Applied Radiology

Drs. Botash and Spirt are with the Department of Radiology, SUNY Upstate Medical University, Syracuse, NY.

Ectopic pregnancy is the leading cause of pregnancy-related death during the first trimester. A woman who has had one ectopic pregnancy is at increased risk for another, as well as for future infertility. Among women with prior ectopic pregnancies the subsequent overall conception rate is approximately 60%. Of those who conceive, only 80% are intrauterine while 20% are repeat ectopic gestations. ¹

The number of hospitalizations for ectopic pregnancy has declined since 1990. ² This may reflect a shift toward treating early, unruptured ectopic pregnancies in an outpatient setting. Outpatient management may include laparoscopic salpingectomy, salpingostomy, or methotrexate therapy. When reviewing the combined data for hospitalization and outpatient diagnosis and treatment, however, the incidence of ectopic pregnancy has increased steadily from approximately 18,000 cases in 1970 to more than 108,000 reported cases in 1992. ² This is a rate of approximately 20 ectopics per 1000 reported pregnancies.

Fortunately, during this same period, the fatality rate associated with this condition has decreased, despite the increasing incidence. Estimates of the death rate from ectopic pregnancy range from 0.5 to 1 death per 1000 ectopic pregnancies, representing a decrease of more than 86% since 1970. ² This is due to a combination of increased clinical awareness, the development of more sensitive and specific pregnancy tests, and improved ultrasound technology, particularly endovaginal scanning. While maternal mortality from complications of ectopic pregnancy has decreased in the last two decades, it still accounts for about 15% of all maternal deaths. ³

Risk factors for ectopic pregnancy

Most ectopic pregnancies occur in the fallopian tube, most commonly the ampullary portion of the tube, followed by the isthmus. Accordingly, the most common predisposing factor in patients with ectopic pregnancy is tubal scarring, usually associated with pelvic inflammatory disease. Up to 30% to 50% of women with ectopic pregnancies have a history of prior salpingitis. ⁴ Salpingitis isthmica nodosa, nodular thickening of the isthmic portion of the fallopian tubes with multiple diverticula, is seen in more than 50% of surgically excised tubal ectopics. ³ The sites of diverticula correlate well with the locations of the ectopic pregnancies.

Other risk factors for the development of ectopic pregnancy include abnormalities of the zygote, endocrine dysfunction that may affect ovum transport, and the presence of an intrauterine contraceptive device (IUD). ⁵ Some investigators believe that an IUD is far more effective in preventing implantation in the uterus than at preventing ectopic implantation. Others feel that the presence of an IUD may increase the incidence of ectopic pregnancy by affecting tubal motility and ovum transport, or by causing tubal infection. Infertile women who undergo in vitro fertilization are also at increased risk due to the high prevalence of tubal damage in these patients, the use of superovulation, and multiple embryo transfer. Ovulation induction also places a patient at increased risk due to the association with multiple pregnancies, including heterotopic implantation (coexistent intra- and extrauterine gestations).

Clinical considerations

The "classic" clinical triad for ectopic pregnancy consists of pain, abnormal vaginal bleeding, and an adnexal mass. However, this presentation is nonspecific and is present in fewer than 50% of patients with ectopic pregnancy. ⁶ Many patients with this presentation are not even pregnant.

As a result of the nonspecific clinical presentation and the varied physical findings, it is important for clinicians to have a high index of suspicion and recognize that any woman in her reproductive years is at risk for developing an ectopic pregnancy.

Pregnancy testing

The development of the radio-immunoassay (RIA) for the serum beta subunit of hCG has advanced the diagnosis and management of ectopic pregnancy. These pregnancy tests offer several advantages over the urine immunologic tests done previously. RIAs are specific for hCG and will not cross-react with other substances. They are also approximately 1000 times more sensitive than immunologic tests and can detect as little as 1 or 2 mIU/mL of hCG. ¹ Therefore, if the RIA is negative, the patient is not pregnant, and an ectopic pregnancy can be excluded. If the RIA is positive, serial quantitative levels of hCG can be measured to help determine whether or not a pregnancy is progressing normally. ¹

When interpreting quantitative levels of hCG, it is important to be aware of the reporting standard being used. The Second International Standard (2IS) was described in the 1960s. Later, a purer standard was introduced as the International Reference Preparation (IRP). HCG levels measured against the 2IS are approximately half those measured against the IRP. ¹ Both standards are currently in use, along with a third international standard that is numerically equivalent to the IRP.

