reserve associated with female reproductive ageing (Broekmans et al., 2006). Ovarian reserve can be defined as the remaining functional capacity of the ovary, reflecting both the quantity and the quality of oocytes that remain (Nikolaou and Templeton, 2004). Ovarian reserve is observed to decline with age as a woman approaches menopause, as a result of atresia and recruitment to ovulation, in addition to a decline in oocyte quality (Nikolaou and Templeton, 2003). It is widely recognized that the onset of menopause and total loss of fertility is preceded by a period of subfertility and, assuming that this period of time is fixed, women with early onset of menopause will also be expected to experience earlier onset of subfertility, recognized as early ovarian ageing (Johnson et al., 2006). Age is not the only factor that contributes to ovarian ageing (Cramer et al., 2002; Gracia et al., 2005; Kinney et al., 2007; Lambert-Messerlian and Harlow, 2006). In the light of increasing maternal age at first birth and the associated increased risk of infertility, knowledge of other factors that can increase the risk of early ovarian ageing and thus affect ovarian reserve becomes more significant to women who wish to make informed lifestyle decisions about factors that may affect their fertility. Cigarette smoking is one such factor that has been associated with premature ovarian failure, indicated by the fact that female cigarette smokers enter the menopause on average 1.5–2 years earlier than non-smokers (Practice Committee of the American Society for Reproductive Medicine, 2004). However, it is not known whether this effect is mediated by a reduction in ovarian follicle number before menopause and there is little evidence available to determine the exact mechanisms that underlie this premature decline in fertility. Ovarian reserve testing provides a means for measuring such an effect. Over the past two decades, a wealth of ovarian reserve tests have been designed that allow the impact of factors that can affect fertility to be measured. These tests are also used to predict outcomes of assisted reproduction technologies and to determine when the most appropriate time to start a family may be. Many ovarian reserve tests are based upon hormonal measurements made at specific times of a woman’s menstrual cycle. Three hormones often used in ovarian reserve testing are inhibin B, FSH and anti-Mu ¨llerian hormone (AMH): each of these is considered useful in the assessment of ovarian reserve. Inhibin B is reduced in women with a diminished ovarian reserve (Erdem et al., 2004) and has been shown to be a more sensitive and specific marker of ovarian reserve than other parameters, including FSH, oestradiol and antral follicle count (Ficicioglu et al., 2003; Seifer et al., 1999). A rise in early follicular FSH concentrations have been shown to reflect a quantitative decline in ovarian reserve (Abdalla and Thum, 2004), and the hormone is regarded as a superior marker of follicle depletion in comparison to age (Toner et al., 1991). As AMH is produced by the granulosa cells of secondary, pre-antral and early antral follicles of the ovary, day 2–3 serum AMH concentrations have a strong correlation with other markers of ovarian ageing such as antral follicle count (van Rooij et al., 2002) and is thought to be an early marker of a decline in ovarian reserve
AL Waylen et al. (de Vet et al., 2002; Tremellen et al., 2005). Serum AMH concentrations also show little inter-cycle variability in women of reproductive age (Fanchin et al., 2005). Conflicting evidence exists regarding the effect of cigarette smoking upon existing measures of ovarian reserve (Kinney et al., 2007). Consequently, the aim of this retrospective data analysis was to determine whether women who have a previous or current history of cigarette smoking are at an increased risk of premature ovarian failure compared with age-matched non-smoking controls.
Materials and methods Data collection An existing database containing data on age, smoking status and serum concentrations of inhibin B, FSH and AMH was used to investigate whether there was a significant effect of cigarette smoking on these hormones. The data had been collected from women who had purchased a test for assessment of ovarian reserve, which involved a blood sample taken on day 2 or 3 of the menstrual cycle. Women all had cycle length of 27–34 days, with a minimum of 1 month between last taking contraceptive pills and testing. Permission to use anonymous data for research purposes had been given and recorded at the time of testing.
Hormone assays Hormone concentrations stored in the database had been determined from venous blood samples taken on day 2 or 3 of the menstrual cycle. Inhibin B samples were assayed in duplicate using a commercial ELISA kit (Oxford Bio-Innovation, Oxford, UK) according to the manufacturer’s protocol. The functional sensitivity of the assay was 15.6 pg/ml. The intra- and inter-assay variability were <7%. Serum FSH concentrations were determined using a two-site immunoradiometric assay principle (Oxford Bio-Innovation). The functional sensitivity of FSH was 0.11 mIU/ml and intra- and inter-assay variability were <4% and <8%, respectively. Anti-Mu ¨llerian hormone samples were assayed in duplicate using a commercial ELISA kit (Oxford Bio-Innovation) according to the manufacturer’s protocol. The sensitivity of the assay was 0.017 ng/ml. The intra- and inter-assay variability were <5 and 8%, respectively. There is currently no international reference standard for AMH. Using this assay, less than 5 pmol is low, 5– 15 is normal and over 15 is high, with the conversion factor for ng/ml to pmol/l of approximately · 7.
