Season of birth influences the timing of menopause
Angelo Cagnacci1,6,
Francesco S. Pansini2,
Alberto Bacchi-Modena3,
Nunzio Giulini5,
Gioacchino Mollica2,
Domenico De Aloysio4,
Eugenio Vadora3,
Annibale Volpe1 for the Emilia-Romagna Operative Group for the Menopause (GOERM)
Department of Obstetrics, Gynaecology and Paediatric Sciences, Gynaecology Units of Policlinico of 1Modena, 2 Ferrara, 3 Parma and 4 Bologna, and 5 Division of Obstetrics and Gynecology of Cattolica, Italy
6 To whom correspondence should be addressed at: Dipartimento Integrato Materno Infantile, Unità di Ginecologia e Ostetricia, Policlinico di Modena, via del Pozzo 71, 41100, Modena, Italy. Email: cagnacci{at}unimore.it
 |
Abstract
|
---|
BACKGROUND: Seasons may influence prenatal growth and future fertility. This study investigated whether season and month of birth influenced the timing of menopause in a group of women attending three Italian menopause clinics. METHODS and RESULTS: Age at menopause of 2822 post-menopausal women (>12 months of amenorrhoea) was stratified by month and season of birth. Mean age at menopause was 49.42 years (SEM: 0.78 years). Menopause occurred earlier for women born in the spring (age 49.04±0.15 years) than in the autumn (49.97±0.14 years). The earliest menopause was found in women born in March (48.9±0.25 years) and the latest in women born in October (50.3±0.25 years). The effect of season of birth on age at menopause remained even when considering factors that in our analysis were capable of significantly interfering with the timing of menopause, such as age at menarche, body mass index, smoking habit, level of education and type of job. CONCLUSIONS: Taking into consideration the retrospective design of the study, and a possible recall bias, the present data seem to suggest that environmental factors linked to seasons are capable of interfering with the timing of a woman's ovarian exhaustion by an action exerted in the prenatal period.
Key words:
menopause/season/rhythms/fertility/environment
 |
Introduction
|
---|
A woman's fertility is limited by the amount of oocytes formed during prenatal life and by the rate of their degeneration (Adashi, 1991
). Accordingly, the final number of primordial cells may be influenced by many factors including those capable of influencing embryofetal development. Environmental and/or social factors linked to seasons, by an action exerted in prenatal life, were reported to influence fetal growth (Rantakallio, 1971
; Selvin and Janerich, 1971
; Bantje, 1987
; Weber et al., 1998
; Matsuda et al., 1993
, 1995
; Murray et al., 2000
), human lifespan (Gavrilov and Gavrilova, 1999
; Doblhammer and Vaupel, 2001
) and fertility of both men (Francois et al., 1997
; Huber et al., 2004a
) and women (Smits et al., 1997
; Lummaa and Tremblay, 2003
; Huber et al., 2004b
). This study investigates whether environmental factors linked to seasons may also influence the timing of ovarian exhaustion.
