Early Pregnancy Loss and Recurrent Pregnancy Loss

Pregnancy loss is defined as a non-viable intrauterine pregnancy occurring up to the 20th week of gestation; The most common form is early pregnancy loss, which occurs in the first trimester (up to 12+6 weeks). There is no need to rush the clinically stable patient: once the diagnosis is confirmed, the patient can consider three management options—waiting, drug therapy, and surgical evacuation—within days or even weeks. This review addresses the diagnosis and management of early pregnancy loss, followed by the definition, etiology, and state of evidence of recurrent pregnancy loss.

Clinical Pictures and Initial Evaluation

Bleeding is common in the first trimester, and most pregnant women who experience bleeding do not lose their pregnancy. In an observational study following 550 pregnant women, 117 (21%) had bleeding before the 20th week, resulting in loss in 12% of the entire group; Of the 117 people with bleeding, 59 (50%) experienced loss (UpToDate, pregnancy loss, 2026). Timing matters: in a prospective study of 701 patients, isolated vaginal bleeding at 6–8 weeks was associated with the greatest increase in risk (risk difference 56%; 95% CI 38–75)—but since early pregnancy loss already occurs frequently during these weeks, bleeding is most likely a symptom rather than the cause itself.

Patients are divided into three clinical categories:

• Uncomplicated bleeding and cramping: Hemodynamically stable, low risk of instability according to bleeding volume, no signs of infection. It may be accompanied by a decrease in pregnancy symptoms (remission of breast tenderness and nausea).
• Complicated (hemorrhage and/or infection): The risk of both increases after the 13th week. The risk of severe hemorrhage requiring hospitalization is generally around 1% (UpToDate, pregnancy loss, 2026). While the frequency of intrauterine infection is approximately 15% in the first 12 weeks, it can reach up to 66% in losses between 12 and 24 weeks; especially in those with a long delay between symptom onset and diagnosis, or between diagnosis and expulsion. A patient with a clinically symptomatic infection is not a candidate for expectant monitoring or drug therapy (UpToDate, pregnancy loss, 2026).
• Asymptomatic / incidental: Sensitive pregnancy tests and ultrasonography enable the diagnosis to be made before symptoms begin.

Distinction based on the state of uterine emptying also determines management: complete (empty uterus in a patient with a previously documented intrauterine pregnancy) and incomplete (residual gestational sac or fetal/placental tissue in the uterus). The absence of symptoms does not exclude residual tissue. The term "inevitable abortion" is no longer preferred: visual assessment of cervical dilatation is not reliable in practice — the external os may appear open even in normal pregnancy due to parity, while a cervix that appears closed and without bleeding does not exclude pregnancy loss.

Ultrasonographic Diagnostic Criteria

Once an intrauterine pregnancy is visualized, the absence of an intrauterine pregnancy on any subsequent ultrasonography or the disappearance of previously detected cardiac activity makes the diagnosis of pregnancy loss. Size criteria are used for diagnosis with a single ultrasonography. The Society of Radiologists in Ultrasound multidisciplinary consensus panel has expanded the old criteria (CRL 5 mm and mean sac diameter 16 mm) to CRL 7 mm and mean sac diameter 25 mm—the aim being to practically eliminate the possibility of a false-positive diagnosis (Doubilet et al., N Engl J Med 2013; impact of criteria: Hu 2014, J Ultrasound Med).

Pregnancy week is determined by ultrasonography, not by the last menstrual date; This determination directly affects management options. The first trimester extends to 13+6 weeks (“early” in the early pregnancy loss literature is often defined as 12+6 weeks); second trimester loss occurs between 14+0 and 20+0 weeks (UpToDate, pregnancy loss, 2026).