Statistical analysis All statistical analyses were performed using the Statistical Package for Social Sciences version 14.0 for Windows (SPSS, Chicago, USA). Descriptive statistics were applied to the data to analyse age and hormonal assays for women in each of the smoking categories (never smoked, ex-smoker and current smoker). Mean values and measures of distribution including standard
The effect of cigarette smoking on reproductive hormones
863
deviation and range were calculated for each parameter in all groups. The distribution of the residuals of each parameter was modelled and examined for normality of distribution. Data with a distribution deviating from normal (all hormonal parameters) were logarithmically transformed before analysis and reported in the form of geometric means with any significant differences between groups reported as a geometric mean ratio. Levene’s test was used to assess homogeneity of variance and used to evaluate the assumptions of the parametric tests. Pearson’s correlation coefficient was calculated to determine any correlation between the hormone concentrations and age, and between the hormone concentrations themselves. Demographic data for age was analysed using a one-way analysis of variance (ANOVA) to determine any significant difference between smoking categories. The analysis of hormonal assays with regard to smoking status was carried out using a univariate general linear model (ANCOVA), in which age was entered as a covariate to exclude any confounding effect. Geometric mean ratios were calculated with 95% confidence intervals where a statistically significant difference existed. The level of statistical significance for all tests was determined at P < 0.05.
and current smokers (64.4 pg/ml). The geometric mean ratio for never smokers compared with ex-smokers was 1.19 (95% CI 1.01–1.41, P = 0.038) and for never smokers compared with current smokers was 1.24 (95% CI 1.00–1.52, P = 0.046). There was no statistically significant difference between mean serum FSH concentration between the smoking categories, after adjustment for age (in ANCOVA): F(2,332) = 0.173. ANCOVA revealed no statistically significant difference in mean serum AMH concentration between the smoking categories after adjustment for age: F(2,332) = 0.895. However, a trend towards a decreasing serum AMH concentration with an increasingly active smoking status was observed.
Results Demographic data Hormone concentrations from a total of 335 women were analysed with regard to smoking status. Of these women, 202 were non-smokers, 86 were ex-smokers and 47 were current smokers. The mean age of the total sample was 37.1 years (range 24–48, SD 4.1). Analysis by one-way ANOVA revealed no significant difference in mean age observed between the different smoking groups: F(2,332) = 0.049 (Table 1).
Hormone assay correlations A Pearson’s correlation coefficient was calculated for each of the hormones to determine any correlation between these values and age. All measures were significantly correlated with age: log inhibin B (r = –0.202), log FSH (r = 0.276) and log AMH (r = –0.461); all P < 0.01. Log FSH was negatively correlated with the other hormones: log inhibin B (r = –0.227) and log AMH (r = –0.503). Log AMH was positively correlated with log inhibin B (r = 0.468); all P < 0.001.
Hormone assays and smoking status The geometric mean values and 95% CI for each of the hormonal assays by smoking group were calculated (Table 1). The geometric mean serum concentrations of inhibin B were significantly different between the smoking groups: F(2,332) = 3.371, P = 0.036. The mean serum concentration was highest in the group that had never smoked cigarettes (79.8 pg/ml), compared with the ex-smokers (67.0 pg/ml)
Discussion The results of this retrospective data analysis provide evidence of a significant effect of cigarette smoking upon early follicular-phase serum inhibin B concentration. Day 2–3 serum inhibin B concentrations were significantly lower in women of reproductive age who had either a previous or current history of cigarette smoking, compared with women who had never smoked cigarettes. There was no evidence of a significant effect of smoking upon serum concentrations of FSH or AMH. The significant decrease observed in serum inhibin B concentration is analogous to the fall in inhibin B usually observed with age (Welt et al., 1999), suggesting that cigarette smoking contributes to early ovarian ageing. The findings of the current study are consistent with those of Lambert-Messerlian and Harlow (2006), who reported for the first time that smoking history had a significant negative association with inhibin B concentrations. The validity of the results from the latter study is high: the sample size was large (404 women, of whom 203 had a history of cigarette smoking), the smokers reported a strong history of cigarette smoking (an average of 17 cigarettes daily for an average of 17 years) and inhibin B measurements were based on a mean calculated from at least three blood samples per patient over a period of 18 months. However, other literature regarding the effect of cigarette smoking upon inhibin B is inconsistent, with other researchers failing to find any significant effect of cigarette smoking upon serum concentrations of this hormone (Gracia et al., 2005; Kinney et al., 2007). Cigarette smoking has long been associated with a negative effect on fertility and has been identified by several studies to be a causative factor for an earlier onset of menopause (Practice Committee of the American Society for Reproductive Medicine, 2004). However, the exact mechanism by which cigarette smoking diminishes reproductive capacity is currently unknown. Findings from the current study provide evidence of a direct effect of smoking on ovarian follicles, indicated by the significant effect upon serum inhibin B concentrations. Inhibin B is produced by the granulosa cells of early antral follicles of the cohort recruited in each menstrual cycle and therefore a fall in inhibin B is likely to reflect a reduction in antral follicle count: a finding usually observed with reproductive ageing (Erdem et al., 2004). Therefore, cigarette smoking may lead
864
AL Waylen et al. Table 1
Mean age and serum hormone concentrations according to smoking status.
Sample size Age (years)b Inhibin B (pg/ml)c FSH (IU/l)c AMH (ng/ml)c
Non-smoker
Ex-smoker
Current smoker
ANOVA/ANCOVAa
202 37.1 (4.2) 79.8 (72.8–87.3) 5.1 (4.8–5.5) 1.074 (0.925–1.245)
86 37.0 (4.1) 67.0 (58.2–76.9) 5.3 (4.8–5.8) 0.955 (0.760–1.199)
47 37.0 (3.7) 64.4 (53.5–77.8) 5.3 (4.7–6.0) 0.869 (0.638–1.183)
– 0.049, 3.371, 0.173, 0.895,
NS P = 0.036d NS NS
Values are mean (standard deviation) or mean (95% CI). ANOVA = analysis of variance; ANCOVA = analysis of co-variance. AMH = anti-Mu ¨llerian hormone; FSH = follicle-stimulating hormone; NS = not statistically significant. a ANOVA model used for comparison of mean age. ANCOVA model used with age as covariate for comparison of geometric mean hormone concentrations. F(2,332). b Arithmetic mean. c Geometric mean. d P-value for comparison between non-smokers, ex-smokers and current smokers.
