Emergency physicians frequently evaluate and manage patients with abdominal pain and/or vaginal bleeding during the first trimester of pregnancy. Although a variety of pathologies may cause these symptoms, the primary concern of the physician is to identify conclusively those patients with an ectopic pregnancy. In recent years bedside ultrasound (performed in the ED by the emergency clinician rather than in the radiology department) has become more common, but approaches to diagnosing ectopic pregnancy still vary considerably because of differing availability of ultrasound across departments or at different times of day.

These patients present a diagnostic challenge because they frequently present before an intrauterine or extrauterine pregnancy can be definitively seen with ultrasound. Therefore, many algorithms incorporate a quantitative b-hCG level and apply the principle of the “discriminatory threshold,” generally defined as the level at which the sensitivity of ultrasound approaches 100% for the detection of intrauterine pregnancy. This threshold is commonly reported as ranging from 1,000 mIU/mL to 2,000 mIU/mL, and the presumptive diagnosis of ectopic pregnancy traditionally has been made if an intrauterine pregnancy is not visualized when the serum b-hCG is above this threshold.

In the September 2012 issue of the Annals of Emergency Medicine, ACEP published a clinical policy on the evaluation and management of patients in early pregnancy, which is a revision of a policy written in 2003. The updated policy examines three critical questions faced by emergency clinicians. The first two questions are driven by ambiguities that commonly arise when using the b-hCG and the bedside ultrasound to identify ectopic pregnancy. The third critical question explores the implications of patients receiving methotrexate for confirmed or suspected ectopic pregnancy. Another question that often arises in the management of these patients is whether to administer anti-D immunoglobulin to those who are Rh-negative. However, because no new quality evidence on this topic was identified during the literature search, the patient management recommendations from 2003 remain unchanged and are not discussed further in the 2012 policy.

Recommendations (Level A, B, or C) for patient management are provided based on the strength of evidence using the Clinical Policies Committee’s well-established methodology:

Level A recommendations represent patient management principles that reflect a high degree of clinical certainty

Level B recommendations represent patient management principles that reflect moderate clinical certainty

Level C recommendations represent other patient management strategies based on Class III studies, or in the absence of any adequate published literature, based on consensus of the members of the Clinical Policies Committee

Question 1: Should the emergency physician obtain a pelvic ultrasound in a clinically stable pregnant patient who presents to the ED with abdominal pain and/or vaginal bleeding and a b-hCG level below a discriminatory threshold?

Level A recommendations. None specified.

Level B recommendations. None specified.

Level C recommendations. Perform or obtain a pelvic ultrasound for symptomatic pregnant patients with a b-hCG level below any discriminatory threshold.

In cases where the b-hCG level is below a discriminatory threshold, some clinicians may not perform an ultrasound because they believe that the diagnostic utility of pelvic ultrasound is low. In some settings, the emergency physician may be unable to obtain a comprehensive ultrasound in the radiology department for the same reason. However, studies reviewed for this clinical policy showed that the ultrasound may be diagnostic even if b-hCG is below 1,000 mIU/mL. Clinicians may also defer an ultrasound when the b-hCG is low, under the assumption that this correlates with a small embryo and low risk of tubal rupture if the pregnancy is, in fact, ectopic. However, rupture has been documented at very low b-hCG levels and there is a potential harm in deferring ultrasound imaging. Therefore, this policy recommends that a pelvic ultrasound be performed or obtained for symptomatic patients with a b-hCG level below any “discriminatory threshold.” Because the evidence is taken from Class II and III studies, this is a Level C recommendation.

Question 2: In patients who have an indeterminate transvaginal ultrasound, what is the diagnostic utility of b-hCG for predicting possible ectopic pregnancy?

Level A recommendations. None specified.

Level B recommendations. Do not use the b-hCG value to exclude the diagnosis of ectopic pregnancy in patients who have an indeterminate ultrasound.

Level C recommendations. Obtain specialty consultation or arrange close outpatient follow-up for all patients with an indeterminate pelvic ultrasound.

A majority of patients who have a pelvic ultrasound during their ED evaluation for symptomatic early pregnancy will receive a diagnosis of an intrauterine pregnancy or an abnormal pregnancy (e.g., ectopic or molar, fetal demise). A significant minority, however, (approximately 25%) will have an indeterminate (or nondiagnostic) ultrasound. This rate depends on a number of factors, including the clinical setting, patient characteristics, ultrasound machine and operator, and diagnostic criteria used.

Indeterminate ultrasounds pose a management dilemma for the clinician. In order to answer the critical question above, the authors reviewed studies reporting the initial b-hCG level in patients with a final diagnosis of ectopic pregnancy who have an initial indeterminate ultrasound. Positive and negative likelihood ratios for risk of ectopic pregnancy above and below different b-hCG levels were calculated by the authors from the available data.

Nine Class II studies and five Class III studies examined this issue. Likelihood ratios were variable across studies and generally neither high nor low enough to help with clinical decision making. This suggested that the initial b-hCG cannot be used to exclude the diagnosis of ectopic pregnancy and resulted in Level B and Level C recommendations.

Question 3: In patients receiving methotrexate for confirmed or suspected ectopic pregnancy, what are the implications for ED management?

Level A recommendations. None specified.

Level B recommendations. (1) Arrange outpatient follow-up for patients who receive methotrexate therapy in the ED for a confirmed or suspected ectopic pregnancy. (2) Strongly consider ruptured ectopic pregnancy in the differential diagnosis of patients who have received methotrexate and present with concerning signs or symptoms.

Level C recommendations. None specified.

Methotrexate, administered as a single intramuscular or intravenous dose in the ED, is an alternative to surgical treatment for known or suspected early ectopic pregnancy. Because a single dose of methotrexate is often ineffective, patients may require repeated doses until their b-hCG levels are clearly decreasing. Treatment failure leading to rupture of the ectopic pregnancy is one of the most serious complications of methotrexate therapy; in several cohort studies, more than 20% of patients receiving methotrexate required surgery. Nineteen Class I, II, and III studies were included for discussion. Taken together, these data support a Level B recommendation to arrange outpatient follow-up and to strongly consider ruptured ectopic pregnancy in patients who develop increasing pain and/or signs of hemodynamic instability after methotrexate therapy.

Conclusion

The definitive diagnosis of ectopic pregnancy in the first trimester remains a clinical challenge. There is growing evidence that clinicians should not rely upon a discriminatory threshold to decide which patients to perform an ultrasound on, nor use the b-hCG to try to predict which patients with an indeterminate ultrasound have an ectopic pregnancy. Despite the clinician’s best efforts, a significant minority of patients will not have a definitive diagnosis after their ED workup, and the management and disposition will need to take into account the patient’s clinical picture as well as other contributing factors, including availability of outpatient follow up.

This clinical policy can also be found on ACEP’s Web site www.acep.org and has been submitted for abstraction on the National Guideline Clearinghouse Web site, www.guidelines.gov.

Dr. Bernstein is a senior resident in emergency medicine at Yale-New Haven Hospital, and is the 2011-2013 EMRA Representative to the ACEP Clinical Policies Committee.

Dr. Hahn is an assistant professor in emergency medicine at the Mount Sinai School of Medicine in New York, and is a member of the ACEP Clinical Policies Committee.