β-hCG kinetics: single measurement is not diagnostic

There is no diagnostic hCG level or trend for pregnancy loss; Only relative changes are helpful. Since hCG levels vary widely at any given week of pregnancy, a single hCG value is not sufficient for diagnosis, and a single value below 4000 mIU/mL should be interpreted with great caution; Typically three series of measurements are required to determine the trend. In a normal pregnancy, average hCG levels plateau at approximately 8–11 weeks and then decline. A significant hCG drop of more than 25% in 48 hours in the presence of uterine bleeding is a strong finding in favor of pregnancy loss and may provide guidance if rapid access to ultrasonography is not available (UpToDate, pregnancy loss, 2026).

The hCG levels at which structures can be seen with a 99% probability were reported as follows in the analysis of 651 pregnancies: 3510 mIU/mL for the gestational sac, 17,716 mIU/mL for the yolk sac, 47,685 mIU/mL for the embryonic pole (UpToDate, pregnancy loss, 2026). Serum progesterone is not used in diagnosis: in a study of 360 people with bleeding in early pregnancy, a threshold of <35 nmol/L gave a 68% positive and 91% negative predictive value for pregnancy loss—with high variability in both normal and abnormal pregnancies (UpToDate, pregnancy loss, 2026).

Management: Three Options

In a clinically stable patient, all three options: expectant management, drug therapy, and surgical evacuation can be used until the 20th week. The main factor determining the choice is the patient's preference and values; Additional medical conditions sometimes highlight an option — for example, surgery may be preferred in a patient using therapeutic anticoagulants, and drug therapy may be preferred in a patient with uterine cavity distortion (myoma, anomaly).

The probability of completion in waiting monitoring varies significantly depending on the type of loss. In a study of 1096 patients, the seventh and fourteenth day completion rates were as follows: 53% and 84% for incomplete miscarriage; 30% and 59% in prolonged loss without expulsion; 25% and 52% in non-viable pregnancy with empty gestational sac. In a study comparing three options, residual pregnancy product was detected in approximately one-third of the patients on ultrasonography performed 10–14 days after tissue passage; While infection rates were similar between groups (waiting 3%, medication 2%, surgery 3%), unplanned hospitalization and unplanned procedures were more common in the waiting group (UpToDate, pregnancy loss, 2026). In the second trimester, ACOG does not recommend expectant monitoring because there is no safety data; Delay rapidly increases life-threatening risks such as disseminated intravascular coagulation.

Medication regimen: mifepristone + misoprostol

Combination therapy is the preferred method if mifepristone is available; this recommendation is consistent with ACOG, World Health Organization, NICE and FIGO (WHO 2022; NICE NG126, 2021). In an observational series of approximately 32,000 patients, combination therapy reduced the need for subsequent uterine aspiration (10.5% vs 14%) and emergency department visits (3.5% vs 7.9%) compared with misoprostol alone (Benson 2024, JAMA Netw Open). Lack of access to mifepristone is not a reason to refuse drug therapy to the patient.

Pain management is offered in all forms of management; A nonsteroidal anti-inflammatory drug is recommended before misoprostol, and routine opioids are usually not required. Most of the side effects are related to misoprostol: in a study comparing it with vacuum aspiration, nausea was reported in 53% of misoprostol recipients and vomiting was reported in 20%; Antiemetics (e.g. ondansetron) are recommended for the patient with nausea (Schreiber 2018, NEJM).

The superiority of hysteroscopic resection over vacuum aspiration for residual product of pregnancy has not been demonstrated: in the study comparing 282 and 281 patients, two-year live birth rates were similar (62.8% vs 67.6%; absolute risk difference −4.8; 95% CI −13.0 to 3.0) and time to conception was unchanged. Routine hysteroscopic resection is not recommended (UpToDate, pregnancy loss, 2026).

Antibiotic prophylaxis

The distinction is clear: routine antibiotics are not recommended for drug therapy — at any week of gestation. On the other hand, in surgical vacuum aspiration, prophylaxis is given 60 minutes before the procedure (UpToDate Grade 1B); Typical administration is a single dose of doxycycline 200 mg (orally or intravenously if intravenous access is available) one hour before the procedure. In a study of more than 3,400 patients with pregnancy loss up to week 22 in low-income countries, overall pelvic infection rates were similar in the antibiotic and placebo groups (4.1% vs. 5.3%), but in analysis by procedure type, the risk of infection was lower in those receiving antibiotics before manual vacuum aspiration (1.3% vs. 4.1%) (UpToDate, pregnancy loss, 2026).

