Diagnosis of Ectopic Pregnancy ============================== þ General Trends in HCG in Ectopics: - "Approximately 50% of women with an ectopic pregnancy present with increasing hCG levels, and 50% present with decreasing hCG values.36 However, 71% of women who receive a diagnosis of an ectopic pregnancy have serial serum hCG values that increase more slowly than would be expected with a viable intrauterine pregnancy or decrease more slowly than would be expected with a miscarriage." "A decrease in the hCG level of 20% or more, 12 to 24 hours after uterine evacuation, suggests that trophoblastic cells were probably removed from the uterus (i.e., that there was a nonviable intrauterine gestation)." [Barnhart, K. T. (2009). "Clinical practice. Ectopic pregnancy." N Engl J Med 361(4): 379-387.] þ Signs of ectopic: - abdominal pain in 90% of cases. [Brennan DF. Ectopic pregnancy - Part I: clinical and laboratory diagnosis. Acad Emerg Med 1995;2:1081-9.] - but may present as syncope. - missed menses or abnormal bleeding in only 75% of cases. [Jehle D, et al. Ectopic pregnancy. Emerg Med Clin North Am 1994;12:55-71.] - of those with moderate/severe pain, lateralizing pain, and cervical motion tenderness, 9/23 had ectopic. - of those with minimal-to-absent midline pain wihtout cervical motion tenderness or above risk factors, only 2/60 had an ectopic. [Dart R, et al. How useful are the history and physical in identifying ectopic pregnancy in patient with first-trimester pain and/or bleeding? (abstract presented at the meeting of SAEM, May 1996.)]. - 10% will have normal pelvic exam. þ Urine vs Serum HCG: - "The only thing that has changed over the years is the progressively increased sensitivity of urine HCG tests which have fallen from detecting 800-1000 IU/l to 25 - 50 IU/l. This allows detection of pregnancy before the first missed menses and corresponds to about the HCG levels expected one week after fertilization (50 IU/l). With urine sensitivities this low, false negative tests should be in the range of about 3%. [Weigel M, Friese K, Schmitt W, Strittmatter HJ, Melchert F. [What is the predictive value of ultrasound diagnosis in suspected extra-uterine pregnancy in routine clinical practice?] Zentralbl Gynakol 1993;115(5):228-32.] Abstract: "As you pointed out last year, the presence of dilute urine is one cause of a falsely negative UCG." [Cartwright PS, et al. Performance of a new enzyme-linked immunoassay urine pregnancy test for the detection of ectopic gestation, Ann Emeg Med 1986; 15(10):1198.] Abstract: "And IV hydration does appear to disproportionately decrease urine as opposed to serum levels." [Holt JA, et al. Effects of IV hydration on levels of hcg in the serum and urine of women with possible ectopic pregnancy. Lab Med 1989;5(3):701.] Abstract: "As we have learned, even a negative serum HCG does not entriely exclude the diagnosis of ectopic pregnancy and, in fact, I received a personal communication from someone today who had done a study and found 3/74 ectopic pregnancies with undetectable serum HCG levels." [Olson CM, Holt JA, Alenghat E, et al. Limitations of qualitative serum beta-HCG assays in the diagnosis of ectopic pregnancy. Journal of Reproductive Medicine 1983;28:838-842.] "The bottom line is that if your index of suspicion is high enough a confirmatory serum level needs to be performed if UCG is negative. (quotes from H. Louzon MD) þ HCG Levels in ectopic pregnancy - Discriminatory of ectopic with single HCG levels (no ultrasound)? + By the time of missed menses (13-14 days post-conception) hCG levels usually range from 50-300; urine hCG assays detect 20-50 mIU/ml, so should detect this. + Some report that a single value of > 35,000 [Marill K, et al. Utility of a single beta hCG measurement to evaluate for ectopic pregnance. (Abstract presented at SAEM May 1996.)] or > 50,000 [Stovall TG, Ling FW. Ectopic pregnancy: Diagnostic and therapeutic algorithms minimizing surgical intervention. J Reprod Med 1993;38:807-812.] distinguish an intrauterine pregnancy (IUP), but ruptured ectopics have been reported with levels from 10 to more than 100,000. [Barnhart K, et al. Prompt diagnosis of ectopic preganancy in an emergency department setting. Obstet Gynecol 1994;84:1010-1015.] - Discrimination of ectopic with serial HCG levels: + hCG titres usually double about every 2 days with IUP. + subnormal increases occur in about 85% of ectopics. [Carson SA, Buster JE. Ectopic pregnancy. N Engl J Med 