to a premature depletion of the follicle pool available for recruitment, causing premature ovarian failure. However, it is still unclear whether cigarette smoking reduces the quality as well as quantity of follicles available for recruitment. Further studies are required to confirm the relationship between cigarette smoking and inhibin B concentrations and to further explain the mechanisms of any such action. The present study found no significant effect of cigarette smoking (either past or current) on early follicular-phase FSH concentrations, despite a wealth of evidence suggesting that such an effect does in fact exist (Cooper et al., 1995; Cramer et al., 2002; El-Nemr et al., 1998; Kinney et al., 2007). Serum FSH concentrations are known to rise with increasing ovarian age, therefore acting as a marker of ovarian ageing (Kline et al., 2005), but this rise is considered to be a late marker of a diminished ovarian reserve (Broekmans et al., 1998; Burger et al., 1999; Kline et al., 2005). It is therefore possible that the inconsistency between the results observed in the present study and the main body of evidence to date may be a reflection of differences in the mean age of women involved in these studies: the mean age of the sample in the present study was 37.1 years (range 24–48) and thus included many young women for whom a smoking related decline in ovarian reserve may not yet be reflected by serum FSH concentrations. No statistically significant difference was observed for day-2–3 serum AMH concentrations in women with a history of smoking compared with those that had never smoked. However, a trend towards a decrease in both previous and current smokers was observed. A recent review found serum AMH concentrations to have only moderate sensitivity in detecting diminished ovarian reserve (Broekmans et al., 2006), which may account for the lack of a definitive finding in this study. Literature to date has shown a significant effect of smoking upon serum AMH concentrations (Freour et al., 2008), but the paucity of research demands future studies to determine whether the trend observed in the present study is also indicative of any significant effect and to further investigate the underlying mechanisms. Consistent with current literature, the hormone profiles generated in this study were all significantly associated with age. FSH was positively correlated with age, reflecting the
increased production of this hormone due to a quantitative diminishment of the ovarian follicular pool (Abdalla and Thum, 2004). Serum concentrations of AMH and inhibin B were negatively correlated with age, consistent with the correlation of these hormones known to be associated with reproductive ageing (Lee et al., 1996; Welt et al., 1999). A significant limitation of this study is the inability to account for demographic and clinical heterogeneity in the data, due to the lack of socio-demographic and clinical data that had been collected for the women from whom the blood samples had been taken. It was therefore not possible to further analyse the data with regard to factors, such as past medical, reproductive and menstrual cycle history or ethnicity. It was also not possible to account for body mass index, alcohol or caffeine intake, all of which have previously been shown to affect reproductive hormones and fertility (Gracia et al., 2005; Kinney et al., 2007). Furthermore, data regarding other potential measures of ovarian reserve including AFC and oestradiol were not included in the database and thus not available for analysis. The data regarding smoking status were also inadequate to enable an in-depth analysis of smoking history: there was no information available on the number of cigarettes smoked per day, duration of smoking habit, time since stopping smoking or passive smoking exposure. It is therefore not possible to comment on the effect of smoking according to the number smoked per day, but rather to report results for smoking history in general at time of hormone measurement. Finally, the smoking status of the women was ascertained by self-report and not confirmed by any other measure, such as urinary cotinine. It is possible that due to the sensitive nature of ovarian reserve testing and the negative effects associated with cigarette smoking upon general health, the women did not divulge their true smoking status. Despite these concerns, findings of a study by Patrick et al. (1994) report the validity of self-reported smoking to be good. Longitudinal prospective studies measuring serum hormonal concentrations including inhibin B are required in cohorts of smoking and non-smoking women of reproductive age to determine the difference in hormone concentrations in these groups throughout the reproductive lifespan. Stud-
The effect of cigarette smoking on reproductive hormones
865
ies should endeavour to collect socio-demographic and clinical information about participants in order to allow researchers to identify and account for heterogeneity and confounding factors, and to ensure generalizability of results to the general population. Findings from such studies will have greater power than retrospective data analyses. In conclusion, this study provides further evidence that cigarette smoking increases the risk of premature ovarian failure in women of reproductive age. The fall in serum concentrations of inhibin B observed with smoking is comparable to that associated with increasing age and therefore indicates a risk of premature ovarian failure. Whilst no significant effect of cigarette smoking was observed for the hormones FSH and AMH, a trend towards a decrease in AMH concentrations was seen and further investigation is required to authenticate this trend. The findings of this study should be reported to women of reproductive age to allow them to make informed decisions about lifestyle choices, therefore preserving their fertility.
