Key points
- •
The rate of ectopic pregnancy in the United States has remained fairly constant since the early 2000s and is approximately 6.6 in 1000 pregnancies in women aged 15 to 24.
- •
Risks of ectopic pregnancy include age, smoking, pelvic inflammatory disease, infertility, and prior tubal surgery. After tubal sterilization, the risk of ectopic pregnancy if a pregnancy occurs is about 30%, reaching 50% if the sterilization technique was tubal desiccation.
- •
Human chorionic gonadotropin (HCG) trends can help diagnose ectopic pregnancies. About 85% of women with an ectopic pregnancy have serum HCG levels lower than in normal pregnancy. In early pregnancy the normal HCG doubling time is 1.4 to 3 days, with 85% of pregnancies demonstrating a 66% HCG increase every 48 hours. An HCG rise less than 53% in 48 hours is 99% sensitive for an abnormal pregnancy.
- •
Ectopic pregnancies can be managed surgically with salpingectomy or salpingostomy. Randomized trial data suggest there is no difference in overall subsequent pregnancy outcomes between women who are treated by salpingostomy versus salpingectomy.
- •
Asymptomatic persistent ectopic pregnancy can be treated expectantly or with methotrexate (MTX). The success of MTX depends on the size and age of the gestation and the initial HCG level. After the methotrexate injection, the HCG level should fall at least 15% between days 4 and 7 and at least 15% weekly thereafter.
Ectopic pregnancy occurs when the fertilized ovum/developing blastocyst implants at a site outside the endometrial cavity. It was probably first described in the year 963 by Albucasis, an Arab writer. In 1876, before the initiation of surgical therapy, the mortality rate from ectopic pregnancy was estimated to be 60%. The first successful operative treatment of ectopic pregnancy was performed in 1883 by Lawson Tait in England. In 1887, he reported that he had performed salpingectomy on four women with ectopic pregnancy and that they all survived.
Epidemiology
The incidence of ectopic pregnancy has been estimated to be between 1% to 2% of all pregnancies . Although the incidence of ectopic pregnancy increased sixfold between 1970 and 1992, it has remained stable since then. In the United States in 1989, the annual ectopic pregnancy rate per 10,000 women aged 15 to 44 was 15.5, similar to that in Finland but higher than the rate in France. The last national data reported by the Centers for Disease Control and Prevention (CDC) showed that the overall incidence of ectopic pregnancy had plateaued to approximately 20 in 1000 pregnancies in the early 1990s. The current incidence of ectopic pregnancy is difficult to estimate from available hospitalization and insurance records because the number of ectopic pregnancy cases requiring inpatient hospital treatment has decreased. The incidence varies among different countries, with rates as high as 1 in 28 and 1 in 40 pregnancies reported in Jamaica and Vietnam, respectively. The risk of ectopic pregnancy associated with assisted reproductive technology is increased compared with the general population, with rates from 0.8% to 8.6 %. Data from the National Assisted Reproductive Technology Surveillance System from 2001 to 2011 showed that the rate of ectopic pregnancy declined from 2% to 1.6% out of 553,577 pregnancies in the United States ( ).
There has been an increasing trend toward treating ectopic pregnancy conservatively (without resorting to salpingectomy) and on an ambulatory basis without overnight hospitalization. With earlier detection of ectopic pregnancy, a steadily increasing percentage of women with this problem are now being treated before tubal rupture occurs, by outpatient laparoscopic procedures or by medical treatment with methotrexate. An analysis of both hospital discharge data and an ambulatory medical care survey revealed that the estimated number of hospitalizations for ectopic pregnancy in the United States declined from nearly 90,000 in 1989 to about 45,000 in 1994. However, in 1992, about half of all women with ectopic pregnancy in the United States were treated as outpatients, and the estimated number of total ectopic pregnancies in this year was 108,000, for a rate of 19.7 per 1000 reported pregnancies. Thus in the United States in 1992, about 2 of every 100 women who were known to conceive had an ectopic gestation. The increased incidence of ectopic pregnancy is thought to be due to two factors: (1) the increased incidence of salpingitis, caused by increased infection with Chlamydia trachomatis or other sexually transmitted pathogens; and (2) improved diagnostic techniques, which enable the diagnosis of unruptured ectopic pregnancy with more precision and earlier in gestation, before asymptomatic resolution of the pregnancy occurs.
