Early Pregnancy Bleeding: A Risk-Based Playbook for Pregnancy of Unknown Location and Ectopic Pregnancy

Introduction

Why the Playbook Has Changed

The older mental model asked, “Is the hCG above the discriminatory zone?” The more useful model asks four questions in sequence: Is the patient clinically stable, what does a high-quality transvaginal ultrasound show, how is hCG changing over time, and what are the patient’s pregnancy goals and follow-up constraints? [2-4]

PUL describes what ultrasound has not yet located. It does not mean ectopic pregnancy, failed pregnancy, or a pregnancy that can safely be ignored. [1-3] The endpoint must be a confirmed intrauterine location, a confirmed ectopic location, or documented resolution to a nonpregnant state. [2,3]

Clinical rule: A new symptom, worsening pain, syncope, shoulder pain, peritoneal findings, hemodynamic change, or increasing hemoperitoneum outranks a previously reassuring hCG curve. [2-4]

1. Stabilize Before You Stratify

The first branch is physiologic, not biochemical. A patient with hypotension, tachycardia out of proportion to circumstances, syncope, peritoneal signs, severe or escalating unilateral pain, or substantial intraperitoneal fluid needs immediate resuscitation, urgent gynecologic consultation, and definitive evaluation for ruptured ectopic pregnancy. [2-4]

A focused initial history should document bleeding severity, pain pattern, syncope or presyncope, prior ectopic pregnancy, tubal surgery, pelvic infection, infertility treatment, intrauterine device status, estimated gestational age, and whether the pregnancy is desired, undesired, or uncertain. [2,3] The absence of a known ectopic risk factor is not reassuring enough to stop evaluation because many ectopic pregnancies occur without an identified risk factor. [3]

For a stable symptomatic patient, the usual initial package is quantitative serum hCG, transvaginal pelvic ultrasound, and clinically indicated blood count and blood-group testing under the applicable local protocol. [2-4] A pelvic examination may clarify cervical bleeding, tissue passage, tenderness, or peritoneal irritation, but it does not replace imaging and follow-up. [2,3]

Initial branch	Findings	Immediate action
Unstable or suspected rupture	Hemodynamic instability, syncope, peritoneal signs, severe escalating pain, or concerning free fluid	Resuscitate, obtain urgent gynecologic/surgical involvement, and do not wait for serial hCG. [2-4]
Stable with definite ectopic pregnancy	Extrauterine gestational sac with yolk sac or embryo, or another definitive ectopic finding	Select surgical, medical, or carefully defined expectant management according to symptoms, anatomy, hCG, contraindications, preferences, and follow-up capacity. [1-3]
Stable with definite intrauterine pregnancy	Definite intrauterine gestational sac with yolk sac or embryo	Evaluate bleeding within an intrauterine-pregnancy pathway while remembering that heterotopic pregnancy remains possible after assisted reproduction or when adnexal findings are concerning. [1,2]
Stable PUL	No definite or probable intrauterine or ectopic pregnancy on transvaginal ultrasound	Begin a documented PUL pathway using serial symptoms, hCG, repeat ultrasound, pregnancy goals, and reliable follow-up. [1-4]

2. Use Transvaginal Ultrasound Early, Not After an Arbitrary hCG Gate

Emergency medicine guidance recommends pelvic ultrasound for symptomatic pregnant patients regardless of the hCG concentration. [4] A low hCG value can explain why a very early intrauterine pregnancy is not yet visible. Still, it does not make ultrasound uninformative because imaging may identify an adnexal mass, hemoperitoneum, an alternative pelvic diagnosis, or a definite pregnancy location. [2-4]

The ultrasound report should use standardized first-trimester terminology and should distinguish definite, probable, and absent evidence of intrauterine or ectopic pregnancy. [1] The 2024 Society of Radiologists in Ultrasound lexicon also classifies implantation in an abnormal intrauterine location, including a cesarean scar, as ectopic pregnancy and recommends the term “cardiac activity” rather than “heartbeat” in early imaging reports. [1]

