Abstract：
OBJECTIVE: To study the association between antimüllerian hormone (AMH) levels and time of pregnancy. Although it has been hypothesized that serum AMH levels may indicate the chance of conception, findings have been mixed. Given that any association is expected to be modest, and it is possible that previous studies have been underpowered, we investigated this relationship in the largest prospective cohort to date.
METHODS: Prospective time-to-pregnancy cohort study.
METHODS: Community.
METHODS: A total of 3,150 US women who had been trying to conceive for <3 months and had purchased a Modern Fertility hormone test.
METHODS: We developed a discrete time-to-event model using a binomial complementary log-log error structure within a generalized additive modeling framework, adjusting for confounding factors such as age, body mass index, parity, smoking status, polycystic ovary syndrome, and others. Sensitivity analyses were performed in women with regular menstrual cycles (21-35 days), who did not report using fertility treatments, using alternate AMH level categories (<0.7, 0.7-8.5, >8.5 ng/mL), and AMH levels as a continuous measure.
METHODS: Primary outcomes included cumulative conception probability within 12 cycles and relative fecundability per menstrual cycle. Conception was defined by a self-reported positive pregnancy test.
RESULTS: Participants contributed 7.21 ± 5.32 cycles, with 1,325 (42.1%) achieving a pregnancy. Women with low AMH levels (<1 ng/mL, n = 427) had a lower chance of natural conception (adjusted hazard ratio [adjHR], 0.77; 95% confidence interval [CI], 0.64-0.94) compared with women with normal AMH levels (1-5.5 ng/mL). There was no difference between high (5.5+ ng/mL) and normal AMH level categories (adjHR, 1.11; 95% CI, 0.94-1.31). The inclusion of AMH improved the model (net reclassification index 0.10 [0.06-0.14]). The instantaneous probability of conception was highest in cycle four across all AMH categories: the probability of natural conception was 11.2% (95% CI, 9.0-14.0) for low AMH levels, 14.3% (95% CI, 12.3-16.5) for normal AMH levels, and 15.7% (95% CI, 12.9-19.0) for high AMH levels. In the regular cycles sensitivity analysis (n = 1,791), the low AMH group had a lower chance of conception (adjHR, 0.77; 95% CI, 0.61-0.97) in the low AMH group compared with normal AMH, and similarly in the continuous model (adjHR, 0.90; 95% CI, 0.85-0.95).
CONCLUSIONS: Low AMH levels (<1 ng/mL) are independently associated with a modest but significant reduction in the chance of conception.
METHODS: Prospective time-to-pregnancy cohort study.
METHODS: Community.
METHODS: A total of 3,150 US women who had been trying to conceive for <3 months and had purchased a Modern Fertility hormone test.
METHODS: We developed a discrete time-to-event model using a binomial complementary log-log error structure within a generalized additive modeling framework, adjusting for confounding factors such as age, body mass index, parity, smoking status, polycystic ovary syndrome, and others. Sensitivity analyses were performed in women with regular menstrual cycles (21-35 days), who did not report using fertility treatments, using alternate AMH level categories (<0.7, 0.7-8.5, >8.5 ng/mL), and AMH levels as a continuous measure.
METHODS: Primary outcomes included cumulative conception probability within 12 cycles and relative fecundability per menstrual cycle. Conception was defined by a self-reported positive pregnancy test.
RESULTS: Participants contributed 7.21 ± 5.32 cycles, with 1,325 (42.1%) achieving a pregnancy. Women with low AMH levels (<1 ng/mL, n = 427) had a lower chance of natural conception (adjusted hazard ratio [adjHR], 0.77; 95% confidence interval [CI], 0.64-0.94) compared with women with normal AMH levels (1-5.5 ng/mL). There was no difference between high (5.5+ ng/mL) and normal AMH level categories (adjHR, 1.11; 95% CI, 0.94-1.31). The inclusion of AMH improved the model (net reclassification index 0.10 [0.06-0.14]). The instantaneous probability of conception was highest in cycle four across all AMH categories: the probability of natural conception was 11.2% (95% CI, 9.0-14.0) for low AMH levels, 14.3% (95% CI, 12.3-16.5) for normal AMH levels, and 15.7% (95% CI, 12.9-19.0) for high AMH levels. In the regular cycles sensitivity analysis (n = 1,791), the low AMH group had a lower chance of conception (adjHR, 0.77; 95% CI, 0.61-0.97) in the low AMH group compared with normal AMH, and similarly in the continuous model (adjHR, 0.90; 95% CI, 0.85-0.95).