Anti-D immunoglobulin: institutions are falling apart

Rh testing is performed in pregnancy loss at 12 weeks and above, and RhD immunoglobulin is recommended for RhD negative patients to prevent alloimmunization. The dose varies depending on the week of pregnancy: 100 mcg (500 IU) between 12 and 18 weeks — if the 100 mcg form is not available, two 50 mcg (250 IU) doses can be given; 300 mcg (1500 IU) after 18 weeks. Guideline differences continue regarding the exact location of the threshold: at least one national guideline recommends the application in pregnancies greater than 70 days, and another institution recommends the application in pregnancies greater than 77 days (UpToDate, pregnancy loss, 2026).

Contraception and subsequent pregnancy

Ovulation may return within days to weeks after early pregnancy loss. All methods, except the intrauterine device, can be initiated as soon as pregnancy loss is diagnosed—before or during drug or surgical management. The intrauterine device can be inserted after tissue expulsion, at the time of uterine evacuation, or by intermediate insertion, typically within four weeks; At six-month follow-up, expulsion rates were similar for post-aspiration and intermediate placement (5.0% vs 2.7%). The prognosis is generally good: in a study of more than 53,000 people who gave birth, 43% had previously experienced at least one first-trimester pregnancy loss, and the likelihood of a subsequent live birth did not differ between medication and expectant monitoring.

Recurrent Pregnancy Loss

Description: differences between institutions have clinical implications

The definition of recurrent pregnancy loss (RPL) varies between institutions; This difference is not an academic detail as it directly determines which couple will be investigated.

In practice, emphasis has shifted toward the two-loss threshold. The rationale is prognostic data: the risk of miscarriage in a subsequent pregnancy is approximately 15% after a single miscarriage, rising to 17–31% after two consecutive miscarriages and 25–46% after three or more miscarriages. In a study covering 156,584 pregnancies, loss rates were 11.6% among those with no prior pregnancy, 19.8% after one loss, 27.7% after two losses, and 41.9% after three or more losses (UpToDate, RPL, 2026). Recurrent loss tends to recur at a similar gestational age, and the risk of recurrence increases with higher gestational age at loss.

RPL is classified into primary (no prior pregnancy reaching viability) and secondary (at least one prior live birth); the prognosis is better in secondary RPL. Prevalence: while approximately 15% of pregnant women experience sporadic clinical pregnancy loss, only 2% experience two consecutive losses and 0.4–1% experience three consecutive losses (UpToDate, RPL, 2026).

Etiology: no cause can be found in half of the cases

Among uterine causes, the anomaly associated with the worst reproductive outcome is the septate uterus: fetal survival in an untreated septate uterus is 6–28%, and the miscarriage rate exceeds 60%. In a systematic review, septoplasty was associated with a reduced miscarriage rate (OR 0.45; 95% CI 0.22–0.90; 12 studies, 796 patients) (UpToDate, RPL, 2026); outcomes did not differ between partial and complete septum. The primary cause of intrauterine adhesions is curettage for pregnancy complications — traumatizing the basal layer, especially within the first four weeks after delivery, triggers adhesion formation. Cervical insufficiency is a cause of recurrent second-trimester loss rather than early pregnancy loss. Maternal age is an independent and strong determinant: clinically recognized miscarriage rates are around 9–17% at ages 20–30, 20% at age 35, 40% at age 40, and 80% at age 45 (UpToDate, RPL, 2026).