The rate of ectopic pregnancy increases with increasing age . However, because of the lower pregnancy rate in older women, overall only about 11% of ectopic pregnancies in the United States occur in women aged 35 to 44, whereas more than half, 58%, occur in women aged 25 to 34 years. Most ectopic pregnancies occur in multigravid women. Only 10% to 15% of ectopic pregnancies occur in nulligravid women, whereas more than half occur in women who have been pregnant three or more times. In the United States, the rates of ectopic pregnancy are similar in each section of the country, but the rates are higher for nonwhite women compared with white women. About 3% of all reported pregnancies in nonwhite women aged 35 to 44 in the United States are ectopic.
Mortality
Even with the increased use of surgery and blood transfusions and earlier diagnosis , ectopic pregnancy remains a major cause of maternal death in the United States today. From the last CDC report , 6% of pregnancy-related mortality during the period of 1991 to 1999 was due to ectopic pregnancies . In the United States, 876 deaths were attributed to ectopic pregnancy between 1980 and 2007. Ectopic pregnancy is the most common cause of maternal death in the first half of pregnancy. The ectopic pregnancy–to–mortality ratio has declined by 57% from the period 1980 to 1984 to 2003 to 2007, from 1.15 to 0.5 ( ). The mortality ratio was 6.8 times higher for African American women than for white women and 3.5 times higher for women older than 35 years compared with women younger than 25 years. Of the 76 deaths among women hospitalized with ectopic pregnancy between 1998 and 2007, 70% of the ectopic pregnancies were located in the fallopian tubes. Unmarried women of all races have a 1.7 times greater chance of dying of ectopic pregnancy than married women. Overall the risk of death from ectopic pregnancy is about 10 times greater than the risk of childbirth and more than 50 times greater than the risk of legal abortion.
The major cause of mortality from ectopic pregnancy is blood loss. Most cases of mortality (70%) result from gestations in the tube, and the other 30% are interstitial cornual or abdominal gestations. Because the overall incidence of ectopic pregnancy occurring in these latter locations is slightly less than 4%, interstitial and abdominal ectopic pregnancies have about a five times greater risk of being fatal. About three-fourths of the women with fatal ectopic pregnancies initially developed symptoms and died in the first 12 weeks of gestation. Of the remaining one-fourth who developed symptoms and died after the first-trimester, 70% had interstitial or abdominal pregnancies. Patient delay in consulting a physician after development of symptoms accounted for one-third of the deaths, whereas treatment delay resulting from misdiagnosis contributed to the half the deaths.
Etiology
Factors contributing to the risk
The major factor contributing to the risk of ectopic pregnancy is salpingitis. Salpingitis results from infectious causes such as pelvic inflammatory disease (PID) or inflammatory causes such as endometriosis. Its morphologic sequelae account for about half of the initial episodes of ectopic pregnancy. However, in about 40% of cases the cause cannot be determined and is presumed to be a physiologic disorder resulting from the delay of passage of the embryo into the uterine cavity. Ovulation from the contralateral ovary has been implicated as a cause of the delay of blastocyst transport, and it has been suggested that contralateral ovulation occurs in about one-third of tubal pregnancies, although this has not been confirmed.
Another possibility in the etiology of ectopic pregnancy is a hormonal imbalance; an elevated circulating level of either estrogen or progesterone can alter normal tubal contractility. An increased rate of ectopic pregnancies has been reported in women who conceive with physiologically and pharmacologically elevated levels of progestogens . The latter condition can be produced locally with a progestogen-releasing intrauterine device (IUD) and systemically with progestin-only oral contraceptives. Iatrogenic, physiologically increased levels of estrogen and progesterone occur after ovulation induction and the use of assisted reproductive technology (ART) with either clomiphene citrate or human menopausal gonadotropins, and an increased rate of ectopic pregnancies has been reported in women conceiving after each of these treatment modalities.
Another possible cause is an abnormality of embryonic development. Although aneuploidy has been found to be prevalent in ectopic pregnancies, it may not be higher than the normal rate of aneuploidy and is unlikely to be a cause of ectopic pregnancies. Inherited genetic abnormalities most probably are not a cause of ectopic pregnancy either. Also, there is no increased incidence of ectopic pregnancy among first-degree relatives.
Several epidemiologic studies have indicated that cigarette smoking is associated with about a twofold increased risk of ectopic pregnancy, even when the data were controlled for the presence of other risk factors. The risk of ectopic pregnancy was directly related to the number of cigarettes smoked per day, with a fourfold increased risk noted among women who smoked 30 or more cigarettes per day. Known risk factors for ectopic pregnancy, presented as odds ratios and attributable risk, are depicted in Table 17.1 ( ).
