The present study represents the first study to establish nomograms that predict the number of 2PN zygotes for the personalization of the FSH starting dose in women with poor ovarian response undergoing IVF treatment with PPOS by using several combinations including BMI, AMH level, and AFC. The nomogram we developed may be clinically useful and seems a promising tool to individualize the treatment upon ovarian reserve in poor ovarian responders reducing the inter-operator variability derived from clinicians’ clinical experience during their daily clinical practice.
The main characteristic of the PPOS protocol is the use of progesterone for example medroxyprogesterone acetate (MPA) to block the rise of LH to produce more follicles during the follicular phase.[4, 7] The PPOS protocol breaks away from the convention of inhibiting LH surge in the traditional down-regulation protocol of relying on Gonadotropin releasing hormone (GnRH) analogues. Because of the asynchrony between the endometrium and embryo with PPOS stimulation, all the oocytes/embryos should be cryopreserved for later transfer[2]. There is also some evidence that PPOS protocol produced more mature oocytes and embryos for cryopreservation and achieved high-quality oocytes and satisfactory pregnancy outcomes.[7, 32] In this trial, we included patients with abnormal ovarian reserve test ( with AFC < 5 follicles or AMH < 1.1 ng/ml). The criteria for this trial were under the widely accepted Bologna criteria. These POR patients provide a good model for investigating better clinical outcome of PPOS. Different types of therapeutic approaches have been reported to improve the cycle outcomes in patients with poor ovarian response, but still there is no consensus on one stimulation protocol with compelling advantage over another. Several reports confirm that the cycle outcomes of patients with POR cannot improve by increasing gonadotropin doses, as the assumption that number of oocytes retrieved may be a function of daily gonadotropin dose is controversial. The possible etiologies of POR could be explained by limited ability to recruit a wave of follicles and different sensitivity of early antral follicles to FSH [23–26]. Studies have shown that poor responder patients were able to recruit an extra oocyte or two by increasing the exogenous FSH dose, with the possible improvement in IVF outcome[27]. However, mature oocyte may determine the outcomes of IVF, not the number of retrieved oocyte. A study demonstrated that a significantly higher rate of immature oocytes after COS in women over forty years-of-age as compared to younger women[2, 28]. Recently, a prospective controlled study had shown that PPOS protocols can overcome premature ovulation and not adversely affect the quality of oocytes for poor responders. The profound suppression of endogenous FSH and LH at the stage of follicular recruitment could be avoided during PPOS, then a better egg retrieval could be obtained. [7, 9, 11] Meanwhile, PPOS can induce an increase in basal plasma FSH by using hMG and FSH surge occurred on the trigger day. The FSH surge before ovulation can promote the development of the follicles and effectively improve the developmental potential of follicles [1]. The clinical study by Huang showed that the MII oocyte, fertilization, and high-quality embryo rates in the PPOS group were significantly higher than those in the antagonist group (p < 0.05)[11]. Therefore, the PPOS protocol may be a better ovarian stimulation regime for poor responder patients. The main problem of POR was not only the depletion of the oocytes but the decrease in oocyte quality. The controlled ovarian hyperstimulation generally results in unsatisfactory oocyte yield for POR during IVF-ET cycles. Moreover, the different sensitivity of early antral follicles to FSH led follicles to develop non-synchronously in different timing of COS with PPOS. It is critical to obtain more high-quality oocytes for later embryo transfer in POR patients[29]. Even if POR patients go forward with very few follicles, the quality of oocyte retrieved is high enough, so that fertilization and embryo transfer will be successful. On the contrary, amounts of retrieved oocytes with lots of which are discarded because of their poor-quality will lead to a bad result[30]. A major problem of the poor responder patients not only is the fewer oocytes recovered number but the oocyte quality is diminished, for these patients produce unsatisfactory oocyte during IVF cycles[29]. Previous studies confirmed that the fertilization rate (2PN) was not affected by the increasing age of the women in IVF or ICSI. This means that 2PN has less variation in IVF procedures[31, 32]. Therefore, we selected 2PN zygotes as the criterion for successful oocytes retrieval in the present study.
