Early Pregnancy Failure: How Can We Confidently Diagnose Nonviable Pregnancies?
By Rebecca H. Allen, MD, MPH
Assistant Professor, Department of Obstetrics and Gynecology, Warren Alpert Medical School of Brown University, Women and Infants Hospital, Providence, RI
Dr. Allen reports no financial relationships relevant to this field of study.
Synopsis: According to new guidelines from the Society of Radiologists in Ultrasound, more conservative cutoffs are recommended for diagnosing early pregnancy failure. For example, a crown-rump length of ≥ 7 mm without a heartbeat or a mean sac diameter of ≥ 25 mm and no embryo are required before the diagnosis of pregnancy failure with 100% confidence can be made.
Source: Doubilet PM, et al. Diagnostic criteria for nonviable pregnancy early in the first trimester. N Engl J Med 2013;369:1443-1451.
The diagnosis of early pregnancy failure is an important call to make as a provider. The consequences of inaccurate diagnosis in the setting of a desired pregnancy are extreme — namely, possible interruption of a normal pregnancy. Diagnosing pregnancy early in gestation is now more common, yet managing equivocal findings of viability on transvaginal ultrasound are also challenging. Many practices routinely obtain an early dating ultrasound on their obstetric patients. In addition, first-trimester bleeding is a common presentation to medical offices and emergency departments. In an area where radiology, obstetrics and gynecology, emergency medicine, and family medicine intersect, recommendations from various specialty societies on the diagnosis of early pregnancy failure have differed in the past. When I was a resident, we learned the 5-10-20 rule. In this mnemonic, one should expect to see a heartbeat by the time the crown-rump length measured 5 mm, a yolk sac if the mean gestational sac diameter was 10 mm, and an embryo if the mean gestational sac diameter was 20 mm. If these structures were not seen, then the diagnosis of miscarriage was made.
In early pregnancy, there is a predetermined and predictable order of development: the gestational sac appears at 5 weeks, the yolk sac at 5.5 weeks, and the embryo can be seen at 6 weeks with a variation of ± half a week.1 When pregnancies deviate from these expected findings, the diagnosis is suspicious for early pregnancy failure. But how can one be absolutely certain? It turns out that the evidence behind the guidelines or "rules of thumb" promulgated by various subspecialty organizations is not robust.2 It is surprising that for a diagnosis of such importance, we have failed to conduct appropriately powered studies. Many of these standard guidelines were based on older studies with small numbers of patients. For example, the American College of Radiologists’ guideline from 2009 states that an embryo should be seen by a mean gestational sac diameter of 16 mm and a heartbeat should be seen in an embryo > 5 mm.1 The studies behind these criteria had a total combined subject number of 47 women!
Fortunately, several systematic reviews published in 2011 in the journal Ultrasound in Obstetrics and Gynecology have clarified what we know and what we don’t know.3,4,5,6 In response to these articles, the Society of Radiologists in Ultrasound (SRU) convened a multispecialty panel on Early First Trimester Diagnosis of Miscarriage and Exclusion of a Viable Intrauterine Pregnancy in October 2012. This group reviewed the literature and issued more conservative recommendations to avoid falsely diagnosing a miscarriage when the pregnancy could be viable. These are similar to new guidelines issued by the American College of Radiologists and the Royal College of Obstetricians and Gynaecologists.7,8 The American College of Obstetricians and Gynecologists has yet to weigh in on this issue.
First, the authors reviewed the cutoff for cardiac activity detection in an embryo. Although cardiac activity is usually seen in an embryo of any crown rump length, the old criteria of 5 mm was based on such small numbers of patients that the 95% confidence intervals for the specificity of this finding are very large. In fact, several studies have documented subsequent viable pregnancies in embryos of 5 mm and 6 mm that did not have a heartbeat visualized. Furthermore, the inter-observer variation in the measurement of crown-rump length is ± 15% even among experienced sonographers.6 Therefore, the SRU concluded that a cutoff of 7 mm is more appropriate and would yield a specificity and positive predictive value of 100%. When the crown-rump length is less than 7 mm and there is no heartbeat, this is considered suspicious, but not diagnostic, of pregnancy failure.
Second, the authors looked at mean gestational sac diameter (the average of the sagittal, transverse, and anteroposterior diameters of the sac). Older studies suggested cutoffs of 16 mm or 20 mm for failed pregnancy if there was no embryo present. These, too, are based on a small number of subjects and have wide confidence intervals. In addition, the issue of interobserver variability comes into play.6 Therefore, the SRU recommends that a mean sac diameter of ≥ 25 mm and no embryo is diagnostic of miscarriage while a mean sac diameter of 16-24 mm and no embryo is suspicious.
The SRU document then outlines time-based criteria with which miscarriage can be diagnosed. They state that the absence of an embryo with a heartbeat ≥ 14 days after a scan that showed a gestational sac without a yolk sac is diagnostic of pregnancy failure. In addition, the absence of an embryo with a heartbeat ≥ 11 days after a scan that showed a gestational sac with a yolk sac is diagnostic of a pregnancy failure. Other findings that are suspicious for, but not diagnostic of, pregnancy failure include enlarged yolk sac (> 7 mm), empty amnion, and small gestational sac in relation to the size of the embryo (< 5 mm) difference between mean sac diameter and crown-rump length. For any suspicious findings, the authors recommend follow-up ultrasound in 7-10 days.
In essence, this article is a plea for clinicians to be more stringent in their diagnosis of miscarriage. The misdiagnosis of miscarriage has been the subject of national health system review in the United Kingdom and Ireland.1 Early Pregnancy Assessment Units, dedicated to the care of women in early pregnancy, are very common in those countries as well as Canada.9 Even with those specialized units, reviews found that improvements in diagnosis could be made. If the diagnosis of miscarriage is unsure and/or the woman requests a repeat ultrasound, this should be ordered. It is rare that the diagnosis of miscarriage by ultrasound is taking place in an emergency situation. Therefore, with a stable patient, there is plenty of time to be sure of the diagnosis before intervening medically or surgically.2 As clinicians, we should first "do no harm," as the Hippocratic Oath says.
- Bourne T, Bottomly C. Fertil Steril 2012;98:1091-1096.
- Thilaganathan B. Ultrasound Obstet Gynecol 2011; 38:487-488.
- Jeve Y, et al. Ultrasound Obstet Gynecol 2011;38:489-496.
- Abdallah Y, et al. Ultrasound Obstet Gynecol 2011; 38:497-502.
- Abdallah Y, et al. Ultrasound Obstet Gynecol 2011;38:503-509.
- Pexsters A, et al. Ultrasound Obstet Gynecol 2011;38:510-515.
- American College of Radiology Appropriateness Criteria. First Trimester Bleeding. Available at: www.acr.org/~/media/ACR/Documents/AppCriteria/Diagnostic/FirstTrimesterBleeding.pdf. Accessed October 22, 2013.
- National Institute for Health and Care Excellence. Guideline # 154. Ectopic pregnancy and miscarriage: Available at: www.nice.org.uk/guidance/CG154. Accessed October 22, 2013.
- Edey K, et al. Clin Obstet Gynecol 2007;50:146-153.