The systematic search retrieved 7766 titles after removal of duplicates. One hundred eighty-two eligible studies had their full texts reviewed. One hundred five studies met our inclusion criteria and were included into the qualitative meta-analysis. A further forty-one studies that did not report per woman data were excluded, leaving a total of sixty-four eligible studies (N = 57,988 women) for quantitative meta-analysis. Study identification and selection process is shown in Fig. 1.
For fresh COS cycle, three studies reported progesterone monitoring during the start of the menstrual cycle, forty-three studies reported monitoring during day of trigger, three studies reported monitoring during egg collection day and three studies reported monitoring progesterone during the luteal phase. For FET cycles, one study reported monitoring on the day of trigger in modified NC-FET, two studies reported progesterone monitoring in NC-FET on the day before ovulation and nine studies reported monitoring progesterone during luteal phase in natural cycle FET with and without progesterone supplementation and medicated HRT cycle (Table 1). Assessment for bias using NOS is shown by Supplementary Table S2.
Table 1
Tables of Included Studies
Author/ Year
|
Country
|
Study duration
|
Study design
|
Type of cycle
|
Total number (patient/cycle)
|
Threshold/ reason for choosing
|
Day of ET
|
Conclusion
|
Fresh COS cycle - Basal follicular phase
|
Hamdine et al, 2014
|
Netherlands
|
Mar’09 to Jul’11
|
prospective
|
IVF/ICSI
|
158/ 158
|
P > 1.5ng/ml/ literature
|
Day 3
|
LBR, OPR and CPR NS
|
Mahapatro & Radhakrishan, 2017
|
India
|
Jan’13 to Mar’14
|
retrospective
|
ICSI
|
151/ 151
|
P > 1.5ng/ml / literature
|
Day 2–3
|
LBR and CPR NS
|
Mutlu et al, 2017
|
Turkey
|
Dec’14 and Feb’16
|
prospective
|
ICSI
|
464/ 464
|
P ≥ 0.65ng/ml/ ROC analysis
|
Day 2,3 or 5
|
Similar OPR and CPR
|
Fresh COS cycle - Pre trigger
|
Bosch et al, 2003
|
Spain
|
NA
|
prospective
|
IVF/ICSI
|
81/ 81
|
P > 1.2ng/ml / ROC analysis
|
Day 3
|
CPR ↓
|
Martinez et al, 2003
|
Spain
|
Jul’2 to Jan’03
|
retrospective
|
IVF/ICSI
|
377/ 377
|
P > 0.9ng/ml / ROC analysis
|
Day 2–3
|
CPR and MR NS
|
Anderson et al, 2006
|
Belgium
|
Feb’04 to Dec’04
|
RCT
|
IVF
|
731/ 731
|
P > 4nmol/L (1.25ng/ml) / literature
|
Day 3
|
COC ↑, OPR ↓
|
Seow KM et al, 2007
|
Taiwan
|
Jan’03 to Jan’05
|
prospective
|
IVF/ICSI
|
95/ 95
|
P ≥ 1.2ng/ml / literature
|
Day 2–3
|
CPR NS
|
Lee F et al, 2008
|
China
|
Mar’03 to Apr’07
|
retrospective
|
IVF/ICSI
|
223/ 223
|
P > 2.0ng/ml / arbitrary
|
Day 2–3
|
CPR ↓
|
Li R et al, 2008
|
China
|
Jul’06 to Dec’06
|
prospective
|
IVF/ICSI
|
251/ 251
|
P > 3.97nmol/L (1.25ng/ml) / sensitivity-specificity analysis
|
Day 3
|
CPR ↓ in fresh cycle, CPR NS in FET
|
Kiliçdag et al, 2009
|
Turkey
|
Oct’04 to May’08
|
retrospective
|
ICSI
|
1045/ 1045
|
P > 1.1ng/ml / sensitivity-specificity analysis
|
Day 3
|
LBR, OPR and CPR ↓
