Pregnancy of Unknown Location

Peter Kovacs, MD, PhD


February 05, 2019

Pregnancy of Unknown Location

The corpus luteum is responsible for the hormonal preparation of the endometrium for implantation. Its life span is about 14 days. When implantation occurs, human chorionic gonadotropin (hCG) secreted by trophoblast cells rescues the corpus luteum, and its activity is maintained until around week 8-9 of pregnancy when steroid synthesis shifts to the placenta.[1]

Measurement of hCG is used to confirm pregnancy. During an uncomplicated early pregnancy, the hCG level keeps rising and doubles every 48-72 hours.[2] By around week 5, an intrauterine sac is visible; and by week 7, an embryo with heartbeat can be seen.[3]

Some early pregnancies are complicated by pain and/or bleeding. These could be signs of a threatened miscarriage, actual miscarriage, or an ectopic pregnancy. When bleeding or pain complicates an early pregnancy, it is important to assess the location and viability of the pregnancy. In some cases, despite a positive pregnancy test, the pregnancy cannot be located on transvaginal ultrasound. These cases are termed "pregnancy of unknown location" (PUL). Although overall 1%-2% of pregnancies are extrauterine, the risk is significantly higher when the pregnancy is classified as PUL (7%-30%).[3] It is important to be able to classify these cases as low- or high-risk for ectopic pregnancy.

PUL Risk-Assessment Protocols

A recent meta-analysis[4] compared various protocols for the risk assessment of PUL. Data from 37 prospective and retrospective studies involving 23,802 cases compared various protocols for the management of PUL. Four different protocols were compared:

  • Single hCG cut-off (1000 IU/L);

  • hCG ratio (hCG at 48 hours/hCG at baseline (0.87-1.66);

  • Progesterone cut-off: 10 nmol/L; and

  • M4 model (multinomial logistic regression model based on hCG at presentation and hCG ratio.

The following predictive performances were obtained using area under the curve (AUC), with 95% confidence intervals (CI) (Table).

Table. Performance of PUL Diagnostic Protocols

Outcome Single hCG AUC (95% CI) hCG ratio AUC (95% CI) Progesterone AUC (95% CI) M4 model AUC (95% CI)
Ectopic pregnancy 0.42 (0.00-0.99) 0.69 (0.57-0.78) 0.69 (0.54-0.81) 0.87 (0.83-0.91)
Intrauterine pregnancy 0.71 (0.00-1.00) 0.97 (0.87-0.99) 0.89 (0.0.76-0.96) 0.86 (0.39-0.99
Failed PUL 0.59 (0.00-1.00) 0.87 (0.76-0.94) 0.87 (0.76-0.94) 0.84 (0.15-1.00)

These data show that the M4 model performed best at predicting ectopic pregnancy and it also performed well at predicting intrauterine pregnancy and failed PUL.


Old wisdom holds that every pregnancy should be considered extrauterine until intrauterine location is confirmed. This is true despite only 1%-2% of pregnancies being extrauterine.[3] Known risk factors for ectopic pregnancies include previous ectopic pregnancy, tubal or pelvic surgery, a history of pelvic infections, smoking, and use of assisted reproductive technology. In these cases, more careful monitoring of the pregnancy from early stages should occur.

Pain and vaginal bleeding can be signs of threatened miscarriage, incomplete miscarriage, or extrauterine pregnancy but also may occur in a normal intrauterine pregnancy. Ultrasound is often helpful to establish the proper diagnosis. In some cases, however, the scan will not detect either an intrauterine sac or an extrauterine mass. These cases are termed PUL. The outcome of PUL could be one of five scenarios: normal intrauterine pregnancy, extrauterine pregnancy, biochemical pregnancy loss, missed abortion, or persistent PUL. Some require intervention whereas others do not. It's important to assess a woman's individual risk for these outcomes. Correct identification of an ectopic pregnancy prior to rupture would enable proper medical/surgical care.

Various prognostic tools have been evaluated (single hCG, serial hCG, progesterone measurement; regression models using these parameters; other serum markers).[5]

On the basis of 20,000-plus cases of PUL, the M4 model was the most predictive for ectopic pregnancy and performed well in identifying intrauterine pregnancies as well as biochemical losses. This model can aid in the proper diagnosis of PUL.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.