卵巢高反應者試管嬰兒療程之安全及病人友善方案：單獨使用促性腺激素釋放激素破卵之評估、根據血中黃體素數值判斷自然週期解凍植入時間

Abstract

研究背景： 試管嬰兒治療主要分為兩個部分：排卵刺激療程與解凍胚胎植入療程。對於卵巢高反應者來說，僅使用性腺激素釋放素促進劑（GnRH agonist）進行破卵後，進行全胚胎冷凍週期，可以將卵巢過度刺激症候群（OHSS）的風險降至最低，但也有發生不良破卵反應（suboptimal trigger）的風險。文獻指出療程第一天血中LH數值（Basal LH）可做為是判斷是否發生suboptimal trigger的指標。 全胚冷凍後，自然週期解凍植入可降低早期妊娠的出血、妊娠中子癲前症發生率。也因為卵巢有排卵，藥物使用遠低於賀爾蒙解凍週期。目前判斷排卵日的主流方法為監測血中LH陡增與超音波監測主要濾泡消失。需要短期密切的回診執行抽血與超音波來判斷排卵日以便排定植入日，療程取消率相較賀爾蒙週期、擬自然週期高。

目標與理由： 本論文主要探討兩個主題： 1. 僅使用GnRH agonist進行破卵，除了透過療程第一天血中LH數值（Basal LH），是否也應參考整個療程中最高的LH數值（LH max）來綜合預測不良破卵反應？ 2. 根據血中黃體素數值來決定自然週期解凍植入的時間，並評估此方法的臨床結果、回診次數與療程取消率。

研究方法： 第一個主題是一項回溯性、單一中心的研究。我的假設是拮抗劑療程（Antagonist protocol）或黃體素輔助排卵刺激療程中（PPOS）中，僅使用GnRH agonist進行破卵，除了Basal LH以外，LH max也具有預測不良破卵反應的能力。我的納入標準如下：antagonist protocol或PPOS中，僅使用GnRH agonist進行破卵、採集的卵子細胞數 >10個、採用全部胚胎冷凍。排除標準如下：低性促素的性腺功能低下症（hypogonadotropic hypogonadism）、在療程前長期（超過三個月）使用口服避孕藥的患者。個案分為三組：第一組為整個療程中LH值均小於等於2.2IU/L（Basal LH與LH max均小於等於2.2IU/L）。第二組BASAL LH小於2.2IU/L但LH max大於2.2IU/L。第三組為整個療程中LH值均大於2.2IU/L（Basal LH與LH max均大於2.2IU/L）。 第二個主題也是回溯性、單一中心的研究我的假設是根據血中黃體素數值來決定自然週期解凍植入的時間能提高胚胎著床率。我的納入標準如下：年齡<40歲、身體質量指數（Body Mass Index，BMI）小於等於30 kg/m2、月經規則、正常子宮腔、月經週期介於24到38日、先前有第五天或第六天的囊胚冷凍、一次植入一個或兩個胚胎其中至少一個為中等等級以上的胚胎。排除標準如下：卵子捐贈、先天性子宮畸形、子宮內膜病變、輸卵管水腫、植入日內膜厚度小於7mm、中度或重度的子宮內膜異位症或子宮肌腺症。個案分為兩組：研究組依據血中黃體素數值來決定自然週期解凍植入的時間，對照組則是透過監測血中LH陡增或超音波監測主要濾泡消失來決定解凍植入時間。

研究結果： 第一個主題經由multivariable regression校正可能的confounder後，第一組（整個療程中LH值均小於等於2.2IU/L）仍然是不良破卵反應的獨立預測因子（OR: 6.99; 95% CI 1.035 – 47.274）。第一組（b = −12.72; 95% CI −20.9 to −4.55）和BMI（b = −0.25; 95% CI −0.5 to −0.004）與取卵率呈負相關。第二組和第三組臨床結果無顯著差異。 第二個主題主要結果包括著床率、臨床懷孕率和繼續懷孕率，研究組和對照組之間無顯著差異。次要結果研究組的回診次數2.69±0.74次為明顯低於對照組的3.45±1.31次，p<0.05。研究組的療程取消率5%也明顯低於對照組的22%，p<0.05。

