In total, 1284 FBT cycles were enrolled during the study period. The demographic and basal characteristics of women with live births and non-live births are presented in Table 1. Overall, the LBR per transfer was 44.7%. We observed that patients with a live birth were younger, had a lower basal serum FSH level (P = 0.003), higher proportion of first FET cycle, greater number of blastocysts transferred, and greater number of good-quality blastocysts transferred than those without a live birth (Table 1). Moreover, body mass index, cause of infertility, the type and duration of infertility, basal serum luteinizing hormone and E2 levels, endometrial thickness, and the type of endometrial preparation were not significantly different between those with and without a live birth.
Table 1 Baseline characteristics of the patients
|
Live birth (n = 574)
|
Non-live birth (n = 710)
|
P-value
|
Variables
|
|
|
|
Age (years), n (%)
|
|
|
|
35–37
|
410 (71.4)
|
403 (56.8)
|
<0.001
|
38–42
|
164 (28.6)
|
307 (43.2)
|
|
BMI (kg/m2), n (%)
|
|
|
0.309
|
<18.5
|
25 (4.4)
|
44 (6.2)
|
|
18.5–24.9
|
426 (74.2)
|
509 (71.7)
|
|
≥25
|
123 (21.4)
|
157 (22.1)
|
|
Cause of infertility, n (%)
|
|
|
0.173
|
Female infertility
|
343 (59.8)
|
384 (54.1)
|
|
Male infertility
|
37 (6.4)
|
58 (8.2)
|
|
Mixed infertility
|
121 (21.1)
|
176 (24.8)
|
|
Unknown infertility
|
73 (12.7)
|
92 (13.0)
|
|
Type of infertility, n (%)
|
|
|
0.472
|
Primary
|
159 (27.7)
|
184 (25.9)
|
|
Secondary
|
415 (72.3)
|
526 (74.1)
|
|
Duration of infertility (years)
|
5.5 ± 1.7
|
5.4 ± 2.1
|
0.290
|
Basal FSH (mUI/ml)
|
7.6 ± 2.1
|
8.0 ± 2.4
|
0.003
|
Basal LH (mUI/ml)
|
4.6 ± 2.4
|
4.5 ± 2.3
|
0.367
|
Basal E2 (pg/ml)
|
44.6 ± 22.8
|
43.2 ± 21.3
|
0.257
|
FET cycles rank
|
|
|
0.017
|
1st cycle
|
439 (76.5)
|
501 (70.6)
|
|
2nd cycle
|
135 (23.5)
|
209 (29.4)
|
|
N of blastocyst transferred, n (%)
|
|
|
<0.001
|
1
|
223 (38.9)
|
362 (51.0)
|
|
2
|
351 (61.1)
|
348 (49.0)
|
|
N of good-quality blastocyst, n (%)
|
|
|
<0.001
|
0
|
112 (19.5)
|
282 (39.7)
|
|
1 or 2
|
462 (80.5)
|
428 (60.3)
|
|
Endometrial thickness (mm)
|
9.6 ± 1.8
|
9.4 ± 1.8
|
0.091
|
Endometrium preparation, n (%)
|
|
|
0.829
|
NC
|
89 (15.5)
|
107 (15.1)
|
|
AC
|
485 (84.5)
|
603 (84.9)
|
|
Abbreviations: N Number, BMI Body mass index, NC Natural cycle, AC Artificial cycles
Clinical outcomes
The clinical outcomes of the first and second FET cycles are presented in Table 2. In cycle 1, the LBR (52.3% vs. 39.9%; OR, 1.65; 95% CI, 1.28–2.14) and CPR (64.2% vs. 50.2%; OR, 1.78; 95% CI, 1.37–2.31) were significantly higher in the double blastocyst transfer (DBT) group than in the SBT group. The differences were statistically significant after multiple regression analysis (LBR: aOR, 1.65; 95% CI, 1.26–2.15; CPR: aOR, 1.78; 95% CI, 1.36–2.32). The confounding factors considered in the regression model were maternal age, basal FSH level, and quantity of good-quality blastocysts transferred. There were no statistical differences between the two groups in terms of miscarriage rate. In cycle 2, LBR did not differ significantly between the two groups (44.3% vs. 33.3%; aOR, 1.30; 95% CI, 0.81–2.08) after adjusting for confounding factors (Table. 2).