Odds Ratio | Attributable Risk * | |
---|---|---|
Probable salpingitis | 2 | |
Confirmed salpingitis | 3.5 | |
History of tubal surgery | 3.5 | 0.18 † |
Smoking | 0.35 | |
Ex-smoker | 1.5 | |
1-9 cigarettes per day | 2 | |
10-19 cigarettes per day | 3 | |
≥20 cigarettes per day | 4 | |
Age (years) | 0.14 | |
30-39 | 1.5 | |
≥40 | 3 | |
Spontaneous abortion | 3 | 0.07 |
Elective abortion | 2 | 0.03 |
IUD history | 1.5 | 0.05 |
Previous infertility | 2.5 | 0.18 |
* From Auvergne registry data (Bouyer J, Coste J, Shojaei T, et al. Risk factors for ectopic pregnancy: a comprehensive analysis based on a large case-control, population-based study in France. Am J Epidemiol. 2003;157:185).
† Risk attributable to history of genital infection and tubal surgery together is 0.33.
The major causes of ectopic pregnancy are discussed in more detail next.
Tubal pathology leading to ectopic risk
Disruption of normal tubal anatomy from infection, surgery, congenital anomalies, or inflammatory disease such as endometriosis is a major cause of ectopic pregnancy. The agglutination of the plicae (folds) of the endosalpinx produced by salpingitis can allow passage of sperm but prevent the normal transport of the larger morula. The morula can be trapped in blind pockets formed by adhesions of the endosalpinx. In their 20-year longitudinal study, Weström and colleagues found that nearly half (45.3%) of the women with ectopic pregnancy had a clinical history or histologic findings of a prior episode of acute salpingitis ( ). This is in agreement with the 40% incidence of prior salpingitis found on histologic examination by several groups of investigators in women with ectopic pregnancy ( Fig. 17.1 ).
Weström and colleagues prospectively followed 900 women aged 15 to 34 years who had laparoscopically confirmed acute salpingitis and found that the subsequent ectopic pregnancy rate was 68.8 per 1000 conceptions, yielding a sixfold increase in the risk of ectopic pregnancy . The risk of ectopic pregnancy after acute salpingitis increased both with the number of episodes of infection and with the increasing age of the women at the time of infection. The odds ratios for ectopic pregnancy after two and after three or more episodes of chlamydial infection were 2.1 and 4.5, respectively. Data also suggest that a history of chlamydial infection results in the production of PROKR2 protein, which creates a microenvironment that predisposes to tubal implantation.
Endometriosis
Inflammation and adhesions of the fallopian tubes as a result of conditions such as endometriosis are risk factors for ectopic pregnancy. Fig. 17.2 portrays an example of endometriosis resulting in a left ovarian endometrioma, tubal inflammation, and adhesions, with subsequent clubbed tube and hydrosalpinx. Compared with women without endometriosis, women with endometriosis had two times the risk for ectopic pregnancy (relative risk, 1.9; 95% confidence interval [CI], 1.8 to 2.1) ( ).
Salpingitis isthmica nodosa
Salpingitis isthmica nodosa (SIN) is defined as the microscopic presence of tubal epithelium within the myosalpinx or beneath the tubal serosa ( Fig. 17.3 ). In two histopathologic studies of tubes removed from women with ectopic pregnancy, it was found that about half contained lesions of SIN compared with 5% in a control group. With serial sectioning, it has been determined that SIN is actually a diverticulum or intrauterine extension of the tubal lumen . Associated histologic evidence of chronic salpingitis was seen in only 6% of the tubes, suggesting that SIN was not necessarily the result of infection . The tubal pregnancy has been found to have usually implanted in a portion of the tube distal to the SIN, indicating that mechanical entrapment of the morula is not the mechanism whereby SIN causes tubal gestation. It is possible that SIN itself or the associated tubal anomalies may be responsible for dysfunction of the tubal transport mechanism without anatomic obstruction.
It is likely that adhesions between the tubal serosa and bowel or peritoneum may interfere with normal tubal motility and cause ectopic pregnancy because, as reported, 17% to 27% of women with ectopic pregnancy have had previous abdominal surgical procedures not involving the oviduct. On the other hand, neither endometriosis nor congenital anomalies of the tube have been associated with an increased incidence of ectopic pregnancy.
Tubal surgery
An operative procedure on the tube itself is a cause of ectopic pregnancy whether the tube is morphologically normal, as occurs with sterilization procedures, or abnormal, as occurs with post salpingitis reconstructive surgery. The incidence of ectopic pregnancy occurring after salpingoplasty or salpingostomy procedures to treat distal tubal disease ranges from 15% to 25%, probably because the damage to the endosalpinx remains. The rate of ectopic pregnancy after reversal of sterilization procedures is lower, about 4%, because the tubes have not been damaged by infection. The rate of ectopic pregnancy after sterilization is discussed later in this chapter.
