Abstract
Background
The physical characteristics which are known to affect the ovarian reserve are age, body mass index (BMI), occupational exposures, age at menarche and menstrual cycle length. A correlation between different physical characteristics and the ovarian reserve will help to identify areas which need to be tackled to increase the chances of fertility of women in India.
Methods
In this retrospective, observational study, namely the MOTHER Study, data of women between 18 and 45 years of age, attending the selected fertility centers across different states in India were taken for evaluation. Demographic information along with information on factors potentially related to fertility like age of menarche, menstrual cycle length and occupational factors were collected by review of medical records at screening visit. Most recent AMH assay and antral follicle count (AFC) where the subject has not taken any contraceptives 12 months prior to the test were collected.
Results
Age of woman, years of marriage, years of infertility and smoking have shown effect on ovarian reserve testing like AMH and AFC. The other physical characteristics which were evaluated and considered to affect the ovarian reserve like body mass index BMI, occupational exposures, age at menarche and menstrual cycle length have not shown statistically significant correlation.
Conclusion
Age of woman and years of infertility are inversely proportional to ovarian reserve markers, namely AMH and AFC. Addictions like smoking and alcohol affect ovarian reserve.
Keywords: Age of woman, AFC, AMH, Ovarian reserve, Infertility
Introduction
Infertility is a medical problem that affects 15–20% of couples in reproductive age group [1]. In the recent times infertility has emerged as a serious health problem in India [2].
The primary function of the female ovary is to produce mature and viable oocytes for fertilization. At birth, the ovary contains about two million oocytes. This available reserve of oocytes is termed “the ovarian reserve (OR).” OR refers to the number and quality of oocytes which, at any given age, are available to produce a dominant follicle during the late follicular phase of the menstrual cycle. It has a significant effect on the fertility of the couple because as the ovary ages, the ovarian reserve declines. The determination of OR is important in the assessment and treatment of infertility [1].
The goal of OR testing is to add more prognostic information to the counseling and planning process so as to help couples choose the best available treatment option. [3] It may also help to assess the likely success of the assisted reproductive technique (ART) such as in vitro fertilization (IVF). There are many serum and ultrasound markers which can reflect the ovarian reserve. Of these, there are two markers which have been studied to better reflect ovarian reserve; they are anti-mullerian hormone assessment (AMH) and AFC [4].
Majority of AMH is secreted by the primary and pre-antral follicles. The levels are gonadotropin independent and therefore a more consistent measure of ovarian reserve than FSH. Low AMH has been associated with natural aging and infertility [4–6].
AFC is an ultrasound marker of ovarian aging. Transvaginal ultrasound on day 2–3 of menstrual cycle gives better idea of antral follicle count and thus the ovarian reserve.
The physical characteristics which are known to affect the ovarian reserve are age, body mass index (BMI), occupational exposures, age at menarche and menstrual cycle length [7–9] .
A correlation between different physical characteristics and the ovarian reserve will help to identify areas which need to be tackled to increase the chances of fertility of women in India. India is a poor resource country and all the patients attending the fertility centers cannot afford costly treatments to optimize their fertility. Any factor which can increase their fertility can reduce their financial and emotional burden.
Methods
It is a multicenter, observational and retrospective study of sub fertile women attending the selected fertility centers distributed across India, meeting the following study eligibility criteria. The MOTHER Study has been conducted in nine centers in different zones of India with the support of the Federation of Obstetric and Gynecological Societies of India (FOGSI) and ethics committee approval has been obtained from the Royal Pune Independent Ethics Committee, India with CTRI number CTRI/2018/12/016,772 and was registered on 21/12/2018.
Inclusion Criteria
Women between 18 and 45 years of age;
Not attained menopause;
Unable to conceive in last 12 months (at time of study consent);
Regular menstrual cycles with length of 25–35 days with ≤ 4 days difference between cycles;
Presence of both ovaries;
Willing and able to sign the study informed consent form.
Exclusion Criteria
Presence of gynecological disorders such as menorrhagia or dysfunctional uterine bleeding;
History of ovarian surgery;
Irregular menstruation (interval: less than 21 days or more than 45 days);
History of hypertension (controlled or uncontrolled);
History of taking injectable hormonal contraception in the preceding year or hormonal therapy/herbal product in last 12 months prior to most recent AMH or AFC test;
Malignancy or exposure to cytotoxic drugs and pelvic radiotherapy, endocrinological disorders, including abnormal prolactin, TSH or diabetes.
Total nine study centers with 422 subjects had participated in the study. The centers were from New Delhi, Mumbai, Guwahati, Chennai, Indore, Trichy, Islampur, Bhopal and Tumkur.
On screening visit (day 0) consent was obtained from each enrolled woman. Data was recorded from medical records of the patient available at the study site. Any study specific information missing in the medical records, was collected in pending data collection visit (within 7 days of screening visit).
This second visit was also performed over the telephone and the conversation was recorded on subject medical record.
The medical records were reviewed for a cutoff period of one year prior to the date of informed consent and study specific data were collected as given below:
Height, weight and BMI value.
AMH test result
AFC result.
If most recent data were not available, these measurements were done at screening visit.
-
4.
