This was a retrospective single-center study conducted at Gifu Prefectural Tajimi Hospital (Tajimi, Gifu, Japan). The inclusion period was from January 2014 to December 2018. Initially, cases that met the International Classification of Diseases related to ectopic pregnancy (633.10, 633.11, 633.20, 633.80, 633.81, 633.90) were retrieved from the hospital electronic chart record. A manual check of all records was performed to identify cases of interstitial pregnancies. Among cases of interstitial pregnancies, only unruptured interstitial pregnancies diagnosed in the early first trimester were extracted, and cases with advanced gestation calculated from the last menstrual period at initial evaluation were excluded.

Detailed information about each patient, including demographics, gynecological and obstetrics history, and previous surgery, was retrieved from a review of clinical records and diagnostic images. Concerning the characteristics of interstitial pregnancies, dates of admission and discharge from the hospital, initial level of $\beta$-hCG, findings obtained by ultrasound and magnetic resonance imaging (MRI), presence of embryo and cardiac activity, mode of treatment, complications, post-treatment outcomes, and date of completion of treatment was examined. If the patient was treated surgically, then the surgical records were also evaluated.

Only patients that were observed for at least a year after the completion of treatment were included. Informed written consent was obtained from the patients regarding the use of their data for the study and publication of their disease status and clinical characteristics. Protective measures were taken to conceal the identity of patients. The Institutional Review Board of Gifu Prefectural Tajimi Hospital approved the current study (IRB number: 2020-OG-3).

The datasets generated and analyzed during the current study are not publicly available, as they contain information that could compromise patient confidentiality. However, they are available from the corresponding author [A.T.] upon reasonable request.

Women with suspected ectopic pregnancy were initially evaluated by transvaginal ultrasonography and measurements of the serum $\beta$-hCG level [5, 6]. Typical presentation of interstitial pregnancy could be diagnosed by ultrasonography according to the established criteria for diagnosis: an empty uterine cavity, gestational sac located at least 1 cm from the lateral aspect of the uterus and lateral to the endometrial cavity, and a thin layer (5 mm) of myometrium surrounding the gestational sac [15, 16].

When definitive diagnosis of interstitial pregnancy was difficult to determine but was suspected on initial ultrasonography, MRI was performed to locate the gestational sac. If necessary, three-dimensional computed tomographic angiography (3D-CTA) was performed to assess the degree of uteroplacental neovascularization [5, 6]. After the images of each case were cooperatively evaluated by gynecologists and diagnostic radiologists, a consensus was made on the diagnosis and strategy for individual management.

Criteria for MRI diagnosis of interstitial pregnancy were a thickened endometrium without signs of gestation in the uterine cavity and identification of a gestational sac-like structure within the muscular layer of the uterine cornus adjacent to the uninterrupted junctional zone [5, 6].

When the precise location of a gestational sac-like structure could not be determined even with the addition of MRI, exploratory single-port laparoscopy through an umbilical incision [17] was performed. If isthmic tubal pregnancy was confirmed, thus eliminating the possibility of interstitial pregnancy, laparoscopic excision of the affected fallopian tube was planned. If interstitial pregnancy was confirmed, either laparoscopic excision of gestational products was completed, or diagnostic laparoscopy followed by chemoembolization was performed based on the surgeon’s decision while considering the patients’ wish according to the informed consent. Depending on the site where gestational products were identified, interstitial pregnancy was further classified into proximal [6] or distal interstitial pregnancy [5].

If immediate uterine devascularization with cytotoxic drugs was deemed necessary to remove interstitial gestational products, urgent digital subtraction angiography was performed. After a vascular mass derived from interstitial gestational products was confirmed, the branch of the uterine artery supplying the interstitial pregnancy was selectively catheterized under fluoroscopic guidance [5, 6].

UACE was performed by infusing dactinomycin (0.5 mg/body; Cosmegen; Merck & Co., Inc., Whitehouse Station, NJ, USA) intra-arterially, followed by endovascular embolization using gelatin sponge particles (Serescue; Nippon Kayaku Co., Tokyo, Japan) as reversible embolization materials [5, 6].

After UACE, serial measurements of the serum $\beta$-hCG level were recorded every other day until 50% of the pre-treatment level was achieved, followed by weekly measurements until the serum $\beta$-hCG level decreased below the cut-off level (6 mIU/mL) [5, 6]. Nonlinear disappearance of serum hCG when plotted on a semi-logarithmic graph was judged as a significant delay in the decline of the hCG value, which was managed by the administration of systemic methotrexate (MTX) (50 mg/m${}^2$) [5, 6].

