Recent improvements in the sonographic depiction of uterine and adnexal structures with transvaginal sonography (TVS) in 2D, 3D, and color Doppler ultrasound and refinements and better understanding of quantitative human chorionic gonadotropin (hCG) have enhanced the ability to identify whether an early pregnancy is intra- or extrauterine. In the clinical setting, as many as 30% of women present with pain and/or bleeding in early pregnancy. These two findings raise the suspicion for early pregnancy failure (EPF). Up to 25% of recognized pregnancies end in miscarriage, and 1% to 2% of these will ultimately be diagnosed as ectopic gestations. Although the sonographic findings in early pregnancy can be subtle, a diagnosis of EPF and ectopic pregnancy is possible in most cases when sonographic findings are combined with results of serial hCG assays. In the case of ectopic pregnancy, early diagnosis affords the patient more conservative medical or surgical options and decreases the risk for rupture, which often leads to more radical, emergent surgery.

Despite advances in the diagnostic tests used in early pregnancy, ectopic pregnancy accounts for an estimated 6% of all pregnancy-related deaths from 1991 to 1999 in the United States.¹ A high index of suspicion is required when evaluating women who present with lower abdominal pain and amenorrhea, as reflected by the fact that the diagnosis is overlooked by the initial examining physician in up to 70% of cases.² Expeditious and accurate diagnosis of patients who are suspected of having ectopic pregnancy is important so that timely intervention and proper management can be instituted. If it is recognized early, before tubal rupture, it may be possible to surgically remove the gestational sac by linear salpingostomy or treat with methotrexate therapy, thereby preserving the tube and future chances of achieving pregnancy. Advanced ectopic pregnancies can result in significant damage to tubal architecture, which often leads to salpingectomy. If the remaining tube is compromised, fertility can be significantly decreased as a result. Once a patient has had an ectopic pregnancy, there is a significant chance (about 1 in 3) of recurrence in a future pregnancy.³

Early diagnosis is also important for patients undergoing medical treatment of ectopic pregnancy. The success of medical treatment appears to be greatest with lower β-hCG levels (≤5000 mIU/mL) and absence of embryonic/fetal heart motion. However neither of these parameters is an absolute contraindication to medical (methotrexate) therapy.⁴ The β-hCG value and presence of embryonic/fetal heart motion corresponds to the level of trophoplastic activity of the ectopic gestation. Transvaginal color Doppler sonography (TV-CDS) might have an important role in determining which type of treatment (medical, local methotrexate, or KCl injections) is most appropriate based on the relative vascularity of the choriodecidua within the tube and the presence or absence of embryonic heart motion.

The main role of TVS is to help identify whether the location of a gestation is intra- or extraovarian. The use of TVS has greatly enhanced the sonographic evaluation of patients with suspected ectopic pregnancy over that possible with transabdominal sonography (TAS). Specifically, the presence or absence of an intrauterine gestation can be documented approximately 1 week earlier with TVS than with TAS. In addition, adnexal masses created by ectopic pregnancies can be more frequently detected by TVS.

The additional use of TV-CDS seems to further enhance detection of ectopic pregnancies that might not be apparent on TVS.^5,6 Viable trophoblastic tissue typically produces a vascular ring within the tube, commonly referred to as the “ring of fire” sign, that can be recognized by TV-CDS. This ring can be seen as being separate from the ovary but may be similar to that of a functioning corpus luteum. Typically, the vascular ring of an ectopic is discontinuous versus the concentric ring of a functioning corpus luteum. TV-CDS can also help distinguish a corpus luteum from an ectopic by identifying the location of the vascular ring, which is within the confines of the ovary for a corpus luteum and extraovarian for an ectopic. Not infrequently, an adnexal structure representing the ectopic can be seen in proximity to a corpus luteum. The obstetrical applications of CDS are discussed further in Chapter 12.

Transvaginal CDS may have an important role in determining which type of treatment (medical or surgical) that is most appropriate based on the relative vascularity of the choriodecidua within the tube and the presence or absence of embryonic heart motion.⁷ In fact, TV-CDS may have a role determining which patients should receive methotrexate versus those that might be observed, since the relative amount of trophoblastic activity can be correlated to vascularity. Our recent research shows no significant correlation between the pretreatment vascularity of ectopics relative and their responsiveness to medical treatment. Interestingly, those ectopic pregnancies found to be responsive demonstrated an increase in vascularity and size on serial examinations.⁸

The reader is encouraged to refer to Chapter 3 for further discussion of the concept of “pregnancy of unknown location” relative to the sonographic evaluation of the viability of early intrauterine pregnancies (IUPs).

