Overview of the Management of Tubal Pregnancy

As a result of technologic advances in ultrasonography and immunoassays, our ability to diagnose ectopic pregnancy continues to improve. Combining these modalities allows for early diagnosis and intervention. Medical and conservative surgical treatment of the early ectopic pregnancy can preserve reproductive performance and minimize morbidity. Physicians caring for women of reproductive age need to remain abreast of the new management principles for this condition.

Extrauterine or ectopic pregnancy is defined as a gestation in which the site of nidation occurs in the fallopian tube, ovary, abdomen, or cervix. Tubal pregnancy accounts for 97.7% of ectopic pregnancies. Of tubal pregnancies, the ampulla is the most common site of implantation (80%); 12% of cases are located in the isthmus, 5% in the fimbria, and 2% in the cornua.¹

The incidence of ectopic pregnancy increased fourfold between 1970 and 1987, from 4.5 cases to 16.8 cases per 1000 pregnancies.^2,3 There were 17,800 ectopic pregnancies reported in 1970 and 88,000 in 1987. Although nonwhites consistently have higher rates than whites, both rates show an increase: fourfold for whites (from 4 per 1000 pregnancies in 1970 to 15.4 per 1000 in 1987) and threefold for nonwhites (from 7.2 per 1000 pregnancies in 1970 to 21 per 1000 in 1987). The rate of ectopic pregnancy increases with age regardless of race and is highest for women 35 to 44 years of age.

In contrast to the rising ectopic pregnancy rates, the case-fatality rate has decreased by 95%, from 35.5 per 10,000 ectopic pregnancies in 1970 to 3.4 per 10,000 in 1987.⁴ Women at greatest risk of dying from ectopic pregnancy are those who present with tubal rupture and hemoperitoneum, which occurs in less than 10% of cases.⁵ In patients amenable to conservative treatment with laparoscopy or methotrexate, the mortality rate may be lower. In the United States, ectopic pregnancy accounts for 30 to 40 deaths per year. It is the leading cause of pregnancy-related deaths in the first trimester and accounts for 9% of all pregnancy-related deaths.

As reported by the Centers for Disease Control and Prevention in 1997,⁶ ectopic pregnancy after tubal sterilization is more common than previously thought. Among 10,685 women studied, the risk of ectopic pregnancy within 10 years after sterilization was about 7 per 1000 procedures.

The factors involved in this worldwide increase in incidence have not been fully elucidated. The dramatic increase in sexually transmitted diseases, which began in the United States in the 1960s, is thought to be the major contributing factor. In developed countries, early detection and intervention resulting from the wide availability of sensitive serum human chorionic gonadotropin (hCG) immunoassays and advances in ultrasound also account, in part, for the increase in the reported ectopic pregnancy rate. Any factor that leads, directly or indirectly, to a reduction in tubal motility increases the risk for tubal pregnancy. Acute salpingitis, salpingitis isthmica nodosa, tubal adhesions, tubal surgery, and failed tubal sterilization surgery account for most cases. The formation of proximal tuboperitoneal fistulas, allowing sperm to enter through the distal fimbriated stump, is the most likely mechanism for pregnancy after tubal ligation. Independent risk factors consistently shown to increase the risk are (1) previous laparoscopically proven pelvic inflammatory disease; (2) previous tubal pregnancy or illegal abortion⁷ (Table 1); (3) previous tubal surgery⁷ (Table 2); and (4) certain contraceptive failures⁷ (Table 3). Other rare risk factors include cigarette smoking, ovulation induction, and in utero exposure to diethylstilbestrol (DES)⁷ (Table 4). Transmigration of the ovum may also be a factor because in about one fourth of tubal pregnancies, the corpus luteum is found on the contralateral ovary.⁸

TABLE 1. Previous Pregnancies and Subsequent Probability of Having an Ectopic Pregnancy

(Modified from Stovall TG, Ling FW (eds): Extrauterine Pregnancy: Clinical Diagnosis and Management, pp 27–63, 219–229, New York, McGraw-Hill, 1993)

TABLE 2. Previous Tubal Surgery and Probability That a Subsequent Pregnancy Will Be Ectopic

(Modified from Stovall TG, Ling FW (eds): Extrauterine Pregnancy: Clinical Diagnosis and Management, pp 27–63, 219–229, New York, McGraw-Hill, 1993)

TABLE 3. Percentage of Contraceptive Failures That Result in Ectopic Pregnancy

(Modified from Stovall TG, Ling FW (eds): Extrauterine Pregnancy: Clinical Diagnosis and Management, pp 27–63, 219–229, New York, McGraw-Hill, 1993)

TABLE 4. Various Risk Factors and Probability That a Subsequent Pregnancy Will Be Ectopic.

(Modified from Stovall TG, Ling FW (eds): Extrauterine Pregnancy: Clinical Diagnosis and Management, pp 27–63, 219–229, New York, McGraw-Hill, 1993)

Advances during the past 20 years have resulted in the earlier diagnosis of ectopic pregnancy. The availability of a highly specific immunoassay for hCG and of high-resolution ultrasonography has resulted in the ability to diagnose ectopic pregnancy in its earliest stages.

Human Chorionic Gonadotropin and Ultrasound

In normal pregnancy, hCG is secreted by the fertilized ovum by postconception day 6 and is detectable in maternal serum by day 11. Peak serum hCG levels (median, 108,000 mIU/mL International Reference Preparation) are reached by postconceptional week 8 (10 weeks after last menstrual period). Mean hCG doubling time is 2.2 days at 4 weeks and 3.5 days at 9 weeks. After 12 weeks, hCG levels begin to fall, reaching 25% of first-trimester peak values by 16 weeks. Ultrasound findings strongly correlate with gestational age and hCG levels. An intrauterine gestational sac in a normal pregnancy can be seen by transvaginal ultrasound at 34.8 (±2.2) days, at which time an hCG level of 1398 (±155) mIU/mL is reached. A fetal pole can be seen and an hCG level of 5113 (±298) is reached at 40.3 (±3.4) days. Fetal cardiac activity can be seen and an hCG level of 17,208 (±3772) is reached at 46.9 (±6) days.⁹ Institutional threshold variations exist.

