Diagnosis and management of ectopic pregnancy

Ectopic pregnancy is a high-risk condition that occurs in 1.9 percent of reported pregnancies. The condition is the leading cause of pregnancy-related death in the first trimester. If a woman of reproductive age presents with abdominal pain, vaginal bleeding, syncope, or hypotension, the physician should perform a pregnancy test. If the patient is pregnant, the physician should perform a work-up to detect possible ectopic or ruptured ectopic pregnancy. Prompt ultrasound evaluation is key in diagnosing ectopic pregnancy. Equivocal ultrasound results should be combined with quantitative beta subunit of human chorionic gonadotropin levels. If a patient has a beta subunit of human chorionic gonadotropin level of 1,500 mIU per mL or greater, but the transvaginal ultrasonography does not show an intrauterine gestational sac, ectopic pregnancy should be suspected. Diagnostic uterine curettage may be appropriate in patients who are hemodynamically stable and whose beta subunit of human chorionic gonadotropin levels are not increasing as expected. Appropriate treatment for patients with nonruptured ectopic pregnancy may include expectant management, medical management with methotrexate, or surgery. Expectant management is appropriate only when beta subunit of human chorionic gonadotropin levels are low and declining. Initial levels determine the success of medical treatment. Surgical treatment is appropriate if ruptured ectopic pregnancy is suspected and if the patient is hemodynamically unstable.     

Accuracy of transvaginal ultrasound and serum hCG in the diagnosis of ectopic pregnancy.

Transvaginal ultrasound was performed upon admission of 127 patients with a clinical suspicion of ectopic pregnancy in association with human chorionic gonadotropin (hCG) determination. Failure to visualize with sonography an intrauterine gestational sac with an hCG level superior to 1000 mIU/ml identified 25/42 tubal pregnancies with a positive predictive value of 86% and a specificity of 93%. Abnormal adnexal findings occurred in 95% of the ectopic pregnancies. Extrauterine gestational sacs with or without embryos could be confidently detected in 19 ectopic pregnancies (45%). A complex adnexal mass was seen in 19 cases and yielded a positive predictive value of 90% (19/21). Adnexal gestational sacs and complex masses were seen more frequently in those ectopic pregnancies with an hCG level above 1000 mIU/ml but the difference was not significant (100% versus 78%). Simple adnexal cysts were found more frequently in intrauterine pregnancies, and fluid in the cul-de-sac was also not indicative of ectopic pregnancy (positive predictive value, 29%). Transvaginal ultrasound has a primary role in the diagnosis of ectopic pregnancy. The combined use of uterine and adnexal sonography associated with elevated hCG levels allows a definitive diagnosis in the vast majority of cases at a very early stage, when the chances for a successful conservative treatment are greater.     

Beta-Human Chorionic Gonadotropin Levels and the Likelihood of Ectopic Pregnancy in Emergency Department Patients with Abdominal Pain or Vaginal Bleeding

Strategies for diagnosing ectopic pregnancy that defer endovaginal ultrasound in women with suggestive symptoms and serum beta-human chorionic gonadotropin (beta-hCG) levels less than 1500 mIU/mL ignore the increased risk of ectopic pregnancy in these patients. OBJECTIVE: To quantify this increased risk by establishing and comparing the beta-hCG distributions of symptomatic women with ectopic pregnancies, abnormal intrauterine pregnancies, and normal intrauterine pregnancies. METHODS: The authors reviewed the records of a cohort of women who visited an urban emergency department (ED) during a 34-month period with abdominal pain or vaginal bleeding and non-zero quantitative beta-hCG levels. Explicit criteria were used to determine whether the pregnancy ultimately turned out to be intrauterine and normal, intrauterine and abnormal, or ectopic. Probability distributions were compared using frequency distributions, receiver operating characteristic (ROC) curves, and likelihood ratios. RESULTS: Of 730 ED patients included in the analysis, 96 (13%) had ectopic pregnancies, 253 (35%) had abnormal intrauterine pregnancies, and 381 (52%) had normal intrauterine pregnancies. The beta-hCG distributions of patients with ectopic pregnancies and abnormal intrauterine pregnancies were similar and much lower than the beta-hCG distribution of patients with normal intrauterine pregnancies. A beta-hCG level less than 1500 mIU/mL more than doubled the odds of ectopic pregnancy (likelihood ratio = 2.24). Of the 158 patients with beta-hCG below 1500 mIU/mL, 40 (25%; 95% confidence interval [CI] = 19% to 32%) had ectopic pregnancies, and only 25 (16%; CI = 11% to 22%) had normal intrauterine pregnancies. CONCLUSIONS: In women with pain or bleeding and serum beta-hCG levels less than 1500 mIU/mL, the risk of ectopic pregnancy is substantially increased, while the likelihood of normal intrauterine pregnancy is low.     

