Can Serum β-hCG Values ​​Measured after Embryo Transfer Predict Live Birth ​​in Both Fresh and Frozen Transfers?

By Burak Elmas¹, Serap Topkara Sucu¹, Berrin Goktug Kadoglu¹, Mustafa Ozturk¹, Ozlem Ozturk², Seyit Temel Ceyhan¹

Affiliations

1. Department of Gynaecology and Obstetrics, Division of Reproductive Endocrinology and Infertility, University of Health Sciences Gulhane Training and Research Hospital, Ankara, Turkiye
2. Department of Medical Biochemistry, University of Health Sciences Gulhane Training and Research Hospital, Ankara, Turkiye

doi: 10.29271/jcpsp.2025.10.1335

ABSTRACT
Objective: To determine the threshold β-hCG values ​​that can predict the live birth (LB) rates in in vitro fertilisation-embryo transfer (IVF-ET) cycles, and to compare these serum β-hCG values ​​separately in fresh and frozen embryo transfers.
Study Design: Case-control study.
Place and Duration of the Study: Department of Gynaecology and Obstetrics, University of Health Sciences Gulhane Training and Research Hospital, Ankara, Turkiye, between January 2017 and January 2023.
Methodology: Serum β-hCG values ​​of patients who underwent single embryo transfer on Day 5 were measured on Days 12 and 14 after transfer. The patients were divided into two groups: pregnancies that resulted in live birth (LB (+)) and pregnancies that did not result in live birth (LB (-)). The two groups were compared in terms of serum β-hCG values ​​and β-hCG increase rates on the 12^th and 14^th days after transfer.
Results: The median β-hCG value measured on day 12 in fresh transfers resulting in LB was found to be 277 IU/L. However, this value was found to be 332 IU/L in frozen transfers. When all transfers were considered without discrimination, it was concluded that the β-hCG test could predict LB with a sensitivity of 75% and a specificity of 72% at a value of 197 IU/L.
Conclusion: A cut-off value of 197 IU/L was found to be a strong parameter to predict LB in all transfers, regardless of whether the transfer was fresh or frozen. It will guide clinicians in predicting pregnancy outcomes and counselling patients.

Key Words: β-hCG, Live birth, Fresh transfer, Frozen transfer, In vitro fertilisation.

INTRODUCTION

The global use of assisted reproductive technology (ART) is increasing. Advancements in embryo cryopreservation have improved success rates and reduced complications. The primary indicator of success in ART is the live birth (LB) rate, which has increased significantly with recent advances. The initial predictor of LB rates is the first beta(β)-human chorionic gonadotropin (β-hCG) value measured after embryo transfers.

β-hCG, secreted by syncytiotrophoblasts post-implantation, indicates trophoblast growth.¹ Maternal blood can detect β-hCG as early as six to eight days after fertilisation,² and levels typically double every 48 hours in normal pregnancies.³ Measurements on Day 12 post-transfer can predict foetal heart activity.⁴ Tracking sequential β-hCG levels after embryo transfer aids in predicting intrauterine  pregnancy.

Questions such as "Will this embryo transfer result in a LB?" concern both patients and physicians, and often cause anxiety for couples undergoing in vitro fertilisation (IVF).⁵ Timely and accurate prediction of LB can aid patient counselling and reduce anxiety after implantation. Although early pregnancy serum β-hCG levels are evaluated, post-transfer levels are used to indicate the pregnancy course and LB prediction are often unsatisfactory. Some studies have sought serum markers to predict IVF outcomes;^4-7 however, limitations such as different measurement days and IVF cycle types have created inconsistent results. Few studies provide strong β-hCG cut-offs for LB. Previous comparisons of early β-hCG values and pregnancy rates between fresh and frozen embryo transfers were inconclusive.^6,7 This study therefore aimed to determine the threshold β-hCG values for predicting LB rates, and to compare serum β-hCG values in fresh versus frozen embryo transfers.

METHODOLOGY

This study analysed records from the in vitro Department of Fertilisation, Ankara Gulhane Training and Research Hospital, Ankara, Turkiye. Women who had a single embryo transfer on Day 5 between January 2017 and January 2023 were included.
 