Role of quantitative hCG determinations

In a patient who is clinically stable, it is useful to obtain a baseline hCG level at presentation. Failure to detect a gestational sac when the hCG value exceeds a discriminatory level places the patient at higher risk for an ectopic pregnancy. Using the 2IS, above 1800 mIU/mL an intrauterine gestational sac should be seen on transabdominal scans. ⁷ Transvaginal scanning has lowered the discriminatory level to 1000 mIU/mL (2IS). ⁸

If the hCG is below the discriminatory level and an intrauterine pregnancy (IUP) is not visible, the differential diagnosis includes an early normal or abnormal IUP, recent spontaneous abortion, or an ectopic pregnancy. Depending on the clinical status of the patient, serial quantitative hCG determinations may be obtained, or laparoscopy may be considered. Trophoblastic tissue produces B-hCG 8 days after conception, with a normal doubling time of approximately 48 hours. ⁹ If the hCG levels do not progress normally, demonstrating a 48-hour doubling, an abnormal IUP or ectopic pregnancy will be present. In the setting of a declining hCG level, a recent abortion or nonviable ectopic pregnancy are likely considerations.

Ultrasonography of ectopic pregnancy

In the stable patient with a positive pregnancy test, an ultrasound study should be performed for further evaluation. The initial sonographic examination should be performed transabdominally; if an intrauterine gestation is identified, the diagnosis of ectopic pregnancy is virtually excluded (see below). If transabdominal imaging does not confidently identify an intrauterine gestational sac, a transvaginal scan should be done. Transvaginal sonography can detect a normal intrauterine gestational sac by a menstrual age of 4 to 5 weeks, approximately 1 week sooner than is possible with the transabdominal probe. For patients with an IUP, transvaginal sonography often can identify the source of symptoms, such as a large and/or hemorrhagic corpus luteum cyst. Transabdominal ultrasound may detect an ectopic pregnancy that is located too high to be visualized with the transvaginal probe (figure 1).

The frequency of a coexistent ectopic pregnancy and an IUP (heterotopic pregnancy) in a patient without risk factors is approximately 1 in 7000. ⁹ However, the number of heterotopic pregnancies has been increasing worldwide, in part because a larger number of women are being treated for infertility. Women undergoing assisted reproduction may have a risk as high as 1 in 100. ¹

Ultrasound diagnosis of an early intrauterine pregnancy

Stimulation of the uterine lining with hormones produced by an ectopic pregnancy can result in endometrial changes (decidual reaction) similar in appearance to those seen with an early IUP (figure 2). Intrauterine fluid surrounded by decidual reaction in a patient with an ectopic pregnancy can simulate a gestational sac at transabdominal sonography ("pseudosac"). ¹⁰ The presence of the "double sac sign" is useful to confirm the diagnosis of an IUP; it discriminates a true gestational sac from a pseudogestational sac of ectopic pregnancy before a fetal pole or yolk sac can be recognized. The double sac appearance is caused by two concentric echogenic borders; the echogenic border of the endometrial cavity and the echogenic villi that surround the gestational sac (figure 3). With the advent of endovaginal sonography, reliance on the double sac sign has decreased.