A systematic study should assess the endometrial cavity, each adnexa, the cul-de-sac, and other dependent spaces for free fluid. [1-4] An intrauterine pregnancy generally makes a concurrent ectopic pregnancy unlikely, but it does not eliminate heterotopic pregnancy, especially after ovulation induction or assisted reproductive technology. [2]

3. Use hCG as a Trajectory, Not an Address

A single serum hCG value cannot locate a pregnancy and should not be used by itself to diagnose ectopic pregnancy or nonviability. [2-4,8] Serial measurement is most useful when interpreted with symptoms and ultrasound, ideally using the same laboratory platform when feasible. [2,3]

The 49/40/33% minimum-rise data

In symptomatic patients whose pregnancies ultimately proved to be viable intrauterine pregnancies, the estimated first-percentile 48-hour rise became slower as the initial hCG increased. [5]

Initial hCG	Approximate minimum 48-hour rise observed in viable intrauterine pregnancy
Less than 1,500 mIU/mL	49% [5]
1,500-3,000 mIU/mL	40% [5]
Greater than 3,000 mIU/mL	33% [5]

These values are lower-bound observations, not proof of location and not automatic intervention thresholds. [5,8] Some ectopic pregnancies rise in an apparently reassuring pattern, and a rise below these values increases concern but does not by itself identify where the pregnancy is located. [2,5,8]

The discriminatory level is a probability concept

The discriminatory concept describes the hCG concentration at which an intrauterine gestational sac is expected to be visible with a specified probability under particular imaging and assay conditions. [6,7] In a 2013 study, the modeled hCG concentration at which a gestational sac was visualized 99% of the time was approximately 3,510 mIU/mL. [6] A 2023 study estimated 90% visualization at approximately 2,421 mIU/mL and 99% visualization at approximately 3,994 mIU/mL. [7]

Those estimates do not justify immediate methotrexate use or uterine evacuation when no sac is seen above the selected value. [2,6,7] Equipment, operator skill, multiple gestation, dating uncertainty, uterine anatomy, and assay differences all influence visualization, so the safer use of a high discriminatory value is to trigger expert review and repeat assessment rather than to terminate a potentially desired intrauterine pregnancy. [2,6,7]

NICE hCG-change thresholds answer a different question

NICE uses two hCG measurements as close as possible to 48 hours apart as triage points within its PUL pathway, while explicitly stating that hCG should not be used to determine pregnancy location and that symptoms take priority. [3]

Approximate 48-hour hCG change in the NICE pathway	Suggested pathway meaning and next step
Rise greater than 63%	A developing intrauterine pregnancy is likely, but ectopic pregnancy is not excluded; repeat transvaginal ultrasound in 7-14 days, with earlier scanning considered when hCG is at least 1,500 IU/L. [3]
Fall greater than 50%	The pregnancy is unlikely to continue, but resolution is not yet proven; NICE advises a urine pregnancy test 14 days after the second serum test, with prompt review if it remains positive. [3]
Rise less than 63% or fall less than 50%	Arrange clinical review within 24 hours because ectopic or persisting PUL remains possible. [3]

The 63% NICE rise threshold and the 49/40/33% minimum-rise observations should not be blended into a single universal rule because they were developed for different purposes. [3,5] One is a pathway triage threshold, and the other describes the lower tail of hCG rise among viable intrauterine pregnancies. [3,5]

4. Risk Models Can Organize Follow-Up, but They Do Not Replace Judgment

Validated models such as M6 combine hCG values, and in some versions progesterone, to estimate ectopic-pregnancy risk and assign follow-up intensity. [9,10] In a multicenter external validation cohort, M6 with progesterone achieved an area under the receiver operating characteristic curve of 0.89 and 96% sensitivity for ectopic pregnancy at a 5% high-risk threshold, although performance and calibration varied by center and 10% of eligible participants were lost to follow-up. [9]

An updated M6 model has also evaluated additional clinical factors, but model output remains a triage estimate rather than a diagnosis. [10] A health system should use a model only when the exact version, laboratory timing, risk threshold, communication process, and follow-up actions are locally validated and operationalized. [9,10]