Comparison of implantation and early development of human embryos fertilized in vitro versus in vivo using transvaginal ultrasound

Several reports in the literature suggest delayed implantation of in vitro-fertilized human embryos compared to in vivo-fertilized eggs. The use of high-frequency transvaginal transducers for early detection of pregnancy has allowed the identification of the gestational sac with very low serum human chorionic gonadotropin (beta-hCG) levels. Thus, the present study evaluated whether retarded implantation can be identified using this novel technology. We studied 13 single pregnancies after in vitro fertilization (IVF) and 14 pregnancies after artificial insemination either by husband (n = 6) or donor (n = 8). In the IVF patients, oocytes were retrieved 35 hours after hCG administration. Embryo transfer occurred approximately 48 hours after retrieval. Artificial insemination was performed 24 and 48 hours after hCG administration. Transvaginal ultrasound scans and serum beta-hCG levels were evaluated every 3 days starting day 12 post-hCG administration. Serum beta-hCG levels rose in parallel when in vitro- and in vivo-fertilized embryos were compared. Similarly, there was no difference between groups in the mean time needed to detect early embryonic structures, such as the embryonic sac, yolk sac, and heartbeats, or the growth rate of the gestational sac. In conclusion, there was no difference in detecting implantation and early embryonic development of human embryos fertilized in vivo versus in vitro as ascertained by ultrasound scans and serum beta-hCG levels. An embryonic sac is detected 23-24 days after hCG administration in pregnancies achieved by assisted reproductive techniques.     

The value of endovaginal sonography and urinary human chorionic gonadotropin tests for differentiation between intrauterine and ectopic pregnancy

Effectiveness of vaginal sonography combined with urinary human chorionic gonadotropin (hCG) for identification of ectopic pregnancy (EP) was studied in 107 pregnant women. Eighty-nine women had clinical symptoms suspicious of EP. It was suggested that 18 women carried an increased risk for developing EP. In 63 women endovaginal sonography showed no evidence of intrauterine pregnancy. Fifty-eight of these turned out to be pathological pregnancies. In 44 women endovaginal ultrasonography revealed intrauterine pregnancies. Thirty-two of these turned out to be viable, 10 were not viable and resulted in spontaneous abortions, and 2 turned out to be EP. The sensitivity of vaginal sonography to identify a viable intrauterine pregnancy thus was 81% and its specificity was 97%. The sensitivity and the specificity for endovaginal ultrasonography for identifying EP was 96% and 71%, respectively. Endovaginal ultrasonography demonstrated an intrauterine gestational sac in 54% of the women with urinary HCG as low as 40 IU/L to 500 IU/L. These results show that endovaginal ultrasonography is a sensitive instrument for identifying both early normal intrauterine pregnancies as well as pathological pregnancies.     

