Table 2.
Association between predefined possible predictors and success of expectant management of embryonic and anembryonic miscarriage ≤ 17 days (univariable logistic regression analysis); n = 83
| Variables tested as possible predictors | Success (n = 44) | Failure (n = 39) | P valuea | Odds ratio (95% CI) |
|---|---|---|---|---|
| Biochemical variables | ||||
| s-beta-human chorionic gonadotropin (IU/L)b | 2 missing | 0.01 | 0.958 (0.922–0.995)c | |
| Mean | 9530.6 ± 7483.3 | 18,128.3 ± 21,019.9 | ||
| Median | 8308.0 (107–26,585) | 7990.0 (1006–80,317) | ||
| s-Progesterone (nmol/L)b | 1 missing | 0.003 | 0.965 (0.936–0.995) | |
| Mean | 18.5 ± 10.6 | 34.6 ± 38.2 | ||
| Median | 18.0 (3–41) | 20.5 (4–190) | ||
| Clinical variables | ||||
| Gestational age according to last menstrual period (days) | 2 missing | 3 missing | 0.04 | 1.042 (1.000–1.086) |
| Mean | 77.8 ± 10.7 | 72.4 ± 12.4 | ||
| Median | 78.0 (60–104) | 72.0 (42–94) | ||
| Vaginal delivery | ||||
| Yes | 25 (56.8) | 23 (59.0) | 0.843 | 0.915 (0.382–2.195) |
| No | 19 (43.2) | 16 (41.0) | ||
| Parity | ||||
| Yes | 29 (65.9) | 24 (61.5) | 0.679 | 1.208 (0.493–2.963) |
| No | 15 (34.1) | 15 (38.5) | ||
| Bleeding at inclusiond | ||||
| Moderate/heavy | 10 (22.7) | 3 (7.7) | 0.053 | 3.529 (0.894–13.927)e |
| None/mild | 34 (77.3) | 36 (92.3) | ||
| Pain at inclusion | ||||
| Yes | 13 (29.5) | 11 (28.2) | 0.893 | 1.067 (0.412–2.765) |
| No | 31 (70.5) | 28 (71.8) | ||
| Ultrasound variables | ||||
| Gestational sac | 0.925 | 0.955 (0.365–2.499) | ||
| Not round/oval | 12 (27.3) | 11 (28.2) | ||
| Round/oval | 32 (72.7) | 28 (71.8) | ||
| Gestational sac diameter (mm)f | 1 missing | 0.967 | 0.999 (0.950–1.050) | |
| Mean | 24.1 ± 9.5 | 24.1 ± 7.9 | ||
| Median | 21.4 (4.7–49.0) | 23.2 (10.7–41.0) | ||
| Crown-rump-length (mm) | ||||
| Mean | 7.5 ± 7.0 | 3.6 ± 4.9 | 0.004 | 1.119 (1.028–1.218) |
| Median | 5.6 (0.0–26.0) | 0.0 (0.0–17.0) | ||
| Miscarriage type on ultrasound | ||||
| Embryonic | 32 (72.7) | 19 (48.7) | 0.024 | 2.807 (1.126–6.998) |
| Anembryonic | 12 (27.3) | 20 (51.3) | ||
| Crown-rump-length if embryonic miscarriage (mm) | n = 32 | n = 19 | 0.069 | 1.109 (0.981–1.254) |
| Mean | 10.3 ± 6.2 | 7.4 ± 4.6 | ||
| Median | 9.5 (3.0–26.0) | 5.3 (3.0–17.0) | ||
| Blood flow in presumed intervillous space according to | ||||
| Grey-scale ultrasound | 0.105 | 0.205 (0.023–1.840) | ||
| Yes | 39 (88.6) | 38 (97.4) | ||
| No | 5 (11.4) | 1 (2.6) | ||
| Color Doppler ultrasound | 1 missing | 0.008 | 0.105 (0.013–0.873) | |
| Yes | 35 (79.5) | 37 (94.9) | ||
| No | 9 (20.5) | 1 (2.6) | ||
| Spectral Doppler ultrasound (arterial blood flow) | 4 missing | 4 missing | 0.125 | 0.468 (0.175–1.252) |
| Yes | 23 (57.5) | 25 (71.4) | ||
| No | 17 (42.5) | 10 (28.6) | ||
Data are presented as n (%), mean ± SD, median (range)
CI confidence interval
aLikelihood ratio test
bAfter mid May 2013 beta-human chorionic gonadotropin and progesterone were analyzed in plasma and not in serum, reference intervals are not affected
cOdds ratio is calculated for units of 1000
dBleeding as judged by the trial physician at speculum examination—all women reported vaginal bleeding
eWe consider this result unreliable, because only three patients in the failure group had heavy or moderate bleeding, the unreliable result is also reflected in the odds ratio
fMean of three orthogonal diameters