Human chorionic gonadotropin levels of cesarean scar pregnancy compared with those of other ectopic pregnancies: a retrospective cohort study.

The characteristics of the five patients with CSP diagnosed in 27,528 cycles of FET at our hospital between January 2016 and March 2025 are shown in Table  1 . Among the patients with CSP, one case was identified as a duplicate entry. However, the duplicate cases were not from the same cycle, thus the cycles were included in the statistical analysis. The characteristics of the 11,800 clinical pregnancy cycles and 44 ectopic pregnancy cycles following FET are shown in Table  2 . In ectopic pregnancy, the rates of fallopian tube infertility and cesarean section history were higher than those in clinical pregnancy (fallopian tube infertility, P  = 0.002, history of cesarean section, P  = 0.023). In addition, in unexplained infertility, clinical pregnancy was more common than ectopic pregnancy ( P  = 0.005). Table  3 shows the effect of each characteristic on ectopic pregnancy and CSP in FET by logistic regression analysis. In ectopic pregnancy, tube infertility and cesarean section history showed a significant difference in the logistic regression analysis [(tube infertility, P  = 0.001; odds ratio [OR] 2.823; 95% confidence interval [CI], 1.491–5.342), history of one time cesarean section, P  = 0.005; odds ratio [OR] 2.25; 95% confidence interval [CI], 1.230–4.117), history of ≥ 2 cesarean sections, P  = 0.027; odds ratio [OR] 11.382; 95% confidence interval [CI], 1.505–86.066)]. Meanwhile, a history of myomectomy and ≥ 2 cesarean sections showed significant differences in the logistic regression analysis for CSP: history of myomectomy, P  = 0.003; OR, 14.86; 95% CI, 2.458–89.95 and history of multiple cesarean sections, P  = 0.006; OR, 22.84; 95% CI, 2.418–215.78. Figure 2 ; Table  4 show the serum hCG levels of each of the 44 ectopic and 11,800 clinical pregnancies during the same period. There was a significant difference between CSP and other ectopic pregnancies ( P  = 0.001). The hCG level for clinical pregnancy was 235.6 mIU/mL (136.6–382.4). There was a significant difference in hCG levels between clinical pregnancy and ectopic pregnancy ( P  < 0.001), but this significant difference disappeared when clinical pregnancy was compared with CSP ( P  = 0.607). Bonferroni’s multiple comparison test showed a significant difference between tubal pregnancy and CSP ( P  = 0.044), tubal pregnancy and clinical pregnancy (P = < 0.001), tubal pregnancy and clinical pregnancy (P = < 0.001), and cervix pregnancy and clinical pregnancy ( P  = 0.002), whereas there were no significant differences between tubal pregnancy and others ( P  = 1.000), tubal and cervix pregnancies ( P  = 1.000), cervix pregnancy and others ( P  = 1.000), CSP and others ( P  = 0.630), clinical pregnancy and others ( P  = 0.060), and cervix pregnancy and CSP ( P  = 0.829). Fig. 2 Box plot of serum hCG level among 44 cases of ectopic pregnancy. Abbreviation: hCG: human chorionic gonadotropin, CSP: cesarean scar pregnancy. Note: The number of ectopic pregnancies was as follows: tube, 26; cervix, 10; CSP, 5; and other, 3. a Others include 2 peritoneum pregnancies and 1 ovarian pregnancy Box plot of serum hCG level among 44 cases of ectopic pregnancy. Abbreviation: hCG: human chorionic gonadotropin, CSP: cesarean scar pregnancy. Note: The number of ectopic pregnancies was as follows: tube, 26; cervix, 10; CSP, 5; and other, 3. a Others include 2 peritoneum pregnancies and 1 ovarian pregnancy