 |
Materials and methods
|
---|
A retrospective study was performed as a part of the goals of the Group Emiliano-Romagnolo for the menopause (GOERM). Four university hospitals (Bologna, Ferrara, Modena and Parma) and one hospital division (Cattolica) use an identical electronic database for their daily clinical management of peri- and post-menopausal women. Women seeking advice and/or treatment of peri- and post-menopausal disturbances at the menopause centres of the above hospitals are interviewed and examined. Their data are entered into an electronic database, from which they are retrieved at each following examination. New information is added to each woman's file whenever she is evaluated. When necessary, data of interest are extracted from the database of each single centre, combined with those of the other centres, and then analysed. In this case, for patients visiting between January 1997 and December 2001, the following parameters were extracted from the databases of Ferrara, Modena and Parma: patient number (but not patient name), date of evaluation, month and year of birth, age at menarche, number of pregnancies (including miscarriages and stillbirth), date of last spontaneous menstrual period, menopausal status at evaluation, use of estrogens, progestins or hormonal contraception during the menopausal transition, weight, body mass index (BMI; kg/m2), education, present/past type of job, and smoking habit. Education was divided into low (primary school) and high (high school and university). On the basis of the type of job, women were divided into housewives, workers (skilled or unskilled) and employers/officials. On the basis of their smoking habit, women were considered as ever or never smokers. Data of 14 320 women were retrieved (Figure 1). Of these, 1000 women were excluded because they had undergone surgical menopause or had suffered from diseases or received medicine capable of interfering with the timing of menopause. Another 6200 women in the perimenopausal transition were also excluded. In 4298 of the remaining 7120 women, the timing of the menopause was not clearly identifiable since they had received oral contraceptives or estrogen and/or progestin therapies during the menopausal transition. Accordingly, the analysis was performed in the remaining 2822 women in which physiological menopause, i.e. amenorrhoea for >12 months, was identifiable during the medical interview. The excluded and included women did not differ by season of birth, as evaluated by the
2 goodness of fit application. Of the women included in the final analysis, 1142 were from the database of Ferrara, 963 from the database of Modena and 717 from the database of Parma.
Statistical analysis was performed by the StatView 5.0.1 (1998; SAS Institute, NC) statistical package. Multiple regression analysis was performed to evaluate factors capable of independently determining the timing of menopause. Linear regression models were constructed with age at menopause as dependent variable and age at menarche, number of pregnancies, BMI, smoking habit, education, present/past type of job and season of birth as predictor variables. Seasons (e.g. spring from March 21 to June 21) were entered into the analysis as three dummy variables, present/past type of job was entered into the analysis as two dummy variables, and ever/never smokers or high/low education were entered as single dummy variable. Contingency tables were used to compare proportions of early (
40 years) and late (
55 years) menopause among women born in the different seasons (Jongbloet et al., 1994
). For all analyses, P-values <0.05 were used to reject the null hypothesis. All results are reported as the mean and SEM.
 |
Results
|
---|
No difference was observed among women from the different centres. Women had a mean age of 53.0 (±0.1) years, and BMI of 25.2 kg/m2 (±0.1). Women experienced menopause at 49.42 (±0.78 years) and had been in menopause for 3.0 (±0.02 years).
Multiple regression analysis showed that age at menopause was not related to number of pregnancies. Accordingly, this estimate was removed from the multiple regression model. Age at menopause was determined positively by age at menarche, BMI and high education, and negatively by ever smoking, and workers/employers versus housewives (Table I). Independently of these determinants, age at menopause was conditioned by season of birth (Table I).
When stratified by month of birth, age at menopause, but not age at menarche, showed a sinusoidal variation, with peak value in October (50.3±0.25 years) and a nadir in March (48.9±0.25 years) (Figure 2). Stratification by season of birth showed the earliest menopause for women born in the spring (49.04±0.15 years) and the latest menopause for women born in the autumn (49.97±0.14 years). Intermediate values were observed in women born in the winter and in summer (Figure 2).

View larger version (32K):
[in this window]
[in a new window]
|
Figure 2. Mean (±SEM) age at menopause of 2822 women, stratified by month (filled circles) or season (grey bars) of birth.
|
|
By stratifying women with earlier (
40 years) and later (
55 years) menopause by seasons, we observed that the proportion of women with earlier menopause was similar in winter (3.6%), spring (3.2%), summer (4.4%) and autumn (2.6%). In contrast, the proportion of women with later menopause was higher in autumn (8.1%) than in spring (3.7%; P=0.013) or in the other three seasons combined (4.9%; P=0.008).
