Progesterone & Ectopic Pregnancy

Progesterone & Ectopic Pregnancy

Assays for chorionic gonadotropin (hCG)
and progesterone (PG) represent the immunoassay laboratory's main contribution to the investigation and management of suspected ectopic pregnancy (EP).¹
Most of the other analytes measurable by immunoassay which have sometimes been proposed for use in this context—AFP, inhibin, OM-MA (CA125), placental protein 14
and 17a-OH PG, for example—have generally been marginalized or discredited.²

Whereas hCG's role
is universally acknowledged, that of PG remains controversial. Strongly advocated by some, it has been ignored altogether or dismissed as "redundant" by others.³ In routine practice, when blood work is first sent to the laboratory, PG is frequently one of the tests requested. All too often, however, there may be no clear guidelines as to how the workup for EP should be affected by the PG result. Hence the issue of redundancy: does this test really add anything?

Two approaches
There have been two major ways of exploiting PG determinations in the context of suspected EP:

A) In one algorithm, developed and extensively tested during the past decade or so, PG figures prominently, ahead of quantitative hCG and ultrasound, with a sufficiently high or sufficiently low PG result taken as obviating the need for these.⁴ This two-sided algorithm capitalizes on comparatively extreme PG levels, but leaves open the question whether clinically useful information can be extracted from results within the interval spanned by the cutoffs.⁵

B) In the second approach, a PG result is interpreted relative to a single, rather more central cutoff, either on its own or in relation to the initial hCG level, likewise reduced to the status of a binary value.⁶ Overall, the few studies which have attempted, in this manner, to integrate PG results into
the workup for suspected EP must be regarded as tentative, in part because they have typically discarded information prematurely, by imposing cutoffs
on continuous assay results, and focused on pair-wise relationships, in spite of the many variables involved, but also because there are serious concerns regarding transferability from one setting to another, given differences in prevalence and fundamental patient characteristics.⁷

Accordingly, this article focuses on the first approach. While the second appears to be more common, it has so far been less well developed.

Of course, EP is not always a diagnostic dilemma. A negative "pregnancy test" immediately and decisively advances the diagnostic process.⁸
At the other extreme, a sufficiently acute presentation might dictate transferring a patient to the operating room without waiting on ultrasound examination or laboratory results.⁹

For a woman hemodynamically stable and known to be pregnant, the central questions are: Does she have a normal pregnancy? If not, has it failed, or is it destined to fail? And is the pregnancy intrauterine or ectopic? Thus viability and location represent the fundamental dimensions of the diagnostic problem, corresponding to three basic possibilities: normal intrauterine pregnancy (NIUP), failed intrauterine pregnancy (FIUP: encompassing spontaneous abortions, etc.), and EP.¹⁰ One additional possibility, however, must always be considered, namely, heterotopic pregnancy: the combination of an ectopic implantation with a second, intra- or extrauterine implantation.

The physiological context
In light of their origin and dependencies during early pregnancy, both hCG
and PG are major candidates for the role of endocrine markers of viability, especially during gestational weeks 4 through 8 when the need to confirm or rule out EP most often arises.¹¹

Normally, the developing organism begins to implant itself in the uterine lining (made receptive by the action of PG) approximately 6 days after fertilization—which occurs, during the journey from ovary to uterus through the fallopian tube, 7 or 8 days after rupture of a mature follicle results in the extrusion of an oocyte at ovulation. But rough passage along the way—due, for example, to scarring or obstruction of the fallopian tube associated with tubal surgery or with pelvic inflammatory disease (PID) caused by chlamydial (or gonorrheal) infection—may result in an ectopic pregnancy of by far the most common sort: a tubal implantation. Abdominal, cervical and ovarian implantations can also occur, but together they account for less than 2 percent of ectopic pregnancies after spontaneous conceptions.

hCG, secreted in rapidly increasing amounts by the trophoblast, can be readily detected in the maternal circulation as early as 10 days after fertilization, at a gestational age of about 2.5 weeks. During the early weeks, PG is secreted predominantly by the corpus luteum, which forms from the ruptured follicle on the ovarian surface, though eventually (after two months
or so) the placenta, which develops from the outer layer of the trophoblast, becomes the principal source of this steroid hormone. PG is essential to the continued development of the embryo and placenta. Maintenance of the corpus luteum, in turn, depends on hCG stimulation.

The gestational age-related patterns of these two hormones differ markedly
during the early weeks of pregnancy. Normally, hCG levels double every 2 to 3 days, following a steep, log-linear trajectory,¹² whereas PG levels remain—at least by contrast—relatively constant. In fact, there is room for more definitive characterization of first-trimester PG levels, based on larger studies, modern assay methods, and more precise gestational age estimates, bearing in mind that spontaneous pregnancies might well show major differences in pattern and/or concentration levels compared to pregnancies relying on one form or another of artificial reproductive technology (ART).