We planned a retrospective cohort study to investigate hCG levels in ectopic pregnancies and evaluate whether there were any differences in values depending on the location of ectopic pregnancies. We investigated 27,528 cycles (10,571 patients) of freeze–thawed embryo transfer (FET) at our hospital between January 2016 and March 2025. Embryos were at day 3 cleavage stage or day 5 and day 6 blastocyst stages, derived from either in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). The serum hCG concentrations were investigated from the 12th day to the 17th day after ovulation in natural cycle FET or hormone replacement cycle FET with progesterone supplementation. Patients with a serum hCG concentration of ≥ 1.0 IU/l were included in the study. The implantation site of ectopic pregnancies (fallopian tube, cervix, cesarean scar, peritoneum, and ovary) was diagnosed by confirming the gestational sac using TVUS or MRI. Furthermore, CSP was diagnosed by confirming full or partial implantation of the gestational sac within the scar of a previous cesarean section. In surgical cases, the site was confirmed by postoperative pathological examination. Statistical analysis was performed on the cases diagnosed as clinical intrauterine pregnancy (clinical pregnancy; 11,800 cycles in 8,389 patients) and ectopic pregnancy (44 cycles in 42 patients) (Fig.  1 ). Among the 44 ectopic pregnancy cycles, 40 underwent surgical intervention, and diagnosis was confirmed pathologically. In contrast, one cervical, two tubal, and one ovarian ectopic pregnancy resolved with spontaneous miscarriage; therefore, surgical intervention was not performed. Serum hCG levels were compared according to implantation sites. Fig. 1 Flow diagram for 27,528 FET cycles. Abbreviation: FET: frozen–thawed embryo transfer, hCG: human chorionic gonadotropin, CSP: cesarean scar pregnancy. a Patients with a serum hCG concentration of ≥ 1.0 IU/l were defined as a biochemical pregnancy Flow diagram for 27,528 FET cycles. Abbreviation: FET: frozen–thawed embryo transfer, hCG: human chorionic gonadotropin, CSP: cesarean scar pregnancy. a Patients with a serum hCG concentration of ≥ 1.0 IU/l were defined as a biochemical pregnancy All procedures were conducted in compliance with the ethical standards of the responsible committee on human experimentation (institutional and national) and the 1964 Helsinki Declaration and its later amendments. This retrospective study was approved by the Clinical Ethics Committee of Denentoshi Ladies Clinic (25030731–2). Informed consent was obtained from all participants. On the 3rd day after menstruation, ovarian stimulation was performed. For this, clomiphene citrate was prescribed at 50 mg/day and used with human menopausal gonadotropin and recombinant follicle-stimulating hormone unless the antral follicle count and anti-Müllerian hormone levels showed poor ovarian reserve. Clomiphene citrate alone was administered when ovarian reserve was low. Oocyte retrieval was performed 35–37 h after administering recombinant hCG hormone (Ovidrel, EMD Serono, Boston, MA, USA) or gonadotropin-releasing hormone agonist (Buserecur, Fuji Parma, Tokyo, Japan) when one or more ovarian follicles reached ≥ 18 mm in diameter on ultrasound examination. Spermatozoa were obtained by density gradient centrifugation, and a universal IVF medium with phenol red (Origio, Måløv, Denmark) was used as the fertilization medium. The insemination concentration was set at 1.0–1.5 × 10 5 /mL. After dropping the sample into a Makler chamber, spermatozoa were analyzed using a sperm motility analysis system (SMAS, DITECT, Japan). Embryos were exposed to the global total medium (Astec, Chattanooga, TN, USA). ICSI was performed using standard techniques. The embryos were assessed using a stereoscope (Olympus SZX16; Olympus Corporation, Shinjuku, Japan). Embryo cryopreservation was performed using a vitrification kit. Furthermore, FETs were performed using a hormone replacement cycle (HRC) or a natural cycle (NC) when the endometrial thickness was ≥ 7 mm. The transfer date for blastocysts was 5 days after ovulation or 5 days after the start of luteal support. By contrast, the transfer date for the cleavage stage was 3 days after