Kadar and colleagues¹⁰ are credited with the concept of the discriminatory hCG zone, defined as a range of serum hCG concentration above which a normal intrauterine gestation can be visualized by ultrasound and below which an intrauterine sac usually is not visible. These authorities reported that 94% of intrauterine pregnancies can be visualized by transabdominal ultrasound when the serum hCG level is 6000 to 6500 mIU/mL (Second International Standard) and by vaginal probe ultrasound when the serum hCG level is 1200 to 2000 mIU/mL.

When the hCG value is greater than the institutional threshold value for visualizing an intrauterine gestational sac, and ultrasound reveals no such sac, the diagnosis of ectopic pregnancy virtually is assured in cases in which recent completed abortion is clinically unlikely (sensitivity and specificity more than 97%; positive predictive value and negative predictive value more than 98%).¹¹ The sensitivity and specificity of ultrasonography alone for the detection of ectopic pregnancy are 69% and 99%, respectively.¹²

When the serum hCG concentration is less than the threshold value, and the patient is stable, the use of serial hCG measurements with or without serial ultrasound can help differentiate intrauterine from extrauterine pregnancy. Kadar and associates¹³ reported that, in early gestation, nonviable pregnancy can be predicted by an abnormal rise in the serum hCG level during a 48-hour period. Using the cutoff of a 66% increase in hCG over 2 days, a 15% false-positive rate (i.e., 15% of normal pregnancies fail to have this rise) and a 13% false-negative rate (i.e., 13% of ectopic pregnancies have this normal rate of rise in hCG) is seen. Of 25 presurgical patients with ectopic pregnancy, Cartwright and DiPietro¹⁴ reported that 20 patients showed a plateau or decrease in serum hCG levels during a 48-hour or longer period.

The presumption that ultrasonographic visualization of an intrauterine gestational sac effectively excludes the presence of an ectopic pregnancy is based on the rare clinical finding of combined ectopic and intrauterine pregnancy. In 1948, the calculated incidence of heterotopic pregnancy was 1:30,000, based on the incidence of ectopic pregnancy and dizygotic twinning rates at the time.¹⁵ Because of substantial increases in both rates since that time, the current estimated incidence of heterotopic pregnancy ranges from 1:3,889 to 1:15,000.¹⁶ The diagnosis of combined pregnancy is suggested in patients who continue to complain of symptoms of ectopic pregnancy after an intrauterine pregnancy has been confirmed.¹

Culdocentesis

The presence of intraabdominal blood can be determined rapidly by placing a needle through the posterior fornix and aspirating contents within the rectouterine pouch of Douglas. In conjunction with a positive pregnancy test in the setting of acute pelvic pain, culdocentesis has a high positive predictive value but a low negative predictive value.¹⁷ As a result, culdocentesis is of limited diagnostic value.

Presence or Absence of Villi at Dilation and Curettage

The role of dilation and curettage (D&C) in the diagnosis of ectopic pregnancy has been abandoned by many authorities.¹⁸ Historically, when serum progesterone levels were lower than 5 ng/mL, or when the patient had abnormally rising serum hCG titers and no intrauterine pregnancy noted on transvaginal ultrasound, a D&C was performed to rule out the presence of villi. Because no villi were obtained in any patient with hCG titers greater than 2000 mIU/mL and an empty uterus or cardiac activity was seen in the adnexa by ultrasound, Stovall and colleagues¹⁸ no longer perform a D&C for diagnostic purposes in these patients and instead proceed directly to treatment.

Progesterone

The serum progesterone level reflects production by the corpus luteum stimulated by a viable pregnancy.¹⁹ During the first 8 to 10 weeks of gestation, serum progesterone concentrations change little; as pregnancy fails, however, the levels decrease.²⁰ A serum progesterone level of at least 25 ng/mL (79.5 nmol/L) excludes ectopic pregnancy with 97.5% sensitivity, obviating the need for further testing. A single level of no more than 5 ng/mL (15.9 nmol/L) indicates nonviability with 100% sensitivity, regardless of location, and permits diagnostic uterine curettage to distinguish ectopic pregnancy from spontaneous intrauterine abortion. This protocol can be useful when ultrasound is not available. When progesterone levels are between 5 and 25 ng/mL, the diagnosis must be established by methods previously mentioned.

Conservative surgical treatment of ectopic pregnancy is well established, and laparoscopic salpingostomy is the preferred operative method in unruptured cases. Nonsurgical therapy for ectopic pregnancy, however, may prevent undesired postoperative adhesions that often result from surgical manipulation of the fallopian tubes.²¹ Despite the theoretical benefits of medical management, the notion that early tubal pregnancy disruption without surgery will ultimately increase fertility potential remains speculative.¹ This section considers the current roles of expectant, medical, and surgical management of ectopic pregnancy.

