Individualised gonadotropin dose selection using markers of ovarian reserve for women undergoing in vitro fertilisation plus intracytoplasmic sperm injection (IVF/ICSI)

Background During a stimulated cycle of in vitro fertilisation or intracytoplasmic sperm injection (IVF/ICSI), women receive daily doses of gonadotropin follicle‐stimulating hormone (FSH) to induce multifollicular development in the ovaries. A normal response to stimulation (e.g. retrieval of 5 to 15 oocytes) is considered desirable. Generally, the number of eggs retrieved is associated with the dose of FSH. Both hyper‐response and poor response are associated with an increased chance of cycle cancellation. In hyper‐response, this is due to increased risk of ovarian hyperstimulation syndrome (OHSS), while poor response cycles are cancelled because the quantity and quality of oocytes is expected to be low. Clinicians often individualise the FSH dose using patient characteristics predictive of ovarian response. Traditionally, this meant women's age, but increasingly, clinicians use various ovarian reserve tests (ORTs). These include basal FSH (bFSH), antral follicle count (AFC), and anti‐Müllerian hormone (AMH). It is unclear whether individualising FSH dose improves clinical outcomes. This review updates the 2018 version. Objectives To assess the effects of individualised gonadotropin dose selection using markers of ovarian reserve in women undergoing IVF/ICSI. Search methods We searched the Cochrane Gynaecology and Fertility Group Specialised Register of controlled trials, CENTRAL, MEDLINE, Embase, and two trial registers in February 2023. Selection criteria We included randomised controlled trials (RCTs) that compared (a) different doses of FSH in women with a defined ORT profile (i.e. predicted low, normal, or high responders based on AMH, AFC, and/or bFSH) or (b) an individualised dosing strategy (based on at least one ORT measure) versus uniform dosing or a different individualised dosing algorithm. Data collection and analysis We used standard Cochrane methodological procedures. Primary outcomes were live birth/ongoing pregnancy and severe OHSS. Main results We included 26 studies, involving 8520 women (6 new studies added to 20 studies included in the previous version). We treated RCTs with multiple comparisons as separate trials for the purpose of this review. Meta‐analysis was limited due to clinical heterogeneity. Evidence certainty ranged from very low to low, with the main limitations being imprecision and risk of bias associated with lack of blinding. Direct dose comparisons according to predicted response in women Due to differences in dose comparisons, caution is required when interpreting the RCTs in predicted low responders. All evidence was low or very low certainty. Effect estimates were very imprecise, and increased FSH dosing may or may not have an impact on rates of live birth/ongoing pregnancy, OHSS, and clinical pregnancy. Similarly, in predicted normal responders (10 studies, 4 comparisons), higher doses may or may not impact the probability of live birth/ongoing pregnancy (e.g. 200 versus 100 international units (IU): odds ratio (OR) 0.88, 95% confidence interval (CI) 0.57 to 1.36; I2 = 0%; 2 studies, 522 women) or clinical pregnancy. Results were imprecise, and a small benefit or harm remains possible. There were too few events for the OHSS outcome to enable inferences. In predicted high responders, lower doses may or may not affect live birth/ongoing pregnancy (OR 0.98, 95% CI 0.66 to 1.46; 1 study, 521 women), severe OHSS, and clinical pregnancy. It is also unclear whether lower doses reduce moderate or severe OHSS (Peto OR 2.31, 95% CI 0.80 to 6.67; 1 study, 521 participants). ORT‐algorithm studies Eight trials compared an ORT‐based algorithm to a non‐ORT control group. It is unclear whether live birth/ongoing pregnancy and clinical pregnancy are increased using an ORT‐based algorithm (live birth/ongoing pregnancy: OR 1.12, 95% CI 0.98 to 1.29; I2 = 30%; 7 studies, 4400 women; clinical pregnancy: OR 1.04, 95% CI 0.91 to 1.18; I2 = 18%; 7 studies, 4400 women; low‐certainty evidence). However, ORT algorithms may reduce moderate or severe OHSS (Peto OR 0.60, 95% CI 0.42 to 0.84; I2 = 0%; 7 studies, 4400 women; low‐certainty evidence). There was insufficient evidence to determine whether the groups differed in rates of severe OHSS (Peto OR 0.74, 95% CI 0.42 to 1.28; I2 = 0%; 5 studies, 2724 women; low‐certainty evidence). Our findings suggest that if the chance of live birth with a standard starting dose is 25%, the chance with ORT‐based dosing would be between 25% and 31%. If the chance of moderate or severe OHSS with a standard starting dose is 5%, the chance with ORT‐based dosing would be between 2% and 5%. These results should be treated cautiously due to heterogeneity in the algorithms: some algorithms appear to be more effective than others. Authors' conclusions We did not find that tailoring the FSH dose in any particular ORT population (low, normal, high ORT) affected live birth/ongoing pregnancy rates, but we could not rule out differences, due to sample size limitations. Low‐certainty evidence suggests that it is unclear if ORT‐based individualisation leads to an increase in live birth/ongoing pregnancy rates compared to a policy of giving all women 150 IU. The confidence interval is consistent with an increase of up to around six percentage points with ORT‐based dosing (e.g. from 25% to 31%) or a very small decrease (< 1%). A difference of this magnitude could be important to many women. It is unclear if this is driven by improved outcomes in a particular subgroup. Further, ORT algorithms reduced the incidence of OHSS compared to standard dosing of 150 IU. However, the size of the effect is also unclear. The included studies were heterogeneous in design, which limited the interpretation of pooled estimates. It is likely that different ORT algorithms differ in their effectiveness. Current evidence does not provide a clear justification for adjusting the dose of 150 IU in poor or normal responders, especially as increased dose is associated with greater total FSH dose and cost. It is unclear whether a decreased dose in predicted high responders reduces OHSS, although this would appear to be the most likely explanation for the results.