NICE advises against using serum progesterone as an adjunct to diagnose viable intrauterine or ectopic pregnancy in its serial-hCG PUL pathway. In contrast, the M6P model uses progesterone as a prediction input. [3,9] These are different protocols, and clinicians should not assemble an unvalidated hybrid from selected pieces of each. [3,9]

5. Pregnancy Goals Change the Acceptable Balance of Delay and Intervention

Pregnancy intention does not alter the need to detect ectopic pregnancy, but it changes the consequences of intervening before location and viability are known. [2,11] The plan should document whether the pregnancy is desired, undesired, or uncertain and should be revisited if the patient’s preference changes. [2,11]

Desired or uncertain pregnancy

For a stable patient who wishes to preserve a potentially viable pregnancy, the default is conservative diagnostic follow-up rather than empiric treatment. [2,3] Obtain a baseline quantitative hCG, repeat it near 48 hours, repeat transvaginal ultrasound according to the hCG trajectory and clinical course, and provide direct return instructions that do not depend on the laboratory result. [2-4]

Do not administer methotrexate or perform uterine evacuation solely because hCG is above a chosen discriminatory value and the uterus is empty. [2,6,7] When intervention becomes necessary, the record should explain how a viable intrauterine pregnancy was excluded or why immediate treatment was required for safety. [2]

Undesired PUL

Among abortion-seeking patients with a documented PUL, the Society of Family Planning estimates ectopic-pregnancy incidence at 4-8%, which is higher than in the general abortion-seeking population. [11] For asymptomatic patients judged to be at low ectopic risk who prefer immediate care, the Society recommends offering mifepristone with misoprostol or uterine aspiration without waiting for a visible gestational sac, provided there is a clear and timely plan to confirm pregnancy resolution. [11]

Medication abortion does not treat ectopic pregnancy. [11-13] A baseline quantitative hCG is therefore required for a documented PUL treated with medication, and no-test abortion follow-up should not be imported into this higher-risk pathway. [11]

A retrospective cohort found that same-day medication initiation shortened time to abortion but required robust serial hCG systems to identify ectopic pregnancy and ongoing pregnancy. [12] In a 2024 randomized trial of 1,504 participants before confirmed intrauterine pregnancy, very-early medication abortion was noninferior to delayed treatment for complete abortion, with complete abortion in 95.2% and 95.3%, respectively. [13] Ectopic pregnancy occurred in 1.3% versus 0.8%, and one tubal rupture occurred before diagnosis in the immediate-treatment group, reinforcing that early access and ectopic surveillance must be delivered together. [13]

Immediate-treatment pathway for a low-risk undesired PUL	Required resolution check
Mifepristone and misoprostol	Obtain baseline quantitative hCG; resolution can be defined as at least a 50% decline 48-72 hours after misoprostol or at least an 80% decline 7 days after mifepristone, or 5-10 days after misoprostol. [11]
Uterine aspiration with chorionic villi or gestational sac identified	The finding supports an intrauterine pregnancy; continue care according to pathology, symptoms, and the clinical plan. [11]
Uterine aspiration with neither chorionic villi nor gestational sac identified	Do not label the case ectopic from this finding alone; repeat ultrasound, quantitative hCG, or both, with resolution defined as greater than a 50% decline at 24 hours, greater than 70% at 48 hours, or greater than 80% at approximately 72 hours. [11,14]

Failure to meet a decline threshold is not itself proof of ectopic pregnancy. Still, it requires prompt reassessment for ongoing intrauterine pregnancy, retained intrauterine tissue, ectopic pregnancy, or an incorrect baseline. [11,14]

Persisting PUL

Persisting PUL describes a stable pregnancy that remains unlocated while hCG fails to resolve or follows a plateauing or abnormal pattern. [2,15] This is the subgroup in which diagnostic uterine aspiration, empiric methotrexate, continued expectant management, or surgery may be considered after an individualized review of pregnancy goals, symptoms, hCG pattern, ultrasound findings, contraindications, and follow-up reliability. [2,15]