Observation

Some patients with ectopic pregnancies undergo spontaneous absorption and require no therapy.²² In 1955, Lund²³ reported on 119 women with unruptured tubal pregnancies who were treated with bed rest and frequent observation. Of these 119 patients, 68 (57%) were eventually discharged from the hospital without operation, but 60% required hospitalization for more than 1 month. The remainder required operative intervention for tubal rupture or worsening clinical course. Fernandez and Rainhorn²⁴ reported that, of 14 women with small, unruptured, laparoscopically diagnosed ectopic pregnancies who were managed expectantly, four required subsequent reoperation. Whether a rigid and expensive hospital protocol that includes serial ultrasound examinations and hCG measurements results in better subsequent pregnancy rates than removing the ectopic pregnancy by salpingostomy at first laparoscopy awaits further study. At present, it is considered better to remove an unruptured ectopic pregnancy at the time of first laparoscopy to avoid the additional expense of hospitalization, serial hCG assays, and a second laparoscopy.²⁵

Medical Management

Nonsurgical therapy for ectopic pregnancy may prevent the undesired postoperative adhesions that often result from surgical manipulation of the fallopian tubes. The first case report describing the use of medical therapy for tubal pregnancy appeared in 1982.²⁶ Observational studies in the mid-1980s used, with varying success, methotrexate, prostaglandins, dactinomycin, etoposide, hyperosmolar glucose, anti-hCG antibodies, potassium chloride, and mifepristone (RU 486).¹⁹ Although treatments given systemically have proved practical, several of these agents also have been injected locally into the ectopic gestational sac under laparoscopic or ultrasound guidance or by hysteroscopic intratubal cannulation.

METHOTREXATE.

Methotrexate, a folic acid antagonist, inhibits dihydrofolate reductase, an enzyme necessary for nucleic acid synthesis, and thereby interferes with DNA synthesis and cell multiplication in actively dividing tissue. The efficacy of methotrexate in the treatment of gestational trophoblastic disease made it an attractive candidate for chemotherapeutic use in ectopic pregnancy. Patients treated with methotrexate should be counseled to refrain from alcohol consumption and intercourse and to avoid use of vitamin preparations containing folic acid until complete resolution of the ectopic pregnancy.

Systemic Methotrexate.

The efficacy of methotrexate in the treatment of unruptured ectopic pregnancy using multiple-dose regimens had been well documented.^27,28,29 In 1989, Stovall and associates³⁰ reported on the use of an individualized, multidose regimen in which 1 mg/kg of methotrexate was given intramuscularly on alternating days, with 0.1 mg/kg of leucovorin given on the intervening days until the β-hCG level had dropped by at least 15% in 48 hours or until four doses of methotrexate had been given. Using this regimen, 15% to 20% of patients required only a single methotrexate dose. In 1991, Stovall and colleagues¹⁸ reported on the efficacy of single-dose methotrexate in the treatment of ectopic pregnancy. Of 30 hemodynamically stable patients diagnosed with an unruptured ectopic pregnancy less than 3.5 cm in greatest dimension who desired future fertility, 29 (96.7%) were successfully treated with a single dose of methotrexate (50 mg/m² given intramuscularly) without leucovorin rescue (Table 5). Mean time to resolution of ectopic pregnancy was 36 days. No patient experienced any chemotherapeutic side effects. Six of 30 (20%) reported an increase in lower abdominal pain between days 5 and 10, and 2 required hospitalization overnight for observation. Five of 6 (83%) ectopic pregnancies with cardiac activity were successfully treated with this protocol. The one treatment failure, a patient whose ectopic pregnancy had cardiac activity, required surgical intervention on day 6 because of tubal rupture. Cardiac activity can be identified much earlier in ectopic gestation because of recent advances in transvaginal ultrasonography; hence, the finding of adnexal cardiac activity is now considered a relative contraindication to medical therapy. Caution should be exercised when medically managing ectopic pregnancies with cardiac activity.

TABLE 5. Protocol For Single-Dose Methotrexate For Treatment of Unruptured Ectopic Pregnancy

hCG, quantitative serum β-human chorionic gonadotropin level; CBC, complete blood count; SGOT, serum glutamate-oxaloacetate transaminase; MTX, intramuscular methotrexate, 50 mg/m².
(Modified from Stovall TG, Ling FW, Gray LA: Single-dose methotrexate for treatment of ectopic pregnancy. Obstet Gynecol 1991;77:754)

Critics of medical therapy for ectopic pregnancy point to the need for prolonged follow-up. The follow-up described by Stovall and colleagues, however, is similar to the follow-up hCG monitoring required after any conservative surgical procedure for ectopic pregnancy.

Methotrexate by Direct Injection.

The advantages cited for the direct instillation of methotrexate, laparoscopically or transvaginally under ultrasound guidance, include higher local drug concentrations, less systemic distribution, a smaller therapeutic dose, and less toxicity. Despite the theoretical advantages of direct injection of methotrexate, success rates in practice appear unacceptably low, whereas other outcomes (tubal patency and subsequent fertility) are similar to those seen with systemic methotrexate¹⁹ (Table 6).

TABLE 6. Summary of Reported Success Using Directly Injected and Systemic Methotrexate for Unruptured Ectopic Pregnancy and Subsequent Fertility, 1987-1992

(Modified from Carson SA, Buster JE: Current concepts: Ectopic pregnancy, N Engl J Med 1993; 329:1174)

Wolf and Witt²¹ reported on the outpatient use of laparoscopically directed injection of methotrexate in nine patients with unruptured tubal pregnancies measuring less than 4 cm. Fifteen milligrams (0.6 mL of 25 mg/mL solution) of methotrexate was injected through a 15-cm, 25-gauge spinal needle, using a three-puncture technique to disperse the drug. Daily telephone contact was made with each patient. Eight patients (89%) had complete resolution of the pregnancy in 10 to 33 days; the ninth patient required surgical intervention. Of the five patients with initial hCG levels greater than 1000 mIU/mL, four were treated successfully. The failure among the study patients had an initial hCG level of 17,000 mIU/mL. Wolf and Witt,²¹ and other authorities,³¹ have suggested that ectopic pregnancies with higher hCG values may be less likely to respond to this treatment, although a patient has been reported who had a serum hCG level of 122,000 mIU/mL and a fetus with cardiac activity who responded to vaginal ultrasound-guided intraamniotic methotrexate injection.³²

Methotrexate by Tubal Cannulation.