Ficicioglu, C., Kutlu, T., Demirbasoglu, S., et al., 2003. The role of inhibin B as a basal determinant of ovarian reserve. Gynecol. Endocrinol. 17, 287–293. Freour, T., Masson, D., Mirallie, S., et al., 2008. Active smoking compromises IVF outcome and affects ovarian reserve. Reprod. Biomed. Online 16, 96–102. Gracia, C., Freeman, E.W., Sammel, M.D., et al., 2005. The relationship between obesity and race on inhibin B during the menopause transition. Menopause 12, 559–566. Johnson, N.P., Bagrie, E.M., Coomarasamy, A., et al., 2006. Ovarian reserve tests for predicting fertility outcomes for assisted reproductive technology: the International Systematic Collaboration of Ovarian Reserve Evaluation protocol for a systematic review of ovarian reserve test accuracy. BJOG Int. J. Obstet. Gynaecol. 113, 1472–1480. Kinney, A., Kline, J., Kelly, A., et al., 2007. Smoking, alcohol and caffeine in relation to ovarian age during the reproductive years. Hum. Reprod. 22, 1175–1185. Kline, J., Kinney, A., Kelly, A., et al., 2005. Predictors of antral follicle count during the reproductive years. Hum. Reprod. 20, 2179–2189. Lambert-Messerlian, G.M., Harlow, B.L., 2006. The influence of depression, body mass index, and smoking on serum inhibin B levels in late reproductive-aged women. J. Clin. Endocrinol. Metab. 91, 1496–1500. Lee, M.M., Donahoe, P.K., Hasegawa, T., et al., 1996. Mu ¨llerian inhibiting substance in humans: normal levels from infancy to adulthood. J. Clin. Endocrinol. Metab. 81, 576. Nikolaou, D., Templeton, A., 2003. Early ovarian ageing: a hypothesis. Detection and clinical relevance. Hum. Reprod. 18, 1137–1139. Nikolaou, D., Templeton, A., 2004. Early ovarian ageing. Eur. J. Obstet. Gynecol. Reprod. Biol. 113, 126–133. Office for National Statistics, 2007. Average age of mother at childbirth: Social Trends 33. Www Statistics Gov Uk. Patrick, D., Cheadle, A., Thompson, D., et al., 1994. The validity of self-reported smoking: a review and meta-analysis. Am. J. Public Health 84, 1086–1093. Practice Committee of the American Society for Reproductive Medicine, 2004. Smoking and infertility. Fertil. Steril. 81, 1181– 1186. Seifer, D.B., Scott Jr., R.T., Bergh, P.A., et al., 1999. Women with declining ovarian reserve may demonstrate a decrease in day 3 serum inhibin B before a rise in day 3 follicle-stimulating hormone. Fertil. Steril. 72, 63–65. Toner, J.P., Philput, C.B., Jones, G.S., et al., 1991. Basal follicle stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil. Steril. 55, 784–791. Tremellen, K.P., Kolo, M., Gilmore, A., et al., 2005. Anti-mullerian hormone as a marker of ovarian reserve. Aust. NZ J. Obstet. Gynaecol. 45, 20–24. van Rooij, I.A., Broekmans, F.J., te Velde, E.R., et al., 2002. Serum anti-Mullerian hormone levels: a novel measure of ovarian reserve. Hum. Reprod. 17, 3065–3071. Welt, C.K., McNicholl, D.J., Taylor, A.E., et al., 1999. Female reproductive aging is marked by decreased secretion of dimeric inhibin. J. Clin. Endocrinol. Metab. 84, 105–111.
References Abdalla, H., Thum, M.Y., 2004. An elevated basal FSH reflects a quantitative rather than qualitative decline of the ovarian reserve. Hum. Reprod. 19, 893–898. Broekmans, F.J., Kwee, J., Hendriks, D.J., et al., 2006. A systematic review of tests predicting ovarian reserve and IVF outcome. Hum. Reprod. Update 12, 685–718. Broekmans, F.J., Scheffer, G.J., Bancsi, L.F., et al., 1998. Ovarian reserve tests in infertility practice and normal fertile women. Maturitas 30, 205–214. Burger, H.G., Dudley, E.C., Hopper, J.L., et al., 1999. Prospectively measured levels of serum follicle-stimulating hormone, estradiol, and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J. Clin. Endocrinol. Metab. 84, 4025–4030. Cooper, G.S., Baird, D.D., Hulka, B.S., et al., 1995. Folliclestimulating hormone concentrations in relation to active and passive smoking. Obstet. Gynecol. 85, 407–411. Cramer, D.W., Barbieri, R.L., Fraer, A.R., et al., 2002. Determinants of early follicular phase gonadotrophin and estradiol concentrations in women of late reproductive age. Hum. Reprod. 17, 221–227. de Vet, A., Laven, J.S., de Jong, F.H., et al., 2002. Antimullerian hormone serum levels: a putative marker for ovarian aging. Fertil. Steril. 77, 357–362. El-Nemr, A., Al-Shawaf, T., Sabatini, L., et al., 1998. Effect of smoking on ovarian reserve and ovarian stimulation in in-vitro fertilization and embryo transfer. Hum. Reprod. 13, 2192– 2198. Erdem, M., Erdem, A., Gursoy, R., et al., 2004. Comparison of basal and clomiphene citrate induced FSH and inhibin B, ovarian volume and antral follicle counts as ovarian reserve tests and predictors of poor ovarian response in IVF. J. Assist. Reprod. Genet. 21, 37–45. Fanchin, R., Taieb, J., Lozano, D.H., et al., 2005. High reproducibility of serum anti-Mullerian hormone measurements suggests a multi-staged follicular secretion and strengthens its role in the assessment of ovarian follicular status. Hum. Reprod. 20, 923– 927.
Declaration: The authors report no financial or commercial conflicts of interest. Received 11 August 2009; refereed 2 November 2009; accepted 25 January 2010.
www.sciencedirect.com www.rbmonline.com
ARTICLE
Effect of cigarette smoking upon reproductive hormones in women of reproductive age: a retrospective analysis AL Waylen
a,*
, GL Jones b, WL Ledger
c
a University of Sheffield School of Medicine and Biomedical Sciences, Beech Hill Road, Sheffield S10 2RX, UK; b Health Services Research Section, ScHARR, Regent Court, 30 Regent Street, Sheffield S1 4DA, UK; c Academic Unit of Reproductive and Developmental Medicine, Jessop Wing Hospital, Tree Root Walk, Sheffield S10 2TJ, UK
* Corresponding author. E-mail addresses: [email protected], [email protected] (AL Waylen). Anna Waylon attended the University of Sheffield School of Medicine from September 2003 to July 2009 during which time she completed a Bachelor of Medical Science degree in the Academic Unit of Reproductive and Developmental Medicine. Her research involved investigating the effects of cigarette smoking on female reproductive hormones and on the clinical outcomes of assisted reproduction, and a qualitative assessment of women’s knowledge of factors that could affect their fertility.