Demographic information along with information like age of menarche, menstrual cycle length and occupational factors, etc. was collected and recorded on the case report form.
The study data was analyzed using recommended version of SPSS software. The database created was subjected to linear regression models to examine the association of the ovarian reserve markers AMH and AFC with the physical characteristics like age, BMI, occupational exposures, age at menarche and menstrual cycle length.
Results
In the demographic history most of the women were of 20–45 age. The urban (51.07%) as well as rural (32.35%) population was covered with 16% of semi-urban women. 16% study population was having professional background while almost 51.42% women were home makers. More than 65% women were graduates and had non-vegetarian diet pattern (Table 1).
Table 1.
Demographic history
| Parameters | Frequency (%) |
|---|---|
| Age in years * (n = 422) | 30.81 ± 4.84 20–45 years |
| Locality | |
| Local | 355 (84.12) |
| Immigrant | 67 (15.88) |
| Type of Locality# | |
| Urban | 191 (51.07) |
| Rural | 121 (32.35) |
| Semi urban | 62 (16.58) |
| Occupation | |
| Professional | 70 (16.59) |
| Semi professional | 39 (9.24) |
| Semi | 23 (5.45) |
| Semi skilled | 19 (4.50) |
| Unskilled | 2 (0.47) |
| Unemployed | 48 (11.37) |
| Farmer | 3 (0.71) |
| Laborer | 1 (0.24) |
| Home maker | 217 (51.42) |
| Education | |
| Post graduate | 83 (19.67) |
| Graduate | 203 (48.10) |
| Higher secondary school | 82 (19.43) |
| School 5–10 Y | 41 (9.72) |
| School < 5 Y | 10 (2.37) |
| Illiterate | 3 (0.71) |
| Diet$ | |
| Vegetarian | 126 (33.78) |
| Non Vegetarian | 247 (66.22) |
*Mean ± SD with Range; #Data is missing for 48 cases, $Data is missing for 49 cases
When the behavioral pattern was studied almost 99% women had never smoked (Table 2) nor consumed alcohol, though significant reduction of ovarian reserve was observed in small population of smokers (Table 10).
Table 2.
Behavioral history
| Parameters | Frequency (%) |
|---|---|
| History of smoking | |
| Current | 3 (0.71) |
| Never | 419 (99.29) |
| Years of smoking | |
| 10-Years | 2 |
| 5- Years | 1 |
| History of alcohol^ | |
| Current | 2 (0.48) |
| Never | 419 (99.52) |
^Data is missing for one case
Table 10.
AFC group as an outcome variable with infertility history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Years of marriage | 373 | 1.104 | 0.018 (S) | 1.016–1.199 |
| Years of infertility | 418 | 1.097 | 0.039 (S) | 1.004–1.19 |
| Duration of sexual activity without contraception (Months) | 422 | 1.005 | 0.119 | 0.998–1.013 |
| Type of infertility | ||||
| Primary * | 314 (74.41) | 1 | ||
| Secondary | 108 (25.59) | 1.202 | 0.655 | 0.535–2.698 |
| Cause of infertility | ||||
| Combined factor * | 14 (3.32) | 1 | ||
| Male | 70 (16.59) | 0.582 | 0.650 | 0.056–6.04 |
| Not known | 3 (0.71) | NA | NA | NA |
| Ovarian | 86 (20.38) | 1.71 | 0.623 | 0.20–14.51 |
| Sexual dysfunction | 6 (1.42) | 6.50 | 0.166 | 0.459–91.92 |
| Tubal | 52 (12.32) | 1.083 | 0.945 | 0.11–10.54 |
| Unexplained | 175 (41.47) | 0.79 | 0.831 | 0.094–6.68 |
| Uterine | 16 (3.79) | 0.866 | 0.922 | 0.049–15.27 |
| Treatment for infertility | ||||
| Yes * | 263 (62.32) | 1 | ||
| No | 159 (37.68) | 0.556 | 0.166 | 0.242–1.27 |
| Contraceptive usage in 1Year | ||||
| Yes | 2 (0.47) | |||
| No | 420 (99.53) | NA | ||
*Reference category, NA variable is omitted because of few numbers, S Significant
In the above table, the variables such as years of marriage and years of infertility were associated with the outcome variable. If the marriage year has been increased by one unit then the odds of being in the poor reserved group increases by 1.104 units with P value 0.018 < 0.05 and it is statistically significant. Similarly if the year of infertility has increased by one unit then the odds of being in the poor reserved group increases by 1.097 units with the significant P value 0.039 < 0.05 and all the other variables were not associated with the outcome variable
Primary infertility was the most common complaint and in that ovarian causes, male factor and tubal factor were on the rise after the unexplained variety, together contributing to almost 90.76% causes of infertility. (Table 3) Mean body mass index of study population was 25.75. The obstetric history and menstrual pattern is shown in Tables 4 and 5, respectively (Table 6).
Table 3.