In a multicenter trial, Risquez and colleagues³³ reported resolution of 27 of 31 cases by the instillation of methotrexate using hysteroscopically directed tubal cannulation. The remaining four patients ultimately required surgery. Although encouraging, this approach appears to have no major advantage over other methods, except that methotrexate can be instilled into very small ectopic sacs without actually visualizing them.

PROSTAGLANDIN BY DIRECT INJECTION.

Prostaglandins cause strong tubal muscular contractions and local vasoconstriction. In 1987, Lindblom and associates³⁴ reported treating nine women with small unruptured tubal pregnancies with 0.5 to 1.5 mg of prostaglandin F_2α injected into the affected tube and corpus luteum. The preliminary experience was promising, but a larger study later failed to confirm the results. Moreover, cardiac arrhythmia, cardiopulmonary edema, and gastrointestinal symptoms have been reported after intratubal injection of prostaglandin.³⁵

ELIGIBILITY FOR MEDICAL MANAGEMENT.

Methotrexate is contraindicated in patients who are hemodynamically unstable and in those who have severe anemia, renal insufficiency, active liver disease, leukemia, bone marrow abnormalities, or allergy to methotrexate. Patients with white blood cell counts of less than 4000/mL, hematocrit less than 26%, bilirubin greater than 1.2 mg/dL, aspartate aminotransferase or alanine aminotransferase greater than 70 IU/dL, and serum creatinine levels of 1.4 mg/dL or greater should be excluded from such management.

As mentioned previously, caution should be exercised when medically managing ectopic pregnancies greater than 4 cm or with cardiac activity. Some authorities advise that methotrexate be used only in patients who desire future fertility.

Surgical Management

HISTORICAL.

Before the development of safe and effective surgical management, ectopic pregnancy was managed expectantly, with a mortality rate of 69%.³⁶ In 1884, Tait³⁷ reported on a series of five patients who underwent salpingectomy for the treatment of tubal pregnancy. In 1897, Martin³⁸ reported removing a tubal pregnancy by opening the tube and shelling out its contents, with closure of the incision. No follow-up was provided. Prochownick,³⁹ in 1894, described a similar operation in a residual tube; 2 years later, this patient conceived and delivered at term.

Despite these case reports, extirpative surgery remained the preferred procedure in most cases of ectopic pregnancy. In 1962, Grant⁴⁰ advocated the use of salpingotomy in the treatment of tubal pregnancy only when the contralateral tube was “absent or blocked.” In his study of 141 patients who had undergone conservative surgery for a pregnancy in the remaining tube, 27 (19%) achieved intrauterine pregnancies, and 9 (6%) developed subsequent tubal pregnancies. At the time, this represented an unacceptably high risk of ectopic pregnancy; in 1962, the risk of ectopic pregnancy was 0.03% in the general population and 2.4% among infertility patients.

In 1967, Timonen and Nieminen⁴¹ studied 1085 patients with ectopic pregnancies and noted a slightly higher term pregnancy rate after nonconservative surgery (27.2% versus 23.8%) as well as a decreased risk of recurrent ectopic pregnancy (11.5% versus 15.7%). These investigators concluded that nonconservative therapy was justified in all cases except in “childless women with the contralateral tube severely damaged or completely destroyed.” A high tubal occlusion rate was thought to occur after conservative therapy.

CONSERVATIVE VERSUS NONCONSERVATIVE SURGERY.

A shift in attitude was slow until improved results of conservative surgery were well documented. In 1979, DeCherney and Kase⁴² compared conservative surgery and nonconservative surgery and found no difference in the term viable pregnancy rate (40%). Conservative therapy, however, did not increase the repeat ectopic pregnancy rate (8%). Although there was no improvement in outcome, these investigators strongly urged conservative therapy whenever possible. This lack of apparent benefit is particularly true in the presence of a normal contralateral tube. Langer and coworkers⁴³ correlated pregnancy outcome with contralateral tubal condition. Although the ratio of intrauterine to ectopic pregnancy after conservative surgery is about 6:1, it rises to more than 10:1 in patients with normal contralateral tubes. Sherman and associates⁴⁴ performed a retrospective study and noted that postoperative infertility was associated significantly with coexistent periadnexal or tubal disease as well as with a history of infertility, advanced maternal age, and tubal rupture. These investigators demonstrated that among those patients with the previously listed risks factors, conservative surgery significantly increased the chances of achieving intrauterine pregnancy (76% versus 44%) compared with extirpative surgery. Among those patients with an otherwise noncontributory reproductive history, however, the surgical approach did not alter subsequent pregnancy potential.