ovulation or 3 days after the start of luteal support. In HRC-FET, luteal support was initiated on days 11–19 of the menstrual period and continued until the 9th week of gestation. In addition, estradiol administration usually begins on the 3rd day of the menstrual period. By contrast, a progesterone concentration of ≥ 1.00 ng/mL was defined as ovulation in the NC–FET from the 3rd day after ovulation until the 4th week of gestation. Thawing medium (KITAZATO BioPharma, Tokyo, Japan) was used for embryo thawing. If an appropriately increasing serum hCG level was encountered, transvaginal ultrasound was performed at 5–6 weeks of gestation to determine the presence of a gestational sac, indicating clinical intrauterine pregnancy but excluding ectopic pregnancy. The serum hCG concentration was measured using an AIA ® -CL2400 and CL AIA-PACK ® HCG TEST CUP (TOSOH Corp., Tokyo, Japan). The calibration curve for use with the CL AIA-PACK ® hCG assay has been standardized against the WHO (World Health Organization) 5th International Standard Chorionic Gonadotropin, National Institute for Biological Standards and Control Code 07/364. The characteristics of five patients with CSP are shown in Table 1 . P -values were calculated using chi-square tests to compare the characteristics of ectopic and clinical intrauterine pregnancies, including the rate of ICSI, double embryo transfer, causes of infertility, history of endometrial operation, history of cesarean section, history of myomectomy, day 3 embryo transfer, and day 5 blastocyst transfer (Table 2 ). A logistic regression analysis was performed with ectopic pregnancy and clinical pregnancy as the dependent variables and each characteristic (maternal age, paternal age, rate of ICSI, double embryo transfer, causes of infertility, history of endometrial operation, history of cesarean section, history of myomectomy, day 3 embryo transfer, and day 5 blastocyst transfer) as the independent variable (Table 3 ). Independent variables were based on previous studies [ 10 – 15 ]. The backward elimination method was used to select variables in the predictive logistic models. Differences in the serum hCG levels between the implantation sites of ectopic pregnancy (cervical pregnancy, CSP, tubal pregnancy, others (peritoneal pregnancy and ovarian pregnancy)) and clinical pregnancy were calculated using the Kruskal–Wallis test (Table 4 ). The Kruskal–Wallis test was performed to compare CSP with another ectopic pregnancy, as well as to compare ectopic pregnancies with clinical pregnancy, and CSP with clinical pregnancy. Furthermore, Bonferroni’s multiple comparison test was used to compare ectopic pregnancies and clinical pregnancy. Table 1 Characteristics of the five CSP cycles (Cases 1–5) 1 2 a 3 a 4 5 Maternal age, years at oocyst retrieval 43 41 41 35 31 at FET 43 41 42 35 39 Paternal age, years 36 47 47 35 41 Fertilization method IVF ICSI ICSI IVF ICSI BMI (kg/m 2 ) b 21.5 23.3 22.6 19.5 24.6 Number of embryo transfers single single single single single Causes of infertility male factor male factor male factor unexplained infertility male factor History of endometrial operation No No No No No Number of CS one time one time one time one time two times History of myomectomy Yes Yes Yes No No Condition of transferred embryo Day 6 blastocyst Day 5 blastocyst Day 6 blastocyst Day 5 blastocyst Day 6 blastocyst CSP Cesarean scar pregnancy, FET Frozen–thawed embryo transfer, IVF In vitro fertilization, ICSI Intracytoplasmic sperm injection, BMI Body mass index, CS Cesarean section a B and C are duplicate cases b BMI was measured during the FET cycle Characteristics of the five CSP cycles (Cases 1–5) CSP Cesarean scar pregnancy, FET Frozen–thawed embryo transfer, IVF In vitro fertilization, ICSI Intracytoplasmic sperm injection, BMI Body mass index, CS Cesarean section a B and C are duplicate cases b BMI was