Normal Intrauterine Pregnancy Is Unlikely in ... - Wiley Online Library

116

TRANSVAGINAL US

Dart et al. • ECHOGENIC MATERIAL ON ULTRASONOGRAPHY

Normal Intrauterine Pregnancy Is Unlikely in Patients Who Have Echogenic Material Identified within the Endometrial Cavity at Transvaginal Ultrasonography ROBERT DART, MD, LINDA DART, RN, PATRICIA MITCHELL, RN

Abstract. In first-trimester patients with abdominal pain or vaginal bleeding, pelvic ultrasonography (US) is often performed to assess the status of the pregnancy. Identification of echogenic material within the endometrial cavity in the absence of a gestational sac has been attributed to the presence of either retained products of conception or clotted blood. However, to the authors’ knowledge, no study has directly addressed whether this finding reliably excludes the diagnosis of a normal intrauterine pregnancy (IUP). Objective: To determine whether the identification of echogenic material within the endometrial cavity at transvaginal US excludes the diagnosis of a normal IUP. Methods: Data were collected retrospectively from August 1991 to August 1997 on consecutive urban teaching hospital ED patients with abdominal pain or vaginal bleeding who had a transvaginal US performed during the ED visit that demonstrated echogenic material within the endometrial cavity. Patients were excluded if they had a dilatation and evacuation (D⫹E) procedure performed prior to the

I

N THE first trimester of pregnancy, pelvic ultrasonography (US) is often performed to assess the status of the pregnancy. In the absence of a gestational sac, identification of echogenic material within the endometrial cavity has been attributed either to the presence of retained products of conception or, in the case of ectopic pregnancy, to the presence of clotted blood.1 However, to our knowledge, no study has addressed whether this finding excludes the diagnosis of a normal intrauterine pregnancy (IUP). If this US finding reliably excludes the diagnosis of a normal IUP, it would allow a dilatation and evacuation procedure (D⫹E) to be performed without the fear of placing a viable fetus at risk. In a patient with an abnormal IUP and heavy bleeding,

From the Department of Emergency Medicine, Boston Medical Center, Boston, MA (RD, LD, PM). Received April 30, 1998; revision received September 17, 1998, and October 14, 1998; accepted October 15, 1998. Presented at the New England Regional SAEM meeting, New Haven, CT, April 1998, and the SAEM annual meeting, Chicago, IL, May 1998. Address for correspondence and reprints: Robert Dart, MD, Department of Emergency Medicine, Dowling 1, Boston Medical Center, One Boston Medical Center Place, Boston, MA 02118.

exclusion of a normal IUP by other means [open cervical os, falling quantitative beta – human chorionic gonadotropin (␤-hCG) values, a progesterone value 3,000 mIU/mL] or if they were lost to follow-up. Patients were followed until a final diagnosis was ultimately determined. Results: A total of 83 patients with echogenic material were identified. Of these, three patients were excluded because a D⫹E was performed prior to an IUP being excluded by other means, and two patients were lost to followup. Of the 78 enrolled patients, none had a final diagnosis of normal IUP. Ectopic pregnancy was confirmed in nine patients, five of whom had no adnexal or cul de sac findings to suggest this diagnosis. Conclusion: In symptomatic patients who have echogenic material but no gestational sac visualized within the endometrial cavity at US, the likelihood of a normal IUP is low. Key words: ectopic pregnancy; intrauterine pregnancy; ultrasound; diagnosis; echogenic; transvaginal; endometrial cavity. ACADEMIC EMERGENCY MEDICINE 1999; 6:116 – 120

this procedure could minimize further hemorrhage. In the patient with suspected ectopic pregnancy, a D⫹E provides important diagnostic information. Identification of chorionic villi in the pathology specimen effectively excludes this diagnosis. In the absence of villi, rising or abnormally falling beta – human chorionic gonadotropin (␤hCG) values are highly suggestive of this diagnosis. The purpose of our study was to determine whether the presence of echogenic material within the endometrial cavity, in the absence of a gestational sac, excludes the diagnosis of a normal IUP.