Patient records provided demographic data (age, BMI, obstetric, and medical history), IVF indications, transfer method (fresh or frozen), pre-implantation genetic testing (PGT) status, and β-hCG values measured on Day 12 post-ET (β1), and repeated 48 hours later (β2). Doubling was defined as a 100% increase in β-hCG within 48 hours. The β-hCG increment rate was the percentage difference between β1 and β2. Non-smokers with no comorbidities, who had a single, good-quality embryo transferred and showed β-hCG positivity, were included. Patients with comorbidities, multiple pregnancies, or detected pregnancy abnormalities were excluded from the study.

In vitro fertilisation-embryo transfer (IVF-ET) cycles were performed by the same physicians. Controlled ovarian hyper- stimulation followed the GnRH antagonist protocol. Ovulation was triggered with recombinant hCG (Ovitrelle 250 mg). Intracytoplasmic sperm ınjection (ICSI) was used for all cases. The frozen embryo transfer (FET) protocol involved artificial endometrium preparation with oral estradiol hemihydrate and micronised progesterone. Good quality blastocysts (Gardner score of 4BB or better) were selected for ET.⁸ Luteal phase support included 600 mg/day vaginal micronised progesterone and  25  mg/day  subcutaneous  progesterone.

Serum β-hCG levels of 10 IU/L and above were considered positive, measured by electrochemiluminescence ımmuno-assay (ECLIA) (Beckman Coulter Dxl 800). β-hCG positive patients were categorised into two groups: those resulting in live births (LB (+)) and those with non-viable pregnancies, such as biochemical pregnancy, ectopic pregnancy, or abortion (LB (-)). β-hCG values were compared between these groups and between fresh and frozen transfers to establish a predictive β-hCG cut-off for LB.

Statistical analyses were conducted using SPSS version 22.0. Normality was assessed with histograms, probability plots, and the Kolmogorov-Smirnov and Shapiro-Wilk tests. Homogeneity of variance was assessed with Levene’s test between the LB (+) and LB (-) groups. For non-normally distributed variables, descriptive statistics were reported as medians with interquartile ranges (Q1-Q3). The Mann-Whitney U test was used to compare these parameters among groups. For categorical variables, descriptive analyses included frequencies and percentages. Relationships between categorical variables (such as IVF indication, status of PGT, status of assisted hatching, and doubling of β-hCG) were analysed using the Chi-square test, or Fisher’s exact test if Chi-square assumptions were violated. Receiver operating characteristics (ROC) curve analysis assessed the ability of various parameters to predict failed IVF outcomes. The authors determined the cut-off value using ROC curve analysis. Significant cut-off values were reported with their sensitivity, specificity, and area under the curve (AUC) values. A p-value of less than 0.05 indicated statistical significance.

Figure  1:  Flowchart  of  the  study.

Figure 2: ROC analysis results of β1, β2, and β-hCG increment rate variables. (A)  Fresh  transfers  (B)  Frozen  transfers  (C)  All  transfers.

Table I: The demographic and clinical characteristics of the transfers that resulted/did not result in LB.

Parameters	LB (+) (n : 121)	LB (-) (n : 54)	Total (n : 175)	p-values
Age (years)	30 (27-32)	31 (28-33)	30 (27-33)	0.061¥
BMI (kg/m2)	26 (24-29)	26 (23-30)	26 (24-30)	0.999¥
IVF Indications (n, %)
DOR	10 (8.3)	10 (18.5)	20 (11.4)	0.099*
Male factor	25 (20.7)	5 (9.3)	30 (17.1)
Unexplained infertility	76 (62.8)	33 (61.1)	109 (62.3)
Endometriosis	2 (1.7)	2 (3.7)	4 (2.3)
Tubal factor	3 (2.5)	3 (5.6)	6 (3.4)
PCOS	5 (4.1)	1 (1.9)	6 (3.4)
AH (n, %)	23 (19)	10 (18.5)	33 (18.9)	NAµ
PGT (n, %)	4 (3.3)	2 (3.7)	6 (3.4)	NAµ
β1 (IU/L)	323 (197-527)	102 (44-274)	250 (105-444)	<0.001¥
β2 (IU/L)	954 (498-1436)	305 (94-690)	700 (312-1181)	<0.001¥
Doubling (+)	108 (89.3)	29 (53.7)	137 (78.3)	<0.001£
β-hCG increment rate (%)	166 (117-225)	95 (66-186)	151 (94-217)	<0.001¥
*Fisher Freeman Halton test, ¥Mann-Whitney U test, µFisher’s exact test, £Pearson Chi-square test. LB: Live birth; BMI: Body mass index; DOR: Diminished over reserve; PCOS: Polycystic ovary syndrome; AH: Assisted hatching; PGT: Preimplantation genetic testing; β1: Serum B-hCG value measured on Post-ET 12th Day; β2: Serum β-hCG value measured on Post-ET 14th Day.

