There have been major advances in ultrasound equipment in the past 2 decades. The resulting improved resolution raises the question of whether the sonographic findings of ectopic pregnancy described in the early days of transvaginal sonography still apply today. It is possible that improved resolution has resulted in earlier identification of adnexal abnormalities, with a corresponding change in the distribution of imaging findings. Our study addresses this question and provides an updated understanding of the imaging characteristics of ectopic pregnancy and their relationship to fallopian tube rupture.
We found that the percentage of ectopic pregnancies with a yolk sac or cardiac activity was lower than previously reported. Embryonic cardiac activity was found in fewer than 10% of our patients, similar to a recent report but much lower than in earlier series, in which up to 24% of ectopic pregnancies had cardiac activity. The ectopic pregnancy in more than half of our patients appeared as a non-specific adnexal mass, which may have represented a blood clot in some cases. Our study confirms the previously published important point that, in a small number of patients, free fluid may be the only sonographic finding of ectopic pregnancy.
In the early 1990s, color Doppler imaging was suggested as a valuable adjunct for the diagnosis of ectopic pregnancy. However, we found color Doppler imaging to be of little clinical value: there was no significant relationship between the color pattern and type of adnexal mass, and the addition of color Doppler imaging to the examination did not improve the ability to predict tubal rupture.
Among our patients who underwent surgery within 1 day of the study sonogram, 25.2% proved to have tubal rupture. However, the overall rate of tubal rupture is likely to be much lower, since the women who underwent surgery were not representative of our entire group. Although our study methods do not allow us to determine the overall rate of tubal rupture with ectopic pregnancy, our results do permit us to conclude, as had been noted previously, that sonographic findings are not reliable predictors of rupture. No single adnexal mass appearance or color Doppler characteristic correlates with the presence or absence of tubal rupture. Only the presence of a moderate-to-large amount of free intraperitoneal fluid was significantly associated with a ruptured fallopian tube, but the finding of this amount of free fluid had poor sensitivity, specificity, and positive predictive value for tubal rupture.
The role of hCG in the diagnosis of ectopic pregnancy is widely debated in the literature. Our study, similar to others, shows that hCG levels vary widely in women with ectopic pregnancy, from less than 10 mIU/mL in one of our patients to greater than 100,000 mIU/mLin another, and reinforces the fact that there is no lower-level hCG cutoff value with ectopic pregnancy. The average hCG value increased as the grade of the adnexal mass increased, with the highest average hCG level found in those with cardiac activity, which also confirms other reports. However, the hCG level did not correlate with the presence or absence of tubal rupture, so this serum measurement has little clinical value if an adnexal mass is seen on sonography.
This study had a few limitations. Measurements of nonspecific adnexal masses were somewhat subjective, since the borders of the masses were not always clear. In particular, some of the poorly defined masses may have represented blood clots, which often do not have discrete margins and may be difficult to measure accurately. Another limitation was that some patients had more than 1 sonographic examination performed before the time of diagnosis. We used the transvaginal sonogram obtained closest to the point of treatment, which may have biased this report toward larger and more advanced adnexal findings. Additionally, some of our patients did not have either surgical or pathologic proof of their ectopic pregnancies but were treated medically.