Exploring determinants of cumulative live birth rates in IVF: insights from the EELI study in Lebanon.

A total of 1,243 patients undergoing ART were included in this analysis (Table  1 ). The mean age of the participants was 33.39 ± 5.45, with most women falling in the 30–34 age group (35.96%). The mean age of their spouses was 38.44 ± 6.86. Geographically, a large proportion resided in Mount Lebanon (43.12%), followed by North Lebanon (11.26%) and Keserwan-Jbeil (9.98%). Only 5.07% of Lebanese patients lived abroad. The vast majority of patients were treated for primary infertility (93.24%), with the most common infertility indication being the female factor (58.97%), followed by male factor (21.8%) and mixed causes (17.94%). Table 1 Characteristics of the study population ( N  = 1,243) Socio-demographic characteristics Patient Age (years) Mean (SD) 33.39 (5.45) Median [IQR] 33 [30–37] Min-Max 16–48  < 30 284 (22.85%)  30–34 447 (35.96%)  35–39 330 (26.55%)  40–44 170 (13.68%)  ≥ 45 12 (0.97%) Patient’s spouse Age (years) Mean (SD) 38.44 (6.86) Median [IQR] 38 [34–42] Min-Max 24–69 Governorate  Akkar 54 (4.34%)  Baalbek-Hermel 19 (1.53%)  Beirut 105 (8.45%)  Beqaa 75 (6.03%)  Keserwan-Jbeil 124 (9.98%)  Mount Lebanon 536 (43.12%)  Nabatiye 21 (1.69%)  North Lebanon 140 (11.26%)  South Lebanon 106 (8.53%)  Non-resident 63 (5.07%) Infertility Type  Primary 1,159 (93.24%)  Secondary 84 (6.76%) Infertility Indication  Female 733 (58.97%)  Male 271 (21.8%)  Mixed 223 (17.94%)  Unexplained 16 (1.29%) Clinical and Cycle Characteristics Transfer strategy  Freeze all 562 (45.21%)  First Fresh cycle 681 (54.79%) Season of collection  Autumn 321 (25.82%)  Winter 273 (21.96%)  Spring 313 (25.18%)  Summer 336 (27.03%) Number of collected oocytes Mean (SD) 13.06 (9.47) Median [IQR] 11 [6–17] Min-Max 1–58 Number of mature oocytes Mean (SD) 9.33 (6.85) Median [IQR] 7 [4–12] Min-Max 1–38 Number of fertilized oocytes Mean (SD) 8.12 (5.96) Median [IQR] 7 [4–11] Min-Max 1–35 Number of embryos cleaved Mean (SD) 7.95 (5.92) Median [IQR] 6 [4–11] Min-Max 1–32 Number of blastocysts Mean (SD) 4.35 (4.19) Median [IQR] 3 [1–6} Min-Max 0–22 Number of good-grade blastocysts Mean (SD) 3.58 (2.54) Median [IQR] 3 [2–5} Min-Max 1–13 Number of Frozen Embryo Transfers  0 497 (39.98%)  1 530 (42.64%)  2 160 (12.87%)  3 44 (3.54%)  4 12 (0.97%) Day of transfer a  Cleavage Stage (D3) 235 (18.91%)  Blastocyst Stage (D5) 1,008 (81.09%) Pregnancy Outcomes Beta HCG  Negative 592 (47.6%)  Positive 651 (52.4%) Clinical Pregnancy  No 657 (52.9%)  Yes 586 (47.1%) Live Births  No 667 (53.7%)  Yes 576 (46.3%) a For women who had multiple transfers at multiple stages, this variable captures the stage of embryo development (Day 3 vs. Day 5) at the first available embryo transfer for each woman (either fresh or frozen). This approach was used to avoid outcome-related selection bias and to reflect the initial clinical decision regarding embryo culture duration Characteristics of the study population ( N  = 1,243) a For women who had multiple transfers at multiple stages, this variable captures the stage of embryo development (Day 3 vs. Day 5) at the first available embryo transfer for each woman (either fresh or frozen). This approach was used to avoid outcome-related selection bias and to reflect the initial clinical decision regarding embryo culture duration Regarding treatment protocols, 45.21% underwent a freeze-all strategy, while 54.79% did fresh transfers following ICSI. The most frequent season of oocyte collection was summer (27.03%), followed by autumn (25.82%), spring (25.18%), and winter (21.96%). Regarding ovarian stimulation outcomes, the mean number of collected oocytes was 13.06 (SD: 9.47), with a median of 11 [IQR: 6–17] and a range of 1 to 58. On average, 9.33 (SD: 6.85) oocytes were mature, and 8.12 (SD: 5.96) were fertilized. The number of embryos that cleaved between day 2 and day 3 (J2–J3) had a mean of 7.95 (SD: 5.92), and the mean number of blastocysts formed was 4.35 (SD: 4.19). On average, 3.58 (SD: 2.54) of these blastocysts were graded as good quality. While 39.98% of patients had no frozen embryo transfer, 42.64% had one, 12.87% had two, 3.54% had three, and 0.97% had four frozen embryo transfers. In terms of pregnancy outcomes, 52.4% of patients had a positive beta HCG test, 47.1% achieved a clinical pregnancy, and 46.3% ultimately had a live birth. Figure  2 shows the cumulative probability of achieving a live birth across up to four embryo transfers, stratified