METHODS Study Design. A retrospective chart review was performed to determine the frequency of patients with echogenic material within the endometrial cavity who have a final diagnosis of a normal IUP. This study was approved by the site institution’s human studies committee. Study Setting and Population. The study included patients evaluated from August 1991

ACADEMIC EMERGENCY MEDICINE • February 1999, Volume 6, Number 2

through August 1997 in an urban academic ED with an annual census of approximately 42,000 adult patients. The rate of ectopic pregnancy in our patient population is relatively high; as many as 15% of symptomatic patients in the first trimester are ultimately diagnosed as having this condition.2 The primary site for obstetric care at our instution is an outpatient obstetrics/gynecology clinic. In addition, a number of affiliated neighborhood health centers provide outpatient obstetric care. However, symptomatic patients in the first trimester of pregnancy are often referred directly to the ED or are referred after a clinical evaluation because of the ED’s ready access to emergency US. The emergency radiology area is located next to the ED and is the site where all urgent US examinations are performed. Patients were eligible for the study if they complained of abdominal pain or vaginal bleeding, had a positive serum ␤-hCG, and had a transvaginal US examination that demonstrated echogenic material within the endometrial cavity in the absence of a gestational sac. Patients were excluded if they had a D⫹E procedure performed prior to the exclusion of a normal IUP by other means (␤-hCG >3,000 mIU/mL) and no sac by US,3 falling ␤-hCG values, a progesterone value 1,000 mIU/mL. From January 1995 until the end of the study, 24-hour-per-day in-house US coverage became available and US was performed irrespective of the quantitative ␤-hCG value. From August 1991 to January 1996, all patients without evidence of an IUP by US were admitted for further evaluation. From January 1996 to the end of the study, the decision to admit or discharge patients without evidence of an IUP was left to the discretion of the treating clinicians. Cases were identified in one of three ways.

117

TABLE 1. Definitions Normal intrauterine pregnancy (IUP): a pregnancy carried to term, or an IUP with a fetal heart beat identified at a followup ultrasound examination. Abnormal IUP: 1) falling quantitative beta – human chorionic gonadotropin (␤-hCG) values, presence of an open internal cervical os, a ␤-hCG value >3,000 mIU/mL, or a progesterone value 3,000 mIU/mL, or a progesterone value 66%8 or that fall by >50%9 over a 2-day period are associated with a decreased risk of ectopic pregnancy, neither excludes this diagnosis with absolute certainty.8 – 10 Therefore, most patients require multiple follow-up visits if the diagnosis of ectopic pregnancy is to be confidently excluded using serial ␤-hCG values alone. Since our cohort of patients with echogenic material have a median ␤hCG value of >3,000 mIU/mL, this management

119

approach is clearly dependent on having a reliable patient who has ready access to follow-up. We believe uterine evacuation should be considered at the time of the ED visit in the following circumstances: 1. When the clinical suspicion of ectopic pregnancy based on history and physical examination is high. Although history and physical examination are unreliable in excluding the diagnosis of ectopic pregnancy, the combination of a history of moderate to severe pain along with the presence of cervical motion tenderness or a history of infertility or tubal ligation identifies a patient at increased risk.11 2. When the patient is unreliable or does not have ready access to follow-up. Identification of chorionic villi on gross inspection of the pathology specimen correlates with the final pathology report in about 90% of cases and can be used as a factor in deciding whether the patient can be safely discharged.4 Manual vacuum aspiration using a vacuum locking syringe and an appropriately sized cannula is an excellent option when uterine evacuation is being performed for diagnostic sampling. This procedure has been demonstrated to be safe, does not necessitate cervical dilatation, and can be performed with little or no anesthesia in most cases.12 These advantages result in significant time savings compared with formal D⫹E, making this procedure well suited for use in the ED.

LIMITATIONS AND FUTURE QUESTIONS We have identified a number of limitations to our study. First, the finding of echogenic material was based on the radiology report rather than the hard copy of the US examination itself. This was because the hard copies of many of the older examinations were not available. Characterization based on the hard copy of the actual US examination might have led to a change in the classification of some of these examinations. However, when this finding was identified in the radiology report, it accurately identified an abnormal pregnancy. Second, in practice, it is occasionally difficult to differentiate a thickened endometrium from the presence of echogenic material within the endometrial cavity. The endometrial stripe is also echogenic but has a regular contour, and the surrounding endometrium typically has a homogeneous echotexture. Obviously some examinations show intermediate findings, making it somewhat difficult to tell whether you are visualizing a thickened stripe or discrete echogenic material. Part of the reason for this difficulty is that sometimes a thickened endometrium and echogenic material are representing the same process,