First trimester bleeding evaluation

First trimester bleeding is a common presentation in the emergency room. Ultrasound evaluation of patients with first trimester bleeding is the mainstay of the examination. The important causes of first trimester bleeding include spontaneous abortion, ectopic pregnancy, and gestational trophoblastic disease; 50% to 70% of spontaneous abortions are due to genetic abnormalities. In normal pregnancy, the serum beta hCG doubles or increases by at least 66% in 48 hours. The intrauterine GS should be visualized by TVUS with beta hCG levels between 1000 to 2000 mIU/mL IRP. Visualization of the yolk sac within the gestational sac is definitive evidence of intrauterine pregnancy. Embryonic cardiac activity can be identified with CRL of >5 mm. A GS with a mean sac diameter (MSD) of 8 mm or more without a yolk sac and a GS with an MSD of 16 mm or more without an embryo, are important predictors of a nonviable gestation. A GS with a mean sac diameter of 16 mm or more (TVUS) without an embryo is a sonographic sign of anembryonic gestation. A difference of <5 mm between the mean sac diameter and the CRL carries an 80% risk of spontaneous abortion. Approximately 20% of women with first trimester bleeding have a subchorionic hematoma. The presence of an extra ovarian adnexal mass is the most common sonographic finding in ectopic pregnancy. Other findings include the tubal ring sign and hemorrhage. About 26% of ectopic pregnancies have normal pelvic sonograms on TVUS. Complete hydatidiform mole presents with a complex intrauterine mass with multiple anechoic areas of varying sizes (Snowstorm appearance). Twenty-five percent to 65% of molar pregnancies have associated theca-leutin cysts. Arteriovenous malformation of the uterus is a rare but life-threatening cause of vaginal bleeding in the first trimester. The sonographic findings in a patient with first trimester bleeding should be correlated with serum beta hCG levels to arrive at an appropriate clinical diagnosis.     

The discriminatory zone of beta-hCG for vaginal probes.

Recent developments in both laboratory measurements and ultrasound technology have revolutionized the management of early pregnancy. The discriminatory zone concept is a direct result of these developments. By correlating the serum beta-hCG values to the size of an intrauterine gestational sac, a value can be chosen that corresponds to the threshold of visualization of the sac. If the beta-hCG is above this value, a sac must be seen, and if it is not, aggressive steps should be taken to determine whether the pregnancy is abnormal or ectopic. The discriminatory zone may vary among institutions due to different equipment and assays. Thus, its value should be calculated individually at each institution. A proper discriminatory zone and a management protocol such as the one above can eliminate much of the uncertainty in the management of suspected ectopic and early pregnancies.     

Ectopic pregnancy. Diagnosis by sonography correlated with quantitative HCG levels

Pelvic sonograms were correlated with simultaneous human chorionic gonadotropin (HCG) determinations in 150 women with early intrauterine pregnancy (N = 76) and ectopic pregnancy (N = 74). Of the 76 patients with intrauterine pregnancy (IUP), 55 had HCG levels exceeding 1,800 mIU/ml (Second International Standard), and in each case a gestational sac was identified. In comparison, 35 of 74 (47%) patients with ectopic pregnancy had HCG levels of 1,800 mIU/ml or more, and no case demonstrated a gestational sac. Although six patients (8%) with ectopic pregnancy demonstrated a "pseudogestational sac," no case was confused with a true gestational sac. We conclude that, when the HCG level exceeds 1,800 mIU/ml, an intrauterine gestational sac is normally detected and its absence is evidence for an ectopic pregnancy.     

Intrauterine pseudogestational sac in ectopic pregnancy.

Ultrasonic diagnosis of ectopic pregnancy has been thought to depend on exclusion of intrauterine pregnancy on the basis of absence of an intrauterine gestational sac. Two cases illustrating intrauterine echoes suggesting a gestational sac associated with ectopic pregnancy and a plausible explanation for this phenomenon are presented. Because of this and the rare possibility of coexisting intrauterine and extrauterine pregnancies, the ultrasonographer must be wary of dismissing the possibility of an ectopic pregnancy because of an intrauterine gestational sac, whether real or apparent.     

Ectopic pregnancy in an urban teaching hospital: can tubal rupture be predicted?

We evaluated the medical history, physical examination, and laboratory tests done on 245 patients with laparoscopically proven ectopic pregnancies. The absence of abdominal pain was the only clinically useful negative predictive value (91%) regarding tubal rupture. Although mean levels of serum human chorionic gonadotropin (hCG-beta subunit) were significantly higher in patients with ruptured versus unruptured ectopic pregnancies (16,612 mIU/mL vs 6406 mIU/mL), no breakpoint excluded the possibility of tubal rupture. In fact, one third of ectopic pregnancies in patients with a serum beta-hCG level below 100 mIU/mL were ruptured. We conclude that clinical symptoms and signs are poor predictors of tubal rupture. In addition, absolute values of serum beta-hCG are not helpful in excluding the possibility of rupture.     

Low-resistance endometrial arterial flow in the exclusion of ectopic pregnancy revisited.