by treatment strategy (First Fresh cycles vs. freeze-all). For First Fresh cycles, the initial live birth rate after the first transfer (FT) was 32%, which increased to 39.6% after the second transfer and reached 41.7% by the fourth transfer. The gap between conservative and optimistic CLBR estimates was minimal, indicating low dropout rates and consistent follow-up. For the freeze-all strategy, the success rates were slightly higher overall. After the first transfer, the live birth rate was 38.8%, rising to 48.9% after the second transfer, and reaching 51.4% by the fourth transfer. Notably, freeze-all strategies demonstrated a greater incremental increase in success across subsequent transfers compared to First Fresh cycles. Fig.  2 visually highlights that freeze-all strategies maintain a steady advantage over First Fresh cycles across all transfer stages, with the cumulative live birth probability surpassing 50% by the fourth transfer [See Additional file 1]. Fig. 2 Cumulative live birth rates across up to four embryo transfers, stratified by transfer strategy Cumulative live birth rates across up to four embryo transfers, stratified by transfer strategy The results of the bivariate analysis are addressed in Table  2 . Maternal age was inversely associated with live birth. Compared to women under 30, the odds of live birth significantly decline with increasing age, especially after age 35. A similar inverse trend was observed for paternal age, with significantly lower odds of live birth among men aged 45 and above, compared to men under 30. Geographically, compared to women residing in Akkar, Keserwan-Jbeil, Mount Lebanon, and non-resident patients have significantly lower odds of live birth rates. No significant seasonal effect on live birth in this unadjusted analysis. Similarly, no evidence of a difference in live birth rates between primary and secondary infertility types in the crude analysis. Male factor infertility was associated with higher odds of live birth compared to female factor; however, other types of infertility were not. Table 2 Unadjusted associations between selected characteristics and live birth Variable OR (95% CI) p -value Age Group  < 30 Reference -  30–34 0.94 (0.69–1.26) 0.667  35–39 0.56 (0.40–0.77) < 0.001  40–44 0.21 (0.14–0.33) < 0.001  ≥ 45 0.07 (0.01–0.56) 0.012 Spouse Age Group  < 30 Reference -  30–34 1.19 (0.73–1.93) 0.493  35–39 1.06 (0.66–1.70) 0.816  40–44 0.76 (0.47–1.25) 0.281  ≥ 45 0.46 (0.28–0.78) 0.004 Governorate  Akkar Reference  Baalbek-Hermel 0.50 (0.17–1.44) 0.2  Beirut 0.70 (0.36–1.36) 0.294  Beqaa 0.57 (0.28–1.16) 0.12  Keserwan-Jbeil 0.42 (0.22–0.81) 0.009  Mount Lebanon 0.54 (0.31–0.96) 0.037  Nabatiye 0.92 (0.33–2.54) 0.867  North Lebanon 0.86 (0.46–1.64) 0.655  South Lebanon 0.61 (0.32–1.19) 0.149  Non resident 0.45 (0.22–0.95) 0.036 Season  Winter Reference -  Spring 1.09 (0.79–1.5) 0.599  Summer 0.77 (0.56–1.06) 0.119  Autumn 0.93 (0.67–1.28) 0.647 Infertility Type  Primary Reference -  Secondary 0.77 (0.49–1.21) 0.265 Infertility Indication  Female factor Reference -  Male factor 1.37 (1.03–1.81) 0.028*  Unexplained 1.24 (0.46–3.34) 0.669  Mixed/Other 0.99 (0.74–1.34) 0.955 Transfer strategy  First Fresh cycle Reference -  Freeze All 1.51 (1.21–1.89) < 0.001 Oocytes collected  < 10 Reference -  10–15 2.14 (1.62–2.84)  15 4.54 (3.42–6.03) < 0.001 Mature Oocytes  < 10 Reference  10–15 2.93 (2.21–3.89)  15 5.31 (3.75–7.52) < 0.001 Fertilized oocytes  < 10 Reference -  10–15 3.56 (2.65, 4.79)  15 6.23 (4.12, 9.41) < 0.001 Number of embryos cleaved  < 10 Reference -  10–15 2.28 (1.74, 3.00)  15 5.4 (3.99, 7.3) < 0.001 Number of blastocysts  < 10 Reference -  10–15 6.55 (3.89, 11.02)  15 10.74 (29.95) < 0.001 Number of good-grade blastocysts  1–2 Reference -  3–4 2.71(2.05, 3.59) < 0.001  5–6 6.73 (4.59, 9.88) < 0.001  7 and more 9.65 (6.22, 14.98) < 0.001 Day of transfer  Day 3 Reference -  Day 5 5.54 (3.85–7.95) < 0.001 Unadjusted associations between selected characteristics and live birth Stimulation response indicators were strongly associated with outcomes. Women with 10–15 oocytes retrieved had twice the odds of live birth, and those with more than 15 had more than four times the odds. Similar trends were observed for the number of mature oocytes, fertilized oocytes, and cleaved embryos. Women with 10–15 blastocysts had over 6 times higher odds of livebirth, and those with more than 15 had more than tenfold odds. Having 5 to 6 good-grade blastocysts and 7 or more was strongly