 |
Discussion
|
---|
Fetal development is becoming increasingly recognized as a key determinant of future life events. For example, several studies have shown that reduced fetal growth is associated with a higher risk of non-insulin-dependent diabetes (Lithell et al., 1996
), metabolic syndrome (Valdez et al., 1994
; Fall et al., 1995
; Yarbrough et al., 1998
), neonatal morbidity, neonatal mortality (McIntire et al., 1999
) and death from cardiovascular diseases (Barker et al., 1989
, 1993
). Probably, a reduced prenatal growth is capable of conditioning future fertility in both men and women. In men, it is associated with a less likely possibility of marrying (Phillips et al., 2001
) and a reduced fertility (Francois et al., 1997
). The latter is probably consequent to a reduced number of Sertoli cells, which are necessary for future spermatogenesis (Orth et al., 1988
). In women, a reduced prenatal growth is associated with reduced ovarian fraction of primordial follicles (de Bruin et al., 1998
), reduced ovarian and uterine volume at adolescence (Ibanez et al., 2000
), ovarian hyperandrogenism and anovulation in late adolescence (Ibanez et al., 1999
). Environmental factors linked to seasons, by an action exerted in prenatal life, influence fetal growth (Rantakallio, 1971
; Selvin and Janerich, 1971
; Bantje, 1987
; Matsuda et al., 1993
, 1995
; Weber et al. 1998
; Murray et al., 2000
), human lifespan (Gavrilov and Gavrilova, 1999
; Doblhammer and Vaupel, 2001
) and fertility of both men (Francois et al., 1997
; Huber et al., 2004a
) and women (Smits et al., 1997
; Lummaa and Tremblay, 2003
; Huber et al., 2004b
), including the timing of menarche (Albright et al., 1990
; Jongbloet et al., 1994
; Tavares et al., 2004
). Our data show that the timing of ovarian exhaustion can also be influenced by prenatal factors linked to seasons, as previously suggested (Jongbloet et al., 1994
). Women born between September and December experience the latest menopause, and those born between March and June the earliest. Interestingly, this is similar to the reported seasonal variation of birth weight observed in Northern Ireland (Murray et al., 2000
) and to the effect of season of birth on human lifespan observed in Denmark and Austria (Northern hemisphere) which is exactly mirrored by the pattern observed in Australia (Southern hemisphere) (Doblhammer and Vaupel, 2001
). Depending on the geographic area, the effect of seasons may differ. Accordingly, our present data cannot be generalized. For example, reproductive performance was reported at its minimum in women born in June and July in Austria (Huber et al., 2004b
) and in women born between July and September in The Netherlands (Smits et al., 1997
).
In our analysis, timing of menarche was not dependent on the month/season of birth, but, in contrast to a previous study (van Noord et al., 1997
), it was positively related to the timing of menopause. Perimenopausal and ageing women may have difficulties in precisely recalling the date of their menarche. In addition, we collected only the referred chronological age at menarche, and not the date of menarche. This may have generated random errors and impaired the capability to detect an effect of season of birth on the timing of menarche. The same bias has probably influenced other studies incapable of showing a relationship between menarche and menopause (van Noord et al., 1997
).
It is not known which one of the seasonal environmental factors is capable of influencing the future timing of the menopause by an action exerted during prenatal life. Modifications of environmental temperature and photoperiod may influence fetal growth (Weber et al., 1998
; Murray et al., 2000
), and therefore also future reproductive capabilities. Seasonally linked diet modifications (Godfrey et al., 1996
; Mathews et al., 1999
) or exposure to infections (Finch and Crimmins, 2004
) may also be involved. Seasonal environmental factors may be operative even before conception when, by influencing oocyte maturation, they may contribute to the development of defective embryos and weak adult individuals (Jongbloet et al., 1994
). It has been suggested that such a mechanism can explain the influence of season of birth on the timing of menarche, on the occurrence of menstrual irregularities and on the incidence of earlier and later menopause (Jongbloet et al., 1994
). Beside the effect of season of birth on mean age at menopause, we also confirmed an influence of season of birth on the occurrence of later menopause. We were not able to document a similar effect on the occurrence of earlier menopause, which was instead previously reported in more numerous data sets (Jongbloet et al., 1994
).