Infertility history
| Parameters | Frequency (%) |
|---|---|
| Years of Marriage* (N = 373) | 5.81 ± 3.92 1–28 Years |
| Years of Infertility* (N = 418) | 4.66 ± 3.60 0.5–19 Years |
| Duration of sexual activity without contraception (Months)* (N = 422) | 51.77 ± 44.50 1.5–228 Months |
| Type of infertility | |
| Primary | 314 (74.41) |
| Secondary | 108 (25.59) |
| Cause of infertility | |
| Combined factor | 14 (3.32) |
| Male | 70 (16.59) |
| Not known | 3 (0.71) |
| Ovarian | 86 (20.38) |
| Sexual dysfunction | 6 (1.42) |
| Tubal | 52 (12.32) |
| Unexplained | 175 (41.47) |
| Uterine | 16 (3.79) |
| Treatment for infertility | |
| Yes | 263 (62.32) |
| No | 159 (37.68) |
| Contraceptive usage in 1Year | |
| Yes | 2 (0.47) |
| No | 420 (99.53) |
*Mean ± SD with range
Table 4.
Menstrual and obstetric history
| Parameters | Frequency (%) |
|---|---|
| Age at Menarche (N = 422)* | 12.91 ± 1.16 |
| 10–19 y | |
| Menstrual cycle@ | |
| Regular | 324 (87.10) |
| Irregular | 48 (12.90) |
| Gravida | |
| 0 | 372 (88.15) |
| 1 | 36 (8.53) |
| 2 | 9 (2.13) |
| 3 | 4 (0.95) |
| 6 | 1 (0.24) |
| Para | |
| 0 | 382 (90.52) |
| 1 | 35 (8.29) |
| 2 | 5 (1.18) |
| Living | |
| 0 | 391 (92.65) |
| 1 | 30 (7.11) |
| 2 | 1 (0.24) |
| Abortion | |
| 0 | 346 (81.99) |
| 1 | 54 (12.80) |
| 2 | 15 (3.55) |
| 3 | 4 (0.95) |
| 4 | 1 (0.24) |
| 5 | 1 (0.24) |
| 6 | 1 (0.24) |
Table 5.
Body measurements
| Parameters | N | Mean(SD) | Range |
|---|---|---|---|
| Height (cm) | 422 | 155.48(7.73) | 120–174 |
| Weight (kg) | 422 | 61.3(11.41) | 36–107.6 |
| Body Mass Index(BMI) | 422 | 25.75(5.12) | 12.55–59.5 |
Table 6.
AFC categories
| AFC | Frequency (%) |
|---|---|
| Normal reserve | 390(92.64) |
| Poor reserve | 31(7.36) |
| Total | 421(100) |
The AFC calculated on day 2 or 3 of menstrual cycle was normal in 92.64% cases (390 women). It was poor in 31 cases i.e., 07.31% of study population. Almost similar findings were found in case of AMH estimations which showed normal results in 91.71% cases. Values showing poor ovarian reserve (low AMH) were noted in 8.29% (Tables 7 and 8).
Table 7.
AMH categories
| AMH | Frequency (%) |
|---|---|
| Normal Reserve | 387(91.71) |
| Poor Reserve | 35(8.29) |
| Total | 422(100) |
Table 8.
AFC group as an outcome variable with demographic details
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Age in years | 421 | 1.12 | 0.002 (S) | 1.04–1.208 |
| Locality | ||||
| Local* | 355 (84.12) | 1 | 0.376–2.75 | |
| Immigrant | 67 (15.88) | 1.017 | 0.973 | |
| Type of Locality | ||||
| Urban* | 191 (51.07) | 1 | ||
| Rural | 121 (32.35) | 0.9375 | 0.883 | 0.396–2.214 |
| Semi Urban | 62 (16.58) | 0.388 | 0.218 | 0.086–1.75 |
| Occupation | ||||
| Professional* | 70 (16.59) | 1 | ||
| Semi professional | 39 (9.24) | 1.207 | 0.840 | 0.192–7.55 |
| Skilled | 23 (5.45) | 2.12 | 0.425 | 0.332–13.59 |
| Semi skilled | 19 (4.50) | 4.1875 | 0.097 | 0.77–22.71 |
| Unskilled | 2 (0.47) | NA | NA | NA |
| Unemployed | 48 (11.37) | 2.59 | 0.207 | 0.59–11.42 |
| Farmer | 3 (0.71) | 11.16 | 0.076 | 0.77–160.37 |
| Laborer | 1 (0.24) | NA | NA | NA |
| Home maker | 217 (51.42) | 1.66 | 0.431 | 0.468–5.93 |
| Education | ||||
| Post graduate* | 83 (19.67) | 1 | ||
| Graduate | 203 (48.10) | 0.882 | 0.807 | 0.323–2.406 |
| Higher secondary school | 82 (19.43) | 0.833 | 0.771 | 0.244–2.845 |
| School 5–10 Years | 41 (9.72) | 1.78 | 0.365 | 0.51–6.22 |
| School < 5 Years | 10 (2.37) | 1.425 | 0.755 | 0.153–13.21 |
| Illiterate | 3 (0.71) | 6.41 | 0.151 | 0.505–81.37 |
| Diet | ||||
| Vegetarian* | 126 (33.78) | 1 | ||
| Non Vegetarian | 247 (66.22) | 0.75 | 0.504 | 0.328–1.72 |
*Reference category, S Significant
In the above table, univariate logistic regression was done. The outcome variable is AFC group (Normal / Poor reserved). Here the age is the only factor which is associated with the outcome variable. If there is a one unit increase in age then the odds of being in the poor reserved group increases by 1.12 units with the 95% CI (1.04–1.208) which does not include 1 in the CI. Also the P value is statistically significant. P = 0.002 < 0.05. All the other independent factors do not have any association with the outcome variable
Univariate logistic regression was done. The outcome variable was AFC group (Normal/Poor reserve). Here age was the only factor which was associated with the outcome variable. (Table 9) If there was a one unit increase in age then the odds of being in the poor reserve group increased by 1.12 units with the 95% CI (1.04–1.208) which does not include 1 in the CI. Also the P value is statistically significant (P = 0.002) (P < 0.05). All other independent factors don’t have any association with the outcome variable. When the behavioral pattern was studied almost 99% women had never smoked. As depicted below in Table 10, current smokers were compared with never smokers. The odds of being in the poor reserve group (based on the AF count) is 26.82 times more for the smokers as compared to non-smokers with 95% CI and the P value is 0.008 (P < 0.05), which is statistically significant. Thus, smoking is one of the determinants of reducing ovarian reserve.