In the ACT or NOT randomized trial, active management was more likely than expectant management to achieve resolution without changing the initial strategy. Still, crossover was substantial and a live birth occurred despite conservative nonviability criteria. [15] The results support structured shared decision-making rather than routine empiric treatment for every persisting PUL. [15]

6. When Ectopic Pregnancy Is Confirmed or Highly Suspected

Hemodynamic instability, suspected rupture, significant pain, an ectopic pregnancy with embryonic cardiac activity, a large or anatomically concerning mass, contraindications to methotrexate, or inability to complete follow-up generally favors surgical management. [2,3] Stable selected patients may be candidates for expectant management, systemic methotrexate, or surgery, depending on the complete clinical picture and the protocol being used. [2,3]

Expectant management requires minimal symptoms, a low and declining hCG pattern, no evidence of rupture, and reliable access to serial assessment. [2,3] Methotrexate requires stability, an unruptured ectopic pregnancy or sufficiently high diagnostic certainty, absence of a viable intrauterine pregnancy, evaluation for contraindications and interactions, baseline laboratory assessment, and reliable post-treatment hCG monitoring. [2]

The current U.S. methotrexate injection labeling does not list ectopic pregnancy among its indications and includes boxed warnings for embryo-fetal toxicity and other serious adverse reactions. [16] Its use for ectopic pregnancy is therefore off-label in the United States and should follow a dedicated institutional protocol rather than a generic chemotherapy label or an improvised regimen. [2,16]

A patient who cannot or is unlikely to return for serial hCG testing is not a routine outpatient methotrexate candidate because follow-up is part of the treatment, not an optional add-on. [2,3] Worsening symptoms after methotrexate or during expectant management require urgent reassessment regardless of the previous hCG trend. [2,3]

7. RhD Testing and Rh Immune Globulin: Current Guidance Is Not Uniform

First-trimester RhD practice changed substantially between 2024 and 2026, but organizations do not give identical recommendations. [3,17-19] ACOG’s 2024 Clinical Practice Update suggests forgoing routine Rh testing and Rh immune globulin for abortion or pregnancy loss before 12 weeks, and ASRM’s 2026 position explicitly supports no routine testing or prophylaxis before 12 weeks for bleeding, pregnancy loss, or abortion while allowing individualized shared decisions. [17,18]

SMFM’s 2024 statement recommends offering RhD testing and Rh immune globulin for spontaneous or induced abortion before 12 weeks when doing so is feasible and does not hinder access. [19] NICE’s June 2026 update advises against routine anti-D prophylaxis through 11 weeks 6 days for ectopic pregnancy, miscarriage, or threatened miscarriage, then gives specific recommendations at 12 weeks 0 days through 12 weeks 6 days. [3]

The practical response is to verify the current local protocol, gestational age, exact clinical scenario, and governing professional guidance rather than relying on a remembered blanket rule. [3,17-19] The chart should document the protocol used and any shared decision-making when recommendations diverge. [17-19]

8. Build a Closed-Loop PUL System

A technically correct evaluation can still fail if the patient leaves without a named owner, a scheduled next step, and a mechanism for reviewing results. [2-4,11] Every PUL discharge or outpatient handoff should include the following elements. [2-4,11]

1. Risk status: Record symptoms, hemodynamic status, ultrasound classification, baseline hCG, and the current differential. [1-4]
2. Pregnancy goals: Record desired, undesired, or uncertain status and the management implications discussed. [2,11]
3. Next test and exact timing: State when hCG will be repeated, when ultrasound will be repeated, and who will order and review each result. [2-4]
4. Result ownership: Name the clinician, service, or tracking pool responsible for contacting the patient and escalating abnormal results. [2-4]
5. Return precautions: Provide written 24-hour instructions for increasing pain, shoulder pain, syncope, weakness, heavy bleeding, dyspnea, or any new concern. [2-4]
6. Closure criterion: Continue until a definite pregnancy location is established or hCG resolves according to the applicable pathway. [2,3,11]
7. RhD plan: Document whether testing or prophylaxis was indicated under the current protocol and why. [3,17-19]