predictive of live birth. Finally, transfer timing was a significant factor: Day 5 transfers were associated with markedly increased odds of live birth compared to Day 3. A multivariate logistic regression was conducted to examine predictors of live birth among women undergoing ART (Table  3 ). The model included categorical age, infertility indication, infertility type, number of oocytes, transfer strategy, and day of embryo transfer. Compared to women under 30, those aged 40–44 had significantly lower odds of live birth (OR = 0.36, 95% CI: 0.23–0.59, p  < 0.001) and women aged 45 and above also showed a strong reduction in odds bordering on significance (0.16, 95% CI: 0.02–1.32, p  = 0.089). Higher oocyte yield was associated with increased odds of live birth. Women in the highest oocyte category had almost 2.8 times the odds of live birth compared to the lowest category (less than 10 oocytes) (2.77, 95% CI: 1.99–3.86, p  < 0.001). Embryo transfer on Day 5 was strongly associated with higher odds of live birth (OR = 3.89, 95% CI: 2.63–5.77, p  < 0.001), aligning with clinical evidence favoring blastocyst transfer. Infertility indication and infertility type did not show a statistically significant association with live birth in the adjusted model. After adjusting for important confounders (maternal age, infertility type, infertility indication, and number of oocytes), the choice of ART protocol did not have a meaningful and statistically significant association with live birth outcomes. > Table 3 Multivariate logistic regression analysis of predictors of live birth Variable a aOR (95% CI) Std. Err. z p -value Age (30–34) 1.12 (0.81, 1.55) 0.184711 0.71 0.476 Age (35–39) 0.79 (0.55, 1.13) 0.14571 -1.25 0.212 Age (40–44) 0.36 (0.23, 0.59) 0.089085 -4.13 < 0.001* Age (≥ 45) 0.16 (0.02, 1.32) 0.172485 -1.7 0.089 Indication: Male factor 0.98 (0.72, 1.34) 0.156408 -0.11 0.916 Indication: Unexplained 0.77 (0.26, 2.23) 0.418088 -0.48 0.628 Indication: Combined 0.85 (0.62, 1.18) 0.142856 -0.94 0.347 Infertility type (secondary vs. primary) 1.21 (0.73, 2.02) 0.316646 0.73 0.464 Oocyte count: 10–15 1.44 (1.06, 1.96) 0.225826 2.31 0.021* Oocyte count: >15 2.77 (1.99, 3.86) 0.468842 6.02 < 0.001* Transfer strategy: Freeze all (vs. first fresh cycle) 0.85 (0.65, 1.11) 0.115909 -1.18 0.238 Day of transfer: Day 5 (vs. Day 3) 3.89 (2.62, 5.77) 0.781749 6.76 < 0.001* Constant 0.19 (0.09, 0.39) 0.069635 -4.56 < 0.001 a Reference categories: age < 30 years, female factor infertility, primary infertility, < 10 oocytes, fresh ICSI method, and day 3 embryo transfer *Statistically significant at p  < 0.05 Multivariate logistic regression analysis of predictors of live birth a Reference categories: age < 30 years, female factor infertility, primary infertility, < 10 oocytes, fresh ICSI method, and day 3 embryo transfer *Statistically significant at p  < 0.05 The logistic regression model demonstrated moderate discriminatory ability, with a pseudo-R² of 0.12 and a log-likelihood of − 7 54.97. Model parsimony was supported by acceptable information criteria (AIC = 1535.94, BIC = 1602.57). The classification table indicated an overall accuracy of 67.1%, with the following outcomes: sensitivity of 69.3% (correctly predicting live births for approximately 7 out of 10 women) and specificity of 65.2% (correctly identifying non-live births for about 7 out of 10 women). The model’s positive predictive value was 63.2%, while its negative predictive value was 71.1%. Figure  3 shows the CLBR across up to four embryo transfers, stratified by oocyte yield categories:  15 oocytes. Within each category, outcomes are compared between First Fresh and freeze-all strategies, with both observed and optimistic estimates plotted [See Additional file 2]. Among patients with fewer than 10 oocytes, CLBRs were low overall, with both treatment strategies plateauing near 30% after the second transfer and showing minimal gains beyond that point. For those with 10–15 oocytes, outcomes improved markedly, with freeze-all slightly outperforming First Fresh cycles by the third transfer and CLBRs reaching approximately 50%. In the > 15 oocyte group, both treatment strategies achieved the highest success, with freeze-all strategies reaching more than 60% CLBR by the fourth transfer. Fig. 3 Cumulative live birth rates across up to four embryo transfers, stratified by oocyte yield and transfer strategy (first fresh cycle vs. freeze-all) Cumulative live birth rates across up to four embryo transfers, stratified