measured during the FET cycle Table 2 Characteristics and population of ectopic and clinical pregnancies in 27,528 FET cycles Ectopic pregnancy Clinical pregnancy a P –values b Number 44 11,800 Maternal age, year, median (IQR) at oocyst retrieval 35 (33–38) 36 (32–39) 0.705 at FET 37 (35–39) 37 (33–39) 0.301 Paternal age, year, median (IQR) 38 (34–41) 38 (34–41) 0.903 ICSI c (%) 45.4 56.3 0.099 BMI (kg/m 2 ) 20.5 (19.5–23.2) 20.4 (19.0–22.2) 0.211 Double embryo transfer d (%) 2.2 2.3 0.726 Causes of infertility  Male factor (%) 40.9 34.9 0.252  Fallopian tube (%) 31.8 13.6 0.002  Endometriosis (%) 13.6 12.4 0.475  Unexplained infertility (%) 27.2 47.7 0.005 History of endometrial operation: Yes (%) 2.2 5.9 0.254 History of cesarean section: Yes (%) 45.4 30.4 0.023 History of myomectomy: Yes (%) 9.0 3.2 0.057 Day 3 embryo transfer e : Yes (%) 0 0.6 0.767 Day 5 blastocyst transfer f : Yes (%) 77.2 83.0 0.203 Abbreviations : FET Frozen–thawed embryo transfer, IQR Interquartile range, ICSI Intracytoplasmic sperm injection, BMI Body mass index a Gestational sac was detected using ultrasound b P -values calculated between ectopic pregnancy and clinical pregnancy using chi–square and Kruskal–Wallis tests c No. of ICSI/total no. of FET d No. of double embryo transfer/total no. of FET e No. of day 3 embryo transfer/total no. of FET f No. of day 5 blastocyst transfer/no. of blastocyst transfer Characteristics and population of ectopic and clinical pregnancies in 27,528 FET cycles Abbreviations : FET Frozen–thawed embryo transfer, IQR Interquartile range, ICSI Intracytoplasmic sperm injection, BMI Body mass index a Gestational sac was detected using ultrasound b P -values calculated between ectopic pregnancy and clinical pregnancy using chi–square and Kruskal–Wallis tests c No. of ICSI/total no. of FET d No. of double embryo transfer/total no. of FET e No. of day 3 embryo transfer/total no. of FET f No. of day 5 blastocyst transfer/no. of blastocyst transfer Table 3 Effect of each characteristic on ectopic pregnancy and CSP in FET by logistic regression analysis Ectopic pregnancy CSP P –values OR 95% CI P –values OR 95% CI Maternal age, year, median  at oocyst retrieval 0.228 0.936  at FET 0.442 0.574 Paternal age, year, median 0.165 0.41 ICSI 0.222 0.530 BMI (kg/m 2 ) 0.315 0.441 Double embryo transfer 0.943 0.992 Causes of infertility  Male factor 0.906 0.933  Fallopian tube 0.001 2.823 1.491–5.342 0.982  Endometriosis 0.465 0.986  Unexplained infertility 0.119 0.997 History of endometrial operation: Yes 0.253 0.984 Number of cesarean section  1 0.005 2.25 1.230–4.117 0.086  ≥ 2 0.027 11.382 1.505–86.066 0.006 22.84 2.418–215.78 History of myomectomy: Yes 0.091 0.003 14.86 2.458–89.85 Day 3 embryo transfer 0.993 0.994 Day 5 blastocyst transfer 0.388 0.221 Abbreviations : FET Freeze-thawed embryo transfer, CSP Cesarean scar pregnancy, OR Odds ratio, 95% CI 95% confidence interval, ICSI Intracytoplasmic sperm injection, BMI Body mass index, NA Not applicable Effect of each characteristic on ectopic pregnancy and CSP in FET by logistic regression analysis Abbreviations : FET Freeze-thawed embryo transfer, CSP Cesarean scar pregnancy, OR Odds ratio, 95% CI 95% confidence interval, ICSI Intracytoplasmic sperm injection, BMI Body mass index, NA Not applicable Table 4 Comparison of serum hCG levels among 44 cases of ectopic pregnancy and normal pregnancy serum hCG (mIU/mL) Implanted location Number Median (IQR) P– values b Ectopic pregnancy Cervix 10 58.15(33.0–100.3) CSP 5 c 179.0 (150.8–294.1) P  < 0.001 d Fallopian tube 26 24.3 (8.8–49.7) Others a 3 9.1 (N/A) Clinical pregnancy 11,800 (42.8% e ) 236.3 (136.6–382.4) P  < 0.001 f P  = 0.607 g Abbreviations : hCG Human chorionic gonadotropin, FET Freeze-thawed embryo transfer, IQR Interquartile range, CSP Cesarean scar pregnancy, N/A Not applicable a Others include two peritoneum pregnancies and