Abstract There is continuing debate concerning the relationship between cigarette smoking and premature ovarian failure. The aim
of this retrospective data analysis was to investigate whether smoking has a measurable effect on early follicular serum concentrations of inhibin B hormone, FSH and anti-Mu ¨llerian hormone (AMH) in women of reproductive age. A database containing data on age, smoking status and serum concentrations of inhibin B, FSH and AMH was analysed. Pearson’s correlation coefficient was calculated to determine the correlation between hormone concentrations and age. One-way analysis of variance was used to determine any significant difference in age between smoking categories and a univariate general linear model was used to compare geometric means and geometric mean ratios of hormone concentrations in relation to smoking status. Serum concentrations of inhibin B were significantly lower in women who had ever smoked cigarettes: F(2,332) = 3.371, P = 0.036. There was no statistically significant difference in FSH or AMH concentrations although a trend towards lower AMH concentrations in smokers was observed. This analysis provides evidence of an advancement of ovarian ageing in women who smoke cigarettes and is relevant to women of childbearing age who wish to avoid premature decline in fertility. RBMOnline ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: cigarette, infertility, inhibin B, ovarian reserve, smoking
Introduction There is a trend in the UK towards deferral of childbearing until older maternal age, with the average age of women
giving birth for the first time inside marriage increasing by almost 6 years since 1971 (Office for National Statistics, 2007). Postponement of childbearing and societal changes in family planning have led to a significant increase in the
1472-6483/$ - see front matter ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.rbmo.2010.02.021
862 incidence of unwanted infertility due to the diminished ovarian reserve associated with female reproductive ageing (Broekmans et al., 2006). Ovarian reserve can be defined as the remaining functional capacity of the ovary, reflecting both the quantity and the quality of oocytes that remain (Nikolaou and Templeton, 2004). Ovarian reserve is observed to decline with age as a woman approaches menopause, as a result of atresia and recruitment to ovulation, in addition to a decline in oocyte quality (Nikolaou and Templeton, 2003). It is widely recognized that the onset of menopause and total loss of fertility is preceded by a period of subfertility and, assuming that this period of time is fixed, women with early onset of menopause will also be expected to experience earlier onset of subfertility, recognized as early ovarian ageing (Johnson et al., 2006). Age is not the only factor that contributes to ovarian ageing (Cramer et al., 2002; Gracia et al., 2005; Kinney et al., 2007; Lambert-Messerlian and Harlow, 2006). In the light of increasing maternal age at first birth and the associated increased risk of infertility, knowledge of other factors that can increase the risk of early ovarian ageing and thus affect ovarian reserve becomes more significant to women who wish to make informed lifestyle decisions about factors that may affect their fertility. Cigarette smoking is one such factor that has been associated with premature ovarian failure, indicated by the fact that female cigarette smokers enter the menopause on average 1.5–2 years earlier than non-smokers (Practice Committee of the American Society for Reproductive Medicine, 2004). However, it is not known whether this effect is mediated by a reduction in ovarian follicle number before menopause and there is little evidence available to determine the exact mechanisms that underlie this premature decline in fertility. Ovarian reserve testing provides a means for measuring such an effect. Over the past two decades, a wealth of ovarian reserve tests have been designed that allow the impact of factors that can affect fertility to be measured. These tests are also used to predict outcomes of assisted reproduction technologies and to determine when the most appropriate time to start a family may be. Many ovarian reserve tests are based upon hormonal measurements made at specific times of a woman’s menstrual cycle. Three hormones often used in ovarian reserve testing are inhibin B, FSH and anti-Mu ¨llerian hormone (AMH): each of these is considered useful in the assessment of ovarian reserve. Inhibin B is reduced in women with a diminished ovarian reserve (Erdem et al., 2004) and has been shown to be a more sensitive and specific marker of ovarian reserve than other parameters, including FSH, oestradiol and antral follicle count (Ficicioglu et al., 2003; Seifer et al., 1999). A rise in early follicular FSH concentrations have been shown to reflect a quantitative decline in ovarian reserve (Abdalla and Thum, 2004), and the hormone is regarded as a superior marker of follicle depletion in comparison to age (Toner et al., 1991). As AMH is produced by the granulosa cells of secondary, pre-antral and early antral follicles of the ovary, day 2–3 serum AMH concentrations have a strong correlation with other markers of ovarian ageing such as antral follicle count (van Rooij et al., 2002) and is thought to be an early marker of a decline in ovarian reserve
AL Waylen et al. (de Vet et al., 2002; Tremellen et al., 2005). Serum AMH concentrations also show little inter-cycle variability in women of reproductive age (Fanchin et al., 2005). Conflicting evidence exists regarding the effect of cigarette smoking upon existing measures of ovarian reserve (Kinney et al., 2007). Consequently, the aim of this retrospective data analysis was to determine whether women who have a previous or current history of cigarette smoking are at an increased risk of premature ovarian failure compared with age-matched non-smoking controls.
Materials and methods Data collection An existing database containing data on age, smoking status and serum concentrations of inhibin B, FSH and AMH was used to investigate whether there was a significant effect of cigarette smoking on these hormones. The data had been collected from women who had purchased a test for assessment of ovarian reserve, which involved a blood sample taken on day 2 or 3 of the menstrual cycle. Women all had cycle length of 27–34 days, with a minimum of 1 month between last taking contraceptive pills and testing. Permission to use anonymous data for research purposes had been given and recorded at the time of testing.
Hormone assays Hormone concentrations stored in the database had been determined from venous blood samples taken on day 2 or 3 of the menstrual cycle. Inhibin B samples were assayed in duplicate using a commercial ELISA kit (Oxford Bio-Innovation, Oxford, UK) according to the manufacturer’s protocol. The functional sensitivity of the assay was 15.6 pg/ml. The intra- and inter-assay variability were <7%. Serum FSH concentrations were determined using a two-site immunoradiometric assay principle (Oxford Bio-Innovation). The functional sensitivity of FSH was 0.11 mIU/ml and intra- and inter-assay variability were <4% and <8%, respectively. Anti-Mu ¨llerian hormone samples were assayed in duplicate using a commercial ELISA kit (Oxford Bio-Innovation) according to the manufacturer’s protocol. The sensitivity of the assay was 0.017 ng/ml. The intra- and inter-assay variability were <5 and 8%, respectively. There is currently no international reference standard for AMH. Using this assay, less than 5 pmol is low, 5– 15 is normal and over 15 is high, with the conversion factor for ng/ml to pmol/l of approximately · 7.