1993;329:1174-1181.] - Discrimination of ectopic by HCG levels combined with ultrasound: + See Also: + quick and dirty: 1000-1500 mIU/ml for endovaginal, 6500 for transabdominal. [Brennan EF. Ectopic pregnancy--Part II: diagnostic procedures and imaging. Acad Emerg Med 1995;2:1090-1097.] + "If one chooses a discriminatory level of 4500 nearly 100% of patients with a nondiagnostic US will have an ectopic." [Ankum WM, Van der Veen F, Hamerlynck JV, Lammes FB. Transvaginal sonography and human chorionic gonadotrophin measurements in suspected ectopic pregnancy: a detailed analysis of a diagnostic approach. Hum Reprod Aug 1993;8(8):1307-11.] Abstract: "On the other hand lowering the discriminatory level to 1000 increases the sensitivity (you'll find more ectopics) but the specificity will drop to less than 50%." [Kaplan et al. Ectopic Pregnancy: Prospective Study with Improved Diagnostic Accuracy. Ann Emer Med 1996;28:10-17.] + "A study in the Annals last month looking at the value of hx/physical/US/HCG in diagnosing ectopic pregnancy concluded that in a high risk inner city population, the likelihood of finding an EP was four-fold *greater* if the HCG was less than 1000 than it was at levels over 1000." [???] "What this says, essentially, is that the HCG is worthless as a discriminatory tool in the ED for diagnosis of EP (unless the level is over 4500). Why? Because *above* the discriminatory zone one must suspect EP with a nondiagnostic US (traditional teaching confirmed by numerous studies) AND *below* the discriminatory zone the likelihood of EP is *even higher*! Most practitioners set the level at which EP is considered 'suspicious' (with a nondiagnostic US) somewhere between 1000 and 4500. So how does one proceed, for example, to manage a patient with an HCG level of 252, as above? According to this study this patient should be admitted for further testing and, in fact, all of the patients in this study with measurable serum HCG and nondiagnostic US *were* admitted. Traditional teaching would conclude, by contrast, that US cannot be expected to disclose the presence of a gestational sac with an HCG less than 1000 and that this patient's risk of ectopic would be *less* than had the level been over 1000. As an aside this study found that 9% of patients with EP had *painless* vaginal bleeding and that 36% lacked adnexal tenderness." + "To confuse the issue even further, one study found that patients with EP who had low levels of HCG (72-5000) could often be managed expectantly (without US) as outpatients." [Letterie GS, Hibbert ML, Ramirez EJ. Expectant management of abnormal concentrations of human chorionic gonadotropin during the first trimester of pregnancy. Gynecol Obstet Invest 1991;31(3):176-8.] Abstract: "73% of patients with HCG levels below 1000 had spontaneous resorption of EP in another study and were also managed as outpatients." [Trio D, Strobelt N, Picciolo C, Lapinski RH, Ghidini A. Prognostic factors for successful expectant management of ectopic pregnancy. Fertil Steril Mar 1995;63(3):469-72.] Abstract: "It is somewhat surprising to realize that intrauterine gestations can be detected, in some cases, with serum HCG levels as low as 40!" [Enk L, Wikland M, Hammarberg K, Lindblom B. The value of endovaginal sonography and urinary human chorionic gonadotropin tests for differentiation between intrauterine and ectopic pregnancy. J Clin Ultrasound Feb 1990; 18(2):73-8.] Abstract: + "We sometimes will see patients who become pregnant soon after a normal delivery or abortion. It is useful in these cases to have an idea of what the normal decay curve is for HCG. After induced abortion the level should be undetectable by 40 days" [Thyssen HH, Christensen H, Schebye O, Berget A, Arends J, Larsen SO. [Elimination of human chorionic gonadotropin in serum and urine after uncomplicated induced abortion during the first trimester] Ugeskr Laeger Jul 1992;154(30):2071-2.] Abstract: "and after a normal pregnancy by the third week post-partum." [Haenel AF, Hugentobler W, Brunner S. [The postpartum course of the HCG titer of maternal blood and its clinical relevance] Z Geburtshilfe Perinatol Nov-Dec 1986;190(6):275-8.] Abstract: (quotes from H. Louzon, M.D.) þ False-negative beta HCGs: - are very rare, almost never occurring: [Taylor RN, Padula C, Goldsmith PC. Pitfall in the diagnosis of ectopic pregnancy: immunocytochemical evaluation in a patient with false-negative serum beta-hCG levels. Obstet Gynecol 1988; 71(6 Pt 2): 1035-8.] Tubal pregnancy