The incidence of ectopic pregnancy in the United States is 1% to 2%. Although in the 1970s and 1980s epidemiologic studies demonstrated an increasing incidence of ectopic pregnancy suspected to be a combination of a true increase in prevalence as a result of partially treated salpingitis as well as more frequent diagnosis made possible by improved diagnostic tools, newer data suggests that the rate of ectopic pregnancy in the United States has remained stable over recent years.^3,9,10 At what was debatably the peak of improvement in diagnosis and treatment of ectopic pregnancy from 1970 to 1978, the number of ectopic pregnancies diagnosed nationwide increased by more than double. The death rate, however, decreased by 75% during this period. This is a reflection of an increase in suspicion of ectopic pregnancies by patients and care providers and an improvement in the ability to diagnose this entity in its earliest stages.

In more recent years, the incidence of ectopic pregnancy has remained stable while the treatment approach has changed significantly. Between 2002 and 2007, methotrexate treatment has increased from 11.1% to 35.1% of ectopic pregnancies while the laparoscopic approach has declined from 40% to 33.1%.³ The incidence of ectopic pregnancies is greatest in patients with a history of salpingitis, tubal surgery, ectopic pregnancy, use of assisted reproductive therapy (ART), or pelvic inflammatory disease.^11-13

Notably, there is also a substantial difference in incidence between socioeconomic levels. Incidence of ectopic pregnancy amongst women with commercial insurance in the United States from 2002 to 2007 is estimated to be 0.64% whereas in the United States Medicaid population, this rate was estimated to be 1.4%.¹⁴

The term ectopic pregnancy refers to an implantation of the conceptus outside the uterine (endometrial) cavity. Ninety-five percent of ectopic pregnancies are tubal, and the majority of these occur in the ampullary or isthmic portions of the tube. The remaining 5% of nontubal ectopic gestations occur in the ovary, cervix, C-section hysterotomy site, abdomen, and in the peritoneal and retroperitoneal spaces.

In ampullary ectopic tubal pregnancies, the conceptus implants beneath the epithelium of the fallopian tube to form a fluid-filled gestational sac, which is lined with trophoblastic tissue, in the wall of the tube. Because the fallopian tube has only two thin layers of muscle, the trophoblastic cells that burrow deep into the tubal wall distend it and can eventually cause it to rupture. The gestational sac within the tube of a ruptured ectopic pregnancy is usually surrounded by fluid or blood due to erosion of adjacent vessels. In the vast majority of cases, the separation of the decidua from the wall of the tube causes death of the embryo. It has been theorized that in rare cases of tubal “abortion” out the fimbriated end of the tube, the embryo may survive by reimplantation within the abdomen and reestablishment of the blood supply from the omentum or the mesentery.

Mild uterine enlargement and decidualization of the endometrium, a process in which the endometrial cells enlarge their nuclei relative to cytoplasm, are usually present with an ectopic pregnancy and can occasionally be detected clinically and sonographically. In the absence of a heterotopic pregnancy, if dilatation and curettage (D&C) is performed on a patient with an ectopic pregnancy, only decidua without chorionic villi will be obtained.

Increasingly reported is the “scar ectopic” or implantation that occurs in a previous C-section scar, forming an “incisional” ectopic pregnancy. It has been found that these pregnancies are at risk to develop abnormally invasive placentation.¹⁵

Another possible etiology of recurrent ectopic pregnancies is the transperitoneal migration of sperm or of the fertilized egg into the contralateral tube. Possibly, this would result in delayed and faulty implantation of the trophoblasts into the tubal wall. As the relative contributions of those factors to the development of ectopic pregnancy are better understood, measures that can prevent ectopic pregnancy may be determined.¹⁶

Proposed explanations for development of ectopic pregnancy include delayed fertilization or delayed transit of the fertilized zygote secondary to fallopian tube malfunction; ovulation from the contralateral ovary with delayed passage of the zygote through the tube; obstruction of zygote passage secondary to intratubal adhesions from pelvic inflammatory disease; endometriosis, surgery, and other abdomino-pelvic infection; and altered tubal motility as a result of progestin contraception and abnormal angulation of the tube relative to the uterine cornu.¹⁷ The true contribution of these etiologies is unclear.

The most common presenting symptoms of ectopic pregnancy are pelvic pain, which may be mild and intermittent or persistent and severe, and abnormal vaginal bleeding.¹⁸ Data demonstrate that among women who present with first trimester vaginal bleeding and/or abdominal pain, up to 18% will be diagnosed with an ectopic pregnancy.⁴ These symptoms are, however, very common in early pregnancy and nonspecific in making the diagnosis of an ectopic pregnancy. It often requires serial hCG assays and ultrasound studies to make a definitive diagnosis.