Statistical analysis All statistical analyses were performed using the Statistical Package for Social Sciences version 14.0 for Windows (SPSS, Chicago, USA). Descriptive statistics were applied to the data to analyse age and hormonal assays for women in each of the smoking categories (never smoked, ex-smoker and current smoker). Mean values and measures of distribution including standard
The effect of cigarette smoking on reproductive hormones
863
deviation and range were calculated for each parameter in all groups. The distribution of the residuals of each parameter was modelled and examined for normality of distribution. Data with a distribution deviating from normal (all hormonal parameters) were logarithmically transformed before analysis and reported in the form of geometric means with any significant differences between groups reported as a geometric mean ratio. Levene’s test was used to assess homogeneity of variance and used to evaluate the assumptions of the parametric tests. Pearson’s correlation coefficient was calculated to determine any correlation between the hormone concentrations and age, and between the hormone concentrations themselves. Demographic data for age was analysed using a one-way analysis of variance (ANOVA) to determine any significant difference between smoking categories. The analysis of hormonal assays with regard to smoking status was carried out using a univariate general linear model (ANCOVA), in which age was entered as a covariate to exclude any confounding effect. Geometric mean ratios were calculated with 95% confidence intervals where a statistically significant difference existed. The level of statistical significance for all tests was determined at P < 0.05.
and current smokers (64.4 pg/ml). The geometric mean ratio for never smokers compared with ex-smokers was 1.19 (95% CI 1.01–1.41, P = 0.038) and for never smokers compared with current smokers was 1.24 (95% CI 1.00–1.52, P = 0.046). There was no statistically significant difference between mean serum FSH concentration between the smoking categories, after adjustment for age (in ANCOVA): F(2,332) = 0.173. ANCOVA revealed no statistically significant difference in mean serum AMH concentration between the smoking categories after adjustment for age: F(2,332) = 0.895. However, a trend towards a decreasing serum AMH concentration with an increasingly active smoking status was observed.
Results Demographic data Hormone concentrations from a total of 335 women were analysed with regard to smoking status. Of these women, 202 were non-smokers, 86 were ex-smokers and 47 were current smokers. The mean age of the total sample was 37.1 years (range 24–48, SD 4.1). Analysis by one-way ANOVA revealed no significant difference in mean age observed between the different smoking groups: F(2,332) = 0.049 (Table 1).
Hormone assay correlations A Pearson’s correlation coefficient was calculated for each of the hormones to determine any correlation between these values and age. All measures were significantly correlated with age: log inhibin B (r = –0.202), log FSH (r = 0.276) and log AMH (r = –0.461); all P < 0.01. Log FSH was negatively correlated with the other hormones: log inhibin B (r = –0.227) and log AMH (r = –0.503). Log AMH was positively correlated with log inhibin B (r = 0.468); all P < 0.001.
Hormone assays and smoking status The geometric mean values and 95% CI for each of the hormonal assays by smoking group were calculated (Table 1). The geometric mean serum concentrations of inhibin B were significantly different between the smoking groups: F(2,332) = 3.371, P = 0.036. The mean serum concentration was highest in the group that had never smoked cigarettes (79.8 pg/ml), compared with the ex-smokers (67.0 pg/ml)
Discussion The results of this retrospective data analysis provide evidence of a significant effect of cigarette smoking upon early follicular-phase serum inhibin B concentration. Day 2–3 serum inhibin B concentrations were significantly lower in women of reproductive age who had either a previous or current history of cigarette smoking, compared with women who had never smoked cigarettes. There was no evidence of a significant effect of smoking upon serum concentrations of FSH or AMH. The significant decrease observed in serum inhibin B concentration is analogous to the fall in inhibin B usually observed with age (Welt et al., 1999), suggesting that cigarette smoking contributes to early ovarian ageing. The findings of the current study are consistent with those of Lambert-Messerlian and Harlow (2006), who reported for the first time that smoking history had a significant negative association with inhibin B concentrations. The validity of the results from the latter study is high: the sample size was large (404 women, of whom 203 had a history of cigarette smoking), the smokers reported a strong history of cigarette smoking (an average of 17 cigarettes daily for an average of 17 years) and inhibin B measurements were based on a mean calculated from at least three blood samples per patient over a period of 18 months. However, other literature regarding the effect of cigarette smoking upon inhibin B is inconsistent, with other researchers failing to find any significant effect of cigarette smoking upon serum concentrations of this hormone (Gracia et al., 2005; Kinney et al., 2007). Cigarette smoking has long been associated with a negative effect on fertility and has been identified by several studies to be a causative factor for an earlier onset of menopause (Practice Committee of the American Society for Reproductive Medicine, 2004). However, the exact mechanism by which cigarette smoking diminishes reproductive capacity is currently unknown. Findings from the current study provide evidence of a direct effect of smoking on ovarian follicles, indicated by the significant effect upon serum inhibin B concentrations. Inhibin B is produced by the granulosa cells of early antral follicles of the cohort recruited in each menstrual cycle and therefore a fall in inhibin B is likely to reflect a reduction in antral follicle count: a finding usually observed with reproductive ageing (Erdem et al., 2004). Therefore, cigarette smoking may lead
864
AL Waylen et al. Table 1
Mean age and serum hormone concentrations according to smoking status.
