Abstract & Commentary
Source: Tayal VS, et al. Outcome of patients with an indeterminate emergency department first-trimester pelvic ultrasound to rule out ectopic pregnancy. Acad Emerg Med 2004;11:912-917.
The use of pelvic ultrasound (US) performed by emergency physicians (EPs) has gained acceptance in many emergency departments (EDs) for the evaluation of patients in their first trimester of pregnancy who present with symptoms referable to the pregnancy. Previous literature supports the fact that there is no combination of historical data or physical examination findings that reliably can rule out ectopic pregnancy in this population, and that US, specifically transvaginal ultrasound (TVUS), is the study of choice to make this determination. TVUS is preferable to the transabdominal approach; the pregnancy (normal or abnormal) can be identified at a much earlier stage. Formerly this procedure was performed exclusively by radiologists and their technologist staff; now it is performed frequently by EPs in the ED.
The primary goal of TVUS performed by the EP is to determine whether there is a normal intrauterine pregnancy (IUP). Alternative diagnoses include ectopic pregnancy, molar pregnancy, fetal demise, and indeterminate (i.e., no definitive diagnosis evident on TVUS). Based on prior literature, it would be expected that a definite IUP would be identified in 60-70% of these cases, with an abnormal finding or indeterminate finding in 30-40%. Although treatment algorithms for IUP, definitive ectopic pregnancy, molar pregnancy, and fetal demise are clear and straightforward, the same is not true for the indeterminate group. This study seeks to define this group, present a diagnostic algorithm, and look at the ultimate outcome for this population.
This observational prospective cohort study at Carolinas Medical Center, a large, urban, community teaching hospital with an ED census of more than 100,000, looked at its experience with TVUS for patients in this group during a 13-month period. Pelvic US was performed by EPs on all first-trimester patients. Obstetrics/gynecology consultation was made for all findings, except definite IUP. Definite IUP was defined as a fundal gestational sac with either a visible yolk sac or fetal pole. For all patients with an indeterminate study, final diagnoses were established by the use of patient records, obstetrics/gynecology US reports, laboratory studies, and operative and pathology reports.
During the study period, 1490 patients were examined. Definite IUP was diagnosed in 1037 (70%), 127 (8%) had fetal demise, 24 (2%) had ectopic pregnancy, two (< 1%) had molar pregnancy, and 300 (20%) had indeterminate findings. In this group of 300 with indeterminate findings, the final diagnoses were as follows: 158 (53%) with fetal demise, 88 (29%) with IUP, 44 (15%) with ectopic pregnancy, and unknown outcome in 10 (3%). Of the 44 patients ultimately diagnosed with ectopic pregnancy, 25 (57%) were treated with metho-trexate, and 16 (36%) were treated with laparoscopy. The authors note that there is a higher percentage of patients with ectopic pregnancies treated medically as opposed to surgically in the "indeterminate" group, than in the group diagnosed with ectopic pregnancy at the outset.
Commentary by Andrew Perron, MD, FACEP, FACSM
This is a great study for the person charged with developing a TVUS program at his or her particular institution (as we currently are doing at my hospital).
Tayal and his group are clear in their definitions of what constitutes a definite IUP, indeterminate scan, etc. They also take pains to present an explicit algorithm that leaves little room for individual practice variation, which is good, especially when instituting a protocol where violations can have potentially serious adverse consequences for the patient. This study also can serve as a nice quality-control group for comparison as a TVUS program is implemented, looking at expected percentages of a particular finding vs. observed percentages.
I see two primary limitations to this study, but neither is particularly serious. The first is the 3% of patients lost to follow-up with the initial indeterminate scans. It is not clear to what extent the authors sought to define the ultimate outcome of this group. Did they present to another hospital? Did they present to the morgue? This represents only 10 of the 300 patients—not a large percentage group—but the authors easily could have presented a worst-case calculation, grouping patients lost to follow-up into the ectopic subset. The second limitation is in the added observation regarding medical vs. surgical treatment in patients ultimately diagnosed with ectopic vs. patients immediately diagnosed with ectopic. We know nothing else about those two populations (i.e., Were patients in the latter group more likely to have hypotension? Were patients in the latter group farther along in gestation?). Because we know little about these two groups, comparisons between them may be beyond the scope of the paper.
Dr. Perron, Residency Program Director, Department of Emergency Medicine, Maine Medical Center, Portland, ME, is on the Editorial Board of Emergency Medicine Alert.