120

TRANSVAGINAL US

i.e., retained products of conception. On the other hand, a thickened endometrium is often the initial US finding in an early but normal IUP.13 Therefore, the ultrasonographer should be cautious about reporting the identification of echogenic material if the US findings are not conclusive. Our study design does not provide a way to assess interrater reliability in identifying this US finding. If the interrater reliability is poor, then the clinical utility of this finding would be limited. We cannot definitively state that none of the patients either lost to follow-up or who had a D⫹E prior to exclusion of a normal IUP by other means harbored a normal but early IUP. In addition, given our relatively small sample size, our results are consistent with as many as 7% of patients with these findings harboring a viable IUP. Therefore, we cannot recommend that clinicians who are contemplating uterine evacuation use this US finding as the sole determinant in excluding the diagnosis of a viable IUP until these findings are prospectively validated. In our cohort of patients, additional alternative evidence often was available at the time of the ED visit to exclude the possibility of a viable IUP. The absence of a gestational sac in patients with a ␤hCG value of >3,000 mIU/mL3,14 excludes the diagnosis of a normal IUP with a high degree of certainty. Since the median ␤-hCG value in our study was 3,450 mIU/mL, this finding alone would exclude the diagnosis of a normal IUP in more than half of our study population. In patients with ␤hCG values below 3,000 mIU/mL, a progesterone value can be measured. Since the majority of the progesterone values obtained from our cohort were 38 days or ␤-hCG >3,000 mIU/mL, but without a gestational sac on ultrasonography. Acad Emerg Med. 1997; 4:967 – 71. 4. Stovall TG, Ling FW, Carson SA, Buster JE. Serum progesterone and uterine curettage in the differential diagnosis of ectopic pregnancy. Fertil Steril. 1992; 57: 456 – 7. 5. Nielsen S, Hahlin M. Expectant management of first trimester spontaneous abortion. Lancet. 1995; 345:84 – 6. 6. Stovall T, Ling F, Carson S, Buster J. Nonsurgical diagnosis and treatment of tubal pregnancy. Fertil Steril. 1990; 54: 537 – 9. 7. Dart R, Dart L. Inpatient vs outpatient evaluation of patients with suspected ectopic pregnancy: does it impact on the rate of delayed diagnosis? [abstract]. Acad Emerg Med. 1998; 5:471. 8. Kadar N, Caldwell B, Romero R. A method for screening for ectopic pregnancy and its indications. Obstet Gynecol. 1981; 58:162 – 6. 9. Kadar N, Romero R. Further observations on serial chorionic gonadotropin patterns in ectopic pregnancies and spontaneous abortions. Fertil Steril. 1988; 50:367 – 70. 10. Gretz E, Quagliarello J. Declining serum concentrations of the ␤-subunit of human chorionic gonadotropin and ruptured ectopic pregnancy. Am J Obstet Gynecol. 1987; 156:940 – 1. 11. Dart R, Kaplan B, Cox C, Paul E. How useful are history and physical examination in identifying ectopic pregnancy in patients with first-trimester pain and/or bleeding? Acad Emerg Med. 1996; 3:483. 12. Kizza A, Rogo K. Assessment of the manual vacuum aspiration equipment in the management of incomplete abortion. East Afr Med J. 1990; 67:812 – 22. 13. Spandorfer S, Barnhart K. Endometrial stripe thickness as a predictor of ectopic pregnancy. Fertil Steril. 1996; 66: 474 – 7. 14. Kadar N, Bohrer M, Kemmann E, Shelden R. The discriminatory human chorionic gonadotropin zone for endovaginal sonography: a prospective randomized study. Fertil Steril. 1994; 61:1016 – 20. 15. McCord M, Muram D, Buster J, Arheart K, Stovall T, Carson S. Single serum progesterone as a screen for ectopic pregnancy: exchanging specificity and sensitivity to obtain optimal test performance. Fertil Steril. 1996; 66:513 – 6. 16. Dart R, Dart L, Segal M, Page C, Brancato J. The ability of a single serum progesterone value to identify abnormal pregnancies in patients with beta human chorionic gonadotropin values less than 1,000 mIU/mL. Acad Emerg Med. 1998; 5:304– 9.