On the other hand, for a woman presenting to the emergency room (ER) with symptoms suggestive of an EP, it is rare to have more than a very crude estimate of how far along she is in her pregnancy. Accordingly, we are forced back to a correspondingly simplistic model of gestational age-invariant normal values for PG and similarly invariant normal doubling times for hCG, providing a frame of reference for interpreting a single PG result, or a pair of hCG results obtained approximately two days apart.

In pregnancies seriously compromised by an undesirable implantation site (EP), or by other causes (which might result in miscarriage), the relative failure to thrive is reflected in a rate of increase for hCG which is lower on average than that for a normal pregnancy, though there is considerable overlap. (An increase by at least two-thirds over a 48-hour period is typically quoted as the lower limit of normal.¹³ ) As the trophoblast is both a source of PG and a major influence, via hCG stimulation, on the corpus luteum's production of this hormone, it is natural to expect lower PG levels in many compromised pregnancies, as well as a slower doubling or decline in hCG values.

Origins of the "progesterone-first" strategy
Early studies confirmed this expectation and even suggested little or no overlap between the PG levels found in normal pregnancies, on the one hand, and abnormal pregnancies, on the other, whether intrauterine or ectopic. Thus, in a small, retrospective study (n = 70) by Yeko et al., reported in 1987, a PG cutoff of 15 ng/mL (47.7 nmol/L) cleanly separated all of the EPs, and all but one of the other abnormal pregnancies, from all of the NIUPs.¹⁴ 14. Figure 1: [Yek87]. (See Figure 1.) In spite of these results, anticipating more extensive overlap between viable and nonviable pregnancies, the authors recommended interpreting PG levels in the 10 to 20 ng/mL range (31.8 to 63.6 nmol/L) with caution.

Figure 1. Progesterone levels in early pregnancy. (From Yeko TR, Hughes LH, Rodi IA, Buster JE, Sauer MV. Timely diagnosis of early ectopic pregnancy using a single blood progesterone measurement. Fert Steril 1987;48:1048-50. Reproduced by permission of the publisher, the American Society for Reproductive Medicine.)

Indeed, subsequent work along these lines led to adopting what can be regarded, from one point of view, as a still broader gray zone, extending from 5 to 25 ng/mL (15.9 to 79.5 nmol/L), in a "PG-first" strategy.¹⁵ From another point of view, the 5 and 25 ng/mL cutoffs represent thresholds for subjecting patients to further noninvasive tests based mainly on serial hCG determinations and transvaginal ultrasound (TVUS).

The PG-first algorithm was tested over a five-year period against the
outcomes for more than 3600 women with positive pregnancy tests in
an ED setting in Memphis, Tennessee.¹⁶ Satisfaction with the results led proponents to assert, in the classic textbook edited by Stovall and Ling (1993), that "the role of serum PG measurements in ectopic pregnancy diagnosis has now generally been resolved."¹⁷

The strategy relies on two distinct PG cutoffs, partitioning the subjects into three groups, with the goal of clarifying almost immediately the status of a substantial subset: those patients with very low or very high PG levels.

Very low progesterone levels
Sufficiently low levels are nearly always associated with nonviable pregnancies.¹⁸ Location remains an issue, but whether or not the pregnancy was ectopic can be settled—according to this classic algorithm—by examining material obtained by uterine curettage for trophoblastic and/or embryonic elements.¹⁹

Clearly, this does risk inadvertently terminating a viable intrauterine pregnancy unless the relevant PG cutoff is set below the lowest level compatible with a viable pregnancy. For some years, 5 ng/mL appeared to be a safe choice; this is, after all, low even compared to midluteal phase levels expected for a normal menstrual cycle. Eventually, however, two viable pregnancies with PG levels below 4 ng/mL (12.7 nmol/L) in the Memphis series, and similar cases from other centers, undermined the absoluteness of this cutoff.²⁰ Moreover, studies based on ART pregnancies demonstrated that even lower, "subluteal" PG levels in the early weeks of pregnancy are compatible with successful outcomes—though it appears to be an open question whether this holds true for spontaneous pregnancies as well.²¹

Although uterine curettage is relatively noninvasive, compared to laparoscopy or laparotomy, and can be performed even in an office setting, many would prefer to eliminate its use as a diagnostic tool in favor of genuinely noninvasive procedures, consistent with the overall trends in EP diagnosis and management.²² Especially given that PG determinations can yield at best presumptive evidence for nonviability, it makes sense not to place too great a burden on a single PG result, which could be flawed, in any case: better to seek confirmation by some independent method, such as serial hCG determinations, before taking an irrevocable step. Here, some redundancy—the use of multiple diagnostic methods—can be a virtue, because of the protection it affords.²³ This point of view has its advocates; and indeed it is clearly visible in mature diagrams of the PG-first algorithm, though not in some early and simplified presentations.

Very high progesterone levels
As for the other PG cutoff (25 ng/mL) involved in the two-sided PG-first strategy: sufficiently high levels are nearly always associated with NIUP, absent the classic symptoms of EP.²⁴ In the Memphis series, roughly half the subjects could be reassured on this basis and exempted from ultrasound examinations, hospitalization, and other costly measures.²⁵ Subjects with values below this cutoff could be recalled, if necessary, for further testing, based on having a lower PG, even in the absence of symptoms raising suspicion for EP.