Table 9.
AFC group as an outcome variable with behavioral history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| History of smoking | ||||
| Current | 3 (0.71) | 26.82 | 0.008 (S) | 2.36–304.708 |
| Never* | 419 (99.29) | 1 | ||
| History of alcohol | ||||
| Current | 2 (0.48) | NA | ||
| Never | 419 (99.52) | |||
*Reference category; NA variable is omitted because of few numbers S Significant
In the above table current smokers were compared with never smokers. The odds of being in the poor reserved group (based on the AF count) is 26.82 times more for the smokers as compared to non-smokers with 95% CI and the P value is 0.008 < 0.05, which is statistically significant
The variables such as years of marriage and years of infertility were associated with the outcome variable. If the marriage year has increased by one unit then the odds of being in the poor reserve group increased by 1.104 units with P value = 0.018 (P < 0.05) and it is statistically significant. Similarly, if the year of infertility has increased by one unit then the odds of being in the poor reserve group increases by 1.097 units with the significant P value = 0.039 (P < 0.05). All the other variables were not associated with the outcome variable. (Table 11) The other factors like causes of infertility, whether couple had received treatment for infertility have not shown statistical significance when its relation was studied with AFC.
Table 11.
AFC group as an outcome variable with obstetric history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Gravida | ||||
| 0* | 372 (88.15) | 1 | ||
| 1 | 36 (8.53) | 0.78 | 0.743 | 0.177–3.43 |
| 2 | 9 (2.13) | 3.79 | 0.107 | 0.749–19.18 |
| 3 | 4 (0.95) | 4.42 | 0.205 | 0.44–44.02 |
| 6 | 1 (0.24) | NA | ||
| Para | ||||
| 0* | 382 (90.52) | 1 | ||
| 1 | 35 (8.29) | 1.76 | 0.319 | 0.577–5.37 |
| 2 | 5 (1.18) | 3.413 | 0.280 | 0.36–31.65 |
| Living | ||||
| 0 | 391 (92.65) | NA | ||
| 1 | 30 (7.11) | |||
| 2 | 1 (0.24) | |||
| Abortion | ||||
| 0* | 346 (81.99) | 1 | ||
| 1 | 54 (12.80) | 1.75 | 0.247 | 0.67–4.51 |
| 2 | 15 (3.55) | 1 | 0.99 | 0.12–7.94 |
| 3 | 4 (0.95) | 4.66 | 0.190 | 0.46–46.65 |
| 4 | 1 (0.24) | NA | ||
| 5 | 1 (0.24) | NA | ||
| 6 | 1 (0.24) | NA | ||
*Reference category; NA variable is omitted because of few numbers
In the above table, none of the obstetric parameters were associated with the outcome variable
The other factors which were studied have shown no statistical significant impact on antral follicle count and hence the ovarian reserve.
In our study there was mixed population ranging from those trying to conceive for first time to multigravidas and cases of multiple abortions who were cases of secondary infertility. In Tables 12, 13, none of the obstetric parameters were associated with the outcome variables of antral follicle count and hence ovarian reserve or vice versa.
Table 12.
AFC group as an outcome variable with menstrual history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Age at menarche | 422 | 0.804 | 0.208 | 0.57–1.12 |
| Menstrual cycle | ||||
| Regular* | 324 (87.10) | 1 | ||
| Irregular | 48 (12.90) | 1.42 | 0.494 | 0.517–3.92 |
*Reference category
In the above table, none of the parameters related with menstrual history were associated with the outcome variable
Table 13.
AFC group as an outcome variable with body measurements
| Parameters | N | Odds ratio | P value | 95% CI |
|---|---|---|---|---|
| Height | 422 | 0.99 | 0.682 | 0.946–1.03 |
| Weight | 422 | 0.99 | 0.950 | 0.967–1.03 |
| Body Mass Index(BMI) | 422 | 1.004 | 0.911 | 0.935–1.07 |
In the above table, none of the body measurement parameters were associated with the outcome variable
Age at menarche was also the factor of importance from ovarian reserve point of view but in our study both age at menarche and pattern of menstrual cycle were not associated with the outcome variable.
