Baseline Antral Follicle Count Compared to Mature Oocytes Retrieved During Autologous In Vitro Fertilization Cycles

In vitro fertilization is often one of the final options patients with infertility will turn to during their attempts to create a family before choosing options such as adoption or donor oocytes. Given the vast amount of information on social media around infertility, patients are often left with unrealistic expectations after undergoing an autologous oocyte retrieval. Our aim was to determine the strength of the association between baseline antral follicle count (AFC) and mature oocytes obtained at the time of autologous in vitro fertilization with intracytoplasmic sperm injection (ICSI) to assist in the appropriate counseling of how many mature oocytes patients can anticipate after a single in vitro fertilization cycle.

This is a retrospective chart review of women between the ages 18-41 undergoing in vitro fertilization with ICSI at Brooke Army Medical Center between 2015 and 2022. Approval was obtained from the Institutional Review Board. A total of 362 individuals were included, and AFC at baseline and mature oocytes obtained at the time of oocyte retrieval were recorded. Data were presented as mean ± SD for numerical data or sum ± percentages for categorical data. A univariate linear regression and several multivariate regression models were performed, and a coefficient of determination squared (R2), along with its corresponding 95% CI, was calculated to determine how well the models predict the outcome of the dependent variable, mature oocytes obtained at the time of oocyte retrieval.

The univariate linear regression model using AFC at baseline to predict mature oocyte obtained at time of oocyte retrieval showed an R2 value of 26.7% (95% CI, 19.0-34.8) with P < .001. This implies that one mature oocyte was obtained for every four antral follicles noted at baseline, but this was noted to have a significant right-sided skew (Figure 1). Several multivariate regression models were performed, yet R2 was only altered marginally to 32.5% (95% CI, 24.5-40.6) over the simple univariate linear regression model, and the multivariate regression model R2 value was included in the 95% CI of the univariate linear regression model. Age alone was significantly less predictive with R2 of 8.5% (95% CI, 3.7-14.8).

This study demonstrated that a simplified equation or model with high accuracy to predict mature oocytes collected at the time of in vitro fertilization from a baseline AFC was unable to be obtained. The results of this retrospective study show that there is a wide prediction interval of the number of mature oocytes obtained at the time of oocyte retrieval with every numerical increase in AFC at the baseline ultrasound. Strengths of this study include the large database and broad inclusion of ages and diagnoses that enable application to a broader population. Potential limitations include the restriction to ICSI and limiting to a military facility. This further highlights the complexity of infertility and the need for appropriate counseling of patients concerning expectations and variance in response.