A recent study by Buckley et al., performed in San Diego, California, and published just last year, sheds light on this branch of the PG-first algorithm.²⁶ Using DPC's Coat-A-Count® Progesterone assay, these investigators arrived at a cutoff of 22 ng/mL (70 nmol/L), corresponding to a level just above the highest concentration observed for any woman with EP in the initial phase of their study. The San Diego study was limited to pregnant women in their first trimester presenting to the ER with symptoms suggestive of EP. The results demonstrated that roughly one quarter of these patients could be reclassified as low risk based on a PG result above the cutoff, thus reducing the need for urgent diagnostic evaluation or admission, providing there are no "strong" clinical signs of EP. Relative to statistics from the Memphis series, which relied on an upper limit for PG of comparable magnitude, this represents a much smaller but still significant percentage, suggesting that an initial PG determination can be useful for resource channeling (triage) within a symptomatic population.

In the Memphis setting, as noted above, PG results were obtained on an apparently much broader class of subjects presenting to the ER in their first trimester, judging from the inclusion criteria. Whereas the San Diego study involved either subjects with "abdominal pain or bleeding" or (by another description) subjects with a "chief complaint of abdominal pain with or without vaginal bleeding,"²⁷ there was no stated restriction to women with symptoms of any kind in the Memphis series. This may help to explain the considerably higher percentage of subjects with elevated PG values in the Memphis series.

It also highlights one additional role played by PG determinations in this context, namely, as what might be regarded as a "screen" for EP in relatively asymptomatic women. Unfortunately, the several publications reporting on the Memphis series specify neither the number of asymptomatic women subjected to more intensive evaluation based on a PG result below the (upper) cutoff, nor the percentage of these who were discovered to have an EP.

Remarkably, in spite of the San Diego study's more restrictive inclusion criteria, which might be expected to isolate a group of women at higher risk for EP, the prevalence of EP in this study—7.5% of 716 subjects—was somewhat lower than the 11% rate reported for the much larger Memphis series.²⁸ Both prevalence figures relate to ER settings.

A view from the physician's office
Decisions regarding the management of first-trimester bleeding and cramping, however, must often be made outside the hospital, in settings far removed from modern, high-tech facilities. Accordingly, in a recent (1996) tutorial review on EP published in the British Journal of Hospital Medicine, Daly and Prendiville tried to address the situation of primary care physicians who must frequently decide, on an outpatient basis, whether a woman presenting in early pregnancy with vaginal bleeding should be sent to the ER for ultrasound and close surveillance.²⁹ EP is, after all, potentially life-threatening and remains the leading cause of maternal death. The early signs and symptoms of a threatened abortion may be quite similar, but this condition is cared for more often than not—the authors cite a survey of general practitioners in Ireland—at home.³⁰

Building their discussion of the diagnosis and management of EP around
a simplified version of the PG-first algorithm, Daly and Prendiville proposed that the branch based on a 25 ng/mL cutoff may also be relevant to the office setting: a single PG result, obtained on blood collected by the physician and sent out for testing, should provide an objective basis, within a day of the patient's visit, either for reassuring her as to the continued viability of her pregnancy, or else for referring her to a suitably equipped center for a thorough EP workup.

The authors presented no new data to support this proposal or to judge its consequences: how it might impact, one way or the other, on the number of women sent to the ER for ultrasound examination and other tests, or on the number of EP cases identified at a relatively early stage. Nevertheless, one can speculate that transposing the PG-first strategy from ER to physician's office should tend to enhance rather than degrade its performance, because it is reasonable to expect a lower prevalence for EP in this context.

Thus, given that EP has an incidence of somewhere between 1 and 2
percent in the US and UK,³¹ while miscarriage is far more common—Daly
and Prendiville quote an estimate of 20%—and given that first trimester bleeding can be due to other causes as well, the prevalence of EP for the population in question should indeed be lower, perhaps very much lower, than the 10% or so observed in the Memphis and San Diego ER settings. From this perspective, the physician's office is not radically different from
the ER. Indeed, results from the Memphis ER series have been taken to support the use of PG assays for the "routine screening of first trimester pregnant women in the office setting," though not in populations at high risk for EP.³²

Thus, at a certain stage in the workup for EP, depending on the patient's history and condition, the finding of a highly elevated PG value—above, say,
20 or 25 ng/mL (63.6 or 79.5 nmol/L)—does appear to be highly informative, though the most appropriate cutoff may well depend both on the assay used and on demographic characteristics of the setting.³³

As yet, the interest of PG results below such a cutoff remains controversial. There have been several relevant studies pointing the way towards a resolution, but this is another story.

Notes
1. General reviews: [Gra99], in an issue devoted to EP; [Bre97], [Pis98], [Tay00], [Ten00].

2. Some other analytes: [Mcc93], [Dit00]; [Kau97]; [Sei96], [Dan98], [Tre00]; [Med98], [Pre00]; [Sta94], [Rom95]; [Cho92], [Rom94], [Cas98].