Table 2 Pregnancy outcomes of the First and second FET cycle
Cycle rank
|
|
|
|
|
|
|
Cycle 1
|
|
SBT (n = 426)
|
DBT (n = 514)
|
Crude OR (95% CI)
|
P-value
|
Adjusted OR (95% CI) *
|
P-value
|
|
Live birth rate
|
170/426 (39.9)
|
269/514 (52.3)
|
1.65 (1.28–2.14)
|
<0.001
|
1.65 (1.26–2.15)
|
<0.001
|
|
Clinical pregnancy rate
|
214/426 (50.2)
|
330/514 (64.2)
|
1.78 (1.37–2.31)0
|
<0.001
|
1.78 (1.36–2.32)
|
<0.001
|
|
Miscarriage rate
|
45/214 (21.0)
|
66/330 (20.0)
|
0.94 (0.61–1.44)
|
0.771
|
|
|
|
|
|
|
|
|
|
|
Cycle 2
|
|
SBT (n = 159)
|
DBT (n = 185)
|
Crude OR (95% CI)
|
P-value
|
Adjusted OR (95% CI) *
|
P-value
|
|
Live birth rate
|
53/159 (33.3)
|
82/185 (44.3)
|
1.52 (1.03–2.26)
|
0.037
|
1.30 (0.81–2.08)
|
0.271
|
|
Clinical pregnancy rate
|
65/159 (40.9)
|
107/185 (57.8)
|
1.98 (1.29–3.05)
|
0.002
|
1.63 (1.03–2.60)
|
0.039
|
|
Miscarriage rate
|
12/65 (18.5)
|
26/107 (24.3)
|
1.42 (0.66–3.05)
|
0.371
|
|
|
Abbreviations: SBT Single blastocyst transfer, DBT Double blastocyst transfer, OR Odds ratio, CI Confidence interval
*Adjusted for age, basal FSH, and N of good-quality blastocyst transferred (1 or 2 vs. 0).
Analysis stratified by maternal age
We analyzed the correlation between transferred blastocyst number (2 vs. 1) and LBR in two subgroups stratified by age, i.e., 35–37 years and 38–42 years (Table 3). For patients aged 35–37 years, the LBR of the DBT group was considerably higher than that of the SBT group in the first cycle (aOR, 1.75; 95% CI, 1.27–2.43). However, the LBR of DBT did not increase in the second cycle (aOR, 0.94; 95% CI, 0.52–1.70). For patients aged 38–42 years, the LBR did not improve significantly in the DBT group compared with the SBT group in the first cycle (aOR, 1.45; 95% CI, 0.91–2.31). Meanwhile, we found the DBT group had significantly higher LBR than the SBT group in the second cycle (aOR, 2.27; 95% CI, 1.05–4.90). The confounding factors considered in the regression model were basal FSH level and the number of good-quality blastocysts transferred.
Table 3 Logistic regression analysis for live-birth in DBT cycles compared with SBT cycles stratified by maternal age
Age (years)
|
Cycle rank
|
Group
|
N live-births/N cycles (LBR per cycle, %)
|
Adjusted OR (95% CI) *
|
P-value
|
35–37
|
Cycle 1 (n = 616)
|
SBT
|
121/271 (44.6)
|
Reference
|
0.001
|
|
|
DBT
|
201/345 (58.3)
|
1.75 (1.27–2.43)
|
|
|
Cycle 2 (n = 197)
|
SBT
|
37/84 (44.0)
|
Reference
|
0.936
|
|
|
DBT
|
51/113 (45.1)
|
0.94 (0.52–1.70)
|
|
|
|
|
|
|
|
38–42
|
Cycle 1 (n = 324)
|
SBT
|
49/155 (31.6)
|
Reference
|
0.122
|
|
|
DBT
|
68/169 (40.2)
|
1.45 (0.91–2.31)
|
|
|
Cycle 2 (n = 147)
|
SBT
|
16/75 (21.3)
|
Reference
|
0.036
|
|
|
DBT
|
31/72 (43.1)
|
2.27 (1.05–4.90)
|
|
Abbreviations: N Number, LBR live birth rates, SBT Single blastocyst transfer, DBT Double blastocyst transfer, OR Odds ratio, CI Confidence interval
*Adjusted for basal FSH and N of good-quality blastocyst transferred (1 or 2 vs. 0).
We then calculated the TDR for each age. We found that the TDR was higher in the DBT group (9.1%–20.7%) compared with the SBT group (0.0%–2.0%) for women aged 35–39 years (Fig. 2). The TDR was almost 0% for patients aged 40–42 years in both, cycle 1 and cycle 2.