|
Papanikolaou et al, 2009
|
Belgium
|
May’04 to Feb’05
|
prospective
|
IVF/ICSI
|
482/ 482
|
P > 1.5ng/ml / literature
|
Day 3 or 5
|
CPR ↓ in D3, similar CPR in D5
|
Rezaee et al, 2009
|
Iran
|
1 year (2009)
|
prospective
|
Fresh cycle
|
38/ 38
|
P > 1.2ng/ml / literature
|
Day 2
|
CPR ↑ but NS
|
Seow KM et al, 2010
|
Taiwan
|
Jun’04 to Jun’07
|
prospective
|
IVF/ICSI
|
233/ 233
|
P > 1.2ng/ml / ROC analysis
|
Day 3
|
CPR ↓
|
Elgindy, 2011
|
Egypt
|
Aug’08 to Jun’10
|
prospective
|
ICSI
|
240/ 240
|
P > 1.5ng/ml / ROC analysis
|
Day 3 or 5
|
CPR ↓ in Day 3 embryo, CPR NS in day 5 embryo
|
Lahoud et al, 2011
|
Australia
|
Jan’03 to Dec’03
|
retrospective
|
IVF/ICSI
|
582/ 582
|
P ≥ 1.7ng/ml / arbitrary
|
Day 2,3 or 5
|
CPR & MR NS, LBR ↓ in fresh cycle, similar LBR, CPR & MR in FET
|
Yding Anderson et al, 2011
|
Denmark
|
Aug’03 to Nov’04
|
Secondary data analysis from prospective RCT
|
IVF/ICSI
|
475/ 475
|
P > 1.25ng/ml/ arbitrary
|
NA
|
Similar CPR
|
Huang R et al, 2012
|
China
|
Jan’02, to Dec’07
|
retrospective
|
IVF/ICSI
|
2566/ 2566
|
P > 1.2ng/ml/ arbitrary
|
Day 3
|
LBR ↓
|
Kyrou et al, 2012
|
Belgium
|
Oct’07 to Dec’08
|
prospective
|
IVF/ICSI
|
207/ 207
|
P > 1.5ng/ml / literature
|
NA
|
CPR ↓
|
Papanikolaou et al, 2012
|
Greece
|
Aug’07 to Dec’09
|
RCT
|
IVF/ICSI
|
190/ 190
|
P > 1.5ng/ml / literature
|
Day 2,3 or 5
|
LBR ↓
|
Peng C et al, 2012
|
China
|
Jun’08 to Feb’10
|
retrospective
|
IVF
|
180/ 180
|
P ≥ 1.2ng/ml / literature
|
Day 3
|
CPR NS
|
Ochsenkuhn et al, 2012
|
Germany
|
Jan’06 to Jan’11
|
retrospective
|
IVF/ICSI
|
2555/ 2555
|
P > 1.5ng/ml/ literature
|
Day 5
|
LBR ↓
|
Wu Z et al, 2012
|
China
|
Apr’08 to Apr’09
|
retrospective
|
IVF/ICSI
|
2921/ 2921
|
P ≥ 1.05ng/ml / literature
|
Day 3
|
LBR and CPR ↓ in fresh cycles,
CPR NS in FET
|
Corti et al, 2013
|
Italy
|
Jan’12 to Dec’12
|
retrospective
|
IVF/ICSI
|
204/ 204
|
P > 1.5ng/ml / literature
|
Day 5
|
OPR & CPR ↓
|
Griesinger et al, 2013
|
Germany
|
NA
|
Pooled analysis of 6 RCTs
|
IVF/ICSI
|
1866/ 1866
|
P > 1.5ng/ml/ literature
|
Day 3
|
OPR ↓
|
Orvieto et al, 2013
|
Israel
|
10-year period
|
retrospective
|
IVF
|
2244/ 2244
|
P > 1.5ng/ml / literature
|
NA
|
CPR ↓
|
Papaleo et al, 2014
|
Italy
|
Aug’11 and Jan’12
|
retrospective
|
IVF/ICSI
|
303/ 303
|
P > 1.35ng/ml / ROC analysis
|
Day 3
|
CPR ↓
|
Acet et al, 2015
|
Turkey
|
Nov’12 to Feb’14
|
retrospective
|
IVF/ICSI
|
101/ 101
|
P ≥ 1.3ng/ml / literature
|
Day 5
|
similar LBR, CPR & MR
|
Huang P et al, 2015
|
Taiwan
|
Jan’10 to Dec’12
|
retrospective
|
IVF/ICSI
|
599/ 599
|
P > 1.5ng/ml / literature
|
Day 2,3 or 5
|
LBR and CPR ↑
|
Huang Y et al, 2015
|
China
|
Jan’10 to Oct’14
|
retrospective
|
IVF/ICSI
|
12010/ 12010
|
Day 3, P ≥ 1.5ng/ml; Day 5 P ≥ 1.75ng/ml / arbitrary
|
Day 3 or 5
|
CPR ↓
|
Koo et al, 2015
|
Korea
|
May’12 to Jul’13
|
prospective
|
IVF/ICSI
|
200/200
|
P > 0.9ng/ml/ arbitrary
|
Day 3
|
CPR ↓
|
Singh et al, 2015
|
India
|
