Ectopic pregnancy

Ectopic pregnancy is defined as any pregnancy implanted outside of the endometrial cavity. It is considered the main cause of maternal death in the first trimester of the pregnancy [5], and the attention of the clinician must be on early diagnosis and institution of therapy before tubal rupture. EPs comprise 1–2% of all first trimester pregnancies in the United States, however this small portion accounts for nearly 6% of all pregnancy‐related deaths [5, 6]. Up to 73.9% of women with an EP may be diagnosed on an initial TVS assessment and 94% are diagnosed prior to the need for emergency surgical intervention [7, 8].

Heterotopic pregnancy is another rare condition that occurs when EP is found in conjunction with an intrauterine pregnancy. The incidence of heterotopic pregnancy is higher when the pregnancy is achieved through Assisted Reproductive Technologies (ART), and it estimated to be 1–3 in every 100 ART pregnancies [9, 10] in contrast to 1 in every 7000–30 000 spontaneous pregnancies [11]. The risk is directly correlated with the number of embryos transferred during the In Vitro Fertilization (IVF) process. When four or more embryos are transferred, it is estimated that the risk for an heterotopic pregnancy is as high as 1 in 45 pregnancies [12]. Thus, in clinical practice it is important to remember that a visualized IUP does not exclude the risk of EP in women undergoing IVF treatment. Ovulation induction also is associated with an incidence of heterotopic pregnancy of nearly 0.5–1% [13]. Rare cases of twin tubal pregnancies, with both embryos in the same fallopian tube or one embryo is each fallopian tube have also been reported [14, 15].

Historical perspective

The first known report describing an ectopic pregnancy is from Abulcasis in 963 CE [16]. Duverney, in 1708, was probably the first to describe a heterotopic pregnancy in an autopsy case. Parvey, in 1876, described 22 cases of ruptured tubal pregnancies. Until the end of the nineteenth century, the therapy of ectopic pregnancies was not surgical, and the mortality rates was as high as 60% [17]. In 1884, the British surgeon, Robert Lawson Tait (1845–1899) reported the first salpingectomy for the treatment of an ectopic pregnancy. In 1888, Tait reported only two deaths out of 42 operated cases, a marked improvement for a condition that had been almost always fatal [18].

In 1891, Whitcomb described a case of a ruptured tubal ectopic pregnancy associated with an intra‐uterine pregnancy in a bicornuate uterus. In 1894, Bussieri described a case of an intact ectopic pregnancy in an autopsy of a prisoner after her execution. Interestingly, at the time, it was felt that ectopic pregnancies were the result of an embryo passing through the tube and implanting there secondary to an interrupted coitus. In subsequent years, the number of described cases increased. In 1902, Zinke described 88 cases and in 1904 Simpson reported 113 cases [19].

In 1941, Caffier described 10 cases of salpingostomia rather than salpingectomy for the treatment of ectopic pregnancy. Of these patients, four proceeded to have subsequent IUP.

It was not until the mid‐twentieth century that technological advancements, including the advent of laparoscopy and ultrasound, the radioimmunoassay for the determination of (β‐hCG), and more recently the introduction of IVF have made it possible the early diagnosis of an unruptured ectopic pregnancy and better prognosis for patients at risk for ectopic pregnancy [20].

Today, we have an improved understanding of the natural history of ectopic pregnancies, including the fact that some patients with this condition may indeed experience spontaneous resolution. The therapeutic armamentarium has expanded and now includes the use of conservative surgery, when indicated, and the use of medical therapy with methotrexate (MTX).

Clinical questions

1. How common is ectopic pregnancy?

Ectopic pregnancy remains a frequency condition during a woman’s reproductive life. In the developed world, the incidence of EP is 11–20 per 1000 live births [21–23]. In the developing world, the incidence is thought to be higher, but the data is not clear. Epidemiological studies have consistently reported a sixfold increase in the incidence of ectopic pregnancies between 1970 and 1992 [24]. This increase is thought to be secondary to a higher incidence of pelvic inflammatory disease (PID), more women of reproductive age with the habit of smoking, increased use of ART, and increased awareness of the conditions. In assisted conception populations the incidence is as high as 4% [10].