It is believed that body mass index affects ovarian reserve but from Table 14 it is clear that in our study height, weight (independently) and BMI were not associated with the outcome variable. An cases of ovarian surgery (including drilling) and irregular menstrual cycles were excluded many classic cases of polycystic ovary syndrome were not included so BMI is not found to correlate with antral follicle count of non-metabolic cases of infertility (Table 15).
Table 14.
AMH group as an outcome variable with demographic details
| Parameters | Frequency (%) | Odds ratio | P value | 95% CI |
|---|---|---|---|---|
| Age in years | 422 | 1.18 | 0.0001 (S) | 1.104–1.278 |
| Locality | ||||
| Local* | 355 (84.12) | 1 | ||
| Immigrant | 67 (15.88) | 2.709 | 0.011 (S) | 1.257–5.83 |
| Type of locality | ||||
| Urban * | 191 (51.07) | 1 | ||
| Rural | 121 (32.35) | 1.636 | 0.267 | 0.686–3.9 |
| Semi Urban | 62 (16.58) | 0.832 | 0.783 | 0.224–3.083 |
| Occupation | ||||
| Professional* | 70 (16.59) | |||
| Semi professional | 39 (9.24) | |||
| Skilled | 23 (5.45) | |||
| Semi-skilled | 19 (4.50) | NA | NA | |
| Unskilled | 2 (0.47) | NA | ||
| Unemployed | 48 (11.37) | |||
| Farmer | 3 (0.71) | |||
| Laborer | 1 (0.24) | |||
| Home maker | 217 (51.42) | |||
| Education | ||||
| Post graduate* | 83 (19.67) | 1 | ||
| Graduate | 203 (48.10) | 1.056 | 0.906 | 0.423–2.63 |
| Higher secondary school | 82 (19.43) | 0.55 | 0.366 | 0.156–1.979 |
| School 5–10 Years | 41 (9.72) | 1.17 | 0.808 | 0.323–4.26 |
| School < 5 Years | 10 (2.37) | 1.206 | 0.868 | 0.132–10.95 |
| Illiterate | 3 (0.71) | 5.42 | 0.189 | 0.435–67.61 |
| Diet | ||||
| Vegetarian * | 126 (33.78) | 1 | ||
| Non Vegetarian | 247 (66.22) | 0.696 | 0.400 | 0.300–1.616 |
*Reference category, NA variable is omitted because of few numbers, S Significant
In the above table, univariate logistic regression was done. The outcome variable is AMH group (Normal/Poor reserved). Here the age and the participants’ locality are the only factors which were associated with the outcome variable. If there is a one unit increase in age then the odds of being in the poor reserved group increases by 1.18 units with the 95% CI (1.104–1.278) which does not include 1 in the CI. Also the P value is statistically significant. P = 0.0001 < 0.05. As compared to local residents the odds of being in the poor resident group is 2.709 times more for the immigrant residents. All the other independent factors do not have any association with the outcome variable
Table 15.
AMH group as an outcome variable with behavioral history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| History of Smoking | ||||
| Current | 3 (0.71) | 5.66 | 0.161 | 0.5–64.05 |
| Never * | 419 (99.29) | 1 | ||
| History of Alcohol | ||||
| Current | 2 (0.48) | |||
| Never | 419 (99.52) | NA |
*Reference category; NA variable is omitted because of few numbers, S Significant
In the above table, none of the parameters related with the behavioral history were associated with the outcome variable
After univariate logistic regression with the outcome variable-AMH (Normal/Poor reserve), the age of woman and the participants’ locality were the only factors which were associated with the outcome variable. If there is a one unit increase in age then the odds of being in the poor reserve group increases by 1.18 units with the 95% CI (1.104–1.278) which does not include 1 in the CI. Also the P value is statistically significant. P = 0.0001 (P < 0.05). As compared to local residents the odds of being in the poor reserve group is 2.709 times more for the immigrant residents. All the other independent factors do not have any association with the outcome variable.
In Table 16, none of the parameters related with the behavioral history were associated with the outcome variable.
Table 16.