by oocyte yield and transfer strategy (first fresh cycle vs. freeze-all) Higher oocyte yield was associated with greater cumulative success across transfers. Patients with more than 15 oocytes achieved the highest outcomes, with CLBR exceeding 65% under freeze-all strategies. The gap between optimistic and observed CLBR estimates remained small, which may be linked to minimal dropout and good continuity of care. When stratified by oocyte yield, the relative advantage of treatment methods varied. Among patients with less than 10 oocytes, fresh transfer slightly outperformed freeze-all, while in the 10–15 oocyte group, freeze-all was more advantageous. Both strategies performed well in the group with more than 15 oocytes, with CLBR exceeding 50%. However, when controlling for oocyte yield, no single treatment method showed a consistent superior effect, which aligns with the multivariate analysis findings that transfer strategy was not a statistically significant predictor of live birth. Although the group with more than 15 oocytes had a higher number of observed live births, the log-rank test did not detect significant differences between oocyte yield categories (χ²(2) = 3.22, p  = 0.20). A direct comparison of fewer than 10 versus more than 15 oocytes showed a borderline trend toward improved CLBR and shorter time to live birth in the higher-yield group (χ²(1) = 3.47, p  = 0.062). Across all oocyte categories, first Fresh cycle and freeze-all strategies produced comparable cumulative live birth outcomes over successive embryo transfers. As shown in Fig.  4 , CLBR declined with increasing maternal age. Women younger than 35 achieved the highest cumulative success, with freeze-all strategies surpassing 60% CLBR [ See Additional file 3 ]. In contrast, women aged 40–44 had lower success rates, with few additional births observed beyond the first transfer cycle. The difference between conservative and optimistic CLBRs remained small across all age groups, indicating low dropout rates and high treatment completion. The log-rank test confirmed a statistically significant difference in CLBR between age groups (χ²(4) = 66.02, p  < 0.001), with younger women (< 30 and 30–34) accumulating live births at a significantly faster rate than older groups. Additional log-rank tests comparing First Fresh cycle and freeze-all within each age group revealed no significant differences for women aged < 30, 30–34, 35–39, or 45+, but a significant difference emerged in the 40–44 age group (χ²(1) = 6.77, p  = 0.0093), where freeze all outperformed First Fresh cycle. Fig. 4 Cumulative live birth rates by number of embryo transfers, stratified by maternal age and transfer strategy (fresh vs. freeze-all) Cumulative live birth rates by number of embryo transfers, stratified by maternal age and transfer strategy (fresh vs. freeze-all)

Despite years of compounded crises, ranging from political instability and economic collapse to the Beirut port explosion and the COVID-19 pandemic, Lebanon continues to see encouraging birth rates through assisted reproductive technologies. Our study, the first of its kind in Lebanon to comprehensively analyze CLBR across both fresh and freeze-all cycles, demonstrates that nearly half of patients achieved live births after ART, with outcomes comparable to international benchmarks. Maternal age, oocyte yield, and embryo transfer stage were identified as key predictors of success, while the choice between fresh and freeze-all protocols proved to be best tailored to ovarian response rather than being generally superior. These findings highlight the resilience of Lebanese patients and fertility care systems, even in times of profound national hardship. As the first large-scale analysis in Lebanon, this study provides a critical foundation for future research and clinical optimization of ART protocols in the region. Having realistic expectations of achieving a live birth over multiple transfers is an important tool that helps physicians tailor treatment strategies and improves decision making for IVF patients. Further research should expand beyond clinical parameters to explore environmental factors, dietary habits, and lifestyle influences that may impact infertility and ART outcomes. Incorporating these variables into CLBR predictive models could further improve their clinical relevance and enable clinicians to estimate the likelihood of IVF success even before treatment begins, thereby enhancing patient counseling and personalized care.