The epidemiology of ectopic pregnancy is easier to understand when divided in two distinct entities: ectopic pregnancies due to contraceptive methods failure, which have a low incidence as the modern contraceptive methods have a low failure rate; and ectopic pregnancies due to reproductive failure rate, with a higher incidence [25]. The estimated failure rate of tubal sterilization ranges from 0.1–0.8%, and a third of these pregnancies are ectopic [26].

2. What effects does ectopic pregnancy have on morbidity and mortality?

Improved methods for early diagnosis and treatment have reduced the fatality rate in developed countries [23]. Nonetheless, ectopic pregnancy remains the leading cause of first trimester pregnancy‐related maternal death, with a rate of 0.35 cases per 1000 ectopic pregnancies [16, 27]. However, in countries in development, the maternal death rate is much higher [28].

Ectopic pregnancy is also considered a frequent cause of maternal morbidity. Acute symptoms, including pelvic pain and vaginal bleeding are common. In the long term, chronic pelvic pain, infertility, and psychological issues are frequently present in women with a previous history of ectopic pregnancy [29].

3. What are the risk factors for ectopic pregnancy?

Though many risk factors have been found in association with EP, up to one‐third of cases occur in women without any apparent risk factors [30].

Of the risk factors associated with EP, the highest risk is attributed to tubal pathology that results from PID, especially those caused by Chlamydia trachomatis and Neisseria gonorrheae [31–33]. The exact mechanism is not determined, but in addition to compromise of the tubal architecture, it may be due to a disruption of the tubal microenvironment. The tubal lumen or ostia may be partially obstructed due to the formation of synechiae or tubal torsion due to pelvic adhesions, obstructing the passage of the embryo, although allowing the passage of the sperm due to its smaller size.

Tubal pathology due to previous pelvic surgery is another important risk factor for ectopic pregnancies. This is especially true in women who have undergone surgical procedures for sterilization or fertility restoration, including fimbrioplasty, salpingoplasty, and neosalpigostomy.

A history of multiple sexual partners is also associated with EP, but this correlation is likely due to the higher incidence of PID in women with multiple sexual partners.

It is important to note that the use of contraceptive methods including tubal sterilization, copper and progestin releasing intrauterine devices (IUDs), and progestin‐only contraceptives is associated with an overall lower rate of ectopic pregnancies due to the efficacy of these contraceptive methods. While IUD use lowers EP rate by 10% when compared to women not using contraception, when the IUD fails women are at a higher risk of EP, perhaps due to the higher risk of PID in these patients. In a multicenter study, the World Health Organization concluded that in low risk populations, the risk for PID due to the use of an IUD is temporary and limited to the insertion period [34–37]. Congenital anomalies of the fallopian tubes, including diverticulums, septums, hypoplasia, and accessory ostia may be involved in the cause of an ectopic pregnancy. In addition, when the lumen of the tube is narrowed, for example due to extrinsic compression caused by uterine fibroids, there could be an increased risk for ectopic pregnancies. Other conditions can also predispose to EP including salpingitis isthmica nodosa (SIN) and congenital fallopian tube anomalies secondary to in utero diethylstilbestrol exposure (DES) [37, 38]. DES exposure increases the rate of EP by ninefold [39]. The etiology of SIN is unknown, and occurs when tubal mucosa penetrates the myosalpinx in the isthmic segment of the tube resulting in muscular hypertrophy.

A previous EP increases the risk of subsequent EP by more than 10%, and approximately 9% of women with a single episode of salpingitis have a subsequent EP [33, 40, 41]. This risk may be attributable to the predisposing tubal disorder that led to the first EP.