When the mean sac diameter of the gestational sac reaches approximately 8 to 10 mm, the yolk sac becomes visible consistently by endovaginal scanning. At a sac diameter of 16 mm, an embryo should be seen. The fetal heartbeat is usually visible when the embryo reaches a crown-rump length of 5 mm. ¹

Color and spectral Doppler imaging have also been used to evaluate early pregnancies. A normally developing IUP will demonstrate low impedance peritrophoblastic arterial signal with minimal or absent venous flow. An abnormal IUP also demonstrates peri-trophoblastic arterial flow, but may also demonstrate increased venous flow. Generally, pseudogestational sacs are not associated with either arterial or venous flow. ¹¹

Adnexal findings

The most specific adnexal finding for diagnosing an ectopic pregnancy is visualization of a living extrauterine embryo (figure 4). In a patient without a demonstrable IUP, characterization of the adnexa improves the ability of sonography to predict the presence of an ectopic gestation. The ovary should be identified and used as an anatomic landmark in evaluating for ectopic pregnancy. In most patients, the ovary is located near the ampullary portion of the fallopian tube, which is the most frequent site of ectopic implantation. Despite the low likelihood of finding a heterotopic pregnancy, the adnexa should be carefully examined for masses even in the presence of a documented intrauterine gestation. Although most ectopics are located between the ovary and the uterus, they may implant anywhere in the pelvis (figure 5). Therefore, careful examination of the regions adjacent to the uterine fundus, the cul-de-sac, and the lateral margins of the pelvis is also necessary.

Although an adnexal mass is present in the majority of patients with ectopic pregnancy, it is nonspecific and may have a wide variety of sonographic appearances. An extraovarian adnexal or tubal ring may be visible and is characterized by an anechoic center and an echogenic periphery. ⁹ An adnexal ring, however, is not specific and can be simulated by an ovarian cyst. In some patients, the ring may contain a fetal pole or yolk sac allowing a confident diagnosis of ectopic pregnancy (figure 4A). Hematosalpinx may appear as a heterogeneous or homogeneous solid mass in the adnexal area (figure 6); a gestational sac may not be visualized, particularly if the fallopian tube has ruptured.

Complex intraovarian lesions are unlikely to represent an ectopic pregnancy as the incidence of ovarian pregnancies is less than 1%. ⁹ Similarly, simple ovarian cysts are of low suspicion; these usually represent the corpus luteum.

Color and spectral Doppler evaluation of adnexal masses may aid in the sonographic diagnosis of ectopic pregnancy. As with an intrauterine pregnancy, ectopic tissue is characterized by increased color flow that displays a low-resistance, high-velocity peritrophoblastic flow pattern. ¹² However, it is generally recognized that other pelvic pathology including pelvic inflammatory disease, pedunculated fibroids, or even an exophytic corpus luteum cyst may demonstrate a similar flow pattern. Nonetheless, the use of color Doppler may attract one's attention to a small mass which may otherwise have been overlooked.

Cul-de-sac findings

Despite meticulous scanning, no adnexal mass will be identified in up to one third of patients with ectopic pregnancy. ⁹ In these patients, the presence of free intraperitoneal fluid may suggest the diagnosis. Free fluid is the least specific finding associated with ectopic pregnancy, however. Transvaginal sonography is more sensitive than transabdominal scanning for detecting small amounts of free fluid. Fluid in the cul-de-sac may result from active bleeding from the fimbriated end of the fallopian tube, tubal rupture, or tubal abortion. ⁹ While hemoperitoneum may appear completely anechoic, echogenic or complex cul-de-sac fluid suggests that diagnosis (figure 7). An amorphous solid mass in the cul-de-sac may represent clotted blood.

Unusual forms of ectopic pregnancy

Interstitial pregnancies --Implantation of a pregnancy within the interstitial, or intramural, portion of the fallopian tube is uncommon, accounting for only 2% to 4% of ectopic pregnancies. ⁹ Since the myometrium surrounds only a portion of the expanding gestational sac and a blood supply develops, the sac can enlarge painlessly for a relatively long period of time until uterine rupture results in massive hemorrhage. As a result, the morbidity and mortality of interstitial pregnancies is higher than for tubal ectopics.

An interstitial ectopic pregnancy may be suggested on ultrasound examination if there is an eccentric location of the gestational sac, thinning or absence of myometrium surrounding the sac, and separation of the sac from the endometrial cavity (figure 8). In an attempt to improve the diagnostic ability of endovaginal sonography, the interstitial line sign has been proposed. ¹² This sign describes a thin echogenic line that extends from the central uterine cavity to the periphery of the interstitial gestational sac. This line reportedly represents the interstitial portion of the fallopian tube or the endometrial canal in patients with large interstitial pregnancies. These signs in combination may be useful in the early diagnosis of interstitial ectopic pregnancy.