one ovarian pregnancy b P -values calculated using the Kruskal–Wallis test for hCG levels c Of the five CSP cases, one was a duplicate case, however, since the duplicate cases were not from the same cycle, these were included in the statistical analysis d Comparison of hCG level between CSP and all non–CSP ectopic pregnancies combined e 11,800 clinical pregnancies/27,528 FET cycles f Comparison of hCG level between clinical pregnancy and ectopic pregnancy g Comparison of hCG level between clinical pregnancy and CSP Comparison of serum hCG levels among 44 cases of ectopic pregnancy and normal pregnancy Abbreviations : hCG Human chorionic gonadotropin, FET Freeze-thawed embryo transfer, IQR Interquartile range, CSP Cesarean scar pregnancy, N/A Not applicable a Others include two peritoneum pregnancies and one ovarian pregnancy b P -values calculated using the Kruskal–Wallis test for hCG levels c Of the five CSP cases, one was a duplicate case, however, since the duplicate cases were not from the same cycle, these were included in the statistical analysis d Comparison of hCG level between CSP and all non–CSP ectopic pregnancies combined e 11,800 clinical pregnancies/27,528 FET cycles f Comparison of hCG level between clinical pregnancy and ectopic pregnancy g Comparison of hCG level between clinical pregnancy and CSP Statistical analysis was performed using SPSS version 26.0 software (SPSS, Inc., Chicago, IL, USA). The Kolmogorov–Smirnov test was used to evaluate the normal distribution. The results are expressed as median, interquartile range (IQR), 95% confidence interval (CI), and P -values. Statistical significance was set at P  < 0.05 for all tests.

Ectopic pregnancy in assisted reproductive technology (ART) occurs in 1.6–8.6% of pregnancies [ 1 ] and often requires emergency treatment [ 2 ]. A cesarean scar pregnancy (CSP), an abnormal pregnancy in which the embryo is implanted in the cesarean scar, is a relatively rare pregnancy that has been reported to occur in approximately 1.15% of patients with a history of cesarean section delivery [ 3 ]. As the frequency of cesarean sections increases [ 4 ], the incidence of CSPs may also increase in the future. CSP often requires emergency surgery owing to heavy bleeding, demanding early diagnosis. Han et al. reported that both magnetic resonance imaging (MRI) and transvaginal ultrasound (TVUS) can effectively diagnose CSP [ 5 ], recommending TVUS for the initial suspicion of CSP, followed by MRI for detailed examination. Therefore, early diagnosis of CSP may be challenging when TVUS findings are inconclusive. Although serum human chorionic gonadotropin (hCG) levels in ART are generally lower in ectopic pregnancies than in normal pregnancies [ 6 – 8 ], they may provide additional diagnostic information. Previous studies have reported prior pelvic inflammatory disease and smoking at the time of conception as risk factors for ectopic pregnancy [ 9 ]. Additionally, conflicting findings have been reported regarding developmental stage of transferred embryos (days 3, 5, and 6) and the risk of ectopic pregnancy [ 10 – 12 ]. Therefore, the influence of embryogenesis on ectopic pregnancy remains unclear. Higher body mass index has been reported as a risk factor for CSP [ 13 ], whereas Gull et al. found no such association [ 14 ]. In addition, they reported that smoking in the first trimester, higher parity, previous cesarean section, and preceding cesarean section were independently predictive of CSP [ 14 ]. Contrarily, Liu et al. reported that the number of abortions, time since the previous cesarean section, and number of cesarean sections were risk factors for CSP [ 15 ]. Thus, risk factors for CSP remain unclear. Herein, we aimed to identify the factors associated with CSP to facilitate the early detection of CSP by clarifying differences in serum hCG levels according to implantation site.