3. PG and hCG frequently offered together as an "EP panel": [Ste95], [Ole96], [Buc99]. Random-access analyzers like the IMMULITE and IMMULITE 2000 make it possible to generate timely results for both analytes.

4. First approach (dual cutoff): [Sto92a], [Mcc93], [Sto93a], [Sto93b], [Sto94], [Pis98], [Lip00] and many other publications. Note the overlapping authorship. [Mcc93] reviews early studies in this tradition.

5. Whether clinically useful information can be extracted from results within the interval spanned by the cutoffs: tentatively addressed by [Mcc96], but see [Mol97]. More promising are multiple logistic regression models, wherein PG figures as a continuous variable, e.g. [Tho92], [Ban99]. See #34.

6. Second approach (single cutoff). PG used on its own: [Hah90], [Cow92], [Sto92a], [Mol98]. Gestational age-related PG: [Ste93]. PG used with concurrently obtained hCG: [Ole96]. PG used with serial hCG: [Hah95], [Ste95]. [Mcc93] reviews early studies in this tradition.

7. Issues of transferability, prevalence, patient characteristics (spontaneous vs. ART pregnancies): see #10, #33, #34. Multivariate approaches—multiple logistic regression, neural network—based on continuous rather than binary inputs: see #34.

8. Pregnancy tests, hCG assays: [Lin78], [Lin82], [Tay88], [Cha92], [Mac93], [Col97], [Stu98], [Mas99], [Rot00], [Lin00].

9. Extreme presentations rendering diagnostic tests inappropriate. Cf. "threshold" analysis, as in [Pau80], [Gla91], [Hoz99].

10. The fourth possibility—a viable EP: [Gru94], [Sta96] p.152, [Gar00]; see also [Gou97]. Meta-analysis confirms that PG, interpreted via a single cutoff, is a better marker of viability than location: [Mol98]. Some studies conclude that PG is at least as powerful as serial hCG: [Sto92], [Ole96]. Evidently, however, PG's predictive value depends both on prevalence and on patient characteristics [Mol97], [Per00]. See also #33, #34.

11. PG in the earliest weeks of pregnancy, when gestational age is likely to be known exactly only in ART pregnancies: [Ole91], [Ste93], [Mar95], [Cas98].

12. hCG doubling along a log-linear trajectory: [Kad90], [Kad93].

13. More complex interpretations of serial hCG: [Dar99b]. See also [Bar94].

14. Figure 1: [Yek87].

15. The possibility of conceiving the 10 and 20 ng/mL (31.8 and 63.6 nmol/L) values in [Yek87] either as delimiters of a "gray zone" surrounding a central cutoff or else as two distinct "thresholds" (see #9) or decision levels makes this study a progenitor of both the single and the dual cutoff approaches.

16. Memphis series—more than 3600 subjects over the course of 5 years or so: [Mcc96]. Note that some of the statistics quoted in [Mcc96] are difficult to reconcile with previous publications relating to the Memphis series.

17. The role of PG supposedly resolved: [Hol93] p.111.

18. Low PG nearly always associated with nonviable pregnancies: [Lip00] and various papers by Stovall or associates.

19. Curettage as an element of the classic PG first algorithm: [Sto93a].

20. Viable pregnancies with PG below 4 ng/mL (12.7 nmol/L): [Mcc93].

21. "Subluteal" PG in ART pregnancies: [Azu93], [Sul93]; but see [Mar95].

22. Opposition to curettage: [Sta96]; but see [Lip98], [Dar99a], [Mol99a].

23. Multiple diagnostic methods afford protection: [Gra99].

24. High PG nearly always associated with NIUP, absent classic symptoms of EP: [Lip00] and various papers by Stovall or associates, e.g. [Sto94].

25. In the Memphis series, roughly half could be reassured and exempted from ultrasound: [Sto93a] quotes a figure of 40%.

26. San Diego study published just last year: [Buc00].

27. Inclusion in the San Diego study: [Buc00] describes the criteria in two nonequivalent ways.

28. Prevalence 7.5% in the San Diego study, 11% in the Memphis series: [Buc00], [Mcc96].

29. Tutorial review in BJHM: [Dal96].

30. Survey of Irish GPs: [Pre97]. See also [Dal97].

31. EP incidence for spontaneous pregnancies: [Pis99], [Lea00], [Raj00]. Figures for the past decade in the United States are not readily available. In some parts of Scandinavia (Sweden, Finland), the incidence rate may have stabilized or even declined in this period. See, for example, [Tho95], [Kam00].

32. Memphis results taken as support for routine screening in the office setting: [Car97].

33. Dependence of high cutoff on assay used and on demographic characteristics: [Per00]. The tremendous rise of ART in the past decade or so has had a major impact on the diagnosis and management of suspected EP. Notoriously, one can no longer trust in the rarity of heterotopic pregnancy: [Pis99]; moreover, expectations should probably depend on the type of ART involved. Both the time of presentation and the degree to which one can pinpoint gestational age are likely to differ between spontaneous and ART pregnancies. The use of supplements to support the luteal phase and early pregnancy may eventually lead to renewed interest in laboratory measurements less likely to be affected: [Cho92], [Rom94], [Cas98]. The rise of ART has also impacted clinical studies of EP and their meta-analysis, by making it increasingly difficult to ensure appropriate homogeneity and characterization of study populations.