By the 1980s, conservative surgery had become the preferred procedure for unruptured ectopic pregnancies. As larger series became available for scrutiny, however, it became apparent that the type of procedure selected for tubal pregnancies, whether conservative or extirpative, had little correlation with fertility outcome. Thorburn and colleagues⁴⁵ reported on a series of 148 patients treated for ectopic pregnancy. Although the overall subsequent birth rate was 53.6%, a history of infertility, previous abdominal surgery, previous curettage, previous ectopic pregnancy, and the presence of adnexal adhesions at surgery significantly reduced the subsequent live birth rate. Choice of surgical management appeared to have no impact on subsequent outcome. Tuomivaara and Kauppila⁴⁶ published results on a series of 323 patients desiring pregnancy after an ectopic pregnancy. Nulliparous patients had a significantly lower conception rate than parous women (74% versus 86%). Nulliparous women also had a higher incidence of repeat ectopic pregnancy than multiparous patients (22% versus 9%). Again, it was concluded that surgical management did not influence subsequent fertility. Querleu and Boutteville⁴⁷ stratified 473 patients on the basis of parity and history of infertility. They found that the intrauterine pregnancy rates were higher among conservatively treated patients, but the difference was dramatic when the infertile nulliparous patients were compared with nulliparous patients without a history of infertility (24.1% versus 41.4%). Again, nulliparous patients had a higher incidence of repeat ectopic pregnancy than multiparous patients, irrespective of the surgical approach (22% versus 9%). Pouly and associates⁴⁸ studied 223 patients treated laparoscopically and found that a history of infertility, ipsilateral adhesions, and poor contralateral tubal status reduced the subsequent conception rates and were associated with a higher incidence of recurrence. When a scoring system was applied (Table 7), the subsequent intrauterine pregnancy rate varied inversely with the score, whereas the risk of recurrence varied directly. Based on this scoring analysis, treatment recommendations at the time of initial ectopic pregnancy favored a conservative approach when risk factor analysis was deemed low and favored extirpative surgery with contralateral sterilization followed by in vitro fertilization (IVF) when risk factor analysis was considered high.

TABLE 7. Statistical Weight of the Risk Factors and Therapeutic Score of the Ectopic Pregnancy

Score 0 to 3: laparoscopic conservative treatment
Score 4: laparoscopic salpingectomy
Score 5 or more: laparoscopic salpingectomy and contralateral sterilization
*If the ectopic pregnancy occurred in both tubes, just count “solitary tube.”
†Only one is taken in count.
‡If the tube is blocked or absent, count “solitary tube.”
(Modified from Pouly JL, Chapron C, Mannes H, et al: Multifactorial analysis of fertility after conservative laparoscopic treatment of ectopic pregnancy in a series of 223 patients. Fertil Steril 1991;56:453)

The conservative technique described in many early reports was salpingotomy, which requires closure of the tubal incision. The currently preferred surgical approach is salpingostomy, first described by Tompkins in 1956,⁴⁹ whereby the linear incision is left open. McComb and Gomel⁵⁰ demonstrated in rabbits that healing by secondary intention decreases the risk of secondary tubal obstruction and facilitates better approximation of the mucosal folds.

CONCOMITANT OOPHORECTOMY.

Jeffcoate⁵¹ advocated salpingo-oophorectomy instead of salpingectomy alone, reasoning that if ovulation occurs on the side of a previous ectopic pregnancy, transmigration is more apt to result in recurrent ectopic pregnancy. In light of current IVF availability and the potential for a higher yield of oocytes at retrieval with two ovaries present, concomitant oophorectomy for ectopic pregnancy cannot be supported.

PREOPERATIVE CONSIDERATIONS.

With the availability of sensitive and rapid serum hCG determinations and high-resolution ultrasonography, the initial presentation of ectopic pregnancy more often has shifted from an acute emergency to early diagnosis during the prodrome. Nevertheless, when a patient presents in the emergency room with an acute abdomen and hemodynamic instability, she must be expeditiously managed, limiting therapeutic options and the opportunity to review them with the patient.

Preoperative Discussion.

In discussing surgical options during the preoperative discussion, patient expectations and the desire for future fertility should be addressed. If future child bearing is not desired, extirpative surgery should be employed. Additionally, if the patient is concerned about minimizing the risk of a subsequent ectopic pregnancy, a nonconservative approach may be indicated, with consideration given to contralateral tubal ligation. If the contralateral tube is abnormal, salpingectomy followed by IVF yields higher and earlier probability of term pregnancy than attempting spontaneous conception (30.5% versus 25.6% within 18 months).⁵² In addition to discussing the risks of the procedure, such as bleeding, infection, and potential damage to the viscera, it is important to discuss the possible need for laparotomy in the event of visceral injury, to achieve hemostasis, or if the size of the specimen exceeds the limits for laparoscopic removal.

Laparoscopy Versus Laparotomy.

The opportunity to treat ectopic pregnancy definitively at laparoscopy has expanded the role of the laparoscope from a purely diagnostic tool. Shapiro and Adler⁵³ described the first case of laparoscopic excision of an ectopic pregnancy, and Bruhat and coworkers⁵⁴ first described laparoscopic linear salpingotomy for ectopic pregnancy. If a conservative procedure by laparoscopy is planned, necessary equipment and personnel must be available. A high-flow insufflator (8 L/min CO₂), multiple operative ports, an instrument for incising the tube (laser, needle-tip electrosurgical device, or scissors), instruments for hemostasis (unipolar and bipolar electrosurgical devices), and an irrigation-aspiration instrument capable of removing large volumes of fluid with clot from the pelvis should be available at the outset. Hypovolemic shock and intestinal distention are absolute contraindications to laparoscopy, but tubal rupture in the absence of hypovolemic shock is not. Other considerations include obesity, multiple previous abdominal procedures, and a history of a bleeding diathesis.

The advantages of operative laparoscopy include less postoperative discomfort, better cosmetic results, shorter hospitalization, faster recuperation, earlier return to work, and reduced costs when compared with laparotomy. In experienced hands, the laparoscopic approach should not prolong anesthetic exposure significantly. Reproductive outcome is not significantly different between the two approaches,⁵⁵ although a conservative approach using the laparoscope has been associated with a higher risk of persistent ectopic pregnancy (9%) than seen in those managed by laparotomy (3%).⁵⁶ The surgeon should not attempt at laparoscopy what he or she would not perform at laparotomy.

OPERATIVE CONSIDERATIONS.