History of ectopic pregnancy
Women who have had a prior ectopic pregnancy, even if treated medically or by unilateral salpingectomy, are at increased risk for having a subsequent ectopic pregnancy. The odds of a recurrent ectopic pregnancy among women with history of ectopic pregnancy compared with other pregnant women is almost threefold greater (adjusted odds ratio [AOR], 2.72; 95% CI 1.66 to 2.8) ( ). Of women who conceive after having one ectopic pregnancy, about 25% of subsequent pregnancies are ectopic . The rates of recurrent ectopic pregnancy after single-dose methotrexate treatment, salpingectomy, and linear salpingostomy are 8%, 9.8%, and 15.4%, respectively, among women who conceive ( ).
Diethylstilbestrol exposure
Although it is less commonly encountered today, the incidence of ectopic gestation is significantly greater (four to five times) in women who have been exposed to diethylstilbestrol (DES) in utero and has been reported at the rate of 4% to 5%. This is likely because of abnormal tubal morphology and impaired function of the fimbriae. In women exposed to DES whose hysterosalpingograms demonstrated abnormalities in the uterine cavity, the ectopic pregnancy rate was as high as 13%.
Contraception failure
Tubal sterilization
For several decades, sterilization has been the most popular method of contraception used by couples in the United States. Since the development of laparoscopic surgery, female tubal sterilization is performed about twice as often as vasectomy. In an analysis of the long-term risk of pregnancy after tubal sterilization reported by Peterson and coworkers, it was found that within 10 years after the procedure, the cumulative life table probability of pregnancy was 1.85% . The 10-year failure rate after bipolar electrosurgical sterilization of the tubes was 2.48%, which rose to 5.43% if the sterilization procedure was performed when the woman was younger than 28 years. These investigators reported that for all 143 pregnancies occurring after tubal sterilization, 43 (30.1%) were ectopic pregnancies ( ).
Several investigators have reported that if pregnancy occurred after tubal sterilization with electrosurgical devices (monopolar or bipolar instruments), the ectopic pregnancy rate was as high as 50%. It has been hypothesized that fistulas may develop, allowing sperm to pass into the distal segment of the oviduct and fertilize the egg ( Fig. 17.4 ) ( ). Such fistulas can be demonstrated radiographically in about 11% of women after laparoscopic electrosurgical sterilization. Peterson and colleagues reported that within 10 years after the sterilization procedure, women sterilized with bipolar devices had twice as many ectopic pregnancies compared with those sterilized with metal clips or silicone bands. The overall ectopic pregnancy rate after bipolar sterilization was 1.7%.
Because about one-third of pregnancies that occur after all tubal sterilizations are ectopic, women should be counseled that if they do not experience the expected menses at any time after tubal sterilization before menopause, a test to detect human chorionic gonadotropin (HCG) should be performed rapidly, and if they are pregnant, a diagnostic evaluation to exclude the presence of ectopic pregnancy is necessary. In women who have an ectopic pregnancy after sterilization with the use of electrosurgery, because the site of the fistula usually cannot be determined clinically, salpingectomies should be carried out.
More often tubal sterilizations are now completed with bilateral salpingectomies. Compared with bipolar electrosurgical sterilization, in which a portion of the tube is desiccated, bilateral salpingectomies offer complete removal of the fallopian tube. This offers several advantages, including reduced risk of posttubal fistula formation, diminished risk of tubal stump ectopic pregnancy, decreased risk of poststerilization hydrosalpinx. and opportunistic salpingectomies, which decrease one’s lifetime risk of ovarian cancer ( ).
Intrauterine device
In general, individuals using contraception have a lower incidence of ectopic pregnancy compared with those not using contraception because the likelihood of conceiving is lower. Among contraceptive failures, women using diaphragms or combination oral contraceptives do not have an increased risk of ectopic pregnancy, whereas contraceptive failures with the IUD lead to an increased risk of ectopic pregnancy. Among IUD contraceptive failures, the risk of ectopic pregnancy is approximately 6% for the copper IUD versus 50% for the levonorgestrel IUD ( ) . This is because the progestogen-releasing IUD inhibits tubal contractions, resulting in a higher ectopic rate than the copper IUD. Women who use this method of contraception have about twice the risk of ectopic pregnancy (7.5 per 1000 woman-years) if they conceive compared with women who use no method of contraception (3.5 per 1000 woman-years). Women using IUDs who elect to have their pregnancies terminated should have a histologic examination of the tissue removed from the uterine cavity to be certain the pregnancy was intrauterine.