AMH group as an outcome variable with infertility history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Years of marriage | 373 | 1.038 | 0.426 | 0.945–1.141 |
| Years of infertility | 418 | 1.114 | 0.009 (S) | 1.0269–1.208 |
| Duration of sexual activity without contraception (Months) | 422 | 1.012 | 0.0001 (S) | 1.005–1.019 |
| Type of infertility | ||||
| Primary* | 314 (74.41) | 1 | ||
| Secondary | 108 (25.59) | 1.006 | 0.986 | 0.456–2.222 |
| Cause of infertility | ||||
| Combined Factor* | 14 (3.32) | 1 | ||
| Male | 70 (16.59) | 0.176 | 0.098 | 0.022–1.37 |
| Not Known | 3 (0.71) | NA | NA | NA |
| Ovarian | 86 (20.38) | 0.701 | 0.673 | 0.134–3.646 |
| Sexual Dysfunction | 6 (1.42) | 6 | 0.109 | 0.67–53.68 |
| Tubal | 52 (12.32) | 0.782 | 0.780 | 0.139–4.378 |
| Unexplained | 175 (41.47) | 0.441 | 0.319 | 0.0885–2.204 |
| Uterine | 16 (3.79) | 0.4 | 0.476 | 0.0322–4.959 |
| Treatment for infertility | ||||
| Yes * | 263 (62.32) | 1 | ||
| No | 159 (37.68) | 0.316 | 0.012 (S) | 0.128–0.780 |
| Contraceptive usage in 1Year | ||||
| Yes | 2 (0.47) | |||
| No | 420 (99.53) | NA | ||
*Reference category, NA variable is omitted because of few numbers, S Significant
In the above table, the variables such as years of infertility and duration of sexual activity without contraception were associated with the outcome variable. If the year of infertility has increased by one unit then the odds of being in the poor reserved group increases by 1.114 units with the significant P value 0.009 < 0.05. Also the duration of sexual activity without contraception increases by 1 unit then the odds of reservation increase by 1.012 units with P value 0.0001 < 0.05 and all the other variables were not associated with the outcome variable
In table, the variables such as years of infertility and duration of sexual activity without contraception were associated with the outcome variable. If the year of infertility increases by one unit then the odds of being in the poor reserve group increases by 1.114 units with the significant ‘P’ value of 0.009 (< 0.05). If the duration of sexual activity without contraception increases by 1 unit then the odds of reserve increase by 1.012 units with P value 0.0001 (P < 0.05). All the other variables were not associated with the outcome variable (Table 17).
Table 17.
AMH group as an outcome variable with obstetric history
| Parameters | Frequency (%) | Odds Ratio | P value | 95% CI |
|---|---|---|---|---|
| Gravida | ||||
| 0* | 372 (88.15) | 1 | ||
| 1 | 36 (8.53) | 2.55 | 0.055 | 0.978–6.674 |
| 2 | 9 (2.13) | 1.59 | 0.664 | 0.1925–13.24 |
| 3 | 4 (0.95) | 4.25 | 0.216 | 0.428–42.34 |
| 6 | 1 (0.24) | NA | ||
| Para | ||||
| 0* | 382 (90.52) | 1 | 0.968 | 0.297–3.53 |
| 1 | 35 (8.29) | 1.025 | ||
| 2 | 5 (1.18) | NA | ||
| Living | ||||
| 0 | 391 (92.65) | 1 | ||
| 1 | 30 (7.11) | 1.246 | 0.729 | 0.358–4.33 |
| 2 | 1 (0.24) | NA | ||
| Abortion | ||||
| 0* | 346 (81.99) | 1 | ||
| 1 | 54 (12.80) | 1.75 | 0.211 | 0.725–4.26 |
| 2 | 15 (3.55) | 0.843 | 0.872 | 0.106–6.66 |
| 3 | 4 (0.95) | NA | ||
| 4 | 1 (0.24) | NA | ||
| 5 | 1 (0.24) | NA | ||
| 6 | 1 (0.24) | NA | ||
*Reference category NA variable is omitted because of few numbers
In the above table, none of the obstetric parameters were associated with the outcome variable
In Tables 18 and 19 none of the parameters were associated with the outcome variable.
Table 18.
AMH group as an outcome variable with menstrual history
| Parameters | Frequency (%) | Odds ratio | P value | 95% CI |
|---|---|---|---|---|
| Age at menarche | 422 | 0.979 | 0.892 | 0.725–1.322 |
| Menstrual cycle | ||||
| Regular* | 324 (87.10) | 1 | ||
| Irregular | 48 (12.90) | 1.57 | 0.347 | 0.612–4.029 |
*Reference category
In the above table, none of the parameters related with menstrual history were associated with the outcome variable
Table 19.
AMH group as an outcome variable with body measurements
| Parameters | N | Odds ratio | P value | 95% CI |
|---|---|---|---|---|
| Height | 422 | 0.99 | 0.892 | 0.953–1.041 |
| Weight | 422 | 0.985 | 0.366 | 0.954–1.017 |
| Body mass index | 422 | 0.964 | 0.339 | 0.895–1.038 |
In the above table, none of the body measurement parameters were associated with the outcome variable
Discussion
Various factors of infertility have been discussed and are predicted to affect the ovarian reserve.
Age of menarche, age at marriage and body mass index are the most frequently studied parameters.
The MOTHER study demonstrates that age of the woman and years of infertility have inverse relationship with ovarian reserve tests like anti-mullerian hormone (AMH) and antral follicle count (AFC). In the study by Erica et al. [10] the age of the female correlates with levels of AMH. The comparative study by Hvidman et al. in 2016 on ovarian reserve testing of women with infertility below age 40 with women of the same age with no history of infertility shows no difference in the results of both AMH and AFC. This study concludes that women of the same age group have similar ovarian reserve irrespective of fertility status. [11] The MOTHER study has shown that increasing age in infertile women is associated with reduced ovarian reserve markers P = 0.002 (P < 0.05).
In our study, years of marriage and years of infertility are associated with a poor reserve P value = 0.018 (p < 0.05). This, when studied in depth again correlates with age of women.
The other factors like cause of infertility and couple receiving treatment for infertility do not show statistical significance in relation to AFC and AMH. The exclusion criteria have also affected these observations as most of the classic cases of PCOS on treatment and those approaching POF have been excluded.