Our study of 1,243 ART cycles found that CLBR were strongly influenced by maternal age, oocyte yield, and day of transfer. While the freeze-all strategy initially showed a modest advantage over First Fresh cycle, multivariate analysis indicated that the treatment method itself was not an independent predictor of live birth when adjusting for confounders. Our results found that maternal age is a decisive predictor of ART outcomes. Women under the age of 35 achieved the highest cumulative live birth rates, and a noticeable decline was observed among patients aged 35–39 and particularly after 40 years, where the odds of live birth dropped by nearly 65% compared to younger women. These results align with global evidence in the current literature. Mitochondrial activity, which is essential for egg maturation, and the probability of chromosomal normality both decline as maternal age increases [ 15 , 16 ]. Furthermore, a 2023 meta-analysis of euploid embryo transfers reported significantly higher ongoing pregnancy and live birth rates among women younger than 35 compared to those 35 and older, which highlights the fact age-related decline persists regardless of embryo quality [ 17 ]. Similarly, a large 2025 donor-oocyte study identified the age of 40 as a critical threshold, with live birth rates falling markedly after this age and miscarriage rates rising beyond 43 years, suggesting a contribution of uterine aging independent of egg quality [ 18 ]. Paternal age in our cohort showed a mild, though observable, negative trend on live birth outcomes, with reduced success among men aged over 45 years. This finding aligns with a 2023 systematic review of egg-donation cycles, demonstrating a slight but significant linear decrease in live birth rates with each additional year of paternal age, independent of maternal age [ 19 ]. Another cohort study found male partners over the age of 40 had lower clinical pregnancy and implantation rates despite normal semen parameters and young female partners [ 20 ]. Though the decline in overall live birth rates with paternal age was modest, our data corroborate the notion that paternal age over 40 years may modestly affect ART outcomes. The impact of advanced paternal age on the prevalence of aneuploid embryos are debatable. An increased sperm DNA fragmentation and genetic changes such as mutations and chromosomal numerical and structural alterations have been described in older men [ 21 , 22 ]. Nevertheless, maternal age still remains the principal determinant of success. In our study, oocyte yield emerged as a strong predictor of success. Women with more than 15 retrieved oocytes had nearly 2.8 times higher odds of live birth compared to those with fewer than 10 oocytes. CLBR surpassed 60% in this high-yield group by the fourth transfer, compared to only ~ 30% in women with < 10 oocytes. These findings align with a 2023 systematic review looking at the association between the oocyte number and CLBR, which confirmed a positive correlation between the two across studies [ 23 ]. The benefit, however, plateaus in younger patients ( 35 years) continue to benefit from higher oocyte yields. Our data confirm that patients aged 35 and above with high oocyte retrieval achieved significantly better outcomes, reflecting previous reports which found that maximizing oocyte yield can counterbalance diminished egg quality in older women. Furthermore, our median oocyte retrieval of 11 (IQR: 6–17) is consistent with the optimal range identified in a 2021 meta-analysis, which noted that fresh live birth rates per cycle peak at around 12–18 oocytes [ 24 ]. However, as highlighted in that same review, the cumulative live birth rate (across all fresh and frozen transfers) continues to increase with more eggs, though with diminishing returns. Our observation that women with > 15 oocytes achieved the highest cumulative success supports the evidence favoring moderate-to-high stimulation protocols, particularly for older patients. Our findings showed that Day 5 (blastocyst-stage) transfers were associated with significantly higher odds of live birth compared to Day 3 transfers, consistent with the broader literature. A 2022 Cochrane review of 32 randomized control trials involving 5,821 patients reported that fresh blastocyst-stage transfers achieved higher live birth rates compared to cleavage-stage transfers (OR ~ 1.27) [ 25 ]. This translates to an absolute increase from approximately 31% to 35–41% live birth rates when Day 5 embryos are used. The improved outcomes are attributed to better embryo selection (since embryos failing to reach blastocyst are naturally screened out) and improved synchrony with endometrial receptivity at the time of transfer. However, the Cochrane review also highlighted potential trade-offs, including fewer embryos available for cryopreservation and a higher risk of having no embryo to transfer when a Day 5 strategy is chosen. In our cohort, 18.9% of transfers were performed at Day 3, mostly in patients with fewer embryos. While our cumulative success rates were still higher with Day 5, this supports the view that Day 3 transfers remain a viable option for poor-prognosis patients who may risk having no blastocyst for