Table II: The comparison of serum β-hCG parameters between the groups with and without LB in fresh and frozen transfers.

Variables	Fresh transfers			Frozen transfers
	LB (+) (n : 42)	LB (-) (n : 18)	p-values	LB (+) (n : 79)	LB (-) (n : 36)	p-values
β1 (IU/L)	277 (163-515)	118 (64-361)	0.005	332 (199-575)	96 (35-201)	<0.001
β2 (IU/L)	888 (494-1347)	389 (106-722)	0.005	1000 (499-1455)	179 (77-483)	<0.001
Doubling (+) (n, %)	38 (90.5)	10 (55.6)	0.004	70 (88.6)	19 (52.8)	<0.001
β-hCG increment rate (%)	134 (113-230)	102 (68-242)	0.165	172 (126-224)	93 (62-175)	<0.001
β1: Serum β-hCG value that was measured on Post-ET 12th Day; β2: Serum β-hCG value that was measured on Post-ET 14th Day; LB: Live birth. p-values ​​were obtained using the Chi-square test for doubling (+) and the Mann-Whitney U test for the values ​​in the table.

Table III: The ROC analysis results of serum β-hCG variables for fresh transfers, frozen transfers, and all transfers.

Variables		AUC	CL	Cut-off	p-values*	Sensitivity (%)	Specificity (%)
Fresh Transfers	β1 (IU/L)	0.728	0.581-0.875	³201	0.005	71	67
	β2 (IU/L)	0.733	0.590-0.876	³445	0.005	79	56
	β-hCG increment rate (%)	0.614	0.426-0.802	³106	0.165	83	56
Frozen Transfers	β1 (IU/L)	0.778	0.674-0.882	³194	<0.001	77	75
	β2 (IU/L)	0.813	0.721-0.906	³405	<0.001	82	75
	β-hCG increment rate (%)	0.713	0.601-0.824	³112	<0.001	81	64
All Transfers	β1 (IU/L)	0.759	0.675-0.844	³197	<0.001	75	72
	β2 (IU/L)	0.789	0.711-0.867	³409	<0.001	81	69
	β-hCG increment rate (%)	0.680	0.583-0.777	³111	<0.001	80	61
*p-values were determined by the ROC curve. β1: Serum β-hCG value that was measured on Post-ET 12th Day; β2: Serum β-hCG value that was measured on Post-ET 14th Day.

RESULTS

As shown in the study flow chart (Figure 1), 2,221 IVF cycles and 976 embryo transfers were examined during the study period. A total of 560 patients who underwent a single transfer of a high-quality Day 5 embryo were identified. After excluding patients who smoked or had comorbidities, 186 patients with β-hCG positivity on Day 12 after transfer were identified. Six patients were excluded due to missing data, and five were excluded because of multiple pregnancies. Consequently, 175 patients were included in the analysis, of whom 60 underwent fresh embryo transfer and 115 underwent frozen embryo transfer.

The demographic, clinical, and laboratory data of patients who achieved LB and those who did not are available in Table I. Evaluation of the available data showed that β1, β2 values, doubling, and β-hCG increment rates (%) were statistically different between the groups (p <0.001, Table I).

When fresh transfer and frozen transfer groups were evaluated by subdividing them into LB and stillbirth, statistically significant differences were detected in β1, β2, doubling, and increment rate results between the groups (p <0.001, Table II).

When the groups were examined separately for ROC values, the β2 value in the fresh transfer group had the highest sensitivity, with an AUC of 0.733. In the frozen transfer group, the β2 value again has the highest AUC of 0.813. In this context, the β2 variable had moderate power in determining LB in the fresh transfer group, but demonstrated high power in the frozen embryo transfer group. In the analysis including all transfers, β1, β2, and β-hCG increase percentages had discriminatory power in predicting the rate (p <0.001). Values above 197 for β1 and above 402 for β2 predicted LB (Table III). The ROC analysis graphs are given in Figure 2. The increase in β-hCG levels in frozen transfers was found to be a significant factor of LB, with levels exceeding an increase of 112 units.