The selection of operation is guided by (1) the location of the pregnancy within the fallopian tube, (2) tubal integrity, (3) patient wishes with regard to future fertility, (4) the capabilities of the surgeon, and (5) the availability of special instruments. The ultimate goal in patients desiring future pregnancy is the preservation of a maximal length of functioning tube. Subsequent tubal function is affected by the condition of both tubes. Adhesion formation and tubal damage can be minimized by ensuring hemostasis, atraumatic tissue handling, prevention of serosal drying, and the use of fine, nonreactive sutures.¹

Ampullary Pregnancy.

Ampullary pregnancies are readily managed by linear salpingostomy. Before the incision, desiccation of the serosal vessels along the anticipated incisional line is performed with an electrosurgical device or defocused CO₂ laser beam. Dilute vasopressin solution (10 U in 100 mL) can be injected into the antimesenteric border to decrease bleeding. Using scissors, laser, or electrosurgery, a linear incision is made over the gestational tissue at the area of maximal distention. Myosalpingeal contraction may occur after incision, resulting in spontaneous expression of the products of conception. If spontaneous extrusion does not occur after 3 minutes, gentle external compression may be performed with the aid of forceps (Video Illustration 1). Hydrodissection also may be helpful in dislodging the gestational tissue. The surgeon must resist vigorous evacuation to avoid disruption of the underlying vascular bed. Often, the implantation bed is not immediately hemostatic. Useful measures to restore hemostasis include pressure, injection of vasopressin, needle-tip or bipolar electrocoagulation, and ultimately, suture placement in the mesosalpinx. Alternatively, mesosalpingeal ligation can be accomplished with bipolar electrocoagulation. Once hemostasis is achieved, the incision is allowed to close by secondary intention. Closing the incision may increase the risk of persistent ectopic pregnancy and subsequent adhesion formation.⁵⁷

Isthmic Tubal Pregnancy.

Segmental resection (partial salpingectomy) has been advocated in the setting of an isthmic ectopic pregnancy because salpingostomy at this site was thought to result in tubal occlusion and recurrent ectopic pregnancy rates of up to 50%.⁵⁸ It has been suggested that, in most isthmic pregnancies, implantation is extraluminal, the mucosa is generally not preserved, and the muscularis is destroyed.⁵⁹ Pauerstein and colleagues,⁶⁰ however, reviewed 25 cases of tubal pregnancy and concluded that trophoblastic spread and tubal hemorrhage were predominantly extraluminal in only 6 cases, 5 of which had ruptured. In 1986, Pouly and coworkers⁶¹ reported on the use of salpingostomy to treat isthmic pregnancies. Of 22 patients with isthmic pregnancies treated with salpingostomy, 12 (54.5%) achieved intrauterine pregnancies. The repeat ectopic pregnancy rate was 36.4% (8 patients).

Immediate tubal reanastomosis has been performed with success,⁶² although most surgeons prefer to perform an interval anastomosis after the resolution of tubal edema and the development of normal tissue planes. DeCherney and Boyers⁶³ reported on four conceptions in six patients who underwent delayed anastomosis after segmental resection, including one ectopic pregnancy in the conserved tube.

Distal Tubal Pregnancy.

Distal tubal pregnancies already undergoing spontaneous tubal abortion may be atraumatically removed from the distal end of the fallopian tube with gentle traction. Fimbrial expression, however, particularly when performed by laparoscopy, is associated with a high risk of persistent ectopic pregnancy (10% to 12%)^54,56 and with greater tissue trauma because a portion of the conceptus commonly is located within the fallopian tube wall.

Tubal Rupture.

In the face of tubal rupture and overt hemorrhage, salpingectomy may be necessary to provide rapid treatment and to prevent serious morbidity and mortality. Alternatively, segmental resection may be performed in a hemodynamically stable patient. Ipsilateral oophorectomy, as proposed by Jeffcoate,⁵¹ is not recommended unless the ovary is closely involved with the ectopic pregnancy, vascularity is compromised, or the ovary is otherwise grossly abnormal.⁶⁴

Partial salpingectomy at laparotomy is performed by desiccating or applying ligatures across the tube, both proximal and distal to the extrauterine gestation. The mesosalpinx is ligated along this portion of the tube, and the specimen excised. Laparoscopic salpingectomy is performed in a similar fashion with bipolar electrosurgery and scissors, stapling devices, or endoscopic loop ligatures.

Repeat Ectopic Pregnancy.

Repeat ectopic pregnancy may be an indication for more aggressive therapy (see Table 7). DeCherney and colleagues,⁶⁵ however, reported that 4 of 13 patients managed conservatively delivered at term after two previous ectopic pregnancies.

Tissue Removal.

Often, tissue removal poses special challenges. If small enough, the specimen can be drawn up through an expanded accessory port (10, 12, or 15 mm) with grasping forceps. If the specimen exceeds 5 cm, consideration should be made for morcellation, which can be performed after the pregnancy has been placed into an endoscopic pouch. Care must be taken to evacuate all gestational tissue from the pelvis.

POSTOPERATIVE CONSIDERATIONS

Persistent Ectopic Pregnancy.

To identify persistent ectopic pregnancy, all patients who undergo linear salpingostomy should be followed at least weekly by quantitative serum hCG levels until nonpregnant levels are attained. Residual trophoblastic tissue may continue to proliferate and lead to life-threatening hemorrhage resulting from tubal rupture. The diagnosis of persistent ectopic pregnancy is made when serum hCG levels rise or decline by less than 20% between two consecutive measurements taken 3 days apart.⁵⁶ In most cases, plateauing serum hCG levels do not require further therapy.²⁴ If serum hCG levels continue to rise, however, three treatment options are available:

1. An additional conservative procedure may be considered, but there is a strong possibility that the remaining trophoblastic tissue will be extraluminal and a second operation may again be ineffective in removing all of the trophoblastic tissue.
   