specimens from a patient with undetectable serum beta-hCG levels and a control case with a low beta-hCG titer by radioimmunoassay were analyzed using peroxidase-antiperoxidase immunocytochemical localization of placental proteins. The results indicated that deficient production of beta-hCG by ectopic trophoblast was responsible for false-negative pregnancy tests in one patient. þ Summary for Mgt of ? Ectopic Richard O. Gray MD. wrote: > Do you US all pts without > confirmed IUPs or just those with abdominal pain, or just those with > lateralizing signs? How do you interpret a negative (no IUP) ultrasound > in a women with minimal sxs (or just bleeding without pain)? We have > been much more aggresive in pursuing formal US in situations with an HCG > less than 1,000 since reviewing your study (we don't have a tv probe and > have had several cases of ectopic with a lower hcg) but often end up > sending women out with ectopic vs early pregnancy. > Rich, I recommend US for all hemodynamically stable, first trimester pregnant women with lower abdominal pain or tenderness, or bleeding (or syncope or unexplained back pain). I believe lateralizing tenderness is NOT an accurate predictor and this has been demonstrated in several studies. If tissue passage has been documented or the internal os is open, then I defer the decision to ultrasound to the obstetrician (as ectopic has been rule out clinically). When the US is non-diagnostic, the beta-hCG and past OB history come into play. A few women will have a history of tubal ligation, infertility, prior ectopic, or indwelling IUD and must be admitted to the OB service. If the beta-hCG is over 1,000 (or 5,000 if only transabdominal US available) then I admit to the OB service as their remains a high risk of ectopic. If the beta-hCG is under 1,000 (or 5,000 if transabdominal US), the patient can be sent home with good aftercare instructions and a scheduled appointment in 2 days for a repeat beta-hCG. I know some believe the above method is not cost-effective. However, at Cook County Hospital we have drastically reduced cases of missed ectopic pregnancy since intitiating the above protocol. Also, the cost of ultrasounds has little to do with what is charged to read them. Tom Scaletta, MD ----------------------------- If 14% of your ectopics had HCG < 1000 and from the Kaplan paper (below) where almost a third of women with ectopics had levels < 1,000, why do you feel comfortable sending these patients home. One answer would be that only patients with clinical risk factors for ectopic had ectopics with the HCG < 1000. I doubt this to be true, though. Another agrument would involve costs. What are your thoughts? In case you ask, I am sending 'some' of these patients home with quick follow-up, also. Inconsistent guidelines on the 'some' are HCG<500, only bleeding without pain, mild/minimal pain, or consultant input. This is based on intuition, not data. Dan Daniel M. Joyce, MD EM, SUNY HSC-Syracuse ECTOPIC PREGNANCY: PROSPECTIVE STUDY WITH IMPROVED DIAGNOSTIC ACCURACY Kaplan BC, Dart RG, Moskos M, Kuligowska E, Chun B, Hamid MA, Northern K, Schmidt J, Kharwadkar A. Ann Emerg Med. 1996;28(7):10-17. SUMMARY: This prospective study assesses the accuracy of various laboratory and clinical characteristics in the diagnosis of first trimester ectopic pregnancy. All 481 women who presented to the Boston City Hospital Emergency Department over a thirteen month period with a chief complaint of abdominal pain or vaginal bleeding with a positive serum hCG (obtained in the ED) and uterus smaller than twelve week size were enrolled. Hemodynamically stable patients without signs of inevitable or incomplete abortion underwent either trans-abdominal or trans-vaginal (if the former was inconclusive) ultrasonography. Clinical findings and quantitative á-hCG levels were obtained and recorded. Ultrasound examinations revealing a double-ringed gestational sac and/or yolk sac, fetal pole, or heartbeat were judged as a normal intrauterine pregnancy (IUP), and patients were consequently discharged. If an adenexal sac was found, patients were considered to have a tubal pregnancy and were admitted for surgery. Patients with non-diagnostic ultrasounds underwent repeat studies, dilatation and curettage (D&C), and/or laparoscopy / laparotomy. Forty two patients were ultimately lost to follow-up. Of the remaining 439 patients, 218 (50%) had a normal IUP by ultrasound, 79 (18%) had abnormal IUPs that required D&C, and 127 patients (29%) were admitted to rule out ectopic pregnancy. Of these, 34 (27%) had ultrasonography highly suggestive of ectopic pregnancy. 