In this study, we demonstrated that serum hCG levels in early pregnancy were higher in CSP than in all non-CSP ectopic pregnancies combined. Furthermore, there was no significant difference in hCG levels between CSP and clinical pregnancy. hCG is produced by blastocysts before implantation and subsequently by syncytiotrophoblasts after implantation, becoming detectable in maternal blood 10 days after fertilization [ 16 ]. We hypothesize that the comparatively higher hCG levels observed in the CSP group than in other ectopic pregnancies may reflect the presence of endometrialized tissue within the cesarean section scar, allowing the embryo to develop sufficient syncytiotrophoblasts to secrete hCG. Contrarily, other ectopic implantation sites lack adequate structural and vascular support for normal trophoblastic development, resulting in lower hCG production. Furthermore, in this study, logistic regression analysis revealed that tubal infertility and history of cesarean section were risk factors for ectopic pregnancy. This finding is consistent with the fact that the fallopian tube is the most common implantation site in ectopic pregnancy. Furthermore, a prior cesarean section may affect implantation in the endometrium. Embryo transfer on days 3, 5, or 6 was not associated with the occurrence of ectopic pregnancy. Conversely, a history of myomectomy and a history of ≥ 2 cesarean sections were factors associated with CSP in the FET group, whereas endometrial surgery was not a factor associated with CSP. Consequently, in patients with prior cesarean sections, elevated hCG levels in early pregnancy should prompt careful evaluation for CSP, beginning with TVUS and supplemented by MRI when necessary. As aforementioned, risk factors for CSP include reduced endometrial thickness, multiple embryo transfers, smoking, higher parity, proportion of previous cesarean sections performed in tertiary hospitals, time since the previous cesarean section, number of cesarean sections, and a trial of labor before a previous cesarean Sects [ 13 – 15 ]. Gull et al. and Liu et al. investigated the risk factors associated with CSP in patients undergoing ART [ 14 , 15 ]. Mao et al. reported that, among 75,438 patients who underwent IVF, 64 were diagnosed with CSP in a single-center retrospective study that included both fresh and frozen-thawed embryo transfer [ 13 ]. The possibility that FET affects the onset of CSP cannot be ruled out; hence, we only considered FET in this study. Furthermore, these studies did not evaluate the incidence of CSP after myomectomy. Previous studies on CSP have focused on its examination, risk factor prognosis, treatment, and operation, whereas there are no reports on early diagnosis. Therefore, we investigated whether serum hCG levels could facilitate early diagnosis by comparing CSP and other ectopic pregnancies with normal pregnancies. Serum hCG levels are often used to predict pregnancy outcomes in ART. Low serum hCG levels raise suspicion for biochemical pregnancy loss or ectopic pregnancy because hCG levels are generally lower in ectopic pregnancies than in intrauterine pregnancies following ART. However, in our study, there was no significant difference in serum hCG levels between CSP and clinical pregnancy; hence, careful evaluation with TVUS and serial monitoring of hCG levels throughout early gestation are warranted. Based on these findings, for cases with tubal factors or cesarean section history, it is necessary to consider the possibility of ectopic pregnancy after embryo transfer and to explain to patients that they should seek early surveillance if they experience clinical symptoms such as abdominal pain or vaginal bleeding. Furthermore, we suggest that it is important to confirm clinical history and clinical symptoms, such as bleeding, especially in patients with a history of myomectomy or ≥ 2 cesarean sections. Accordingly, careful evaluation for CSP should be considered. To our knowledge, this is the first study to report differences in serum hCG levels between CSP and other ectopic pregnancies. However, this study has some limitations. First, it was a retrospective, single center study. Furthermore, the analysis of CSP was limited to only five cases, including one duplicate patient from a different treatment cycle. Consequently, the findings should be considered exploratory and limited to identifying potential associations. The very small number of CSP cases also increases the risk of unstable estimates and sparse-data bias. Therefore, further studies including larger numbers of CSP cases are necessary. Second, we investigated only FET cycles in this study; therefore future studies should evaluate fresh embryo transfer after accumulation of additional cases. Lastly, because serum hCG was measured between days 12 and 17 after ovulation or progesterone administration, variation in measurement timing may have influenced the measured hCG levels, as hCG levels increase with advancing gestational age during early pregnancy [ 17 ]. Despite these limitations, this study included a large sample size and was conducted at a single-center, minimizing variability arising from differences in clinical procedures and embryo culture methods.

Serum hCG levels in early pregnancy were higher in CSP than in those with other ectopic pregnancies, although hCG did not clearly distinguish CSP from clinical pregnancy. In patients with a history of cesarean section, a high early pregnancy hCG level without evidence of an intrauterine gestational sac on TVUS should prompt consideration of CSP during clinical assessment.