The next parameter analyzed is the effect of body mass index [BMI] on the ovarian reserve tests. In our study height, weight and BMI are not affecting ovarian reserve tests. This is in contrast to various studies which show significant association between body size and ovarian reserve tests. [10] In the study by Bernardi et al. [12] excessive obesity is associated with a compromised ovarian reserve. The study by H. Irene Su shows that AMH is lower in obese women in comparison to women with normal weight but AFC is not affected by body size [13] . Our study excludes women with PCOS and hence ovulatory infertility due to obesity is excluded. In the study by Rich Edwards, ovulatory infertility is related to overweight women and a sedentary lifestyle [14].
The effect of occupational exposures like moving or lifting heavy objects, level of physical exertion and work shifts on ovarian reserve tests are studied in this study. They do not show any effect on the ovarian reserve tests.
The baseline reproductive characteristics of the study population like age at menarche and menstrual cycle patterns are analyzed in this study. They show no effect on the ovarian reserve tests in our study. In the study by Gizzo et al. the menstrual cycle diary is more predictive than chronological age for determining ovarian reserve and age. In women who are older than 35 years, lengthening of menstrual cycle is considered as a marker of ovarian aging. Chronological age is more accurate for predicting pregnancy [15]. In the study by Erica et al. [10] menstrual cycle length of more than 35 days is associated with increased AMH .The study by Zhu.R et al. shows that AMH and AFC positively correlate with menstrual cycle length (MCL). A longer MCL (more than 35 days and between 25–34 days) shows an association with increased AMH in this study. A shorter MCL is associated with decreased AMH. MCL is associated with age factor also but is not associated with lifestyle factors [16].
Age at menarche does not show any association with ovarian reserve tests in the MOTHER Study. In the study by Ramona et al. age at menarche is associated with a poor ovarian reserve in later ages. In this study by Ramona et al. [17] occurrence of menarche is related to follicular size and speed of follicular recruitment which in turn predicts poor ovarian reserve in later life.
In our study, the behavior of women like smoking and alcohol usage and their effects on ovarian reserve tests were studied. 99% of women are non-smokers and have never consumed alcohol. In this small population of smokers there is a significant reduction in ovarian reserve mainly affecting AMH and to some effect AFC.
MOTHER study has various limitations like not being a comparative study between a fertile and an infertile population of women with respect to ovarian reserve tests. The sample size is not equal from all zones in the country. This was an attempt to find out the fertility potential and factors affecting it in various population zones of India.
Conclusion
Age of woman and years of infertility are inversely proportional to ovarian reserve markers, namely AMH and AFC. Addictions like smoking and alcohol affect ovarian reserve. Further comparative and prospective studies are needed. But MOTHER study has given a clear message to Indian women to avoid late marriages, late conceptions and all addictions.
Acknowledgements
The Federation of Obstetric and Gynecological Society of India (FOGSI). Dr.Jaideep Malhotra, President, FOGSI 2018. Dr.Nandita Palshetkar, President, FOGSI 2019. Dr.Jaydeep Tank, Secretary General, FOGSI
Charmila Ayyavoo
She is the Director of Aditi Hospital & Parvathy Ayyavoo Fertility Centre, Trichy. She is a consultant to the Southern Railway Hospital, Trichy. She was the Chairperson of the Clinical Research Committee, FOGSI in 2016-18. She has won several awards from FOGSI. She is the recipient of Woman’s day award in Trichy in 2020. She is the President of the Menstrual Hygiene Consortium, Tamilnadu. She has conducted health camps for self help group women and awareness programs in 200 schools and colleges. She has arranged for incinerators and napkin vending machines for 25 schools and colleges. She is the Joint Treasurer of the Indian Fertility Society, Tamilnadu chapter. She has edited books for FOGSI and is the coordinator for FOGSI FOCUS books for 2020.
Funding
The Federation of Obstetric and Gynecological Society of India (FOGSI).
Compliance with Ethical Standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical Approval
Royal Pune Independent Ethics Committee, India with CTRI number CTRI/2018/12/016772 and was registered on 21/12/2018.
Informed Consent
Informed consent was obtained from all individual participants involved in the study.