The objective of this study was to evaluate the efficacy of endometrial arterial flow in the exclusion of ectopic pregnancy. From October 1997 to June 1999, 66 women with elevated beta-human chorionic gonadotropin titers and clinical indications of ectopic pregnancy were evaluated by endovaginal sonography. Women with a gestational sac containing an embryo, a yolk sac, or both were excluded from the study. Doppler ultrasonography was performed in the remaining cases when a definite intrauterine pregnancy could not be visualized. In all cases the thermal index was kept to less than 1.0, consistent with as-low-as-reasonably-achievable principles. Trophoblastic flow was defined as a resistive index of less than 0.6 within the endometrium. Statistical analysis was performed using a 2-tailed t test. Twenty women had ectopic pregnancies; 33 had spontaneous pregnancy losses; and 13 had normal intrauterine pregnancies. A total of 29 women had endometrial trophoblastic flow: 11 of 13 with intrauterine pregnancies, 1 of 20 with ectopic pregnancies, and 17 of 33 with spontaneous pregnancy losses. The negative predictive value for the presence of endometrial low-resistance flow for excluding ectopic pregnancy was 97%. The presence of low-resistance arterial endometrial flow can be a useful sign in diagnosing an early intrauterine pregnancy and decreasing the probability that an ectopic pregnancy is present, particularly in patients with otherwise normal ultrasonographic findings.     

Transvaginal ultrasound scanning in early pregnancy and correlation with human chorionic gonadotropin levels.

Pregnancy can be detected early using transvaginal ultrasonography and human chorionic gonadotropin (hCG) measurements. The purpose of this study was to correlate serum hCG levels with transvaginal gestational sac measurements. The mean sac diameter (MSD) and gestational sac diameter (GSD) were calculated as the mean and cube root of the product, respectively, of three sac dimensions taken at right angles. A nonlinear relationship between hCG (natural logarithm) and MSD and GSD was observed. The data were best fitted by a second order polynomial regression model (r2 = 0.98), thereby establishing normal hCG levels for various gestational sac dimensions in early pregnancy.     

The decidua-chorionic sac: a reliable sonographic indicator of intrauterine pregnancy prior to detection of a fetal pole.

Morphologically, the uterus in early intrauterine pregnancy, unlike ectopic pregnancy, contains both decidua and a chorionic sac. In view of this fact, a prospective study was performed to determine the accuracy of the demonstration of an intrauterine decidua-chorionic sac (DCS) in differentiating early or small intrauterine pregnancies (IUPs) without a demonstrable fetal pole from the intrauterine findings of ectopic pregnancy. The DCS was seen in 78 patients with proven IUPs and was not present in 52 consecutive patients with proven ectopic pregnancy (100 per cent accuracy). The decidua-chorionic sac is defined, and its relationships to the "double sac," double decidual sac, pseudosac, "crescent," pseudogestational sac, and the discriminatory human chorionic gonadotropin zone are discussed.     

The timing of a repeat ultrasound examination in the evaluation for ectopic pregnancy

A method is described for timing a repeat ultrasound examination in patients evaluated for ectopic pregnancy in whom the findings are initially nondiagnostic. This method is based on measuring the serum hCG at the time of the initial ultrasound examination, which allows the time interval required for the serum hCG to exceed 6500 mIU/ml to be calculated. When the serum hCG is above this level, the gestational sac of an intrauterine pregnancy can be reliably identified, and failure to do so signifies ectopic gestation. Seven patients referred for ultrasound to rule out ectopic pregnancy had initially nondiagnostic findings. In each case, a correctly-timed repeat examination, demonstrating a gestational sac, enabled the diagnosis to be excluded.     

Serum biochemistry correlates with the size of tubal ectopic pregnancy on sonography.