A missed second hCG is not a neutral event. [2-4,11] It should trigger the same kind of active outreach used for other time-sensitive diagnostic safety nets because ectopic pregnancy remains possible until the pathway is closed. [2-4]

Common Failure Modes and Their Replacements

Failure mode	Why it fails	Safer replacement
Waiting for hCG to cross a threshold before ultrasound	Important ectopic or bleeding findings may be visible at low hCG, and ACEP recommends ultrasound at any hCG in symptomatic patients. [4]	Perform transvaginal ultrasound when clinically indicated and interpret it with symptoms and hCG. [2-4]
Treating the discriminatory level as proof of ectopic pregnancy	The level describes probability of visualization, not location, and rare viable intrauterine pregnancies can be missed. [2,6,7]	Use a conservative level only to trigger expert review, repeat imaging, and closer follow-up. [2,6,7]
Calling a reassuring rise “not ectopic”	Ectopic pregnancies can have apparently normal rises. [2,5,8]	Use hCG to stratify risk, not to certify location. [2,5,8]
Treating every PUL with methotrexate	PUL includes early viable intrauterine pregnancies and resolving pregnancies. [1,2]	Match intervention to stability, pregnancy goals, diagnostic certainty, contraindications, and follow-up capacity. [2,11,15]
Assuming absent villi after aspiration proves ectopic pregnancy	Very early intrauterine tissue can be missed, and pathology or gross examination may be nondiagnostic. [11,14]	Use defined post-aspiration hCG decline thresholds and repeat imaging when indicated. [11,14]
Using no-test abortion follow-up for a known PUL	A documented PUL carries higher ectopic risk. [11]	Obtain baseline hCG and use PUL-specific decline criteria. [11]
Discharging without result ownership	A sound algorithm cannot work when follow-up is lost. [2-4,11]	Assign a named owner, a specific interval, a direct contact process, and a closure criterion. [2-4,11]

Special Situations That Lower the Threshold for Expert Review

Assisted reproductive technology, prior ectopic pregnancy, known tubal disease, an intrauterine device in situ, unusual implantation sites, significant free fluid, or discordant symptoms and laboratory findings should lower the threshold for specialist review. [1,2] A visualized intrauterine pregnancy does not fully exclude heterotopic pregnancy after assisted reproduction, and the adnexa still require evaluation when symptoms or imaging are concerning. [2]

Uncertain dates are common and make calendar-based expectations unreliable. [2,3] When hCG and ultrasound appear discordant, repeat assessment under controlled conditions is usually safer than forcing the case into a diagnosis that the evidence does not yet support. [2,3,6,7]

Evidence Limitations

PUL pathways differ across countries, professional organizations, early pregnancy units, emergency departments, and abortion-care settings. [2-4,11] hCG studies use different assays, populations, endpoints, and follow-up structures, while risk-model performance depends on local prevalence and workflow fidelity. [5-10] Very-early abortion evidence supports immediate care for selected low-risk patients, not unrestricted treatment without ectopic surveillance. [11-14] RhD recommendations remain an area of active guideline divergence. [3,17-19]

 