34. Pointing the way. Several recent studies—e.g. [Bro94], [Ank96], [Buc99], [Dar99b], [Mol99a], [Bou00]—have attempted to assess, all too often in relative isolation, either risk factors, physical examination or ultrasound findings, or laboratory results (primarily single or serial hCG determinations) for their quantitative impact on the probability of EP, using the standard apparatus of evidence based medicine (EBM): not only sensitivity and specificity, but also prevalence, likelihood ratios, and other fundamental concepts. [Gal99] and [Mol99b] are particularly instructive. A few studies have looked at limited combinations of factors, findings or test results from this perspective. In particular, [Ole96] examined the joint value of PG and hCG results obtained at the initial presentation, in a study based on IMMULITE® assays for these two analytes. [Ank96] and [Mol99b] provide reminders of the difficulty of combining information from studies of isolated factors within the simple EBM model of clinical judgment: in brief, one cannot assume that their contributions to revising probability estimates are even approximately "additive"—via Bayes' formula and likelihood ratios—if the factors are correlated (mutually dependent). Modern multivariate methods, including both logistic regression and neural networks, get around this limitation by taking into account the possible relatedness of the factors, and weighting them accordingly. The proper development and validation of such models is far from trivial: see [Har96], for example. But Thorburn and her colleagues, who have regularly made excellent, practical use of multivariate techniques in EP studies, have demonstrated that the application of logistic regression results can be handled just as easily as the output of simpler prediction models: [Tho92]. Other examples of logistic regression models relevant to EP diagnosis and management include [Tho88], [Tho86], [Mol99a], [Str99]. Two such models incorporating PG measurements have shown that this analyte does make a significant contribution even when other parameters are taken into account: [Tho92], [Ban99]. Thus PG cannot be dismissed out of hand as "redundant"!

References
Unique identifiers (UI) are provided to facilitate retrieving the MEDLINE record—full citation and abstract—when available.

[Ank96] Ankum WM, Mol BW, et al. Risk factors for ectopic pregnancy: a meta-analysis. Fertil Steril 1996;65:1093-9. Comment: ibid 1997;67:791-2. UI=96222894.

[Azu93] Azuma K, et al. Is the luteo-placental shift a myth? Analysis of low progesterone levels in successful ART pregnancies. J Clin Endocrinol Metab 1993;77:195-8. UI=93315588.

[Ban99] Banerjee S, et al. The expectant management of women with early pregnancy of unknown location. Ultrasound Obstet Gynecol 1999;14:231-6. Comment: 229-30. UI=20052920.

[Bar94] Barnhart K, et al. Prompt diagnosis of ectopic pregnancy in an emergency department setting. Obstet Gynecol 1994;84:1010-5. UI=95060528.

[Bou00] Bouyer J, et al. Risk factors for extrauterine pregnancy in women using an intrauterine device. Fertil Steril 2000;74:899-908. UI=20511960.

[Bre97] Brennan DF. Diagnosis of ectopic pregnancy. J Fla Med Assoc 1997-98;84:549-56. UI=98166374.

[Bro94] Brown DL, Doubilet PM. Transvaginal sonography for diagnosing ectopic pregnancy: positivity criteria and performance characteristics. J Ultrasound Med 1994;13:259-66. UI=95018513.

[Buc00] Buckley RG, King KJ, Disney JD, Riffenburgh RH, Gorman JD, Klausen JH. Serum progesterone testing to predict ectopic pregnancy in symptomatic first-trimester patients. Ann Emerg Med 2000;36:95-100. UI=20378585.

[Buc99] Buckley RG, King KJ, Disney JD, Gorman JD, Klausen JH. History and physical examination to estimate the risk of ectopic pregnancy: validation of a clinical prediction model. Ann Emerg Med 1999;34:589-94. Comment:[Gal99]. UI=20004516.

[Car99] Carson SA. Reply to [Mol97]. Fertil Steril 1997;67:981. UI=97277534.

[Cas98] Castracane VD, Stewart DR, Gimpel T, et al. Maternal serum androgens in human pregnancy: early increases within the cycle of conception. Hum Reprod 1998;13:460-4. UI=98216936.

[Cha92] Chard T. Pregnancy tests: a review. Hum Reprod 1992;7:701-10. UI=92348703.

[Cho92] Choe JK, et al. Serum progesterone and 17-hydroxyprogesterone in the diagnosis of ectopic pregnancies and the value of progesterone replacement in intrauterine pregnancies when serum progesterone levels are low. Gynecol Obstet Invest 1992;34:133-8. UI=93051708.

[Col97] Cole LA. Immunoassay of human chorionic gonadotropin, its free subunits, and metabolites. Clin Chem 1997;43:2233-43. Comment: ibid 1998;44:2001-2. UI=98102548.