Of 125 women with ectopic pregnancy, eight consecutive cases (6.4%) were identified as interstitial pregnancies (Fig. 1). Gestational age was calculated based on the last documented menstrual period. Four cases with ruptured interstitial pregnancy, which were managed by laparoscopic surgery, were excluded from the current study.

Fig. 1.

Flow diagram of the classification of ectopic pregnancies experienced from 2014 to 2018. Of 125 ectopic pregnancies, unruptured interstitial pregnancy was diagnosed in the early first trimester of pregnancy in four cases. One case was managed by laparoscopic cornuostomy due to concerns about cornual rupture because of thinning of the myometrium (inset, arrow). Two cases were successfully managed solely by uterine artery chemoembolization (UACE), while another case required systemic methotrexate administration after UACE to achieve complete resolution.

Unruptured interstitial pregnancy was diagnosed in four women in the early first trimester of gestation. In one case with thinning of myometrial tissue detected under laparoscopic vision, laparoscopic cornuostomy was performed because of concerns about rupture, and this case was excluded (Fig. 1). The other three cases were initially managed by UACE and were included in the current investigation. The clinical course of each case is described below.

A 30-year-old, gravida 3, para 2 woman (Fig. 2) was referred at 6 weeks and 2 days of gestation because her physician suspected an ectopic pregnancy after she presented with amenorrhea following spontaneous conception.

Fig. 2.

Case 1: Proximal interstitial pregnancy managed by uterine artery chemoembolization (UACE) after diagnosis by magnetic resonance imaging (MRI). (A) Coronal T2-weighted MRI showing suspected gestational products with a diameter of 8 mm (arrow) in the proximal portion of the right interstitial uterine wall adjacent to the junctional zone. (B) Coronal three-dimensional computerized tomographic angiographic image in the arterial phase showing a prominent vascular mass (arrow) being fed by the right uterine artery and anastomosing left uterine artery in the right cornual region. (C) Digital subtraction angiography showing gestational products (arrow) supplied by the right uterine artery. (D) Changes in the serum $\beta$-hCG levels related to treatment. Nineteen days after UACE, the serum hCG value dropped to normal without any significant complications.

At the initial referral, the serum $\beta$-hCG level was 4047 IU/L (Fig. 2D). Coronal T2-weighted MRI (Fig. 2A) revealed a gestational mass with a diameter of 8 mm in the muscular layer of the right interstitium adjacent to the uninterrupted junctional zone. A fetal heartbeat (FHB) was not detected. 3D-CTA (Fig. 2B) revealed a prominent vascular mass (Fig. 2B, arrow) localized in the right interstitial portion with blood supplied from the right uterine artery and the anastomosing left uterine artery.

After MRI-based diagnosis of an unruptured proximal interstitial pregnancy of the right fallopian tube, UACE was performed for immediate devascularization of gestational products (Fig. 2C, arrow). The serum $\beta$-hCG level dropped immediately after UACE (Fig. 2D). Nineteen days after UACE, the serum $\beta$-hCG level decreased to a normal level, and the natural ovulatory cycle resumed without any complications. Subsequent gestation was not reported, with no plans to conceive another child.

A 30-year-old, gravida 2, para 0 woman (Fig. 3) was referred at 6 weeks and 5 days of gestation under suspicion of ectopic pregnancy after conception by ovulation induced with clomiphene citrate followed by hCG injection. At triage, the serum $\beta$-hCG value was 3723 IU/L (Fig. 3C).

Fig. 3.

Case 2: Distal interstitial pregnancy managed by uterine artery chemoembolization (UACE) after laparoscopic diagnosis, followed by a successful pregnancy outcome. (A) Coronal T2-weighted magnetic resonance imaging showing suspected gestational products with a diameter of 8 mm (arrow) in the left lateral angle of the uterine corpus, suggesting a radiological diagnosis of left isthmic pregnancy rather than interstitial pregnancy. (B) Single-port laparoscopic view showing an enlarged left interstitial portion (arrow) protruding from the uterine cornu median to uteroovarian ligament (arrowhead), indicating that gestational products exist in the distal portion of the interstitial fallopian tube. (C) Changes in the serum $\beta$-hCG levels related to treatment. After diagnosis by laparoscopy (L), uterine artery chemoembolization (UACE) was performed, followed by disappearance of the fetal heartbeat (FHB). Thirty-one days after UACE, the serum hCG value had dropped to normal without any significant complications. (D) Hysterosalpingography showed patency of the left fallopian tube (arrow) three months after UACE.