The clinical course of an ectopic pregnancy is related to its site of implantation.¹⁹ The ampullary portion of the tube is the most common location for ectopic implantation. As in other sites, the ectopic pregnancy can expand until it ruptures the wall of the tube. Complete or partial tubal abortion may also occur, with the contents of the sac extruded through the fimbriated end of the tube into the peritoneal cavity. If the fimbriated end of the tube is occluded, hematosalpinx may result. Ectopic pregnancies that occur in the narrow isthmic portion of the tube usually distend it eccentrically and, because of the tube’s small diameter, rupture early in the pregnancy.

Ectopic pregnancy in the interstitial portion of the tube is uncommon (3% to 4% of all ectopic pregnancies), but, like the cornual ectopic, has the highest likelihood of serious potential complications. Because of its location within the muscular portion of the uterus, the cornual pregnancy can survive until 3 to 4 months gestation. Massive bleeding from the uterine arteries and veins can then result because of the proximity to these major vessels.

Chronic ectopic pregnancies may occur, resulting in hematoma formation in the cul-de-sac.²⁰ Such patients usually present with recurrent, intermittent low-grade fever associated with a palpable solid mass. On physical examination, there is usually a firm pelvic mass located in the midline and difficult to separate from the uterus. Culdocentesis may be negative because the blood in the cul-de-sac is clotted. In very rare cases, the embryo and products of conception will undergo dehydration in situ with the formation of a lithopedion pregnancy.

Other rare sites of implantation include intra-abdominal, ovarian, cervical, and extraperitoneal. True advanced abdominal ectopic pregnancies may be difficult to differentiate from normal IUPs the uterus must be defined separately from the amniotic sac and its contents.²¹ Abdominal pregnancies are postulated by some to be the result of reimplantation of an aborted fetus after it passes out the fimbriated end of the tube and reimplants on the mesentery or omentum.²¹ These pregnancies rarely progress to term without symptoms and may first present because of difficulty during the initial stages of labor, growth problems, pain, or absence of labor. Extraperitoneal ectopic pregnancies are quite rare and are probably the result of tubal rupture, with expulsion of the fetus between the leaves of the broad ligament. The rupture occurs between the fimbriated end of the tube (where it is not covered by peritoneum) and the site where the 2 folds of the broad ligament are loosely opposed.²¹ The tubal contents may empty into the soft tissue and mesosalpinx and thereafter remain in that region.

As previously mentioned, another rare type of ectopic involves implantation of the conceptus within a previous C-section hysterotomy site. These can be diagnosed by their unusual position within the uterus at the lower uterine segment, and often have an asymmetric shape. However, these ectopics are often misdiagnosed as cervical or aborting pregnancies.²²

To properly evaluate the vast majority of patients in whom an early ectopic pregnancy is suspected, it is helpful to correlate the sonographic findings with the results of a quantitative serum pregnancy test. In addition, it is important for the sonographer and sonologist to know the type of assay used and its relative sensitivity.

The enzyme-linked immunoassays that detect urine hCG are nonquantitative but very sensitive and may easily be performed in an office or clinic. These tests are useful for determining the presence or absence of a pregnancy and are routinely positive when the serum hCG level is at least 50 mIU/mL (8 to 10 days postconception). They are positive in about 99% of patients with a symptomatic ectopic pregnancy. The enzyme-linked immunoabsorbant assays (ELISAs) detecting the β subunit of the serum hCG molecule are quantitative and most helpful in cases where a problem arises during an early pregnancy, such as suspected ectopic pregnancy or threatened abortion.

All commercially available kits that measure serum hCG now use nonradioactive technology, and the older radioimmunoassays (RIAs) have been replaced. In addition, the earlier confusion over different “international standards” has been resolved, and all kits now use the same standard. It is important, where possible, to use the same assay when following serial tests as there can be significant variation among assays performed in different labs.

The ability to quantitate serum levels of hCG allows the clinician to grossly approximate the gestational age of the pregnancy, assuming it is normal. The hCG level can then be correlated with the sonographic findings in looking for certain developmental “milestones” (Table 4-1). Correlating the serum hCG level with the sonographic findings enables the clinician to evaluate the normalcy of the pregnancy in question. However, there appears to be significant overlap in these values, and they are similarly limited in multiple gestations.

The sonographic “milestones” that are helpful and more commonly used include the delineation of a “chorionic sac” at the fifth week, detection of a yolk sac within the gestational sac of approximately 1 cm at 5 to 6 weeks, and an embryo within a sac of approximately 1.5 mm at 6 weeks.