[Cow92] Cowan BD, et al. Receiver-operator characteristic, efficiency analysis, and predictive value of serum progesterone concentration as a test for abnormal gestations. Am J Obstet Gynecol 1992;166:1729-34. Discussion: 1734-7. UI=92312637.

[Dal96] Daly S, Prendiville W. First trimester pregnancy: ultrasound and biochemical evaluation. Br J Hosp Med 1996;56:225-7. UI=97034045.

[Dal97] Daly S, Prendiville W. The management of failed first trimester pregnancy. Ir Med J 1997;90:52. Comment: 116. UI=97258978.

[Dan98] D'Antona D, etal. Evaluation of serum inhibin A as a surveillance marker after conservative management of tubal pregnancy. Hum Reprod 1998;13:2305-7. UI=98427747.

[Dar99a] Dart R, Dart L, et al. The utility of a dilatation and evacuation procedure in patients with symptoms suggestive of ectopic pregnancy and indeterminate transvaginal ultrasonography. Acad Emerg Med 1999;6:1024-9. UI=99458244.

[Dar99b] Dart RG, Mitterando J, Dart LM. Rate of change of serial beta-human chorionic gonadotropin values as a predictor of ectopic pregnancy in patients with indeterminate transvaginal ultrasound findings. Ann Emerg Med 1999;34:703-10. UI=20044373.

[Dit00] Ditkoff EC, Sauer MV. Diagnosis: hCG and other serum markers. In: Leach RE, Ory SJ, editors. Management of ectopic pregnancy. Malden, MA: Blackwell Science, 2000:30-50.

[Gal99] Gallagher EJ. Application of likelihood ratios to clinical decision rules: defining the limits of clinical expertise. Ann Emerg Med 1999;34:664-7. Comment on:[Buc99]. UI=20004526.

[Gar00] Garrone CM, Broso PR. Tubaric pregnancy—an excellent idea? Med Hypotheses 2000;54:900-2. UI=20327818.

[Gla91] Glasziou P. Threshold analysis via the Bayes' nomogram. Med Decis Making 1991;11:61-2. UI=91238396.

[Gou97] Gould D. Ectopic pregnancy: causes and outcomes. Nurs Times 1997;93(14):53-5. UI=97274520.

[Gra99] Graczykowski JW, Seifer DB. Diagnosis of acute and persistent ectopic pregnancy. Clin Obstet Gynecol 1999;42:9-22. UI=99172950.

[Gru94] Grudzinskas JG, et al. Relocation of ectopic pregnancy to the uterine cavity: a dream or a reality? Br J Obstet Gynaecol 1994;101:651-3. Comment on: 716-7. UI=95034497.

[Hah90] Hahlin M, Wallin A, Sjoblom P, Lindblom B. Single progesterone assay for early recognition of abnormal pregnancy. Hum Reprod 1990;5:622-6. UI=90368915.

[Hah95] Hahlin M, Thorburn J, Bryman I. The expectant management of early pregnancies of uncertain site. Hum Reprod 1995;10:1223-7. UI=95386650.

[Har96] Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996;15:361-87. Tutorial. UI=96243493.

[Hol93] Holley JE, Stovall TG. Clinical presentations. In: Stovall TG, Ling FW, editors. Extrauterine pregnancy; clinical diagnosis and management. New York: McGraw-Hill, 1993:97-118.

[Hoz99] Hozo I, Djulbegovic B. Using the Internet to calculate clinical action thresholds. Comput Biomed Res 1999;32:168-85. UI=99269792.

[Kad90] Kadar N. Diagnosis and treatment of extrauterine pregnancies. New York: Raven Press, 1990.

[Kad93] Kadar N, et al. A prospective, randomized study of the chorionic gonadotropin-time relationship in early gestation: clinical implications. Fertil Steril 1993;60:409-12. UI=93387493.

[Kam00] Kamwendo F, et al. Epidemiology of ectopic pregnancy during a 28 year period and the role of pelvic inflammatory disease. Sex Transm Infect 2000;76:28-32. UI=20277162.

[Kau97] Kaur M, et al. Markedly elevated serum alpha fetoprotein is an indicator of extrauterine pregnancy. Am J Med Genet 1997;69:112-3. UI=97219604.

[Lan92] Serum progesterone in the diagnosis of ectopic pregnancy. Lancet 1992;340:583. Editorial. UI=92382292.

[Lea00] Leach RE. Ectopic pregnancy: trends and risks. In: Leach RE, Ory SJ, editors. Management of ectopic pregnancy. Malden, MA: Blackwell Science, 2000:1-13.

[Lin00] Lindstedt G, Janson PO. Problems of interpretation in connection with measurements of chorionic gonadotropin. Lakartidningen 2000;97:1554-6. In Swedish. UI=20234024.

[Lin78] Lindstedt G, et al. Serum-chorionic-gonadotropin in normal and ectopic pregnancy. Lancet 1978;i:108-9. UI=78091022.

[Lin82] Lindstedt G, Lundberg PA, Janson PO, Thorburn J. Biochemical diagnosis of ectopic pregnancy. Scand J Clin Lab Invest 1982;42:201-10. UI=83040838.