Sample size Age (years)b Inhibin B (pg/ml)c FSH (IU/l)c AMH (ng/ml)c
Non-smoker
Ex-smoker
Current smoker
ANOVA/ANCOVAa
202 37.1 (4.2) 79.8 (72.8–87.3) 5.1 (4.8–5.5) 1.074 (0.925–1.245)
86 37.0 (4.1) 67.0 (58.2–76.9) 5.3 (4.8–5.8) 0.955 (0.760–1.199)
47 37.0 (3.7) 64.4 (53.5–77.8) 5.3 (4.7–6.0) 0.869 (0.638–1.183)
– 0.049, 3.371, 0.173, 0.895,
NS P = 0.036d NS NS
Values are mean (standard deviation) or mean (95% CI). ANOVA = analysis of variance; ANCOVA = analysis of co-variance. AMH = anti-Mu ¨llerian hormone; FSH = follicle-stimulating hormone; NS = not statistically significant. a ANOVA model used for comparison of mean age. ANCOVA model used with age as covariate for comparison of geometric mean hormone concentrations. F(2,332). b Arithmetic mean. c Geometric mean. d P-value for comparison between non-smokers, ex-smokers and current smokers.
to a premature depletion of the follicle pool available for recruitment, causing premature ovarian failure. However, it is still unclear whether cigarette smoking reduces the quality as well as quantity of follicles available for recruitment. Further studies are required to confirm the relationship between cigarette smoking and inhibin B concentrations and to further explain the mechanisms of any such action. The present study found no significant effect of cigarette smoking (either past or current) on early follicular-phase FSH concentrations, despite a wealth of evidence suggesting that such an effect does in fact exist (Cooper et al., 1995; Cramer et al., 2002; El-Nemr et al., 1998; Kinney et al., 2007). Serum FSH concentrations are known to rise with increasing ovarian age, therefore acting as a marker of ovarian ageing (Kline et al., 2005), but this rise is considered to be a late marker of a diminished ovarian reserve (Broekmans et al., 1998; Burger et al., 1999; Kline et al., 2005). It is therefore possible that the inconsistency between the results observed in the present study and the main body of evidence to date may be a reflection of differences in the mean age of women involved in these studies: the mean age of the sample in the present study was 37.1 years (range 24–48) and thus included many young women for whom a smoking related decline in ovarian reserve may not yet be reflected by serum FSH concentrations. No statistically significant difference was observed for day-2–3 serum AMH concentrations in women with a history of smoking compared with those that had never smoked. However, a trend towards a decrease in both previous and current smokers was observed. A recent review found serum AMH concentrations to have only moderate sensitivity in detecting diminished ovarian reserve (Broekmans et al., 2006), which may account for the lack of a definitive finding in this study. Literature to date has shown a significant effect of smoking upon serum AMH concentrations (Freour et al., 2008), but the paucity of research demands future studies to determine whether the trend observed in the present study is also indicative of any significant effect and to further investigate the underlying mechanisms. Consistent with current literature, the hormone profiles generated in this study were all significantly associated with age. FSH was positively correlated with age, reflecting the
increased production of this hormone due to a quantitative diminishment of the ovarian follicular pool (Abdalla and Thum, 2004). Serum concentrations of AMH and inhibin B were negatively correlated with age, consistent with the correlation of these hormones known to be associated with reproductive ageing (Lee et al., 1996; Welt et al., 1999). A significant limitation of this study is the inability to account for demographic and clinical heterogeneity in the data, due to the lack of socio-demographic and clinical data that had been collected for the women from whom the blood samples had been taken. It was therefore not possible to further analyse the data with regard to factors, such as past medical, reproductive and menstrual cycle history or ethnicity. It was also not possible to account for body mass index, alcohol or caffeine intake, all of which have previously been shown to affect reproductive hormones and fertility (Gracia et al., 2005; Kinney et al., 2007). Furthermore, data regarding other potential measures of ovarian reserve including AFC and oestradiol were not included in the database and thus not available for analysis. The data regarding smoking status were also inadequate to enable an in-depth analysis of smoking history: there was no information available on the number of cigarettes smoked per day, duration of smoking habit, time since stopping smoking or passive smoking exposure. It is therefore not possible to comment on the effect of smoking according to the number smoked per day, but rather to report results for smoking history in general at time of hormone measurement. Finally, the smoking status of the women was ascertained by self-report and not confirmed by any other measure, such as urinary cotinine. It is possible that due to the sensitive nature of ovarian reserve testing and the negative effects associated with cigarette smoking upon general health, the women did not divulge their true smoking status. Despite these concerns, findings of a study by Patrick et al. (1994) report the validity of self-reported smoking to be good. Longitudinal prospective studies measuring serum hormonal concentrations including inhibin B are required in cohorts of smoking and non-smoking women of reproductive age to determine the difference in hormone concentrations in these groups throughout the reproductive lifespan. Stud-
The effect of cigarette smoking on reproductive hormones
865
ies should endeavour to collect socio-demographic and clinical information about participants in order to allow researchers to identify and account for heterogeneity and confounding factors, and to ensure generalizability of results to the general population. Findings from such studies will have greater power than retrospective data analyses. In conclusion, this study provides further evidence that cigarette smoking increases the risk of premature ovarian failure in women of reproductive age. The fall in serum concentrations of inhibin B observed with smoking is comparable to that associated with increasing age and therefore indicates a risk of premature ovarian failure. Whilst no significant effect of cigarette smoking was observed for the hormones FSH and AMH, a trend towards a decrease in AMH concentrations was seen and further investigation is required to authenticate this trend. The findings of this study should be reported to women of reproductive age to allow them to make informed decisions about lifestyle choices, therefore preserving their fertility.