A discriminatory zone for the level of serum hCG has been described for discerning an IUP from an ectopic or nonviable IUP by means of TVS. This varies somewhat depending on the sonographer’s expertise and equipment used. In addition, there seems to be a wide range of hCG values in normally developing IUPs, making absolute values not as predictive as has been previously reported.²³ With an experienced sonographer, this level is between 1500 and 2000 mIU/mL. This means a viable singleton IUP should be visible sonographically by the time the serum hCG is 1500 to 2000 mIU/mL or more. An important exception to this is multiple gestation. The absence of an intrauterine gestational sac when the serum level is above the discriminatory zone is highly suspicious for an ectopic pregnancy or for a nonviable IUP. Even when an adnexal mass is visualized, this may simply be a self-contained hemorrhagic corpus luteum cyst. Measuring the thickness of the endometrial stripe is also of limited diagnostic value. Some have observed that normal, early IUPs tend to be associated with a thickened (>6 mm) endometrium, whereas the stripe is thinner with an ectopic pregnancy or spontaneous abortion.¹⁸ When ambiguity persists, serial hCG determinations should be drawn to look for a normal or abnormal progression, and the sonogram should be repeated once the level has risen above the discriminatory zone.

It must be emphasized, however, that these criteria represent guidelines, not absolute end points.²⁴ It is possible for a viable IUP to demonstrate a low hCG level and/or slow progression. Conversely, a normal rise in the hCG level may sometimes be associated with an ectopic pregnancy.²⁴ Also, a multiple gestation or a heterotopic pregnancy may show an uncharacteristically elevated hCG level for any given gestational age.

The amount of hCG produced by an ectopic pregnancy is generally less than that by a viable IUP of the same gestational age,²⁵ which may be due to an unfavorable location for trophoblast proliferation or a difference in the implantation process. This fact is useful, however, only if the date of conception is known. It should be pointed out that the majority of these patients presented with the serum hCG level below the discriminatory zone, as defined at the time when the study was conducted.

The level of serum hCG tends to be roughly proportional to the size of a tubal pregnancy and the extent of trophoblastic differentiation. A ruptured tubal pregnancy tends to be associated with a higher level than one that has not ruptured. The range of serum hCG levels for any given situation, however, is so broad that this observation has little clinical relevance.

Sonographic visualization of an intrauterine sac when the serum hCG is below the discriminatory zone may signify an abnormal gestation (Figure 4-1). A nonviable IUP may appear this way. Also, there is the “pseudogestational sac,” which is sometimes associated with an ectopic pregnancy. A pseudogestational sac lacks the “double-sac” sign and is smaller and more irregular than a true gestational sac at a comparable gestational age. The reader is encouraged to realize the limitations of the use of the term “double-sac,” since it can be misleading.

Serial determinations of the serum hCG level have proven useful in the clinically stable patient when ambiguity persists after a nondiagnostic TVS findings have been correlated with a single quantitative hCG. The first known value is placed on the standard line, and subsequent values are plotted accordingly. It is apparent that most patients showed a plateau or fall in the level during the period of preoperative evaluation. This plateau or fall is diagnostic of a nonviable pregnancy when it occurs at levels below 3000 mIU/mL during at least a 48-hour period. It does not, however, distinguish between a nonviable intrauterine and an ectopic pregnancy. It is also apparent that some ectopic pregnancies may show an initial “normal” rise in the level of hCG. This normal rise, however, is usually short-lived, and an abnormal progression often develops.

Serial hCG determinations are also essential after treatment of an ectopic pregnancy by either medical or surgical means. A plateau or rise in the level may be the first indication of a persistent ectopic pregnancy indicating the need for further treatment.²⁶ Furthermore, a negative hCG may signify a resolution of the ectopic pregnancy, which may precede normalization of any sonographic findings.

An important term used to designate nondiagnostic TVS findings which represent a “pregnancy of unknown location” in those patients in whom an intrauterine or extrauterine pregnancy cannot be diagnosed. A repeat TVS in 2 to 5 days and a repeat hCG after 48 hours is often recommended to determine the exact nature of the pregnancy. In normal IUPs, hCG increases by an average of 66% in 48 hours. Not all IUPs have this expected increase, and up to 15% of normal IUPs will have less than the expected rise.²⁶ Recent data, however, demonstrate that an increase in hCG of less than 53% in 48 hours confirms an abnormal pregnancy with 99% sensitivity.²³