transfer. As the review notes, cumulative live birth rates across both fresh and frozen transfers may converge between Day 3 and Day 5 strategies, a trend also reflected in our data where cumulative outcomes were ultimately comparable after multiple transfers. In our unadjusted analysis, freeze-all strategies showed slightly higher success compared to First Fresh cycle (CLBR: 51.4% vs. 41.7% by the fourth transfer). However, after adjusting for confounders such as maternal age and oocyte yield, the transfer strategy itself was not an independent predictor of live birth. This suggests that the apparent benefit of freeze-all is driven largely by patient characteristics, particularly ovarian response. Recent studies support this individualized approach. A large analysis of 7,236 ICSI cycles reported that freeze-all strategies yield superior cumulative live birth rates in normal responders (10–15 oocytes: 40.5% vs. 36.6%) and hyper-responders (> 15 oocytes: 52.2% vs. 47.7%) compared to fresh transfers [ 26 ]. These results align with our subgroup findings, where patients with > 15 oocytes benefitted most from freeze-all strategies. This advantage is likely due to improved uterine receptivity when transfer occurs in a non-stimulated cycle, alongside the ability to transfer cryopreserved high-quality embryos sequentially. In contrast, for patients with low ovarian yield (< 10 oocytes), our data showed minimal differences between the two protocols. This aligns with a 2025 multicenter randomized control trial in China that found fresh transfer significantly outperformed freeze-all in low-prognosis women (40% vs. 32% live birth rate). Our findings confirm that maternal age, oocyte yield, embryo transfer stage, and time selection are key determinants of live birth in ART. Younger women, particularly those under 35, achieved the highest CLBR, while older women benefitted most from higher oocyte yields and blastocyst-stage transfers. Although freeze-all strategies offered a slight unadjusted advantage, this was not significant after accounting for confounders, supporting the current evidence that treatment should be individualized. This study has several limitations that should be acknowledged. First, it was conducted at a single ART center in Beirut, which may limit the generalizability of our findings to other clinics or regions. The study design also restricts our ability to control for all potential confounders. Second, key clinical and lifestyle variables, such as body mass index, smoking status, AMH levels, and genetic screening of embryos, were not tested for the purpose of this study, which may have influenced ART outcomes. Additionally, psychological stress, socio-economic status, and treatment discontinuation reasons were not evaluated, despite their known impact on fertility success. Finally, the study period coincided with severe national crises in Lebanon, including economic collapse, political unrest, and the aftermath of the Beirut port explosion, which may have indirectly affected patient behaviors and access to care, but these contextual factors were not systematically measured.

A study was conducted within the framework of the Exposures in Lebanese Infants (EELI) study, using data derived from 1769 couples who underwent their first ovarian stimulation cycles performed at the Saint Joseph Fertility center located in Beirut, Lebanon from June 16, 2020 till June 30, 2024. Cycles with missing information or no response to ovarian stimulation or that started after June 30, 2024 were excluded from the study. Cycles aiming to cryopreserve embryos for medical reasons ( N  = 53) or transfer embryos after preimplantation genetic testing ( N  = 274) were also not included in the study. We collected data from cycles with at least one egg fertilized and at least one embryo formed and found suitable for transfer (Fig.  1 ). A total of 1243 patients was included in the analysis. Patients were divided into groups based on whether they underwent fresh embryo transfer with or without subsequent FETs ( N  = 681) or only FETs in case of “freeze all” ( N  = 562). Women were followed up with for at least one year, until their first live birth delivery or interruption of their treatment, whichever occurred first. Women with live births were removed from any further analysis. The data set was de-identified prior to analysis. Written informed consent was obtained from all participants, who were assured confidentiality, privacy, and anonymity of their information and identity. Ethical approval for this study was granted by the Hotel-Dieu de France University Hospital’s Ethics Committee. Fig. 1 Flow chart representing the selection process of the study patients Flow chart representing the selection process of the study patients Patients first underwent standard infertility assessments and were asked to perform tests and ultrasounds before starting IVF treatments. For the IVF cycles, hormonal treatments were given to female patients to stimulate follicular growth. Dosages were adjusted according to the hormonal response and the anthropometrics of patients. Patients with an Anti-Mullerian Hormone (AMH) level above 1 ng/ml were prescribed 200–300 IU of recombinant FSH (GONAL