Recommended diagnostic workup panel

Minimum evaluation consists of complete medical, surgical, genetic and family history and physical examination. The week and nature of each pregnancy (anembryonic pregnancy, live embryo seen) are recorded in the history — the week of loss narrows down the etiology. Recommended starter panel for lab and imaging:

• Uterine evaluation—sonohysterography preferred; It is more accurate than hysterosalpingography and provides more information than sonography alone. Three-dimensional ultrasonography is reliable in the diagnosis of anomaly because it shows both the cavity and the outer contour. Hysteroscopy is the standard in intrauterine pathology, but cannot distinguish septate from bicornuate uterus unless accompanied by laparoscopy; MRI is superior in this distinction.
• Antiphospholipid syndrome—anticardiolipin antibody (IgG and IgM) and lupus anticoagulant twice, at least 12 weeks apart. The justification for repeating is that low-moderate positivity is due to viral disease and can return to normal. Diagnosis also requires testing for anti-β2 glycoprotein I antibodies.
• Thyroid — TSH and TPO antibody.
• Karyotype — the karyotype of the parent and, if possible, the karyotype of the abortive material. This is the last test ordered because the likelihood of abnormalities is low; It is done if previous examinations are negative. If the aborted material does not grow in culture, array comparative genomic hybridization (array CGH) can be used because it does not require dividing cells; At least one institution prefers array CGH to karyotype.

RPL is a severe source of stress for couples; in a study of 301 women, the rate of depression was four times higher (8.6% vs. 2%) and severe stress twice as high (41% vs. 23%) among those with RPL. Screening for depression is part of the evaluation.

Treatment: what works, what doesn't

In recurrent pregnancy loss with a known cause, treatment is directed towards the cause: uterine septum, intrauterine adhesion and submucous myoma are corrected hysteroscopically; hyperprolactinemia, thyroid disorder, and diabetes are treated medically. Couples with detected chromosomal abnormalities are referred to genetic counselling. Aspirin and heparin improve pregnancy outcome in the presence of antiphospholipid syndrome—but this benefit cannot be extrapolated to women without antiphospholipid antibodies.

Clinical Summary

• Diagnosis: Definitive criteria by single ultrasonography — CRL ≥7 mm and no cardiac activity, or mean sac diameter ≥25 mm and no embryo. CRL 5–7 mm and sac 16–25 mm are uncertain areas and require monitoring. Pregnancy week is determined by ultrasonography, not by the last menstrual period.
• β-hCG: A single value is not diagnostic; Three serial measurements to assess the trend. In the presence of bleeding, a >25% decrease in 48 hours is in favor of pregnancy loss.
• Emergency: Hemodynamic instability or sepsis → surgical evacuation even if ultrasonographic criteria are not fully met. The patient with symptomatic infection is not a candidate for waiting or drug therapy.
• Management: Expectant management 70–80% at 14 days; mifepristone 200 mg + misoprostol 800 mcg 91.2% on day 30 (misoprostol alone 75.8%); vacuum aspiration is close to 100%. The patient makes the choice.
• Antibiotic: No routine prophylaxis in drug therapy; 60 minutes before the procedure in vacuum aspiration (doxycycline 200 mg single dose).
• Anti-D: <12 weeks — WHO and ACOG do not recommend routine testing/RhIG, SMFM does. ≥12 weeks — test + RhIG: 100 mcg (500 IU) at 12–18 weeks, 300 mcg (1500 IU) at >18 weeks.
• RPL definition: ASRM, ESHRE, and SOGC define it as two losses; RCOG defines it as three losses. The risk of miscarriage in the subsequent pregnancy is ~15% after one loss, 17–31% after two losses, and 25–46% after three losses.
• RPL workup: Sonohysterography · anticardiolipin + lupus anticoagulant (twice, 12 weeks apart) + anti-β2 glycoprotein I · TSH + TPO antibody · lastly, parental and products-of-conception karyotyping.
• RPL treatment: Aspirin + heparin is effective in antiphospholipid syndrome. In unexplained RPL, progesterone is not recommended for most patients, heparin is ineffective, and immunotherapy and glucocorticoids are not recommended. Even when the workup is normal, the live birth probability is 77%.

← All scientific guides