Cervical pregnancies --Cervical pregnancies are uncommon and account for less than 1% of all ectopic pregnancies. ⁹ Predisposing conditions in the genesis of cervical ectopics include in vitro fertilization, prior uterine curettage, fibroids, an indwelling IUD, and previously treated Asherman's syndrome. ⁹ The diagnosis can be suggested when a well-formed gestational sac that contains a yolk sac, embryo, or cardiac activity is identified within the cervix. In the past, patients with cervical ectopic pregnancies were managed with hysterectomy because of the risk of uncontrollable hemorrhage. Now with the earlier diagnosis of cervical pregnancy afforded by endovaginal sonography, treatment with local potassium chloride injection, systemic methotrexate treatment, or preoperative uterine artery embolization before dilation and evacuation can be used to preserve future fertility. ⁹

Conclusion

Ectopic pregnancy can present a difficult diagnostic dilemma. It is important for the referring clinician to maintain a high index of suspicion in evaluating women of reproductive age. Sensitive pregnancy tests that are specific for the beta subunit of hCG will distinguish the subset of patients who may harbor an ectopic from those that are not pregnant. The use of high-resolution endovaginal sonography allows for the earlier detection of an intrauterine gestation, which permits exclusion of ectopic pregnancy with a high degree of certainty. In the absence of an IUP, identification of a mass in the adnexa and/or complex fluid in the cul-de-sac increases the probability that an ectopic pregnancy is present. Serial quantitative hCG analysis and follow up sonography can assist the clinician in managing the stable patient. AR

References

1. Filly RA: Ectopic pregnancy. In: Callen PW, (ed): Ultrasonography in Obstetrics and Gynecology, ed 3, pp 641-659. Philadelphia, Saunders, 1994.

2. Centers for Disease Control: Ectopic pregnancy--United States, 1990-1992. MMWR 44:46-48, 1995.

3. Bohm-Velez M, Mendelson EB, Freimanis MG: Transvaginal sonography in evaluating ectopic pregnancy. Semin Ultrasound CT/MR 11:44-58, 1990.

4. Berry SM, Coulaim CB, Hill LM, et al: Evidence for contralateral ovulation in ectopic pregnancy. J Ultrasound Med 4:293-295, 1976.

5. Schwartz RO, Di Pietro DL: B-hCG as a diagnostic aid for suspected ectopic pregnancy. Obstet Gynecol 56:197-203, 1980.

6. Spirt BA, Berlow ME, Gordon LP, Weil LR: Ectopic pregnancy: Sonographic-pathologic correlations. RadioGraphics 4:821-848, 1984.

7. Nyberg DA, Filly RA, Mahony BS, et al: Early gestation: Correlation of HCG levels and sonographic identification. AJR 144:951-954, 1985.

8. Nyberg DA, Mack LA, Laing FC, Jeffrey RB: Early pregnancy complications: Endovaginal sonographic findings correlated with human chorionic gonadotropin levels. Radiology 167:619-622, 1988.

9. Frates MC, Laing FC: Sonographic evaluation of ectopic pregnancy: An update. AJR 165:251-259, 1995.

10. Nyberg DA, Laing FC, Filly RA, et al: Ultrasonographic differentiation of the gestational sac of early intrauterine pregnancy from the pseudogestational sac of ectopic pregnancy. Radiology 146:755-759, 1983.

11. Emerson DS, Cartier MS, Alticri LA, et al: Diagnostic efficacy of endovaginal color Doppler flow imaging in an ectopic pregnancy screening program. Radiology 183:413-420, 1992.

12. Ackerman TE, Levi CS, Dashefsky SM, et al: Interstitial line: Sonographic finding in interstitial (cornual) ectopic pregnancy. Radiology 189:83-87, 1993.