[Lip00] Lipscomb GH, Stovall TG, Ling FW. Nonsurgical treatment of ectopic pregnancy. N Engl J Med 2000;343:1325-1329. UI=11058678.

[Lip98] Lipscomb GH, et al. Aspiration and microscopy in the diagnosis of ectopic pregnancy. Int J Gynaecol Obstet 1998;63:153-8. UI=99072141.

[Mac93] Maccato ML, et al. Ectopic pregnancy with undetectable serum and urine beta-hCG levels and detection of beta-hCG in the ectopic trophoblast by immunocytochemical evaluation. Obstet Gynecol 1993;81:878-80. UI=93226331.

[Mar95] Marcus SF, et al. The prediction of ectopic pregnancy after in-vitro fertilization and embryo transfer. Hum Reprod 1995;10:2165-8. UI=96077469.

[Mas99] Mashburn J. Ectopic pregnancy: triage do's and dont's. J Nurse Midwifery 1999;44:549-57. UI=20099829.

[Mcc93] McCord ML, Stovall TG, Wild RA. Serum markers for ectopic pregnancy. In: Stovall TG, Ling FW, editors. Extrauterine pregnancy; clinical diagnosis and management. New York: McGraw-Hill, 1993:119-35.

[Mcc96] McCord ML, Muram D, Buster JE, Arheart KL, Stovall TG, Carson SA. Single serum progesterone as a screen for ectopic pregnancy: exchanging specificity and sensitivity to obtain optimal test performance. Fertil Steril 1996;66:513-6. Comment: [Mol97]. UI=96413455.

[Med98] Meden H, Fattahi-Meibodi A. CA 125 in benign gynecological conditions. Int J Biol Markers 1998;13:231-7. UI=99245591.

[Mol97] Mol BW, Hajenius PJ, Ankum WM, et al. Cut-off level - what do you want it to be? Fertil Steril 1997;67:980-1. Comment on: [Mcc96]. UI=97277534.

[Mol98] Mol BW, Lijmer JG, Ankum WM, et al. The accuracy of single serum progesterone measurement in the diagnosis of ectopic pregnancy: a meta-analysis. Hum Reprod 1998;13:3220-7. UI=99068773.

[Mol99a] Mol BW, Hajenius PJ, et al. Can noninvasive diagnostic tools predict tubal rupture or active bleeding in patients with tubal pregnancy? Fertil Steril 1999;71:167-73. UI=99131836.

[Mol99b] Mol BW, van Der Veen F, Bossuyt PM. Implementation of probabilistic decision rules improves the predictive values of algorithms in the diagnostic management of ectopic pregnancy. Hum Reprod 1999;14:2855-62. UI=20018228.

[Ole91] O'Leary P, Boyne P, Flett P, et al. Longitudinal assessment of changes in reproductive hormones during normal pregnancy. Clin Chem 1991;37:667-72. UI=91235385.

[Ole96] O'Leary P, Nichols C, Feddema P, et al. Serum progesterone and human chorionic gonadotrophin measurements in the evaluation of ectopic pregnancy. Aust N Z J Obstet Gynaecol 1996;36:319-23. UI=97038166.

[Pau80] Pauker SG, Kassirer JP. The threshold approach to clinical decision making. N Engl J Med 1980;302:1109-17. UI=80165393.

[Per00] Perkins SL, Al-Ramahi M, Claman P. Comparison of serum progesterone as an indicator of pregnancy nonviability in spontaneously pregnant emergency room and infertility clinic patient populations. Fertil Steril 2000;73:499-504. UI=20156300.

[Pis98] Pisarska MD, Carson SA, Buster JE. Ectopic pregnancy. Lancet 1998;351:1115-20. UI=98322034.

[Pis99] Pisarska MD, Carson SA. Incidence and risk factors for ectopic pregnancy. Clin Obstet Gynecol 1999;42:2-8. UI=99172949.

[Pre00] Predanic M. Differentiating tubal abortion from viable ectopic pregnancy with serum CA-125 and beta-human chorionic gonadotropin determinations. Fertil Steril 2000;73:522-5. UI=20156304.

[Pre97] Prendiville W, et al. The management of first trimester miscarriage by general practitioners in Ireland. Ir J Med Sci 1997;166:3-6. UI=97210344.

[Raj00] Rajkhowa M, et al. Trends in the incidence of ectopic pregnancy in England and Wales from 1966 to 1996. BJOG 2000;107:369-74. UI=20201616.

[Rom94] Romer T, Wodrig W. Value of serum progesterone and 17-hydroxyprogesterone determination in differential diagnosis of abnormal early pregnancy. Zentralbl Gynakol 1994;116:276-9. In German. UI=94295327.

[Rom95] Romer T, et al. Value of serum placental protein 14 determination in differential diagnosis of abnormal early pregnancy. Geburtshilfe Frauenheilkd 1995;55:490-2. In German. UI=96076327.