Our present data seem to indicate that women born in autumn develop better during their prenatal life and are born with a higher number of oocytes than women born in spring. An alternative explanation might be that early mortality is highest among children born in autumn, thus selecting the fittest for survival. However, the findings of Doblhammer and Vaupel (2001)
do not support this explanation. The present study has limitations. The study is retrospective, and mainly based on data collected during a medical interview. Accordingly, a recall bias is possible. Furthermore, the study was not performed in the general population but in women referred to menopause centres of one Italian region. Accordingly, generalization of the present results may be incorrect. Many events may play a role in timing of menopause. We controlled for several of them but we cannot exclude that our analysis was partial and/or insufficient. Several items were categorized in a way which may have led to inaccurate considerations. However, taking into consideration these limitations, the data seem to suggest notable effects of the month/season of birth on the length of a woman's fertile life, further supporting a role for environmental factors in regulating adult reproductive life and the timing of its termination.
 |
References
|
---|
Adashi L (1991) The ovarian life cycle. In Yen SSC and Jaffe RB (eds) Reproductive Endocrinology, 3rd edn. W.B. Saunders Company, Philadelphia, pp. 181237.
Albright DL, Voda AM, Smolensky MH, Hsi BP and Decker M (1990) Seasonal characteristics of and age at menarche. Chronobiol Int 7, 251258.[ISI][Medline]
Bantje H (1987) Seasonality of births and birthweight in Tanzania. Soc Sci Med 24, 733739.[CrossRef][ISI][Medline]
Barker DJP, Osmond C, Golding J, Kuh D and Wadsworth MEJ (1989) Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. Br Med J 298, 564567.[ISI][Medline]
Barker DJP, Osmond C, Simmonds SJ and Wield GA (1993) The relation of small head circumference and thinness at birth to death from cardiovascular disease in adult life. Br Med J 306, 422426.[ISI][Medline]
de Bruin JP, Dorland M, Bruinse HW, Spliet W, Nikkels PGJ and te Velde ER (1998) Fetal growth retardation as a cause of impaired ovarian development. Early Hum Dev 51, 3946.[CrossRef][ISI][Medline]
Doblhammer G and Vaupel JW (2001) Lifespan depends on month of birth. Proc Natl Acad Sci USA 98, 29342939.[Abstract/Free Full Text]
Fall CHD, Osmond C, Barker DJP, Clark PMS, Hales CN, Stirling Y and Meade TW (1995) Fetal and infant growth and cardiovascular risk factors in women. Br Med J 310, 428432.[Abstract/Free Full Text]
Finch CE and Crimmins EM (2004) Inflammatory exposure and historical changes in human life-spans. Science 305, 17361739.[Abstract/Free Full Text]
Francois I, de Zegher F, Spessens C, D'Hooghe T and Vanderschuren D (1997) Low birth weight and subsequent male subfertility. Pediatr Res 42, 899901.[Abstract]
Gavrilov LA and Gavrilova NS (1999) Season of birth and human longevity. J Anti-Aging Med 2, 365366.
