Hormones and Tobacco: Joint Effects on Reproduction

Institution:	Public Health Institute
Investigator(s):	Shanna Swan, Ph.D.
Award Cycle:	1994 (Cycle 3)	Grant #: 3RT-0093	Award: $263,611
Subject Area:	Epidemiology
Award Type:	Research Project Awards

Abstracts

Initial Award Abstract

The purpose of this study is to examine hormonal mechanisms for the relationship between exposure to tobacco smoke and reproductive function. Several lines of evidence suggest that smoking may alter production or metabolism of steroid hormones and thus may affect fertility and the maintenance of a viable pregnancy. We are examining these hypotheses by comparing parameters of menstrual function, time to conception and occurrence of spontaneous abortion (SAB) in healthy women who are exposed to mainstream cigarette smoke, to environmental tobacco smoke(ETS) or who are unexposed to either. For this purpose, we conducted a two-year follow-up of women who participated in the Women's Reproductive Health Study, a prospective study conducted by the California Department of Health Services in collaboration with Kaiser Division of Research, in which women collected daily urine samples for an average of 5.6 menstrual cycles. Under this grant, 378 women were reinterviewed. In this sample we identified 205 clinically recognized pregnancies since urine collection began, of which 27 resulted in spontaneous abortion. The identification of subclinical pregnancies by hCG assay is just now being completed due to laboratory problems. We have developed computerized algorithms for assessing the ovulatory status of all menstrual cycles based on urinary steroid metabolites (E1C and PdG) and the day of ovulation when appropriate. Each cycle which meets useability criteria (mostly based on completeness) has been classified as to the presence or absence of six menstrual endpoints. These include short (9%) or long (6%) cycle length, short luteal phase (5.4%), long follicular phase (5%), long menses (8%) and anovulatory segments (about 7%). Three samples per women of pooled urine have been assessed for cotinine level. Duplicate assays for urine cotinine showed the assay to be highly reliable (correlation coefficient 0.99). There was also a strong correlation between cotinine and self-reported smoking; the median cotinine levels (ng/ml) in women reporting 0, <1, 1-10 and >10 cigarettes per day on average were 0.3, 4.8, 594 and 1490, respectively. Non-smokers reporting ETS exposure had intermediate cotinine levels between those unexposed and smokers. Laboratory delays (hormone and cotinine) led to delays in analysis, which will be on-going to the extent possible. Preliminary analysis shows an associated between smoking a pack of cigarettes or more per day and short cycles, as well as anovulatory cycles.

Publications

Cigarette smoking and menstrual cycle function
Periodical: American Journal of Epidemiology			Index Medicus:
Authors: Windham GC, Swan SH, Elkin E, Waller K, Fenster L,Schaefer C				ABS
Yr: 1996	Vol:	Nbr:	Abs: S6	Pg: 143

Cigarette smoking and menstrual cycle function
Periodical: American Journal of Epidemiology			Index Medicus:
Authors: Windham GC, Swan SH, Elkin E, Waller K, Fenster L,Schaefer C				ABS
Yr: 1996	Vol:	Nbr:	Abs: S6	Pg: 143