One‐third of all cases of ectopic pregnancies are associated with smoking [33], and smoking has been found to be an independent risk factor for EP. There is a dose–effect relationship present, and the risk is higher when a woman uses more than 20 cigarettes a day [42]. Though the mechanism is unknown, several mechanisms have been proposed, including delayed ovulation, altered tubal and uterine motility, impaired tubal ciliary motility, and impaired immunity [31].

Ectopic pregnancy is more common in women with infertility, even in the absence of tubal disease [43]. Assisted reproductive technologies are particularly associated with EP, with a risk of 2–5%, and it may be higher when tubal disease is present [31]. One study found that women taking clomiphene citrate doubled their risk of ectopic pregnancy from 3–6% [44]. Similarly, controlled ovarian stimulation with gonadotropins has been associated with an increased risk of ectopic pregnancy [45, 46].

4. What causes ectopic pregnancy?

Ectopic pregnancies occur when there is a disruption to the blastocyst migration through the fallopian tube or when there are conditions that promote early implantation. The exact mechanisms are not well‐established. Of note, ectopic pregnancies are unique to the human species, and perhaps to higher primates. Because of this species‐specificity, there is no good animal model to study this condition.

Approximately 93–98% of EPs occur within the fallopian tube. Of these, 70% implant in the ampulla, followed by the isthmus, fimbria, and cornual/interstitial locations. [47]. Up to 7% of EPs are located outside the fallopian tube [47]. These sites include the ovary, cervix, peritoneal cavity, and prior cesarean scar. The fallopian tube lacks a submucosal layer; thus a zygote is able to quickly invade the epithelium and the rapidly proliferating trophoblast often invades as far as within the muscularis, and may even reach the tubal serosa. Commonly, the developing embryo in an EP is absent or poorly developed.

The patognomonical finding is an ectopic gestation is the presence of chorionic villi in the lumen or in the wall of the fallopian tube or another extra‐uterine site. The pathology evaluation of these villi revealed it to be normal or with hyaline degeneration. However chromosomal abnormality is not likely an important etiology of ectopic pregnancy [48, 49]. In one study in which chorionic villi were karyotyped from 30 viable surgically excised ectopic gestations, the rate of karyotype abnormality was no different from that of controls with intrauterine pregnancies [49].

Chronic salpingitis is observed in up to 90% of ectopic pregnancies [50]. Infection produces an inflammatory response that damages the tubal ciliary epithelium, potentially disrupting the embryonic transport through the fallopian tube or may result in the formation of intra‐tubal synechiae, thus contributing to closure of fallopian tube. PID can promote adhesion formation with adjacent pelvic organs, disrupting the anatomy of the fallopian tube, and potentially also resulting in altered tubal transport.

We previously discussed the higher incidence of EP in patients undergoing ART. One theory for this higher rate is that the medications used to stimulate ovarian follicle development result in high levels of progesterone and estradiol that may slow tubal peristalsis and promote uterine relaxation, thus promoting early embryo implantation. Women with tubal infertility undergoing IVF are at even higher risk of EP. This is one of the reasons physicians may recommend removal of diseased tubes before undergoing IVF treatment [51].

Oocyte or embryonic anomalies may also participate in the pathogenesis of ectopic pregnancy. An abnormal embryo may have sub‐optimal transport through the fallopian tube. Although not necessarily related to chromosome abnormalities, it could be speculated that this could also be associated with the fact that women older than 35 years of age have a higher rate of EP [52].

In relation to the uterus, the endometrium undergoes decidualization and the endometrial mucosa may present an atypical aspect. This has been called the Arias–Stella reaction, and it is characterized by increased cellular volume, hyperchromatosis, pleomorphism, and increased mitotic activity [53]. In 5–10% of ectopic pregnancies a decidual cast may be passed and often mistaken for products of conception. This occurs because the abnormal pregnancy does not produce enough progesterone to maintain the decidua. Notably, on pathologic examination only decidua will be seen and chorionic villi will be absent. This may often be confused with a miscarriage.