Ficicioglu, C., Kutlu, T., Demirbasoglu, S., et al., 2003. The role of inhibin B as a basal determinant of ovarian reserve. Gynecol. Endocrinol. 17, 287–293. Freour, T., Masson, D., Mirallie, S., et al., 2008. Active smoking compromises IVF outcome and affects ovarian reserve. Reprod. Biomed. Online 16, 96–102. Gracia, C., Freeman, E.W., Sammel, M.D., et al., 2005. The relationship between obesity and race on inhibin B during the menopause transition. Menopause 12, 559–566. Johnson, N.P., Bagrie, E.M., Coomarasamy, A., et al., 2006. Ovarian reserve tests for predicting fertility outcomes for assisted reproductive technology: the International Systematic Collaboration of Ovarian Reserve Evaluation protocol for a systematic review of ovarian reserve test accuracy. BJOG Int. J. Obstet. Gynaecol. 113, 1472–1480. Kinney, A., Kline, J., Kelly, A., et al., 2007. Smoking, alcohol and caffeine in relation to ovarian age during the reproductive years. Hum. Reprod. 22, 1175–1185. Kline, J., Kinney, A., Kelly, A., et al., 2005. Predictors of antral follicle count during the reproductive years. Hum. Reprod. 20, 2179–2189. Lambert-Messerlian, G.M., Harlow, B.L., 2006. The influence of depression, body mass index, and smoking on serum inhibin B levels in late reproductive-aged women. J. Clin. Endocrinol. Metab. 91, 1496–1500. Lee, M.M., Donahoe, P.K., Hasegawa, T., et al., 1996. Mu ¨llerian inhibiting substance in humans: normal levels from infancy to adulthood. J. Clin. Endocrinol. Metab. 81, 576. Nikolaou, D., Templeton, A., 2003. Early ovarian ageing: a hypothesis. Detection and clinical relevance. Hum. Reprod. 18, 1137–1139. Nikolaou, D., Templeton, A., 2004. Early ovarian ageing. Eur. J. Obstet. Gynecol. Reprod. Biol. 113, 126–133. Office for National Statistics, 2007. Average age of mother at childbirth: Social Trends 33. Www Statistics Gov Uk. Patrick, D., Cheadle, A., Thompson, D., et al., 1994. The validity of self-reported smoking: a review and meta-analysis. Am. J. Public Health 84, 1086–1093. Practice Committee of the American Society for Reproductive Medicine, 2004. Smoking and infertility. Fertil. Steril. 81, 1181– 1186. Seifer, D.B., Scott Jr., R.T., Bergh, P.A., et al., 1999. Women with declining ovarian reserve may demonstrate a decrease in day 3 serum inhibin B before a rise in day 3 follicle-stimulating hormone. Fertil. Steril. 72, 63–65. Toner, J.P., Philput, C.B., Jones, G.S., et al., 1991. Basal follicle stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil. Steril. 55, 784–791. Tremellen, K.P., Kolo, M., Gilmore, A., et al., 2005. Anti-mullerian hormone as a marker of ovarian reserve. Aust. NZ J. Obstet. Gynaecol. 45, 20–24. van Rooij, I.A., Broekmans, F.J., te Velde, E.R., et al., 2002. Serum anti-Mullerian hormone levels: a novel measure of ovarian reserve. Hum. Reprod. 17, 3065–3071. Welt, C.K., McNicholl, D.J., Taylor, A.E., et al., 1999. Female reproductive aging is marked by decreased secretion of dimeric inhibin. J. Clin. Endocrinol. Metab. 84, 105–111.
References Abdalla, H., Thum, M.Y., 2004. An elevated basal FSH reflects a quantitative rather than qualitative decline of the ovarian reserve. Hum. Reprod. 19, 893–898. Broekmans, F.J., Kwee, J., Hendriks, D.J., et al., 2006. A systematic review of tests predicting ovarian reserve and IVF outcome. Hum. Reprod. Update 12, 685–718. Broekmans, F.J., Scheffer, G.J., Bancsi, L.F., et al., 1998. Ovarian reserve tests in infertility practice and normal fertile women. Maturitas 30, 205–214. Burger, H.G., Dudley, E.C., Hopper, J.L., et al., 1999. Prospectively measured levels of serum follicle-stimulating hormone, estradiol, and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J. Clin. Endocrinol. Metab. 84, 4025–4030. Cooper, G.S., Baird, D.D., Hulka, B.S., et al., 1995. Folliclestimulating hormone concentrations in relation to active and passive smoking. Obstet. Gynecol. 85, 407–411. Cramer, D.W., Barbieri, R.L., Fraer, A.R., et al., 2002. Determinants of early follicular phase gonadotrophin and estradiol concentrations in women of late reproductive age. Hum. Reprod. 17, 221–227. de Vet, A., Laven, J.S., de Jong, F.H., et al., 2002. Antimullerian hormone serum levels: a putative marker for ovarian aging. Fertil. Steril. 77, 357–362. El-Nemr, A., Al-Shawaf, T., Sabatini, L., et al., 1998. Effect of smoking on ovarian reserve and ovarian stimulation in in-vitro fertilization and embryo transfer. Hum. Reprod. 13, 2192– 2198. Erdem, M., Erdem, A., Gursoy, R., et al., 2004. Comparison of basal and clomiphene citrate induced FSH and inhibin B, ovarian volume and antral follicle counts as ovarian reserve tests and predictors of poor ovarian response in IVF. J. Assist. Reprod. Genet. 21, 37–45. Fanchin, R., Taieb, J., Lozano, D.H., et al., 2005. High reproducibility of serum anti-Mullerian hormone measurements suggests a multi-staged follicular secretion and strengthens its role in the assessment of ovarian follicular status. Hum. Reprod. 20, 923– 927.
Declaration: The authors report no financial or commercial conflicts of interest. Received 11 August 2009; refereed 2 November 2009; accepted 25 January 2010.