OBJECTIVE: To investigate whether there is a correlation between serum biochemistry (human chorionic gonadotropin (hCG), CA 125, progesterone and estradiol) and the common sonographic findings (blob sign, bagel sign or extrauterine gestational sac with cardiac activity) or size of a tubal ectopic pregnancy, and whether there is a difference in serum biochemistry between women with a tubal ectopic pregnancy who are hemodynamically unstable (tachycardia, hypotension, falling hemoglobin levels and/or acute severe abdominal pain) and those who are hemodynamically stable. METHODS: This was a prospective cohort study of 106 women with a tubal ectopic pregnancy. We noted transvaginal ultrasound examination findings including adnexal mass size, and the serum levels of hCG, CA 125, progesterone and estradiol. The data were analyzed retrospectively. RESULTS: The mean maternal and gestational ages were 30.7+/-5.7 years and 44+/-4.2 days, respectively. There was no correlation between serum markers and common sonographic findings. However, in the presence of the bagel sign on ultrasound, hemodynamic stability was more common (P=0.03). The mean serum hCG concentrations in tubal ectopic pregnancies<20 mm, 20-40 mm and >40 mm in size were 2225.3+/-3166.9, 4124.8+/-6121.4, and 11 011.8+/-12 670.1 IU/mL, respectively (P<0.001). Serum hCG, CA 125 and estradiol values were well correlated with adnexal mass size; for CA 125 this correlation was linear. There was no difference in serum biochemistry between hemodynamically stable and hemodynamically unstable women. CONCLUSION: Common sonographic findings of tubal ectopic pregnancy do not correlate with serum biochemistry. High levels of CA 125, hCG or estradiol may suggest a larger adnexal mass in women with uncomplicated tubal pregnancies. Hemodynamically stable and hemodynamically unstable women do not differ in their serum biochemistry.     

Classification of tubal gestations by transvaginal sonography.

Transvaginal sonographic images of tubal pregnancies were correlated with the surgical findings in 191 patients having this condition. The sonographic appearance of the ectopic gestations was classified by the absence or presence of structures such as a 'tubal ring' containing the yolk sac, embryonic structures, heart activity, a sonolucent or irregularly echogenic gestational sac, dilated Fallopian tube with amorphous content, fluid in the pelvis and an empty uterus. The following classifications were made: Type Ia (n = 43) A well-defined 'tubal ring' and a beating heart with or without discrete embryonic or extra-embryonic structures.Type Ib (n = 48) A 'tubal ring' containing embryonic and/or extra-embryonic structures without heart beats. Type II (n = 64) An ill-defined or thin tubal wall containing sonolucent or an irregularly echogenic core but not embryonic or extra-embryonic structures. Type III (n = 28) Free pelvic fluid and an empty uterus in patients with positive serum beta-hCG levels. The outline of the tube cannot be visualized. Surgery revealed unruptured tubal pregnancies in 90 patients of the combined groups of Types Ia and Ib. In one additional patient, a tubal rupture was found. In the Type II patients, 26 tubal pregnancies with blood clots in the tube but no evidence of bleeding into the pelvis, and 38 tubal ruptures or abortions were diagnosed. All Type III patients had ruptured tubal pregnancies or bleeding tubal abortions. In eight patients (4.2%), the only sonographic finding was an empty uterus, and these cases were erroneously diagnosed as not having an ectopic pregnancy (false negatives). There were two false-positive cases in which a tubal ring was detected, and this was related to an hemorrhagic corpus luteum. When used for diagnosing tubal pregnancy, the transvaginal scanning technique (together with beta-hCG in Type III cases) carries a sensitivity of 95.8% and specificity of 99.9%. It is an invaluable tool in the diagnostic work-up and management of patients with suspected ectopic gestation.     

Serial Transvaginal Sonographic Findings of Cervical Ectopic Pregnancy Treated With High‐Dose Methotrexate

Objective. We analyzed transvaginal sonographic findings from patients with cervical ectopic pregnancies treated with high‐dose methotrexate (MTX). Methods. This was a retrospective analysis of cervical pregnancies diagnosed in our institution from 1996 through 2006. We divided the cases into an MTX treatment group and a surgical treatment group. We included cases treated with high‐dose MTX alone. We analyzed 9 cervical ectopic pregnancies treated with MTX, which was injected intravascularly at 100 mg/m² plus 200 mg/m² in 500 mL of a normal saline solution with folinic acid rescue. The gestational sac sizes and serum human chorionic gonadotropin (hCG) levels were periodically monitored to determine the resolution status. Results. Fifty cervical pregnancies were diagnosed during the study period. Thirty cases were treated with MTX, and 20 were treated with surgical procedures. Among the 30 cases in the MTX treatment group, 9 had high‐dose MTX injection without surgical procedures. Cervical mass regression appeared at a median of 40 (range, 10–88) days after treatment, whereas the serum hCG level decreased at a median of 14 (range, 9–17) days after treatment. The median time to complete regression of the cervical mass was 86 (range, 48–141) days, and the median time to complete regression of the serum hCG level was 68 (range, 19–143) days. Cervical pregnancy was noted as a gestational sac at first but coexisted with a mixed echoic lesion 19 days after treatment. At 33 days after treatment, the cervical pregnancy was completely replaced by the mixed echoic lesion. Conclusions. Resolution of the cervical mass on sonography lagged far behind resolution of the serum hCG level. The cervical mass evolved from a gestational sac into a mixed echoic lesion on serial transvaginal sonography.     