EP rupture is usually spontaneous. Early rupture, 6–8 weeks, occurs with isthmic implantations due to the small diameter size of the fallopian tube at this location. Ampullar ruptures occur later around 8–12 weeks because it is more easily distensible compared to the isthmus. Interstitial ruptures occur even later at 12–16 weeks as the myometrium provides more room for the embryo to develop. Interstitial rupture is the most dangerous because proximity to uterine and ovarian vessels can result in massive hemorrhage.

5. How is the diagnosis of ectopic pregnancy made?

The early diagnosis of an ectopic pregnancy is important as it decreases the risk of tubal rupture and ameliorates the success rates of conservative therapy. Special attention must be given to patients at risk for ectopic pregnancy, including patients with previous history of ectopic pregnancy, tubal surgery (including tubal ligations), infertility, use of assisted reproductive technologies therapy, history of PID, presence of endometriosis, use of IUD, and history of smoking.

In patients with delayed menses, vaginal bleeding, and/or pelvic pain are possible indicators of an ectopic pregnancy. In these cases, it is important to maintain close follow‐up until a final diagnosis is confirmed.

In patients that are hemodynamically stable, it is possible to proceed with non‐invasive diagnostic steps, including transvaginal ultrasound and β‐hCG measurements. Conversely, in hemodynamic unstable patients, it may be necessary to resort to surgical diagnosis, either with a laparoscopy or laparotomy. In certain cases, a uterine curettage may be employed with the goal to verify the presence of intra‐uterine villi, confirm a non‐viable intra‐uterine pregnancy. Culdocentesis, to verify the presence of intra‐pelvic blood, is rarely performed in today’s clinical practice.

Historically the triad of pain, vaginal bleeding or spotting, and delayed or missed menstruation raised suspicion of EP in women of child bearing age [54]. Decades ago women often presented with symptoms of acute abdomen and/or hypovolemic shock secondary to ectopic rupture. Today, with the advent of commercially available urine pregnancy tests in combination with early utilization of TVS and serial measurements of plasma β‐hCG, women are diagnosed earlier in the course of the disease. In fact, most patients with EP deny symptoms of abdominal pain or this is a late finding [55]. Up to 10% are asymptomatic and one in three women have no clinical signs [55].

Information regarding date of LMP, date of first positive home pregnancy test, dates of positive urine and/or blood pregnancy tests, date of HCG administration for ovulation triggering or date of oocyte retrieval if the pregnancy resulted from infertility treatment are all important information in the evaluation of patient at risk for ectopic pregnancy.

The patient’s history must include information regarding the onset, volume, and duration of vaginal bleeding as well as nature, intensity, and location of the pelvic pain. Early presentation symptoms of EP are subtle or may even be absent. It is common for patients to believe they are carrying a normal pregnancy or having a miscarriage. Up to 30% of patients with ectopic pregnancies have no vaginal bleeding [56].

Because the symptoms of EP are non‐specific, it may be misdiagnosed as other gynecological, gastrointestinal, or urological disorders. Common conditions that may present similarly to EPs include appendicitis, salpingitis, ovarian cyst rupture, miscarriage, adnexal torsion, urolithiasis, and urinary tract infection [57].

Other symptoms that are not specific for EP include nausea, vomiting, and diarrhea. Symptoms of ruptured EP are the same as those of an acute abdomen: abdominal distension, tenderness, peritoneal signs, and shock. Therefore the diagnosis of EP should be considered for all reproductive age women who present with acute abdominal pain or GI symptoms [23].

The overall likelihood of ectopic pregnancy is 39% in a patient with abdominal pain and vaginal bleeding, but no other risk factors. The probability increases to 54% if the patient has other risk factors [58]. It is important to note that 30–50% of patients with ectopic pregnancies may not bleed [59]. Conversely, only 29% of pregnant patients presenting to a hospital emergency room with bleeding or pain, findings of peritoneal irritation, and cervical motion tenderness turned out to have ectopic pregnancies in one series. Even though this was fourfold higher than the 7.7% prevalence of ectopic pregnancies among the entire cohort of pregnant patients presenting with pain and bleeding, the majority of pregnancies in the high‐risk group were still intrauterine [59].