F, Merck or PUREGON, Organon). Women with AMH lower than 1 ng/ml were given recombinant FSH along with LH ratio 2:1 (MERIONAL, IBSA or MENOPUR, Ferring or PERGOVERIS, Actoverco). Additionally, the Antagonist (CETROTIDE, Merck or ORGALUTRAN, Organon) or Progestin-primed ovarian stimulation (PPOS) (DUPHASTON, Abbott) protocols were adopted to control LH surge. After monitoring the follicular size via ultrasound, when two or more follicles reached 15 to 20 mm and the peak of estradiol reached at least 500 pg/ml, ovulation was induced (OVITRELLE, Merck) around 36 h before egg retrieval (day 0). Eggs were collected under sedation before being treated and injected using the ICSI technique. The samples were then cultured at 37 C, 5.5% CO2 and 5.5% O2 and monitored for fertilization (day1) and embryo development (day 2 till day 5 or 6). For fresh cycles, embryos were transferred at cleavage stage (day 3) or at blastocyst stage (day 5 or 6). At cleavage stage, embryos were graded according to the BLEFCO system [ 12 ] while blastocysts were graded according to Gardner scoring system [ 13 ]. The supernumerary viable embryos of good grade were cryopreserved via vitrification method (IRVINE, FUJIFILM Irvine Scientific) and stored in liquid nitrogen tanks for future embryo use. Luteal phase support was provided by the prescription of ESTROFEM, Novo Nordisk, and vaginal progesterone. For “freeze all” cycles, all embryos were cryopreserved for transfer during subsequent cycles. Frozen embryo transfers followed either a supplemented cycle (ESTROFEM, Novo Nordisk prescribed with vaginal progesterone 600–800 mg/day) or a natural cycle (OVITRELLE, Merck, trigger followed by a daily dose of DUPHASTON, Abbott). Transfers were majorly performed using guided catheters. The number of embryos transferred was based on each patient’s case. In the majority of cycles, one or two embryos were transferred, in accordance to standard clinical practice. If clinically indicated, the transfer of up to three embryos was permitted. Assisted hatching was applied before embryo transfers. Each transfer cycle is considered a unique treatment event. Demographic and ART data were collected from the medical records of the fertility center. The collected characteristics included the age of women and men, type of infertility (primary or secondary), and cause of infertility (male factor, female factor, mixed factors, unexplained infertility, or other reasons) during the first IVF cycle. We also recorded the date of egg retrieval to determine the season and the couples’ addresses to identify the patients’ geographical distribution across the nine Lebanese governorates (Mount Lebanon, Beqaa, Beirut, North Lebanon, South Lebanon, Nabatieh, Keserouan-Jbeil, Akkar, and Baalbek-Hermel). The primary outcome was the CLBR (delivery of at least one live baby [ 14 ]) per woman following a fresh cycle (fresh transfer and subsequent frozen-thawed embryo transfers) or a freeze all cycle (frozen-thawed embryo transfers only) over four transfer cycles, as few women underwent more than four transfers. Both conservative and optimal cumulative live birth were estimated. Cumulative live birth rates were also stratified into groups according to female age at the start of the treatment (< 30, 30–34, 35–39, 40–44, ≥ 45) and the number of oocytes retrieved ( 15 oocytes). The patients’ data are displayed in counts and percentages for categorical variables and in means, standard deviations, medians, interquartile ranges, and minimum and maximum values for continuous variables. Bivariate associations with live birth were assessed using simple logistic regression models to estimate crude odds ratios (ORs) and 95% confidence intervals (CIs). Variables examined included maternal age, transfer strategy (first fresh vs. freeze-all), and number of oocytes retrieved. Multivariate logistic regression models were used to estimate adjusted odds ratios (aORs), controlling for relevant confounders. Model performance and goodness-of-fit were evaluated using pseudo-R², Wald-chi square statistics, and classification statistics, including sensitivity, specificity, and overall accuracy. To assess treatment outcomes, we calculated both conservative and optimistic CLBR. The conservative CLBR was defined as the proportion of patients who achieved at least one live birth up to and including a specific embryo transfer cycle, divided by the total number of patients who initiated their first ovarian stimulation during the study inclusion period. The optimistic CLBR assumes that patients who have not yet achieved a live birth still achieve one in the future. It is estimated using the Kaplan–Meier survival analysis, treating live birth as the event of interest and censoring patients at their last recorded transfer cycle if no live birth occurred. This approach accounts for loss to follow-up and varying numbers of transfer cycles among patients. Differences in cumulative live birth probabilities across age groups and oocyte yield categories were assessed using the log-rank test. Confidence intervals (95% CI) for proportions were calculated using standard errors based on the binomial distribution. All analyses were conducted using STATA software, version 12.