Godfrey K, Robinson S, Barker D, Osmond C and Cox V (1996) Maternal nutrition in early and late pregnancy in relation to placental and fetal growth. Br Med J 312, 410414.[Abstract/Free Full Text]
Huber S, Fieder M, Wallner B, Iber K and Moser G (2004a) Effects of season of birth on reproduction in contemporary humans: brief communication. Hum Reprod 19, 445447.[Abstract/Free Full Text]
Huber S, Fieder M, Wallner B, Moser G and Arnold W (2004b) Brief communication: birth month influences reproductive performance in contemporary women. Hum Reprod 19, 10811082.[Abstract/Free Full Text]
Ibanez L, de Zegher F and Potau N (1999) Anovolutaion after precocious pubarche: early markers and time course in adolescence. J Clin Endocrinol Metab 84, 26912695.[Abstract/Free Full Text]
Ibanez L, Potau N, Enriquez G and de Zegher F (2000) Reduced uterine and ovarian size in adolescent girls born small for gestational age. Pediatr Res 47, 575577.[Abstract/Free Full Text]
Jongbloet PH, Kersemaekers WM, Zielhuis GA and Verbeek AL (1994) Menstrual disorders and month of birth. Ann Hum Biol 21, 511518.[ISI][Medline]
Lithell HO, McKeigue PM, Berglund L, Mohsen R, Lithell U-B and Leon DA (1996) Relation of size at birth to non-insulin dependent diabetes and insulin concentrations in men aged 5060 years. Br Med J 312, 406410.[Abstract/Free Full Text]
Lummaa V and Tremblay M (2003) Month of birth predicted reproductive success and fittness in pre-modern Canadian women. Proc R Soc Lond 270, 23552361.[CrossRef][ISI][Medline]
Mathews F, Yudkin P and Neil A (1999) Influence of maternal nutrition on outcome of pregnancy: prospective cohort study. Br Med J 319, 339343.[Abstract/Free Full Text]
Matsuda S, Sone T, Doi T and Kahyo H (1993) Seasonality of mean bithweight and mean gestational period in Japan. Hum Biol 65, 481501.[ISI][Medline]
Matsuda S, Hiroshige V, Furuta M, Doi T, Sone T and Kahyo H (1995) Geographic difference in seasonal variation of mean birthweight in Japan. Hum Biol 67, 641656.[ISI][Medline]
McIntire DD, Bloom SL, Casey BM and Leveno KJ (1999) Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 340, 12341238.[Abstract/Free Full Text]
Murray LJ, O'Reilly DPJ, Betts N, Patteron CC, Smith GD and Evans AE (2000) Season and outdoor ambient temperature: effects on birth weight. Obstet Gynecol 96, 689695.[Abstract/Free Full Text]
Orth JM, Gunsalus GL and Lamperti AA (1988) Evidence from Sertoli cell-depleted rats indicates that spermatid number in adults depends on numbers of Sertoli cells produced during perinatal development. Endocrinology 122, 787794.[Abstract]
Phillips DJW, Handelsman DJ, Eriksson JG, Forsen T, Osmond C and Barker DJP (2001) Prenatal growth and subsequent marital status: longitudinal study. Br Med J 322, 771.[Free Full Text]
Rantakallio P (1971) The effect of a Northern climate on seasonality of births and the outcome of pregnancies. Acta Pediatr Scand 218, 5167.
Selvin S and Janerich WT (1971) Four factors influencing birthweight. Br J Prev Soc Med 25, 1216.[ISI][Medline]
Smits LJ, Van Poppel FW, Verduin JA, Jongbloet PH, Straatman H and Zielhuis GA (1997) Is fecundability associated with month of birth? An analysis of 19th and early 20th century family reconstitution data from The Netherlands. Hum Reprod 12, 25722578.[Abstract]
Tavares CH, Barbieri MA, Bettiol H, Barbieri MR and De Souza L (2004) Monthly distribution of menarche among schoolgirls from a municipality in Southern Brazil. Am J Hum Biol 16, 1723.[CrossRef][ISI][Medline]
Valdez R, Athens MA, Thompson GH, Bradshaw BS and Stern MP (1994) Birthweight and adult health outcomes in abiethnic population in the USA. Diabetologia 37, 624631.[CrossRef][ISI][Medline]
van Noord PAH, Dubas JS, Dorland JS, Boersma H and te Velde E (1997) Age at natural menopause in a population-based screening cohort: the role of menarche, fecundity, and lifestyle factors. Fertil Steril 68, 95102.[CrossRef][ISI][Medline]
Weber GW, Prossinger H and Seidler H (1998) Heights depend on month of birth. Nature 391, 754755.[CrossRef][ISI][Medline]
Yarbrough DE, Barrett-Connor E, Kritz-Silverstein D and Wingard DL (1998) Birth weight, adult weight, and girth as predictors of the metabolic syndrome in postmenopausal women: the Rancho Bernardo Study. Diabetes Care 21, 16521658.[Abstract]
Submitted on October 9, 2004;
resubmitted on February 16, 2005;
accepted on April 1, 2005.