Ectopic pregnancy treatment using pelviscopic or vaginosonographically-guided intrachorionic injection of methotrexate and POR 8.

It is believed by some that the basis of modern first-trimester pregnancy care should involve screening for ectopic pregnancy using vaginal ultrasonography. This has led to this condition being diagnosed in most cases at an earlier stage than has previously been possible. This advance in diagnostic ability has introduced the concept of conservative management for this condition.The authors describe the conservative management of nine cases of ectopic pregnancy using intrachorionic injection of either methotrexate alone or in combination with POR 8. In all cases less than 9 weeks had passed since the last menstrual period, and the ectopic pregnancies treated under ultrasound control were all thought to be viable. Five cases were treated using pelviscopy, whilst in another four the needle was guided using vaginal ultrasonography. There were no procedure-related complications.Following therapy, the women were followed up using serial levels of serum beta-hCG. In seven cases there was a gradual reduction to normal levels. There were two treatment 'failures'. In both these cases methotrexate alone, rather than in combination with POR 8, was used for injection. Although formal tests of tubal patency have not been performed in all cases, one woman in the series has since conceived and has a normal intrauterine pregnancy.It is concluded that vaginal instillation of methotrexate is an alternative treatment for early ectopic pregnancy. It should be limited to viable ectopic pregnancies whose chorionic cavities are less than 2.0 cm in diameter. A plan for the management of ectopic pregnancy is presented, involving selection for both ultrasound-guided injection and pelviscopic surgery. Laparotomy is not considered to be a treatment option in the majority of cases.     

Conservative management of live tubal pregnancies by ultrasound guided potassium chloride injection and systemic methotrexate treatment

Unruptured live tubal ectopic pregnancies are often managed surgically. Significantly elevated beta-hCG levels in these patients make treatment with methotrexate ineffective. However, achieving cardiac asystole via sonographically guided injection of potassium chloride (KCl) along with systemic methotrexate can improve treatment outcome. We describe the successful conservative management of 3 cases of unruptured tubal pregnancy with cardiac activity and significantly elevated beta-hCG levels. Under sonographic guidance, KCl was injected into the fetal heart to achieve cardiac asystole, and patients concurrently received a systemic methotrexate injection. The resolution of ectopic pregnancy was achieved and surgery was avoided in all 3 cases. Conservative management may thus be an option for patients with live ectopic pregnancies.     

Accuracy of transvaginal ultrasonography for detection of hematosalpinx in ectopic pregnancy.

The transvaginal approach has significantly improved the accuracy of ultrasonography for the detection of ectopic pregnancy. However, there has been limited emphasis given to determining the sensitivity of ultrasonography when a hematosalpinx was used as a specific finding to identify an ectopic pregnancy. The sensitivity of transvaginal ultrasonography was evaluated for the detection of a hematosalpinx defined as an "echogenic homogeneous or inhomogeneous, rounded or elongated structure" in a group of patients with surgically proven ectopic pregnancy. Retrospectively, transvaginal ultrasonography showed a hematosalpinx in 16 out of 18 (88.8%) tubal pregnancies. In 6/6 (100%) patients with a ruptured tube and 10/12 (83.3%) patients with an unruptured tube, a hematosalpinx was detected sonographically. A gestational sac with a live embryo was seen in 26.3% of these patients. The significance of identifying a hematosalpinx, predictability of rupture and implication in the treatment of ectopic pregnancy are discussed.