72 other patients had non-diagnostic ultrasounds, while nineteen lacked an initial study. Of this subset of 91 patients, 21 (23%) were found to have ectopic pregnancies, 58 (64%) had abnormal IUPs, and twelve (13%) had normal IUPs. Quantitative á-hCGs were also analyzed. Patients with a á-hCG of less than 1,000 mIU/ml were found to have a four fold risk of having an ectopic pregnancy. However, 62% of patients with ectopic pregnancies had á-hCGs greater than 1,000 mIU/ml and 20% of patients with normal IUPs had á-hCG less than 1,000 mIU/ml. Most previous studies had shown the opposite trend: i.e.: a á-hCG level >1,000 with a non diagnostic ultrasound suggested an ectopic pregnancy. Finally, initial clinical impressions were compared with the final diagnosis. 91% of women with documented ectopic pregnancies complained of pain, while the remaining 9% presented with bleeding and no pain. Over one ectopic in 5 (21%) presented with pain and no bleeding. In addition, twenty of the 56 women (36%) with documented ectopic pregnancy were without adenexal tenderness. (Anthony Yang, MD) --------------------------- Dan, It is true in the abstract Beth and I published that 14% (22/162) of the ectopic preganancies had both an indeterminate ultrasound and a beta-hCG < 1,000 mIU. Of these, 32% (7/22) were ruptured. I remember that nearly every one of these had a concerning clinical feature illuminated by the retrospectoscope (i.e., hemodynamic instability, historical risk factor, severe anemia, etc.). I believe few hemodynamically stable, first-trimester preganant woman presenting with pain and/or bleeding, without historical risk factors for ectopic preganancy, with a non-diagnostic TV ultrasound, and with a beta-hCG under 1,000 mIU have ectopic pregnancies. Some are though. The real risk of sending home a potential ectopic, of course, is that the patient will rupture before diagnosis. Very few ectopics that are sent home intact will rupture in the intervening two days. No system short of admitting everyone is full-proof. I think that the "ultrasound, beta-hCG, historical risk factors method" is an sufficiently accurate. I agree with you that subjective intuition plays a role. In Beth's paper nearly 80% of those with intial US were diagnostic. Of those non-diagnostic, I am not sure how many ectopic preganancies had beta-hCG >1,000 and no historical risk factors. I will ask Beth. I suspect the discussion will soon turn to what rate of ectopics being sent home (if any) is reasonable. It will be interesting to use data from published studies and a decision analysis model to study this question. Perhaps there are enough centers interested enough to begin a multi-center study. I still think that some institutions are not obtaining ultrasound on all "at risk" patients. This, I suspect, is responsible for most missed ectopic pregnancy cases nationwide. Some get a quantitative beta-hCG first and only ultrasound those over a certain threshold. Scary! Tom Scaletta, MD Of 177 ectopic preganancies which I retrospectively reviewed at San Francisco General Hospital, 124 had an initial ultrasound. Of these 71% were either diagnostic (12%) or suggestive (59%) for ectopic pregnancy. Sixty-five (37%) of the total had beta-hCG under 1,000 mIU. Forty-two of these had an ultrasound. Of these 71% were either diagnostic (5%) or suggestive (66%) for ectopic pregnancy. Of note some of the ultraounds were transabdominal only and still we demonstrated diagnostic accuracy (at least suggesting ectopic pregnancy). Beth and I also reported 2 cases in our combined data (n=297) of ectopic pregnancy with beta-hCG over 100,000 mIU (0.7%) [NEJM (330(10):713)]. This was published as a letter to the editor in response to a review article written by Buster and Carson, two gynecologists, who claimed ectopics could be excluded beta-hCG levels over 100,000 mIU. They did recommend ultrasound as an initial screen but neglected to address the further work-up/disposition patients with indeterminate ultrasounds. þ Progesterone levels for diagnosis of ectopic: - progesterone decreases in nonviable pregnancies - progesterone < 5 ng/ml indicates nonviability (100% specific) while levels of > 25 ng/ml usually indicate viability (97.5% sensitive) - but nonviable does not equal ectopic [Carson SA, Buster JE. Ectopic pregnancy. N Engl J Med 1993;329:1174-1181.]