Hormonal alterations
Ovarian stimulation
As occurs with exogenous progesterone administration, if increased levels of exogenous or endogenous estrogens are present shortly after the time of ovulation, the incidence of ectopic pregnancy is increased. Several investigators have reported that the ectopic pregnancy rate is about 1.5% for conceptions that occur after ovulation has been induced with clomiphene citrate. The ectopic rate in pregnancies occurring after ovulation with human menopausal gonadotropins (HMGs) has been reported to range between 3% and 4%. Fernandez and colleagues, in a case-control study, found the risk of ectopic pregnancy was increased about fourfold among ovulatory women treated with controlled ovarian hyperstimulation—clomiphene citrate, HMG, or a combination of both—for unexplained infertility ( ). These reports indicate that increased levels of estrogen and progesterone interfere with tubal motility and increase the chance of an ectopic pregnancy.
In vitro fertilization
Ectopics occur more often after in vitro fertilization and embryo transfer. The reason for this increased incidence likely is due to one or more of several factors: increased sex steroid hormone levels, the presence of proximal tubal disease (although the ratio is similar in women with normal tubes), and flushing an embryo directly into the tube. Data from the National ART Surveillance System identified that 1.7% of more than 550,000 pregnancies were ectopic between the period of 2001 and 2011. The ectopic pregnancy rate was also noted to be associated with multiple embryo transfer. The rate of ectopic pregnancy was 1.6% when one embryo was transferred compared with 1.7%, 2.2%, and 2.5% when two, three, or four or more embryos were transferred, respectively ( ).
Sites of ectopic pregnancy
Most ectopic pregnancies occur in the tube. In Breen’s series and more recent series ( ; ), 97.7% of the ectopic pregnancies were tubal, 1.4% were abdominal, and less than 1% were ovarian or cervical ( Fig. 17.5 ). The majority of tubal pregnancies, 70% to 81%, were located in the ampullary portion of the tube , being about equally divided between the distal and middle third of the tube. About 12% of tubal gestations occur in the isthmus and 5% to 11% in the fimbrial region. Although Breen considered pregnancies located in the cornual area of the uterus to be uterine in origin, they are in fact pregnancies implanted in the interstitial portion of the tube. About 2% of all ectopic pregnancies are interstitial, and these are often associated with severe morbidity because they become symptomatic later in the gestation, are difficult to diagnose, and often produce massive hemorrhage when they rupture ( Fig. 17.6 ) ( ). A true cornual pregnancy is one located in the rudimentary horn of a bicornuate uterus, which is quite rare. In a review of 240 true cornual pregnancies reported by O’Leary and O’Leary, about 90% of them ruptured with massive hemorrhage ( ).
About 1 in 200 ectopic pregnancies are true ovarian pregnancies . The four criteria that fulfill the diagnosis of ovarian pregnancy originally described by Spiegelburg are as follows:
- 1.
The tube and fimbria must be intact and separate from the ovary.
- 2.
The gestational sac must occupy the normal position of the ovary.
- 3.
The sac must be connected to the uterus by the ovarian ligament.
- 4.
Ovarian tissue should be demonstrable in the walls of the sac.
Many women with ovarian pregnancies are believed to have a ruptured corpus luteum cyst, and the correct diagnosis was made during the surgical procedure only 28% of the time. The hemorrhagic mass (ovarian ectopic) should be located adjacent to the corpus luteum, never within it. Ovarian pregnancy is also associated with profuse hemorrhage, with 81% of reported to have a hemoperitoneum greater than 500 mL. Nevertheless, most can be successfully treated by ovarian resection and not oophorectomy.
Most abdominal pregnancies occur secondary to tubal abortion with secondary implantation in the peritoneal cavity ( Fig. 17.7 ). On rare occasions a primary abdominal pregnancy may occur, and all of the following three criteria originally set forth by Studdiford must be present:
- 1.
The tubes and ovaries must be normal, with no evidence of recent or past injury.
- 2.
There must be no evidence of a uteroplacental fistula.
- 3.
The pregnancy must be related only to the peritoneal surface, and it must be early enough in gestation to eliminate the possibility of secondary implantation after primary tubal nidation ( ).
An unusual type of primary abdominal pregnancy occurs when implantation is in the spleen or liver, producing massive intraperitoneal hemorrhage.