Physical examination in patients at risk for ectopic pregnancy should include an assessment of the volume of vaginal bleeding, presence of abdominal/pelvic tenderness, presence of adnexal mass and cervical motion tenderness, uterine size, and hemodynamic status. Tenderness to palpation may be elicited on abdominal and bimanual examination. EPs of older gestational age may produce mass effect and push the uterus to one side.

Symptoms of dizziness, lightheadedness, and syncope should raise suspicion for intra‐abdominal bleeding from a ruptured EP. Blood accumulation in the rectouterine cul‐de‐sac may present as posterior vaginal fornix bulging. Women with hemoperitoneum may complain of symptoms of diaphragmatic irritation characterized by referred pain to the neck, shoulder or scapula which worsens with maximal inspiration.

Only 10% of patients with an ectopic pregnancy have a palpable adnexal mass, and up to 10% have negative pelvic examinations [60].

There are significant limitations in using medical history and physical examination in the diagnosis of ectopic pregnancy. Therefore, decision–analysis studies have determined that diagnostic algorithms using a combination of pelvic transvaginal ultrasound and β‐hCG offered the most accurate means of diagnosing ectopic pregnancy [59].

The presence of an intrauterine gestation on ultrasound almost always rules out an ectopic pregnancy, since the incidence of heterotopic pregnancies has been reported as 1/30 000 spontaneous conceptions. In pregnancies achieved through assisted reproductive technologies, heterotopic gestation must always be a consideration as reported incidence is much higher, approaching 1/100–1/1000 pregnancies [59].

Unless the patient is hemodynamically unstable as in cases of ruptured ectopic pregnancies, or the diagnosis is definite as when products of conception are seen at the external os or in the vagina during pelvic examination, an initial ultrasound evaluation should be undertaken. Transvaginal sonography provides an accurate diagnosis of pregnancy status for most intrauterine pregnancies [61].

A transvaginal ultrasound can visualize an intra‐uterine gestational sac when the gestational age is approximately five to six weeks [62]. When the gestational age is unknown, the β‐hCG values may help determine the gestational age and improve the interpretation of the transvaginal ultrasound findings [63–65]. The discriminatory value (also called the “discriminatory zone”) of the β‐hCG is defined as the value above which a gestational sac must be identified by the ultrasound, when the pregnancy is intra‐uterine and normal. The discriminatory values for transvaginal ultrasounds are usually between 1500 and 2000 mIU ml⁻¹. The discriminatory value for abdominal ultrasound is usually around 6500 mIU ml⁻¹ [66].

However, these values are also largely dependent on ultrasound equipment resolution and examiner experience. Moreover, inter‐assay variability also contributes to made it very difficult to establish a universally reproducible discriminatory β‐hCG value [67], and therefore these values are often defined by each institution.

When the β‐hCG value is above the discriminatory zone, an intra‐uterine pregnancy (gestational sac) must be seen by an ultrasound. The absence of an intra‐uterine pregnancy when the β‐hCG is above the discriminatory zone indicates an unviable pregnancy, but it cannot distinguish between an ectopic pregnancy or a miscarriage [68]. The presumptive diagnosis of an ectopic pregnancy in these cases can be incorrect in more than 50% of the cases [69]. However, it is also important to take into consideration the cases of multiple gestation, in which the β‐hCG values are higher than when compared to singleton gestations [70].

Care must be taken to confirm the presence of a true gestational sac, rather than a pseudosac. A true gestational sac is eccentrically placed and adjacent to the central echogenic endometrial stripe, reflecting implantation of the conceptus in the endometrial tissue rather than in the endometrial cavity. A pseudosac is a collection of fluid within the endometrial cavity and can be seen in ectopic pregnancies [71].