DISCUSSION

The most desired outcome after ART is pregnancy, and the happiest outcome for a couple suffering from infertility is to take home a live baby. The study found that serum β-hCG levels after transfer can determine LB, with a value of 197 IU/L. However, when fresh and frozen transfers were considered separately, the median β-hCG value differed between the two groups. In frozen transfers, the β-hCG percentage increase was an indicator of LB, with a cut-off value above 112%.

Several studies in the literature evaluating β-hCG levels and increases following ET have reported values ranging from 30-360 mIU/ml to be associated with pregnancy outcomes.^3,6,7,9 In a 2011 study,⁴ the β-hCG value measured on Day 12 after transfer was highly predictive of LB, with a cut-off of 80 IU/L. However, the study was based on Day 3 embryos. In their study, Grin et al. reported β-hCG cut-off values predictive of LB as 211 IU/L for fresh transfers and 182 IU/L for frozen transfers.⁹ The results of their study, with a larger sample size, were found to be close to the cut-off values ​​of the current study. The higher sensitivity observed in this study may be attributed to the large number of patients. However, unlike Grim et al., who included patients with embryos transferred on different days, the present study included only patients whose embryos were transferred on Day 5.

The study focuses on LB rate and its relationship with clinical pregnancy rate and ongoing pregnancy. It uses only good-quality single embryo transfers on Day 5 to minimise possible variability. In previous studies, Day 3 and Day 5 embryos were evaluated together, with β-hCG values reported to be higher in Day 5 embryos.^10,11 However, patients undergoing cleavage transfer may have had poor embryo quality, which could contribute to overall lower β-hCG production. The two extra days in vivo development for Day 3 compared to Day 5 embryo may be associated with different trophoectoderm maturation. Furthermore, embryo growth in the laboratory is not analogous to the complex environment of the intrauterine environment.^12,13

β-hCG measured on Day 14 predicted LB at a higher rate than β-hCG measured on Day 12. However, the β-hCG percentage increase was an important indicator in frozen transfers rather than fresh transfers. In a previous study, the change in β-hCG between Days 14 and 16 was reported as 211% in fresh transfers and 245% in frozen ones. However, these increases were unexpectedly detected in the non-viable group, and their clinical significance was reported to be limited.⁶

One surprising result of the present study was that the median β-hCG value measured on Day 12 in fresh transfers was 277 IU/L, and in frozen transfers, the value of 332 IU/L was associated with LB. No clear consensus was detected on the subject in the literature. Some studies reported that there was no difference in β-hCG values between fresh and frozen transfers.^4,14 However, live births were achieved with higher β-hCG levels in frozen transfers compared to that in fresh transfers, similar to the present study.⁹ This study is limited by its retrospective design and sample size.

CONCLUSION

The most important parameter in predicting LB, the primary outcome in IVF applications, remains β-hCG measured after transfers. A cut-off value of 197 IU/L was determined as the most powerful predictor of LB rate in all transfers, regardless of whether the transfer was fresh or frozen. These findings may guide clinicians in predicting pregnancy outcomes at an earlier stage.

ETHICAL APPROVAL:
The study was conducted in line with the principles of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of Ankara Gulhane Training and Research Hospital, Ankara, Turkiye (21 November 2023/ Issue: 2023/281).

COMPETING INTEREST:
The authors declared no conflict of interest.

AUTHORS’ CONTRIBUTION:
BE: Conception, design of the work, and drafting.
STS: Analysis and interpretation of the data.
BGK: Drafting of the work and critical revision of the manu-script.
MO: Acquisition, analysis, and interpretation.
OO: Contributed to data collection.
STC: Design of the work.
All authors approved the final version of the manuscript to be published.