The prognosis for fetal survival in abdominal pregnancy is poor (11%) and is difficult to diagnose. Once the diagnosis is established, a laparotomy with removal of the fetus should be performed immediately to prevent a possible fatal hemorrhage. An adjunctive option is to administer methotrexate. On occasion, when the placenta is tightly adherent to bowel and blood vessels, it should be left in the abdominal cavity. In such instances the placental tissue usually resorbs. Retained placenta in the abdomen can lead to symptoms of abdominal pain and intermittent fever that may last for many months, as well as possible partial bowel obstruction and abscess formation. Therefore, if it is surgically feasible, removal of the entire placenta is the ideal goal. Furthermore, partial removal of the placenta may result in massive hemorrhage. Therefore, given all of these risks, it is important to carefully consider the best surgical approach because the decision making is challenging and critical.
The four pathologic criteria for the diagnosis of cervical pregnancy as reported by Rubin and colleagues are as follows ( Fig. 17.8 ):
- 1.
Cervical glands must be present opposite the placental attachment;
- 2.
The attachment of the placenta to the cervix must be intimate;
- 3.
The placenta must be below the entrance of the uterine vessels or below the peritoneal reflection of the anteroposterior surface of the uterus; and
- 4.
Fetal elements must not be present in the corpus uteri.
The usual characteristic clinical findings of cervical pregnancy are uterine bleeding after amenorrhea without cramping pain, a softened cervix that is disproportionately enlarged, complete confinement and firm attachment of the products of conception to the endocervix, and a closed internal os.
Most cervical pregnancies are associated with previous cervical or uterine surgery, such as curettage or cesarean delivery . The differential diagnosis is difficult and includes incomplete abortion, placenta previa, carcinoma of the cervix, and a degenerating leiomyoma. Although cervical ectopic pregnancies previously have been associated with a high mortality because of massive hemorrhage, now, with better methods of diagnosis and treatment, death is rare. In the past, more than half of women with a cervical pregnancy required a hysterectomy for treatment, and this was nearly always necessary if the pregnancy had advanced beyond 18 weeks. There have been several case reports in which a cervical pregnancy was successfully treated by systemic methotrexate. Other case reports have shown that after angiographic uterine artery embolization, evacuation of the pregnancy can be easily performed transcervically with minimal blood loss. Transvaginal ultrasound–guided injections of potassium chloride directly into the gestational sac have successfully terminated pregnancies, as has the local injection of methotrexate, with or without uterine artery embolization.
Cesarean scar ectopic pregnancy is a rare but potentially serious complication of early pregnancy. In this type of ectopic pregnancy, the gestational sac is located in the previous cesarean scar and is surrounded by myometrium and connective tissue ( Fig. 17.9 ). It occurs in about 1 in 2000 pregnancies and 6% of ectopic pregnancies among women with previous cesarean deliveries ( ). In a large series of 268 cesarean scar ectopic pregnancies, Sadeghi and colleagues reported four cases of uterine rupture, including one case of fetomaternal death at 38 weeks’ gestation ( ). The incidence does not appear to correlate with the number of cesarean deliveries. It is believed that the mechanism for implantation is related to the migration of the embryo through a small defect in the previous incision site or a microscopic fistula within the scar. Adenomyosis, in vitro fertilization, previous dilation and curettage, and manual removal of the placenta are also reported as risk factors.
Another uncommon form of ectopic gestation is combined intrauterine and extrauterine (heterotopic) pregnancy (94% tubal and 6% ovarian). Heterotopic pregnancy is traditionally considered a rare occurrence, with an incidence between 1 in 16,000 or 1 in 30,000 pregnancies. With the increase of use of ovulation-inducing agents and ART, the overall incidence of heterotopic pregnancy has risen to approximately 1 in 3900 pregnancies. In a series of all registered ART pregnancies in the United States from 1999 to 2002, the incidence of heterotopic pregnancy was 1.5 per 1000 ART pregnancies . The incidence has been reported to be higher when tubal damage was present or when four or more embryos were transferred.
A chronic ectopic pregnancy occurs when the intraperitoneal hemorrhage associated with tubal abortion or rupture is relatively minor and ceases spontaneously, but the ectopic gestation neither resolves completely nor implants and continues to develop as an abdominal pregnancy. The trophoblast continues to secrete HCG in small amounts, with the circulating levels less than 1000 mIU/mL in 50% of cases and less than 100 mIU/mL in 20% of cases. In one series, about 6% of all surgically treated ectopic pregnancies in one institution were classified as chronic. The most common (72%) gross pathologic finding was dense adhesions produced by the inflammatory response to the trophoblast. These adhesions attach omentum and bowel to the site of the ectopic pregnancy. In one-third of the cases, a collection of clotted blood or old hematoma was present. It has been reported that because of the extensive disease, it was necessary to perform a hysterectomy in 25% of cases, and for chronic ovarian ectopic pregnancies, an oophorectomy was necessary in 60% of patients.