A gestational sac with a positive FHB in the left adnexal region adjacent to the uterus (data, not shown) was detected upon ultrasonographic examination. Coronal T2-weighted MRI revealed a gestational mass-like structure with a diameter of 8 mm (Fig. 3A, arrow) at the left lateral angle of the uterine corpus, which led to a radiological diagnosis of left isthmic pregnancy rather than interstitial pregnancy.

A single-port laparoscopic view revealed an enlarged left interstitial portion (Fig. 3B, arrow) medial to the utero-ovarian ligament (Fig. 3B, arrowhead) protruding from the interstitial portion, which indicated that gestational products exist in the distal interstitial portion of the fallopian tube.

Since early rupture was deemed improbable, UACE was performed for devascularization of gestational products. One day after UACE, there was a decrease in the serum $\beta$-hCG level, and the FHB could not be detected. Thirty-one days after UACE, the serum $\beta$-hCG level dropped below the normal range and no complications were reported. Three months after UACE, hysterosalpingography showed that patency of the left fallopian tube was restored (Fig. 3D, arrow).

Two years later, spontaneous conception occurred (side of ovulation unknown), followed by an uneventful vaginal birth at 38 weeks and 3 days of gestation resulting in a healthy, 2972 g female infant with Apgar scores of 9 and 10 at 1 and 5 minutes, respectively.

A 42-year-old, gravida 2, para 0 woman (Fig. 4) was referred at 6 weeks and 4 days of gestation under suspicion of right interstitial pregnancy with a positive FHB after conception by in vitro fertilization and embryo transfer.

Fig. 4.

Case 3: Distal interstitial pregnancy managed by uterine artery chemoembolization (UACE) after laparoscopic diagnosis, followed by systemic methotrexate administration. (A) Coronal T2-weighted magnetic resonance imaging showing suspected gestational products with a diameter of 7 mm (arrow) covered by a thin layer of myometrium in the right lateral angle of the uterine corpus. Radiological diagnosis indicated distal interstitial pregnancy, although isthmic tubal pregnancy could not be excluded. (B) A single-port laparoscopic view showing an enlarged right cornual portion indicated that gestational products exist in the distal portion of the interstitial fallopian tube. (C) Digital subtraction angiography showing gestational products (arrow) supplied by the right uterine artery. (D) Changes in serum $\beta$-hCG levels related to treatment. After a diagnosis by laparoscopy (L), UACE was performed, followed by disappearance of the fetal heartbeat (FHB). Although an immediate temporary decline in the serum $\beta$-hCG value was noted, re-elevation of the serum $\beta$-hCG value demanded an additional four rounds of cytotoxic therapy with methotrexate. Thereafter, the serum $\beta$-hCG level dropped below the normal range 51 days after UACE without any complications.

At the initial examination, the serum $\beta$-hCG value was 6244 IU/L (Fig. 4D). Coronal T2-weighted MRI (Fig. 4A) revealed a gestational mass-like structure with a diameter of 7 mm (Fig. 4A, arrow) covered with a thin layer of myometrium at the right lateral angle of the uterine corpus.

Radiological diagnosis strongly indicated a distal interstitial pregnancy, although isthmic tubal pregnancy could not be completely excluded. Thereafter, single-port laparoscopy revealed a swollen right interstitial portion (Fig. 4B, arrow), which confirmed the diagnosis of right distal interstitial pregnancy, and pelvic adhesion due to peritoneal endometriosis was identified.

Since immediate rupture was deemed improbable, UACE was performed for devascularization of gestational products (Fig. 4C, arrow). One day after UACE, there was an immediate decrease in the serum $\beta$-hCG level, and the FHB could not be detected. However, due to re-elevation of the serum $\beta$-hCG value, systemic MTX administration was performed four times, resulting in the serum $\beta$-hCG level dropping below the normal range at 51 days after UACE without any complications.

In this case, infertility therapy was resumed four months after UACE but was eventually terminated due to a complication of severe tubo-ovarian abscess at transvaginal oocyte retrieval.