REFERENCES

1. Schumacher A, Zenclussen AC. Human chorionic gonadotropin-mediated immune responses that facilitate embryo implantation and placentation. Front Immunol 2019; 10:2896. doi: 10.3389/fimmu.2019.02896.
2. Wang Q, Zhang R, Jia M, Luo L, Ding C, Li T, et al. Serum human chorionic gonadotropin measured 7 days following day 3 embryo transfer might predict pregnancy outcome in IVF. Gynecol Endocrinol 2017; 33(1):62-6. doi: 10.1080/ 09513590.2016.1203894.
3. Porat S, Savchev S, Bdolah Y, Hurwitz A, Haimov-Kochman R. Early serum beta-human chorionic gonadotropin in pregnancies after in vitro fertilization: contribution of treatment variables and prediction of long-term pregnancy outcome. Fertil Steril 2007; 88(10):82-9. doi: 10.1016/j. fertnstert.2006.11.116.
4. Lawler CC, Budrys NM, Rodgers AK, Holden A, Brzyski RG, Schenken R. Serum beta human chorionic gonadotropin levels can inform outcome counseling after in vitro fertilization. Fertil Steril 2011; 96(2):505-7. doi: 10.1016/j. fertnstert.2011.05.071.
5. Xu H, Ouyang N, Li R, Tuo P, Mai M, Wang W. The effects of anxiety and depression on in vitro fertilisation outcomes of infertile Chinese women. Psychol Health Med 2017; 22(1):37-43. doi: 10.1080/13548506.2016.1218031.
6. Shamonki MI, Frattarelli JL, Bergh PA, Scott RT. Logarithmic curves depicting initial level and rise of serum beta human chorionic gonadotropin and live delivery outcomes with in vitro fertilization: An analysis of 6021 pregnancies. Fertil Steril 2009; 91(5):1760-4. doi: 10.1016/j.fertnstert.2008. 02.171.
7. Sung N, Kwak-Kim J, Koo HS, Yang KM. Serum hCG-b levels of postovulatory day 12 and 14 with the sequential application of hCG-b fold change significantly increased predictability of pregnancy outcome after IVF-ET cycle. J Assist Reprod Genet 2016; 33(9):1185-94. doi: 10.1007/ s10815-016-0744-y.
8. Kljajic M, Sayme N, Krebs T, Wagenpfeil G, Baus S, Solomayer EY, et al. Zygote diameter and total cytoplasmic volume as useful predictive tools of blastocyst quality. Geburtshilfe Frauenheilkd 2022; 83(1):97-105. doi: 10.1055/a-1876-2231.
9. Grin L, Indurski A, Leytes S, Rabinovich M, Friedler S. Trends in primeval β-hCG level increment after fresh and frozen-thawed IVF embryo transfer cycles. Gynecol Endocrinol 2019; 35(3):261-6. doi: 10.1080/09513590. 2018.1519789.
10. Shibli Abu Raya Y, Bilgory A, Aslih N, Atzmon Y, Shavit M,Estrada D, et al. High initial β-hCG predicts IVF outcomes accurately and precludes the need for repeated measurements. Endocr Connect 2023; 12(11):e230189. doi: 10.1530/EC-23-0189.
11. Oron G, Esh-Broder E, Son WY, Holzer H, Tulandi T. Predictive value of maternal serum human chorionic gonadotropin levels in pregnancies achieved by in vitro fertilization with single cleavage and single blastocyst embryo transfers. Fertil Steril 2015; 103(6):1526-31.e2. doi: 10.1016/j.fertnstert.2015.02.028.
12. Hughes LM, Schuler A, Sharmuk M, Schauer JM, Pavone ME, Bernardi LA. Early β-hCG levels predict live birth after single embryo transfer. J Assist Reprod Genet 2022; 39(10):2355-64. doi: 10.1007/s10815-022-02606-w.
13. Dahiya M, Rupani K, Yu SL, Fook-Chong SMC, Siew Fui DC, Rajesh H. Embryo transfer day does not affect the initial maternal serum β-hCG levels: A retrospective cohort study. Eur J Obstet Gynecol Reprod Biol 2017; 212:75-9. doi: 10.1016/j.ejogrb.2017.03.015.
14. Reljic M, Knez J, Vlaisavljevic V. Human chorionic gonadotropin levels are equally predictive for pregnancy outcome after fresh and vitrified-warmed blastocyst transfer. J Assist Reprod Genet 2013; 30(11):1459-63. doi: 10.1007/s10815-013-0099-6.