The sensitivity and specificity of a transvaginal ultrasound to accurately diagnose an ectopic pregnancy depends upon the criteria used to establish the diagnosis. If stringent criteria are used, like for example the extra‐uterine presence of embryonic heart activity or a gestational sac containing a yolk sac or embryo, the sensitivity is low, ranging from 20.1% to 64.6%, and the negative predictive value of the test is also low. Alternatively, if less stringent criteria are used, like for example any adnexal mass other than a simple cyst, with or without cul‐de‐sac fluid, the sensitivity of ultrasound improves to 69–84.4%, and the negative predictive value improves to 95%, at the expense of minimal losses in specificity and positive predictive value [72, 73]. If ultrasound demonstrates neither an intrauterine pregnancy nor an adnexal mass in early pregnancy, further evaluation is required and an ectopic gestation must be considered. The presence of an intrauterine gestation on ultrasound almost always rules out an ectopic pregnancy, since the incidence of heterotopic pregnancies has been reported as 1/30 000 spontaneous conceptions [11].

Knowledge of the temporal appearance of embryonic structures as visualized by transvaginal ultrasound is essential to the correct diagnosis of a viable intra‐uterine pregnancy. For example, an intrauterine gestational sac can often be imaged by a transvaginal ultrasound approximately 4.5 – 5 weeks after the first day of the LMP, and should almost always be detectable by 5½ weeks gestation [73]. Failure to image an intrauterine gestation should be interpreted with caution when LMP is the reference, since dating can be incorrect due to recall bias and delayed ovulation. Initially, the diameter of the gestational sac may only be 2–3 mm, and will increase by 1 mm d⁻¹ in early pregnancy, and a yolk sac should be imaged at a gestational sac diameter of 8–10 mm, or approximately one week after the appearance of the gestational sac. This correlates to 5.5–6.5 weeks gestation [74]. Embryonic cardiac activity adjacent to the yolk sac is detected next. Fetal cardiac activity can be detected prior to six weeks gestation in some instances, and should be confirmed by 46 days gestational age and a mean sac diameter of 16 mm in almost all viable gestations [67].

In reproductive aged patients, the symptoms of vaginal bleeding and/or pelvic pain should always raise suspicion for a possible pregnancy, and β‐hCG measurements in blood or urine should be obtained to confirm or rule‐out a pregnancy. In conjunction with transvaginal ultrasound, quantitative β‐hCG assays are integral in the diagnosis of ectopic pregnancy. Serial quantitative β‐hCG measurements are an integral part of the clinical follow‐up of a pregnant patient with an early pregnancy and trimester vaginal bleeding and/or pain until it is determined if the pregnancy is viable, or if the presenting symptoms are due to a miscarriage or an ectopic pregnancy. Similarly, serial β‐hCG measurements following a diagnosis of a miscarriage or an ectopic pregnancy are essential to confirm resolution of these problems.

When the gestational age is unknown, β‐hCG values may help in the determination of the gestational age. Moreover, quantitative β‐hCG values serve as a reference when interpreting the results of the transvaginal ultrasound. The trophoblast begins to secrete hCG into maternal blood upon implantation. For this reason β‐hCG can be detected in maternal serum even prior to the anticipated menses, and 2½ weeks prior to the ability of ultrasound to consistently image a gestational sac [67]. As already mentioned, the “discriminatory zone” of the β‐hCG values, is the β‐hCG level above which an intrauterine gestational sac, if the pregnancy is normal and intrauterine, will always be detected by ultrasound.

6. What if the location of the pregnancy remains unknown following ultrasound?

Although initially there was confusion in the literature regarding the definition for a “Pregnancy of Unknown Location” (PUL), today the term is generally applied to describe women with a positive pregnancy test who have no evidence of either an intra‐uterine pregnancy or an ectopic pregnancy on transvaginal ultrasound [72

Stay updated, free articles. Join our Telegram channel

Join