Footnotes
Charmila Ayyavoo is the Director of Aditi Hospital and Parvathy Ayyavoo Fertility Centre in , Usman Ali Street, TVS Tolgate, Trichy-620020, Tamil Nadu, India. Asha Baxi is a MS, FRCOG, FICOG degree holder in Disha Fertility and Surgical Centre, E-30, Saket Nagar, Indore, Madhya Pradesh, 452018, India. Gorakh Mandrupkar is a MD in Mandrupkar Clinic, More Nagar, Near ST Stand, Islampur (Dist- Sangli), Maharashtra, 415409, India . Sonia Malik is a MD in Southend Fertility and IVF, Max Smart Super Speciality Hospital, Press Enclave Road, Malviya Nagar, Saket, New Delhi, 110017, India. Jayam Kannan is MD, DGO, FICOG in Garbba Rakshambigai Fertility Center, Jayam Hospital, No. 4, 6th Cross Street, United India Colony, Kodambakkam, Chennai, Tamil Nadu, 600024, India. Deepak Goenka is MD, FICOG, FIAOG in Institute of Human Reproduction, Bharalumukh, Guwahati, Assam, 781009, India. Nagesh Kumar is DGO in Janani Hospital and IVF Centre, Gandhi Nagar, Tumkur, Karnataka, 572101, India. Monica Singh is MD, DNB, FMAS, FICOG in Bhopal Test-tube-baby and Endoscopy Centre, E-1/13A, Arera Colony, Near No. 7 Stop, Bhopal, Madhya Pradesh, 462016, India. Sharda Kulkarni is MD, DGO, FRCOG in Siddhilife Assisted Reproduction Center, Borival West Mumbai, 40009, India. Kanchan Murarka Agarwal is DGO, FICOG, FIAOG in Institute of Human Reproduction, Bharalumukh, Guwahati, Assam, 781009, India. Sonam Baxi is MS in Disha Fertility and Surgical Centre, E-30, Saket Nagar, Indore, Madhya Pradesh, 452018, India.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Roudebush WE, Kivens WJ, Mattke JM. Biomarker of ovarian reserve. Biomark Insights. 2008;3:259–268. doi: 10.4137/BMI.S537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jaiswar SP, Natu SM, Sankhwar PL, et al. Prediction of poor ovarian response by biochemical and biophysical markers: a logistic regression model. J ObstetGynecol India. 2015;65(6):411–416. doi: 10.1007/s13224-014-0639-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Practice Committee of the American Society for Reproductive Medicine: Testing and Interpreting Measures of Ovarian Reserve: A Committee Opinion. Fertility and Sterility 2015; 103(3): 0015–0282 [DOI] [PubMed]
- 4.Durlinger A. Regulation of ovarian function: the role of anti-mullerian hormone. Reproduction. 2002;124(5):601–9. doi: 10.1530/rep.0.1240601. [DOI] [PubMed] [Google Scholar]
- 5.Fanchin R. 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. 2005;20(4):923–927. doi: 10.1093/humrep/deh688. [DOI] [PubMed] [Google Scholar]
- 6.Wachs DS. Serum anti-mullerian hormone concentrations are not altered by acute administration of follicle stimulating hormone in polycystic ovary syndrome and normal women. J ClinEndocrinolMetab. 2007;92:1871–1874. doi: 10.1210/jc.2006-2425. [DOI] [PubMed] [Google Scholar]
- 7.Gnoth C, Godehardt E, Frank-Herrmann P, et al. Definition and prevalence of subfertility and infertility. Hum Reprod. 2005;20(5):1144–1147. doi: 10.1093/humrep/deh870. [DOI] [PubMed] [Google Scholar]
- 8.Haller- Kikkatalo K. The prevalence and phenotypic characteristics of spontaneous premature ovarian failure a general population registry-based study. Hum Reprod. 2015;30(5):1229–1238. doi: 10.1093/humrep/dev021. [DOI] [PubMed] [Google Scholar]
- 9.Tucker EJ. Premature ovarian insufficiency: new perspectives on genetic cause and phenotypic spectrum. Endocr Rev. 2016;37(6):609–635. doi: 10.1210/er.2016-1047. [DOI] [PubMed] [Google Scholar]
- 10.Marsh EE, et al. Novel correlates between antimüllerian hormone and menstrual cycle characteristics in African-American women (23–35 years-old) Fertil Steril. 2016;106(2):443–450. doi: 10.1016/j.fertnstert.2016.04.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hvidman HW. Infertile women below the age of 40 have similar anti-Müllerian hormone levels and antral follicle count compared with women of the same age with no history of infertility. Hum Reprod. 2016;31(5):1034–1045. doi: 10.1093/humrep/dew032. [DOI] [PubMed] [Google Scholar]
- 12.Bernardi Lia A. Relationship between obesity and anti-Müllerian hormone in reproductive-aged African American women. Obes J. 2017;25(1):229–235. doi: 10.1002/oby.21681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Irene Su H, Sammel MD. ScD Body size affects measures of ovarian reserve in late reproductive age women. Menopause. 2008;15(5):857–861. doi: 10.1097/gme.0b013e318165981e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rich-Edwards JW. Physical activity, body mass index, and ovulatory disorder infertility. Epidemiology. 2002;13(2):184–190. doi: 10.1097/00001648-200203000-00013. [DOI] [PubMed] [Google Scholar]
- 15.Gizzo S, Andrisani A, Noventa M, et al. Menstrual cycle length: a surrogate measure of reproductive health capable of improving the accuracy of biochemical/sonographical ovarian reserve test in estimating the reproductive chances of women referred to ART. ReprodBiolEndocrinol. 2015;13:28. doi: 10.1186/s12958-015-0024-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhu R. Antimüllerian hormone, antral follicle count and ovarian volume predict menstrual cycle length in healthy women. ClinEndocrinol. 2017;390(10093):510–520. doi: 10.1111/cen.12984. [DOI] [PubMed] [Google Scholar]
- 17.Perhar R, Shukla R, Mishra A. Age at menarche as a predictor of diminished ovarian reserve: a new approach. Int J ReprodContraceptObsGynaecol. 2015;4(1):235–238. [Google Scholar]