[Rot00] Rotmensch S, Cole LA. False diagnosis and needless therapy of presumed malignant disease in women with false-positive human chorionic gonadotropin concentrations. Lancet 2000;355:712-5. Erratum: ibid 2000;356:600. Comments: ibid 2000;355:671, 1724-6. UI=20166543.

[Sei96] Seifer DB, Lambert-Messerlian GM, Canick JA, et al. Serum inhibin levels are lower in ectopic than intrauterine spontaneously conceived pregnancies. Fertil Steril 1996;65:667-9. UI=96370411.

[Sta94] Stabile I, Olajide F, Chard T, Grudzinskas JG. Circulating levels of placental protein 14 in ectopic pregnancy. Br J Obstet Gynaecol 1994;101:762-4. UI=95034532.

[Sta96] Stabile I. Ectopic pregnancy: diagnosis and management. Cambridge: Cambridge University Press, 1996.

[Ste93] Stern JJ, et al. Early diagnosis of ectopic pregnancy using receiver-operator characteristic curves of serum progesterone concentrations. Hum Reprod 1993;8:775-9. UI=93301086.

[Ste95] Stewart BK, et al. Biochemical discrimination of pathologic pregnancy from early, normal intrauterine gestation in symptomatic patients. Am J Clin Pathol 1995;103:386-90. UI=95243191.

[Sto92a] Stovall TG, Ling FW, Andersen RN, Buster JE. Improved sensitivity and specificity of a single measurement of serum progesterone over serial quantitative beta-human chorionic gonadotrophin in screening for ectopic pregnancy. Hum Reprod 1992;7:723-5. UI=92348706.

[Sto92b] Stovall TG, Ling FW. Clinical decision making: elevate or reduce to a science? Fertil Steril 1992;57:465. Discussion: 466. Comment on: ibid 1991;55:1033-8. UI=92137401.

[Sto93a] Stovall TG, Ling FW. Ectopic pregnancy. Diagnostic and therapeutic algorithms minimizing surgical intervention. J Reprod Med 1993;38:807-12. UI=94087675.

[Sto93b] Stovall TG, Ling FW. Combined algorithm for the diagnosis of ectopic pregnancy. In: Stovall TG, Ling FW, editors. Extrauterine pregnancy; clinical diagnosis and management. New York: McGraw-Hill, 1993:193-202.

[Sto94] Stovall TG. Medical management of ectopic pregnancy. Curr Opin Obstet Gynecol 1994;6:510-5. UI=95201177.

[Str99] Strandell A, Thorburn J, Hamberger L. Risk factors for ectopic pregnancy in assisted reproduction. Fertil Steril 1999;71:282-6. UI=99140928.

[Stu98] Sturgeon CM, McAllister EJ. Analysis of hCG: clinical applications and assay requirements. Ann Clin Biochem 1998;35:460-91. Comment: ibid 1999;36:259-60. UI=98346028.

[Sug00] Sugantha SE, Webster S, Sundar E, Lenton EA. Predictive value of plasma human chorionic gonadotrophin following assisted conception treatment. Hum Reprod 2000;15:469-73. UI=20122402.

[Sul93] Sultan KM, et al. Viable term pregnancy despite 'subluteal' serum progesterone levels in the first trimester. Fertil Steril 1993;60:363-5. UI=93339474.

[Tay00] Tay JI, et al. Ectopic pregnancy. BMJ 2000;320:916-9. Erratum: ibid 2000;321:424. UI=20206507.

[Tay88] Taylor RN, et al. Pitfall in the diagnosis of ectopic pregnancy: immunocytochemical evaluation in a patient with false-negative serum beta-hCG levels. Obstet Gynecol 1988;71:1035-8. UI=88233526.

[Ten00] Tenore JL. Ectopic pregnancy. Am Fam Physician 2000;61:1080-8. UI=20168677.

[Tho86] Thorburn J, Philipson M, Lindblom B. Background factors of ectopic pregnancy. II. Risk estimation by means of a logistic model. Eur J Obstet Gynecol Reprod Biol 1986;23:333-40. UI=87106275.

[Tho92] Thorburn J, Bryman I, Hahlin M. Distinction between early normal intrauterine pregnancies and pathological pregnancies by means of a logistic model. Hum Reprod 1992;7:120-2. UI=92202331.

[Tho95] Thorburn J. Ectopic pregnancy. The 'epidemic' seems to be over. Lakartidningen 1995;92:4701-6. In Swedish. UI=96116709.

[Tre00] Treetampinich C, et al. Maternal serum inhibin A concentrations in early pregnancy after IVF and embryo transfer reflect the corpus luteum contribution and pregnancy outcome. Hum Reprod 2000;15:2028-32. UI=20425044.

[Yek87] Yeko TR, Gorrill MJ, Hughes LH, Rodi IA, Buster JE, Sauer MV. Timely diagnosis of early ectopic pregnancy using a single blood progesterone measurement. Fertil Steril 1987;48:1048-50. UI=88055621.

Home - Search - Site Map - Contact Us
About DPC - Medical Conditions - Technology - Immunoassay Products - Financial - Employment