Infertility describes a condition where an individual or couple is unable to conceive due to several factors, including age, medical history, reproductive profile, or other diagnostic findings. Couples having unprotected intercourse with no pregnancy after several attempts should seek the evaluation of a physician–within one year for women under the age of 35 and within 6 months in case of older individuals [ 1 ]. Patients with fertility issues resort to reproductive care centers and undergo assisted reproductive techniques (ART) in attempts to achieve pregnancy. One of these techniques is In-Vitro Fertilization (IVF), which has become one of the most widely used methods to assist infertile couples. Due to the high demand for reproductive assistance, researchers are currently focusing on providing quality reproductive services, especially considering the high cost, effort, and time required by the patients [ 2 ]. At the same time, patients find themselves eager to know their chances of success with IVF, particularly after several attempts. The success of ART is usually determined by pregnancy rates, which depend on the measurement stage: per egg retrieval, per embryo transfer cycles, or per patient undergoing IVF. Nevertheless, the cumulative live birth rates (CLBRs) have become a better indicator of IVF effectiveness and a helpful tool for patients and physicians [ 3 ]. CLBRs are usually estimated using two different methods. The optimistic CLBR approach assumes that patients who have already discontinued IVF treatment following the first failed attempt still have the same chance of giving birth as those who have not. On the other hand, the conservative CLBR approach assumes that dropouts after failure will not achieve live birth. Many studies reported both approaches to avoid overestimating success rates and to provide a more realistic picture of patients’ outcomes [ 4 , 5 ]. Results are usually reported per egg retrieval in order to reflect the real-life probability of achieving pregnancy from a single stimulation cycle or per transfer to account for the embryo selection and transfer protocols [ 6 ]. However, there remain significant gaps in the current literature regarding the success rates of both approaches, especially for some vulnerable populations. Mainly, low- and middle-income countries are found to be underrepresented in studies on IVF research. For instance, infertility among Lebanese couples is poorly explored. One cross sectional study addressed to the general Lebanese public used a survey to examine the participants’ perceptions towards infertility and test their awareness regarding the possible causes and the available solutions for fertility issues. According to the survey responses, of the 609 respondents, most male participants believed that infertility was a valid ground for divorce and of female origin. In addition, individuals with a higher educational level were more open to the possibility of adoption or IVF [ 7 ]. Another study enrolling 952 Lebanese women focused on the socio-demographic factors and medical conditions associated with higher rates of infertility. The results showed that advanced age, endometriosis, and polycystic ovarian syndrome were some of the factors increasing the risk of infertility. On the other hand, having a higher level of education was associated with lower odds of infertility [ 8 ]. Moreover, one retrospective study on 547 women that underwent 736 fresh embryo transfer cycles over one year of enrollment calculated the cumulative live birth rate up to and including six consecutive IVF cycles. Results showed that the CLBR reached 67.9% after six IVF cycles, compared to 56.9% after three cycles only [ 9 ]. Given the growing emphasis on optimizing IVF outcomes and minimizing treatment risks, there has been a noticeable shift in clinical practice toward the increased use of frozen embryo transfers (FETs). Comparing the outcomes of both fresh and frozen embryo transfers is quite important since freeze all cycles and FETs have become more popular and sought after over the past years due to concerns that hormone stimulation may negatively affect endometrial receptivity or increase risks such as ovarian hyperstimulation syndrome. Recent data from Lebanese patients across multiple transfer cycles may provide meaningful insights when compared to international registries and populations. There are currently several international data registries gathering data from ART centers, such as the European Society of Human Reproduction and Embryology (ESHRE) and the Centers of Disease Control and Prevention and Society for Assisted Reproductive Technology (SART) [ 10 , 11 ]. These international registries report data on ART outcomes, offer valuable insights into global trends, monitor practice and improve IVF services. However, the lack of data collection from low- and middle-income countries, leads to a limited understanding of region-specific challenges and success rates. For example, infertility remains underexplored in Lebanon, where only a handful of studies have examined public perceptions, socio-demographic factors, and clinical outcomes related to IVF. Therefore, the objective of this study is to assess the demographic characteristics of a group of Lebanese infertile patients attending an ART center in Beirut, along with the outcomes of their first IVF cycles. We evaluated CLBR across multiple transfer cycles, comparing fresh and freeze-all transfer strategies and using both conservative and optimistic approaches. Furthermore, we stratified cumulative live birth results into groups by female age and number of retrieved eggs in order to assess the impact of these factors on pregnancy outcomes.

Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.