The majority of reproductive specialists agree that nowadays AMH and AFC are considered as two reliable markers for predicting ovarian response[13]. Indeed AMH and AFC have a strong correlation and are actually measure the same thing: ovarian reserve[14]. AMH is produced exclusively by granulosa cells of the developing preantral and antral follicles and reflects the overall amount of the granulosa cells and is thought to be a direct surrogate of the number of growing follicles in follicular pool. Serum AMH level does not exhibit significant intra- and inter-cycle variability and has the advantage of its operator-independence. Serum AMH level declines by 5.6% per year and is affected by factors such as follicle size, granulosa cell volume and genetic characteristics. AMH has low sensitivity and specificity for the prediction of successful achievement of pregnancy in IVF[33, 34]. AFC much more immediately represents the number of growing primordial follicles remaining in the ovary and reflects the ovarian follicular patrimony. AFC has a great advantage that it could be detected at the same moment in which clinicians examine the patient and an inherent shortcoming of high inter-observed variability. Although AFC has a high specificity for predicting a poor response, its sensitivity is low[12, 35]. BMI was another significant parameter predicting the FSH starting dose. It is known today that weight and body mass are a highrisk factor for menstrual dysfunction and anovulation. Obese women have a lower chance of conception following assisted reproductive techniques (ART). Obesity has a negative impact on the woman´s reproductive system through various pathways, including impaired ovarian, follicular development, quantitative and qualitative development of the oocyte, fertilization, and embryo development and implantation. Increased BMI significantly reduces the chance of clinical pregnancy in IVF[36]. Obese women require higher amounts of gonadotropins respond poorly to ovarian stimulation and more days to achieve follicular maturation. It is well documented that increased BMI is associated with low oocyte retrieval and poor embryo quality[37, 38]. The influence of age, AMH level, AFC level and BMI on the FSH starting dose was examined through multiple regression analysis in present study. our study demonstrated no relationship between age and FSH starting dose of POR patients. AMH level, AFC and BMI were important parameters to predict ovarian response to exogenous FSH.
Clinicians could assess markers of ovarian reserve and choose the best therapy in order to improve IVF outcomes. The ovarian reserve markers recently used include age, BMI, FSH, AMH and AFC et al[22]. One single ovarian reserve marker may be insufficient to predict ovarian response to exogenous FSH, a combination of multiple markers may improve the accuracy in predicting ovarian response to gonadotropins to optimize the FSH starting dose in IVF/ICSI cycles at individual level[15, 22]. Even though it is well-recognized that the female reproductive capacity is decreasing with the age increasing and the prevalence of POR increases with age, the prevalence of dominant follicles does not differ with age and women's age has no impact on the summed score for cleavage stage embryo quality.[9, 13, 31] The previous findings shown that the fertilization rate (2PN) was not affected by women's age in IVF or ICSI[31]. The most frequent cause of poor ovarian response may be diminished ovarian reserve in both older women and younger women [10]. Prediction of starting dose had no concern with age in patients with poor ovarian response in the present study. Our data analysis showed that the main FSH starting dose prescribed on clinical experience was 225 IU (41.22%), while the nomogram-based prescription of FSH starting dose was mainly either 187.5 IU or 206.25 IU (total percentage of the two was 88.52%). This suggests that when faced with patients with poor ovarian response, our doctors often worried that the initial dose will not be sufficient, and gradually reduce it at a later stage. But according to the nomogram, the starting dose can be reduced to decrease the financial burden and risk of side effects in patients. As the application of the nomogram could lead to more accurate starting dose.
In conclusion, the individual FSH starting threshold dose for ovulation induction in patients with poor ovarian response can be predicted based on easily available three variables detected in clinical practice: AMH, AFC and BMI. An FSH dosage nomogram to predict the FSH starting dose was constructed based on these predictive factors.