2. A salpingectomy may be performed, but the potential fertility-sparing effects of the first procedure will be lost.
   
3. Systemic methotrexate is effective in treating persistent trophoblastic tissue and, despite the absence of large series confirming its efficacy, it may represent the best clinical option.
   

Adjuvant Methotrexate Therapy.

Graczykowski and Mishell⁶⁶ reported on a nonblinded, randomized controlled series on the use of adjuvant methotrexate therapy in 116 patients undergoing conservative treatment of an unruptured ectopic pregnancy. The patients in the treatment arm received 1 mg/kg of methotrexate by intramuscular injection within 24 hours of the operative procedure. One of 54 patients (1.9%) in the methotrexate prophylaxis group and 9 of 62 (14.5%) in the control group had persistent ectopic pregnancies. The results were statistically significant (P < 0.05). The relative risk of developing persistent ectopic pregnancy after prophylactic methotrexate was 0.13 (95% confidence interval: 0.02 to 0.97). Three patients in the prophylaxis group reported side effects; one complained of mild stomatitis and two of mild gastroenteritis lasting 1 to 2 days. In all three, the side effects resolved spontaneously and did not require medical intervention.

Patients undergoing salpingectomy rarely have persistent trophoblastic activity. Routine postoperative measurement of serum hCG levels is not required.

Rh Factor.

All Rh-negative, unsensitized women with ectopic pregnancies should receive Rh immunoglobulin at a dose of 50 μg if the gestation is of less than 12 weeks' duration and 300 μg otherwise. Grimes and associates⁶⁷ reported that, because hospitalization typically is unscheduled and not preceded by Rh screening, most Rh-negative women with ectopic pregnancies in the United States do not receive Rh(D) immunoglobulin. The magnitude of the risk of sensitization is unknown but is estimated to vary from no risk at 1 month to about 9% at 3 months' gestation. Because of the potential benefits and lack of risk, administration of Rh immunoglobulin should be undertaken in all Rh-negative unsensitized women with ectopic pregnancy.²⁵

Although operative laparoscopy is associated with substantially fewer complications than laparotomy, there remains an irreducible minimal degree of morbidity intrinsic to surgery and anesthesia. Although not yet standard treatment in many centers, medical therapy can greatly reduce this morbidity, and consequently there is increasing interest in its use. To supplant surgery, however, medical therapies must match the success rates, low complication rates, and subsequent reproductive potential achieved with laparoscopic operations. This appears to have happened.¹⁹

With reliable early nonoperative diagnosis, choices for treatment include expectant, medical, or operative management. Factors influencing which choice to pursue depend on issues such as medical benefit, surgeon expertise, minimization of morbidity, costs for services, patient desire for future fertility, and efficacy rates for each approach.

The costs associated with ectopic pregnancy were estimated to surpass $1 billion in 1990 in the United States.⁶⁸ Efforts toward reducing morbidity and costs have rapidly changed the standard of care for surgically managed ectopic pregnancy from laparotomy to laparoscopy. A major driving force behind this change has been recent studies indicating a substantial cost reduction with laparoscopy.

Laparoscopy Versus Laparotomy

An analysis from Sweden reported on the cost-effectiveness of laparoscopy versus laparotomy for the treatment of laparoscopically diagnosed ectopic pregnancy.⁶⁹ Operative approach was randomly assigned after the diagnosis of ectopic pregnancy was made. Cost estimates for the two methods incorporated operative, inpatient, and follow-up care, including all costs associated with consequences of the original treatment, but excluded nonmedical economic costs, such as lost wages, sick leave benefits, and child care, which are expected to be higher for laparotomy. The initial procedure eliminated trophoblastic activity in 81% of 52 laparoscopy patients, compared with 95% of 57 laparotomy patients. Second-line therapy involved methotrexate chemotherapy or reoperation. Mean simulated costs (Swedish kroner converted to US dollar: 8 kroner equals $1) for the overall laparoscopy strategy were $3057, compared with $4087 for laparotomy (P = 0.03). The main factor for the observed cost difference was a 2.9-day longer mean postoperative inpatient stay for the laparotomy patient; this aspect alone generated an average cost difference of $786.

In a study by Foulk and Steiger,⁵⁶⁸ laparoscopy incurred a 25% cost reduction for treatment of ectopic pregnancy even when the cases in which laparoscopy was converted to laparotomy were included (P = 0.0003). When the analysis was broken down by components, the longer length of operative time for laparoscopy resulted in a 37% increase in costs associated with surgery compared with laparotomy (P < 0.001). These costs, however, were more than offset by a 60% reduction in costs associated with a shorter length of stay after laparoscopy (P < 0.001). Based on unit costs at Memorial Medical Center at Long Beach ($321 per hour for operating room use and $252 per day for floor use), the cost of about 47 minutes of surgical time is equivalent to the cost for each postoperative day the patient stays, and vice versa. Equivalent costs between laparoscopy and laparotomy are noted when discharge occurs on postoperative day 2 for the laparotomy patient. The least costly of all operative management schemes was laparotomy with discharge on postoperative day 1.

