Miscarriage (Mis)management part 1: Evaluation and Diagnosis

This is a guest post by Dr. Julia Dmytryshyn.

Miscarriages are one of the most poorly managed presentations in emergency medicine, but this is not because emergency medicine physicians manage them poorly. This is largely due to the current methods of miscarriage diagnosis and management, which inherently lead to patient uncertainty, anxious waiting periods, and psychological distress. Part one of this two-part review will briefly summarize the most recent ultrasound guidelines for diagnosis of failed pregnancy and the utility of serum progesterone in the evaluation of bleeding in early pregnancy. Part two will review management and patient counselling.

There’s not much reassurance in telling someone they must wait 7 days for a repeat ultrasound to confirm that their pregnancy is non-viable. The diagnostic challenge that threatened spontaneous abortions presents leads to delayed care and increased anxiety. Setting clearer diagnostic pathways will hopefully minimize this. Different ultrasound cut-off values for diagnosing early pregnancy failure have been published by various national organizations. In attempts to minimize the risk of false positive diagnoses, leading to potential termination of wanted pregnancies, stricter diagnostic guidelines have been developed.

The consensus ultrasound findings that are 100% specific for non-viable pregnancy are:

• no fetal heart rate when crown-rump length is greater than 7 mm
• absence of an embryo in a gestational sac greater than 25 mm in diameter
• absence of an embryo with a fetal heart rate 11 days after a scan showing a gestational sac with a yolk sac or 2 weeks after a scan showing a gestational sac without a yolk sac (1)

Stringent criteria are necessary, especially because diagnostic inaccuracy is compounded by inter-observer variability when measuring crown-rump length and mean sac diameter. However, these criteria increase the proportion of inconclusive scans.

Serum progesterone is another tool that has been proposed to predict fetal viability among symptomatic patients in early pregnancy. A recent paper published in the Canadian Journal of Emergency Medicine concluded that over 90% of patients with bleeding in early pregnancy can be diagnosed with a non-viable pregnancy if their serum progesterone level is less than 6.3 ng/mL or with a viable pregnancy if the level is above 20-25 ng/mL. This systematic review and meta-analysis conducted by Ghaedi and colleagues included a total of 15,878 patients, of whom 49.5% were confirmed to have a viable pregnancy and 50.5% were confirmed to have a non-viable pregnancy. They evaluated the performance of different progesterone cut-off values and found that to detect non-viability, values less than 6.3 ng/mL had a sensitivity of 73.1% (95% CI 55.9, 85.3) and a specificity of 99.2% (97.2, 99.8). To detect viability, values greater than 20-25 ng/mL had a sensitivity of 91.3% (95% CI 84.0, 95.4) and a specificity of 75% (57.4, 86.9) (2).

Can we use the serum progesterone level to change our management of these patients? One of the problems with this, is that of the 50.5% of pregnancies that were confirmed to be non-viable, 21.7% of these were ectopic pregnancies (2). An editorial published a month later in the same journal argued that serum progesterone levels in the emergency department should not change the care of patients with first trimester bleeding, due to the fact that a serum progesterone level cannot distinguish an intrauterine pregnancy from an ectopic pregnancy (3).

The other issue is that a large percentage of serum progesterone levels will fall within the middle range (greater than 6.3 but less than 20 ng/mL), where this result won’t help us. Drawing on one of the largest trials included in the review as an example, of the 3700 patients enrolled, about 1100 had a low progesterone and were confirmed non-viable, and another 450 had a high progesterone and were confirmed viable. About 1900 patients (51%) had progesterone levels that fell in the middle range (4). This might end up adding more testing, cost, and time to ED visits without benefit to about half our patients.

Therefore, although perhaps not for widespread adoption at this point, a serum progesterone level is an additional piece of information that can help to frame patients’ expectations (with a progesterone level over 25 suggesting a 90% chance of viable pregnancy), to reduce some uncertainty, expedite referrals, and reduce return visits to the ED. However, a serum progesterone level should not change management when concern for ectopic pregnancy remains, and the costs of this test should be considered before incorporating it into routine practice.

References

1. American College of Obstetricians and Gynecologists. (2018). ACOG Practice Bulletin No. 200: early pregnancy loss. Obstetrics and gynecology, 132(5), e197-e207.
2. Ghaedi, B., Cheng, W., Ameri, S., Abdulkarim, K., Costain, N., Zia, A., & Thiruganasambandamoorthy, V. (2022). Performance of single serum progesterone in the evaluation of symptomatic first-trimester pregnant patients: a systematic review and meta-analysis. Canadian Journal of Emergency Medicine, 24(6), 611-621.
3. Tunde-Byass, M., & Varner, C. E. (2022). Serum progesterone levels in the emergency department should not change the care of patients with first trimester bleeding. Canadian Journal of Emergency Medicine, 24(6), 559-560.
4. McCord, M. L., Muram, D., Buster, J. E., Arheart, K. L., Stovall, T. G., & Carson, S. A. (1996). Single serum progesterone as a screen for ectopic pregnancy: exchanging specificity and sensitivity to obtain optimal test performance. Fertility and sterility, 66(4), 513-516.

---

You can find more guest posts here.