Histopathology
When the morula implants in the tube, it does not grow mainly in the tubal lumen as has been assumed for many years. A review of the pathology of tubal gestation found that after implanting on the mucosa of the endosalpinx, the trophoblast invaded the lamina propria and then the muscularis of the tube and grew mainly between the lumen of the tube and its peritoneal covering. Growth occurred both parallel to the long axis of the tube and circumferentially around it. As the trophoblast invades vessels, retroperitoneal tubal hemorrhage occurs that is mainly extraluminal but may extrude from the fimbriated end and create a hemoperitoneum before tubal rupture ( Fig. 17.10 ) ( ).
The stretching of the peritoneum covered by this hemorrhage results in episodic pain before the final perforation into the peritoneal cavity. Rupture occurs when the serosa is maximally stretched, producing necrosis secondary to an inadequate blood supply.
Hemoperitoneum is nearly always found in advanced ruptured ectopic pregnancy other than that which is cervical in origin. Usually there is a combination of clotted and unclotted blood in the peritoneal cavity. The unclotted blood does not clot on removal from the peritoneal cavity because it originates from lysis of blood that has previously coagulated, similar to what occurs during menstrual bleeding. The hematocrit value of this nonclotting blood is nearly always greater than 15%, such a finding being reported in 98% of specimens obtained by culdocentesis in a series of ectopic pregnancies. Historically, at the time of laparotomy for a ruptured ectopic pregnancy, about half of the women have less than 500 mL of hemoperitoneum, one-fourth between 500 and 1000 mL, and one-fifth more than 1000 mL.
When the tube is removed and examined histologically, inflammatory cells are nearly always seen . These include plasma cells, lymphocytes, and histiocytes. The presence of chorionic villi, which are often degenerated or hyalinized, and nucleated red cells establish the diagnosis of ectopic pregnancy. Decidual reaction in the tube is uncommon.
Because of limited space or inadequate nourishment, the trophoblastic tissue of most ectopic pregnancies does not grow as rapidly as that of pregnancies within the uterine cavity . As a result, HCG production does not increase as rapidly as in a normal pregnancy, and although steroid production of the corpus luteum is initiated, elevated progesterone levels cannot be maintained. Thus initially the endometrium becomes decidualized because of continued progesterone production by the corpus luteum. Sometimes the secretory cells of the endometrial glands become hypertrophied with hyperchromatism, pleomorphism, and increased mitotic activity, as originally described by Arias-Stella ( Fig. 17.11 ). The Arias-Stella reaction can be confused with neoplasia, but it is not unique for ectopic pregnancy because it can occur with an intrauterine pregnancy (IUP) and after ovarian stimulation with clomiphene citrate. In a histologic study of the endometrium in 84 women with ectopic pregnancies, 40% of cases had secretory endometrium, with the remainder being about equally divided among the findings of proliferative endometrium, decidual reaction, and Arias-Stella reaction. When progesterone levels fall as a result of insufficient HCG, endometrial integrity is no longer maintained and it breaks down, producing uterine bleeding. Sometimes nearly all the decidua is passed through the cervix in an intact way, producing a decidual cast that may be clinically confused with a spontaneous abortion ( Fig. 17.12 ).
Symptoms
Among women with risk factors for ectopic pregnancy, with the use of early hormonal testing and vaginal sonography, it is now often possible to establish the diagnosis of ectopic pregnancy before symptoms develop. However, symptoms often develop when intraperitoneal bleeding occurs from extrusion of blood through the fimbriated end of the tube in cases of tubal pregnancies or from disruption of overlying tubal, ovarian, or myometrial tissue from rupture of the gestational sac.
The most common symptoms of ectopic pregnancy are abdominal pain, absence of menses, and irregular vaginal bleeding. Abdominal pain is nearly a universal symptom of intraperitoneal bleeding, but its characteristics are similar with different causes of bleeding. Before rupture occurs, the pain may be characterized as only a vague soreness or be colicky in nature. Its location may be generalized, unilateral, or bilateral. Shoulder pain occurs in about one-fourth of women with ruptured ectopic pregnancy as a result of diaphragmatic irritation from the hemoperitoneum. During rupture of the tube, the pain usually becomes intense. Syncope occurs in about one-third of women with tubal rupture. Other symptoms that occur after tubal rupture include dizziness and an urge to defecate.
The majority of women with ectopic pregnancy fail to have menses at the expected time but have one or more episodes of irregular vaginal bleeding when the decidual endometrial tissue is sloughed. The interval of amenorrhea is usually 6 weeks or more. The bleeding is usually characterized as spotting but may simulate menstrual bleeding. It is rarely as heavy as that which occurs in spontaneous abortion. About 5% to 10% of women with an advanced ectopic pregnancy will note passage of a decidual cast, as noted previously (see Fig. 17.12 ).