References

1. Rodgers SK, Horrow MM, Doubilet PM, et al. A Lexicon for First-Trimester US: Society of Radiologists in Ultrasound Consensus Conference Recommendations. Radiology. 2024;312(2):e240122. DOI. PubMed.
2. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 193: Tubal Ectopic Pregnancy. Obstet Gynecol. 2018;131(3):e91-e103. DOI. PubMed. Official guidance (accessed August 19, 2026).
3. National Institute for Health and Care Excellence. Ectopic pregnancy and miscarriage: diagnosis and initial management. NICE guideline NG126. Updated June 17, 2026. Official guideline (accessed August 19, 2026).
4. Hahn SA, Promes SB, Brown MD. Clinical Policy: Critical Issues in the Initial Evaluation and Management of Patients Presenting to the Emergency Department in Early Pregnancy. Ann Emerg Med. 2017;69(2):241-250.e20. DOI. PubMed. ACEP resource page (accessed August 19, 2026).
5. Barnhart KT, Guo W, Cary MS, et al. Differences in Serum Human Chorionic Gonadotropin Rise in Early Pregnancy by Race and Value at Presentation. Obstet Gynecol. 2016;128(3):504-511. DOI. PubMed.
6. Connolly AM, Ryan DH, Stuebe AM, Wolfe HM. Reevaluation of Discriminatory and Threshold Levels for Serum beta-hCG in Early Pregnancy. Obstet Gynecol. 2013;121(1):65-70. DOI. PubMed.
7. Park KE, Latack KR, Vestal NL, Awadalla MS. Association of HCG Level with Ultrasound Visualization of the Gestational Sac in Early Viable Pregnancies. Reprod Sci. 2023;30(12):3623-3628. DOI. PubMed.
8. van Mello NM, Mol F, Opmeer BC, et al. Diagnostic Value of Serum hCG on the Outcome of Pregnancy of Unknown Location: A Systematic Review and Meta-analysis. Hum Reprod Update. 2012;18(6):603-617. DOI. PubMed.
9. Christodoulou E, Bobdiwala S, Kyriacou C, et al. External Validation of Models to Predict the Outcome of Pregnancies of Unknown Location: A Multicentre Cohort Study. BJOG. 2021;128(3):552-562. DOI. PubMed.
10. Kyriacou C, Christodoulou E, Bobdiwala S, et al. Updating M6 Pregnancy of Unknown Location Risk-Prediction Model Including Evaluation of Clinical Factors. Ultrasound Obstet Gynecol. 2024;63(3):408-418. DOI. PubMed.
11. Nippita S, Cansino C, Goldberg AB, et al. Society of Family Planning Clinical Recommendation: Management of Undesired Pregnancy of Unknown Location and Abortion at Less Than 42 Days of Gestation. Contraception. 2025;150:110865. DOI. PubMed. Official guidance (accessed August 19, 2026).
12. Goldberg AB, Fulcher IR, Fortin J, et al. Mifepristone and Misoprostol for Undesired Pregnancy of Unknown Location. Obstet Gynecol. 2022;139(5):771-780. DOI. PubMed.
13. Brandell K, et al. Randomized Trial of Very Early Medication Abortion. N Engl J Med. 2024;391(18):1685-1695. DOI. PubMed.
14. Baldwin MK, Bednarek PH, Russo J. Serum Human Chorionic Gonadotropin Decline Following Aspiration Abortion in Pregnancies of Unknown Location. Contraception. 2021;103(2):113-115. DOI. PubMed.
15. Barnhart KT, Hansen KR, Stephenson MD, et al. Effect of an Active vs Expectant Management Strategy on Successful Resolution of Pregnancy Among Patients With a Persisting Pregnancy of Unknown Location: The ACT or NOT Randomized Clinical Trial. JAMA. 2021;326(5):390-400. DOI. PubMed.
16. U.S. National Library of Medicine. DailyMed: Methotrexate injection, solution. Current prescribing information. DailyMed (accessed August 19, 2026).
17. American College of Obstetricians and Gynecologists. Rh D Immune Globulin Administration After Abortion or Pregnancy Loss at Less Than 12 Weeks of Gestation: ACOG Clinical Practice Update. Obstet Gynecol. 2024;144:e140-e143. DOI. PubMed.
18. American Society for Reproductive Medicine. Position Statement on Rho(d) Immune Globulin Administration in the First Trimester. Published July 1, 2026. Official statement (accessed August 19, 2026).
19. Society for Maternal-Fetal Medicine, Prabhu M, Louis JM, Kuller JA, SMFM Publications Committee. Society for Maternal-Fetal Medicine Statement: RhD Immune Globulin After Spontaneous or Induced Abortion at Less Than 12 Weeks of Gestation. Am J Obstet Gynecol. 2024;230:B2-B5. Official statement (accessed August 19, 2026).