結論： 根據我們的研究結果，發現基礎LH較低但療程中LH數值有超過2.2IU/L的患者可以僅使用GnRH agonist破卵，其臨床結果與整個療程LH數值皆有超過2.2IU/L的患者相似。另外，全胚冷凍後，採用根據黃體素數植來決定植入時間的自然週期可降低病患回診次數與療程取消率，同時臨床結果與傳統方法無明顯差異。

Background/What is known already There are two main parts of an IVF cycle: controlled ovarian stimulation and embryo transfer. For high responders, the safest approach to prevent OHSS is to use a GnRH agonist trigger followed by a freeze-all strategy. Current evidence suggests that the basal LH level may predict the risk of a suboptimal trigger. However, we are curious whether the LH level during controlled ovarian stimulation could also predict a suboptimal trigger. After embryo freezing, a true natural cycle frozen-thawed embryo transfer may lead to favorable obstetric and perinatal outcomes. Determining the timing of ovulation is a critical factor in deciding the embryo transfer schedule. The most commonly used methods to identify ovulation are detecting the serum LH surge or observing the collapse of the dominant follicle. However, these approaches require frequent follow-up visits and are associated with a high cancellation rate.

Objective and rationale The present two studies aimed to explore two main objectives. First, we examined the value of LH max in predicting suboptimal triggers, particularly in individuals with a low basal LH level but a high LH max level. Second, we compared clinical outcomes between a progesterone monitoring protocol and protocols based on the LH surge or the collapse of the dominant follicle in true natural cycle frozen-thawed embryo transfer.

Study Design The first study is a retrospective study. We hypothesize that LH max have a better prediction of GnRH agonist suboptimal trigger. Our inclusion criteria are PPOS or GnRH antagonist protocol triggered only with GnRH agonist. Patients with hypogonadotropic hypogonadism or with prior long-term oral contraceptive pill (OCP) usage before starting the COS cycle also excluded. Subjects were divided into three groups, depending on basal LH level and LH max level. Group 1 included a cycle where basal LH level and LH max were both ≤2.2 IU/L. Group 2 included a cycle where basal LH level ≤2.2 IU/L and LH max level >2.2 IU/L. Group 3 included a cycle where basal LH level and LH max were both >2.2 IU/L. The second study is also a retrospective study. We hypothesize that basing the timing of embryo transfer on progesterone levels in true natural cycle frozen-thawed embryo transfer could lead to a higher clinical pregnancy rate. The inclusion criteria were as follows: women under 40 years of age, with a body mass index (BMI) ≤ 30 kg/m²; a previous cycle involving blastocyst cryopreservation; a normal intrauterine cavity; and regular menstrual cycles with cycle lengths between 24 and 38 days. All patients underwent one or two autologous frozen-thawed blastocyst transfers, with at least one blastocyst of intermediate or better quality. The exclusion criteria were as follows: use of donor eggs or sperm, presence of hydrosalpinx, endometrial thickness < 7 mm on the day of embryo transfer, or moderate to severe endometriosis or adenomyosis. Cycles were divided into two groups based on the method used to determine the timing of embryo transfer (ET): Group 1: Timing based on serum progesterone levels. Group 2: Timing based on serum LH surge or dominant follicle collapse.

Results After adjusting the possible confounders, multivariable regression analysis showed that cycles with consistently low LH levels remains an independent predictor of suboptimal response. Cycles with consistently low LH levels and BMI were negatively associated with oocyte yield rate. Patients with low basal LH but high LH max levels had similar clinical outcomes compared to those with high LH max levels through COS. In true natural cycle frozen-thawed embryo transfers, pregnancy outcomes were similar in both groups. However, the group whose embryo transfer timing was based on serum progesterone levels had significantly fewer monitoring visits and cycle cancellations. Conclusion The maximum LH level may serve as an indicator of LH reserve and could be a more reliable predictor of achieving an optimal oocyte yield compared to relying solely on basal LH levels. This approach aims to minimize the risk of obtaining a suboptimal oocyte yield and to improve overall treatment outcomes. Additionally, by adopting a monitored progesterone protocol in tNC-FET, pregnancy outcomes are comparable to those of other protocols, while significantly reducing both the number of visits and the cancellation rate, making it more patient-friendly