The use of TVS has greatly enhanced the accuracy of sonographic evaluation of patients with suspected ectopic pregnancy over that possible with TAS.^28,29 In particular, the presence or absence of an intrauterine gestation can be documented or excluded at an earlier stage (approximately 1 week) than with TAS. Most importantly, transvaginal transducer/probes typically allow accurate and definitive inclusion or exclusion of an IUP by demonstration of an intrauterine gestational sac. Transvaginal sonography can also be used to demonstrate an extrauterine gestational sac, corpus luteum, or both. TAS can be used to evaluate these parameters but is, in general, less accurate or definitive. In addition, adnexal masses and/or typical collections of intraperitoneal fluid created by ectopic pregnancies can more frequently be detected and identified by TVS. The use of TVS with highly sensitive pregnancy tests has markedly enhanced the ability to detect ectopic pregnancies over techniques and tests available in the recent past. A very high degree of accuracy now exists in the ability to establish the presence or exclude the possibility of an ectopic pregnancy.^28,30 (Please refer to the discussion in Chapter 3 for a more detailed description of normal and abnormal first trimester pregnancies.)

As mentioned previously, TV-CDS may be a useful adjunct to TVS in that the “vascular ring” of the ectopic pregnancy can be visualized. Functioning corpora lutea may also have this appearance on TV-CDS and tend to have a more complete rim than that seen in ectopics. In most cases, the vascular ring of an ectopic pregnancy is sparser (discontinuous) than that of a functioning corpus luteum. Additionally, most nonviable ectopic pregnancies may not demonstrate flow.

With methotrexate treatment, most ectopic pregnancies demonstrate increased vascularity as defined by the number of colorized pixel elements in the tubal ring. However, a recent case series showed no significant difference in the TV-CDS in responsive or nonresponsive ectopics. An increase in flow may reflect vasodilatation that occurs with effective treatment. CDS can be used to delineate the boundary of the extraovarian mass of the ectopic pregnancy from the ovary itself. Three-dimensional (3D) sonography may provide an additional means to depict the anatomic proximity of an ectopic pregnancy to the ovary.

On both TAS and TVS, sonographic examinations should begin by delineation of the uterus in its long axis. One should carefully evaluate the endometrium for the presence or absence of a gestational sac or decidual thickening. Once the uterus is adequately evaluated, the adnexal region should be carefully examined. If possible, both ovaries should be identified because many ectopic pregnancies are associated with coexisting corpus luteum. On a transverse transvaginal scan, the relative position of the proximal segment of tube can be approximated by recognition of several anatomic landmarks. These include delineation of the round ligament as it courses directly anterior to the tube near the uterine fundus, and the location of the interstitial portion of the tube by its proximity to the endometrium, which invaginates into the uterine cornu on a transverse scan in the region of the tubal ostia. If color Doppler is used, the pulse repetition frequency should be low to maximize detection of slow flow. Acquisition of multiplanar images is needed for 3D reconstruction. The newer acquisition of 3D images can best be shown in the semicoronal plane so as to depict the area of the tube relative to the ovary.

The sonographic findings that are encountered in a patient with ectopic pregnancy differ according to the developmental stage of pregnancy in which the patient is examined, and to some degree whether or not rupture has occurred. One should be aware that the sonographic findings also depend on what type of transducer/probe is used. The following discussion is organized into uterine, adnexal, and peritoneal sonographic findings, even though multiple findings are usually present.

Uterus

In most ectopic pregnancies, the uterus contains thickened endometrium layer(s) due to its thickening related to decidualization (Figure 4-2). Particularly with TVS, the increased fluid content of the decidualized endometrium can sometimes be appreciated due to enhanced through transmission distal to this layer. In more advanced ectopic pregnancies, fluid or blood may be present within the decidualized endometrium, simulating the appearance of an early gestational sac. In some cases, before sloughing of the decidua, a hypoechoic interface beneath the decidua can be seen that represents hemorrhage between the necrotic decidua and inner myometrium. In contradistinction to normal IUPs, where the gestational sac is spherical and well defined, the pseudogestational sac created by sloughing decidua found in some advanced ectopic pregnancies is more irregular and has angulated borders.³¹ For a more detailed discussion of the sonographic changes that occur within the uterus in early IUP, refer to Chapter 3. As opposed to the decidualized endometrium in normal IUP, the decidualized endometrium of ectopics usually demonstrates little or no diastolic flow on TV-CDS. The myometrium typically shows a poorly vascularized or “cold” pattern. The waveform from the decidualized endometrium of the ectopic pregnancy demonstrates little or no diastolic flow as compared with the decidua of an early IUP. Tiny (a few millimeters) cysts can be seen within the decidua, and they have been reported to correspond to areas of decidual necrosis.³²