CONCLUSIONS: Low AMH levels (<1 ng/mL) are independently associated with a modest but significant reduction in the chance of conception.
摘要：
目的：研究AMH与妊娠时间的关系。虽然有人假设血清抗苗勒管激素（AMH）水平可能表明受孕的机会，调查结果好坏参半。鉴于任何协会预计都是适度的，以前的研究可能动力不足，我们在迄今为止最大的前瞻性队列中调查了这种关系.
方法：前瞻性妊娠时间队列研究。
方法：3,150名试图怀孕不到3个月并购买了现代生育激素测试的美国女性。
方法：我们在广义加性建模框架内利用二项式互补对数-对数误差结构开发了离散时间-事件模型，调整混杂因素，如年龄，BMI,奇偶校验，吸烟状况,PCOS,和其他人。敏感性分析在月经周期正常（21-35天）的女性中进行，没有报告使用生育治疗的人，使用替代AMH类别（<0.7，0.7-8.5，>8.5ng/mL），和AMH作为一个连续的措施。
方法：主要结局包括12个周期内的累积受孕概率和每个月经周期的相对受胎率。概念是由自我报告的阳性妊娠试验定义的。
结果：参与者贡献了7.21±5.32个周期，1,325（42.1％）实现怀孕。低AMH（<1ng/mL，n=427）与AMH正常（1-5.5ng/mL）的女性相比，自然受孕的机会较低（调整后的危险比（adjHR0.77，95CI0.64，0.94，p=0.009）。高（5.5ng/ml）和正常AMH类别之间没有差异（adjHR1.11，95％CI0.94，1.31，p=0.2）。AMH的纳入改善了模型(净重新分类指数0.10[0.06-0.14);P<0.001)。在所有AMH类别中，第4周期的瞬时受孕概率最高：低AMH的自然受孕概率为11.2％（95％CI9.0，14.0），正常AMH的14.3%（95%CI12.3，16.5），高AMH为15.7%（95CI12.9，19.0）。在常规周期敏感性分析(n=1,791)中，与正常AMH相比，低AMH组的受孕机会较低（adjHR0.7795％CI0.61，0.97，p=0.028），在连续模型中也是如此(adjHR0.90；95%CI0.85-0.95，p<0.0001)。
结论：低AMH水平（<1ng/ml）与受孕机会的适度但显着降低独立相关。
方法：前瞻性妊娠时间队列研究。
方法：3,150名试图怀孕不到3个月并购买了现代生育激素测试的美国女性。
方法：我们在广义加性建模框架内利用二项式互补对数-对数误差结构开发了离散时间-事件模型，调整混杂因素，如年龄，BMI,奇偶校验，吸烟状况,PCOS,和其他人。敏感性分析在月经周期正常（21-35天）的女性中进行，没有报告使用生育治疗的人，使用替代AMH类别（<0.7，0.7-8.5，>8.5ng/mL），和AMH作为一个连续的措施。
方法：主要结局包括12个周期内的累积受孕概率和每个月经周期的相对受胎率。概念是由自我报告的阳性妊娠试验定义的。
结果：参与者贡献了7.21±5.32个周期，1,325（42.1％）实现怀孕。低AMH（<1ng/mL，n=427）与AMH正常（1-5.5ng/mL）的女性相比，自然受孕的机会较低（调整后的危险比（adjHR0.77，95CI0.64，0.94，p=0.009）。高（5.5ng/ml）和正常AMH类别之间没有差异（adjHR1.11，95％CI0.94，1.31，p=0.2）。AMH的纳入改善了模型(净重新分类指数0.10[0.06-0.14);P<0.001)。在所有AMH类别中，第4周期的瞬时受孕概率最高：低AMH的自然受孕概率为11.2％（95％CI9.0，14.0），正常AMH的14.3%（95%CI12.3，16.5），高AMH为15.7%（95CI12.9，19.0）。在常规周期敏感性分析(n=1,791)中，与正常AMH相比，低AMH组的受孕机会较低（adjHR0.7795％CI0.61，0.97，p=0.028），在连续模型中也是如此(adjHR0.90；95%CI0.85-0.95，p<0.0001)。
结论：低AMH水平（<1ng/ml）与受孕机会的适度但显着降低独立相关。
