Table 4.
Evaluation of current guidelines using data from study
| Source of guidance | Criteria proposed to definitively diagnose miscarriage | Positive predictive value (%, 95% CI) | Specificity (%, 95% CI) |
|---|---|---|---|
| Initial scan criteria: | |||
| ACR, NICE, RCOG, RANZCOG, SRU (Doubilet et al) | Presenting without an embryo or yolk sac, and mean gestational sac diameter ≥25 mm | 12/12 (100, 73.5 to 100) | 364/364 (100, 99.0 to 100) |
| ACR, NICE, RCOG, RANZCOG, SRU (Doubilet et al) | Presenting with an embryo without heart activity, and crown-rump length ≥7 mm | 17/17 (100, 80.5 to 100) | 110/110 (100, 96.7 to 100) |
| Repeat scan criteria: | |||
| NICE | Presenting as an empty gestational sac with mean sac diameter <25 mm or with embryo (no heart activity) with crown-rump length <7 mm, and returning after at least seven days: no definitive diagnosis (that is, no embryo heart activity visible) | 530/549 (96.5, 94.8 to 97.8) | 1074/1093 (98.3, 97.4 to 98.9) |
| ACR | Presenting as an empty gestational sac, and gestational sac still empty 7-10 days later | 75/77 (97.4, 91.9 to 99.5) | 86/88 (97.7, 92.9 to 99.5) |
| NICE | Presenting with an embryo (no heart activity) with crown-rump length <7 mm, and still no embryo heart activity visible after at least seven days | 140/140 (100, 97.4 to 100) | 103/103 (100, 96.5 to 100) |
| RCOG | In event of doubt repeat scan in at least one week | Not applicable | |
| SRU (Doubilet et al) | Presenting with an empty gestational sac, and no embryo heart activity visible after at least 14 days | 71/72 (98.6, 93.7 to 99.9) | 180/181 (99.5, 97.5 to 99.9) |
| SRU (Doubilet et al) | Presenting with a gestational sac and yolk sac, and no embryo heart activity visible after at least 11 days | 85/85 (100, 95.8 to 100) | 537/537 (100, 99.3 to 100) |
ACR=American College of Radiology; RCOG=Royal College of Obstetricians and Gynaecologists; NICE=National Institute for Health and Care Excellence; RANZCOG=Royal Australia and New Zealand College of Obstetricians and Gynaecologists; SRU=Society of Radiologists in Ultrasound.
Modified Jeffreys method used for confidence intervals when percentages equalled 100% (or 0%). Standard Jeffreys methods used otherwise.