Signs
The most common presenting sign in a woman with symptomatic ectopic pregnancy is abdominal tenderness, which, together with adnexal tenderness elicited at the time of the bimanual pelvic examination, is present in nearly all women with an advanced or ruptured ectopic pregnancy. It is possible to palpate an adnexal mass in half of the women, and about one-third have some degree of uterine enlargement that is nearly always smaller than a normal 8-week intrauterine gestation except when an interstitial gestation is present. Tachycardia and hypotension can occur after rupture if blood loss is profuse, but temperature elevation is an uncommon finding, present in only about 5% to 10% of women with tubal rupture, and is rarely greater than 38° C.
Differential diagnosis of symptomatic ectopic pregnancy
The diagnosis is usually obvious for women with the classic symptoms of ruptured ectopic pregnancy: a history of irregular bleeding followed by sudden onset of pain and syncope accompanied by signs of peritoneal irritation. However, before rupture the symptoms and signs are nonspecific and may also occur with other gynecologic disorders. Entities often confused with ectopic pregnancy include salpingitis, threatened or incomplete abortion, ruptured corpus luteum, appendicitis, dysfunctional uterine bleeding, adnexal torsion, degenerative uterine leiomyoma, and endometriosis.
In the past, studies have found that women with an ectopic pregnancy were seen multiple times before a correct diagnosis was made. Because of the possibility of a fatal outcome from undiagnosed ruptured ectopic pregnancy, it is essential that the diagnosis of ectopic pregnancy be considered in any woman of childbearing age with abdominal pain and irregular uterine bleeding even if she has had a previous tubal sterilization procedure or is using an effective method of reversible contraception.
Ectopic pregnancy should be suspected in any woman who develops the symptoms listed earlier, particularly if she has previously had a pelvic operation, especially tubal surgery, either a tubal reconstructive procedure or a sterilization procedure. Other risk factors include one or more episodes of salpingitis, a previous ectopic gestation, current use of a progesterone-releasing IUD, use of a progestin-only oral contraceptive, use of pharmacologic methods of ovulation induction, and a history of infertility. In any woman with the symptoms of ectopic gestation, the diagnosis is facilitated by a quantitative assay for HCG and pelvic ultrasonography and can be established and treated by laparoscopy or laparotomy. Culdocentesis and measurement of serum progesterone levels have also been used for diagnostic assistance. Before the development of pelvic vaginal ultrasound, the finding of nonclotting blood at the time of culdocentesis, especially if the hematocrit was more than 15%, was of great assistance in establishing the diagnosis of ruptured ectopic pregnancy. With the use of high-resolution pelvic ultrasound, the presence of intraperitoneal fluid can be easily visualized and culdocentesis is no longer routinely done.
Procedures used for the diagnostic evaluation of the asymptomatic or mildly symptomatic woman with suspected ectopic pregnancy
Human chorionic gonadotropin
About 85% of women with ectopic pregnancy have serum HCG levels lower than those seen in normal pregnancy at a similar gestational age. However , a single quantitative HCG assay cannot be used to diagnose ectopic pregnancy because the actual dates of ovulation and conception are often not known. Even if the date of ovulation is known, 2.5% of women with normal gestations will have HCG levels lower than the normal 95% confidence limits. Furthermore, low HCG levels are also found in women with various stages of spontaneous abortion, conditions that must be considered in the differential diagnosis. Intact HCG and free β-HCG levels were measured in a large group of women in early pregnancy who presented with symptoms of ectopic pregnancy. Although mean levels of intact HCG and free β-HCG were significantly lower in the group of women with ectopic pregnancy and those who aborted than in those with viable intrauterine pregnancies, the individual HCG levels among the three conditions overlapped too much to devise a cutoff level for diagnostic purposes ( ).
Fig. 17.13 , as constructed by Barnhart, shows the expected changes (increases) in HCG levels in women with an intrauterine pregnancy and in spontaneous abortion . Ninety-nine percent of normal intrauterine pregnancies have an increase of at least 53% in 2 days, which is less than the rise that was previously accepted (approximately 66%). This rate of increase should be similar in single or multiple gestations ( ). In women with an ectopic pregnancy, the rate of rise in HCG can mimic an intrauterine pregnancy 21% of the time and can mimic a spontaneous abortion 8% of the time . Note the overlap in this increase or decrease in HCG levels as depicted in Fig. 17.13 .