Medical Versus Surgical Management

Several studies attest to the cost-effectiveness of primary methotrexate therapy for ectopic pregnancy. Laparoscopic salpingostomy, although well accepted as treatment for small unruptured ectopic pregnancies, is associated with high costs related to operating room expense, anesthesia services, and surgeons' fees. Data on the medical management of ectopic pregnancy with methotrexate suggest that the efficacy of such an approach is equivalent to that of laparoscopic salpingostomy.^18,70 Assuming that both treatment strategies would result in identical clinical outcomes, Alexander and colleagues⁷¹ compared costs of methotrexate and laparoscopy in the treatment of ectopic pregnancy. Diagnostic costs were considered identical. Success rates for single-injection methotrexate ranged from 58% to 91%.⁷¹ Retreatment with additional methotrexate doses increased the overall nonsurgical success rate to 72% to 94%. Thus, 6% to 28% of patients initially treated with methotrexate ultimately require surgery to achieve resolution of ectopic pregnancy. For laparoscopic salpingostomy, the average initial success rate from 17 series with a total of 976 patients is 94.5% (range, 83% to 100%).^61,71 Thus, up to 17% of patients require retreatment for persistent ectopic pregnancy, pain, or bleeding. When preservation of fertility is not an issue, salpingectomy can be performed, with initial cure rates approaching 100%. Using best-case and worst-case comparisons, Alexander and colleagues⁷¹ demonstrated over a wide range of performance estimates that the initial use of methotrexate as opposed to laparoscopic salpingostomy would result in a substantial cost savings ($1124 to $2536 per patient) in the management of the small unruptured ectopic pregnancy. Based on an annual ectopic pregnancy rate of 88,000 and an estimate that 45% of all patients with ectopic pregnancy are eligible for initial treatment with methotrexate,⁷² a potential annual direct cost savings of $43 to $97 million could be achieved when methotrexate is used as initial therapy in appropriate cases. Other authorities have estimated even greater annual savings.⁷³

One measure of long-term morbidity in the patient treated for ectopic pregnancy is future reproductive potential. A review of published reports on laparoscopic salpingostomy revealed a tubal patency rate of 86%, a pregnancy rate of 66%, and a repeat ectopic pregnancy rate of 23%.⁷¹ Although fewer data are available on methotrexate use, reported outcomes are similar, with a tubal patency rate of 81% and a pregnancy rate of 70%, but with a repeat ectopic pregnancy rate of 11%, less than that seen with laparoscopy.^74,75

Future Reproductive Potential

The overall subsequent conception rate in women with ectopic pregnancies is about 60%. Slightly less than half of these pregnancies result in another ectopic or spontaneous abortion, so only about one third of women with ectopic pregnancies have subsequent live births. These general figures are modified by several factors, particularly age, parity, evidence of contralateral tubal disease, and whether the tube is ruptured. The subsequent fertility rate is significantly higher in parous women younger than 30 years. If the ectopic pregnancy is a women's first pregnancy, her subsequent conception rate is only about 35%. On the other hand, women with high parity (more than three pregnancies) who develop an ectopic pregnancy have a relatively high rate of conception (80%). The subsequent conception rate is lower in women who have a history of salpingitis and in those who have gross evidence of damage to the opposite oviduct as a result of previous salpingitis. Future fertility is significantly higher in women who have unruptured tubal pregnancies than in those who have ruptured ectopic pregnancies; hence, early diagnosis with serial hCG and ultrasound is desirable. In a study by Sherman and associates,⁴⁴ although only 65% of patients with ruptured ectopic pregnancies subsequently conceived, the conception rate in women with unruptured tubal pregnancies was 82%. In 1993, Ory and O'Brien⁷⁶ published a retrospective cohort study comparing pregnancy rates after radical or conservative surgical treatment at laparotomy in patients with their first ectopic pregnancies. All patients were actively trying to conceive and had at least 3 years of follow-up. When patients were stratified by history of infertility, no significant difference in pregnancy rate was noted between the radical and conservative treatment groups. Of the 63 patients without a prior history of infertility, 43 (68%) conceived. In contrast, only 5 of 25 (20%) patients with a history of infertility conceived (4 of 16 in the conservative group and 1 of 9 after radical treatment). Despite the inability to generalize these results to laparoscopic treatment, these investigators concluded that a prior history of previous infertility is the single most important factor influencing future fertility potential in these patients. Patients with a prior history of infertility have a poor subsequent fertility prognosis regardless of the procedure chosen and should be advised to consider IVF.

Repeat Ectopic Pregnancy

The rate of repeat ectopic pregnancy after a single ectopic pregnancy ranges from 8% to 20%, with a mean of 15%. Only about one of three nulliparous women who have an ectopic pregnancy ever conceives again (35%), and about one third have another ectopic pregnancy (13%). After two ectopic pregnancies, infertility rates as high as 90% have been reported.²⁵

In 1987, DeCherney and Diamond⁷⁷ reported on a series of 79 patients with ampullary tubal ectopic pregnancies managed with laparoscopic linear salpingostomy. Follow-up revealed that, of 69 patients actively trying to conceive, 43 (62%) conceived. Seven (16%) of these were repeat ectopic pregnancies—four contralateral and three ipsilateral to the previous ectopic pregnancy. Ten (23%) of 43 patients aborted. The viable pregnancy rate was 38% (26 of 69), comparable to that seen after other techniques.

Historically, salpingectomy was introduced as a means of reducing the high mortality associated with ectopic pregnancy. As the world experience grew, salpingotomy and then salpingostomy were heralded as equally effective in the treatment of unruptured ectopic pregnancies. Conservative procedures provide a means for retaining fertility potential in patients in whom the contralateral tube is absent. High-resolution ultrasonography and sensitive hCG determinations permit earlier diagnosis. Therapeutic laparoscopy has resulted in a reduction in operative morbidity, hospital stay, cost, and recovery time. As reports of larger series of patients become available, it is clear that the choice of surgical procedure, whether conservative or extirpative, whether by laparoscopy or laparotomy, is not a significant determinant of subsequent fertility. Pregnancy outcome after ectopic pregnancy appears to be more dependent on factors such as a history of infertility, previous abdominal surgery, and the presence of adnexal adhesions. In some patients, extirpative surgery for ectopic pregnancy with concomitant contralateral tubal ligation followed by IVF may provide the best available treatment option.