Jan’12 to Jul’14
|
retrospective
|
IVF/ICSI
|
681/ 681
|
P > 1.0ng/ml / ROC analysis
|
Day 3 or 5
|
CPR ↓
|
Tsai Y et al, 2015
|
Taiwan
|
Jan’00 to Dec’12
|
retrospective
|
IVF/ICSI
|
1508/1508
|
P > 1.94ng/ml/ ROC analysis
|
Day 3 or 5
|
LBR, OPR and CPR ↓
|
Demir et al, 2016
|
Turkey
|
Jan’12 to Jun’14
|
prospective
|
ICSI
|
201/ 201
|
P > 2ng/ml / arbitrary
|
Day 3 or 5
|
CPR NS
|
Healy et al, 2016
|
USA
|
2011 to 2013
|
retrospective
|
IVF/ICSI and FET
|
608/ 608
|
P ≥ 2ng/ml / literature
|
Day 3 or 5
|
LBR ↓ in fresh cycle, LBR similar in FET
|
Ashmita et al, 2018
|
India
|
Jan’16 to Dec’16
|
prospective
|
IVF/ICSI
|
235/ 235
|
P > 1.5ng/ml/ arbitrary
|
Day 3
|
CPR ↓
|
Simon et al, 2019
|
France
|
Sep’12 and Jul’17
|
retrospective
|
IVF/ICSI
|
1399/ 1399
|
P > 1.10ng/ml / arbitrary
|
Day 2–3
|
CPR ↓
|
Wu et al, 2019
|
China
|
Jan’08 to Mar’11
|
retrospective
|
IVF/ICSI
|
2351/ 2351
|
P > 1.0ng/ml in low ovarian response/ arbitrary; P ≥ 2.0ng/ml in intermediate ovarian response/ arbitrary
|
Day 3
|
LBR and CPR↓ in low and intermediate ovarian response
|
Lee C et al, 2020
|
Taiwan
|
Feb’11 to Oct’16
|
retrospective
|
IVF/ICSI
|
337/ 337
|
P > 1.5ng/ml / literature
|
Day 3
|
LBR ↓, CPR and MR NS
|
Yu Y et al, 2020
|
China
|
2013 to 2017
|
secondary analysis of 3 RCTs
|
IVF/ICSI and natural cycle/HRT FET
|
5137/ 5137
|
P > 1.14ng/ml / ROC analysis
|
Day 3 or 5
|
LBR & CPR in FET ↑ than fresh cycle
|
Benmachiche et al, 2021
|
Denmark
|
2014 to 2016
|
secondary data analysis of RCT
|
IVF/ICSI
|
328/ 328
|
P > 1.3ng/ml / arbitrary
|
Day 2–3
|
CPR & LBR NS
|
Mahran et al, 2021
|
Egypt
|
Oct’16 to May’18
|
prospective
|
IVF/ICSI
|
200/ 200
|
P > 1ng/ml / ROC analysis
|
Day 3 or 5
|
CPR NS
|
Mirta et al, 2021
|
India
|
Jan’13 to Jun’16
|
retrospective
|
IVF/ICSI
|
273/ 273
|
P > 1.5ng/ml / literature
|
Day 2–3 or Day5-6
|
CPR and MR NS
|
Yang et al, 2021
|
China
|
Jun’13 and Sep’20
|
retrospective
|
IVF/ICSI
|
1254/ 1254
|
P ≥ 0.9ng/ml/ ROC analysis
|
Day 3 or 5
|
LBR, CPR & MR NS
|
Jiang W et al, 2022
|
China
|
Jan’16 to Oct’16
|
retrospective
|
IVF/ICSI
|
2550/2550
|
P > 1.5ng/ml/ literature
|
Day 5
|
LBR and CPR ↓
|
Kong N et al, 2022
|
China
|
Jan’18 to Dec’20
|
retrospective
|
IVF
|
1951/1951
|
P > 1.5ng/ml/ literature
|
Day 3 or 5
|
LBR, CPR and MR NS
|
Zhao et al, 2022
|
China
|
Jan’20 to Apr’21
|
retrospective
|
IVF/ICSI
|
455/ 455
|
P ≥ 1.0ng/ml/ arbitrary
|
Day 3
|
CPR ↓
|
Fresh COS cycle - Day of transvaginal oocyte retrieval
|
Niu Z et al, 2008
|
China
|
May’05 to May’07
|
NA
|
ICSI
|
289/ 289
|
P > 11.7ng/ml sensitivity-specificity analysis
|
Day 3
|
OPR and CPR NS
|
Nayak et al, 2014
|
USA
|
Feb’10 and May’12
|
prospective
|
IVF/ICSI
|
186/ 186
|
P > 12ng/ml / arbitrary
|
Day 3
|
CPR ↓, MR NS
|
Tulic et al, 2020
|
Serbia
|
Jan’15 to Dec’15
|
prospective
|
IVF/ICSI
|
164/ 164
|
P ≥ 2ng/ml / ROC analysis
|
Day 2–3
|
LBR ↓
|
Fresh COS cycle – luteal phase
|
Kim et al, 2017
|
S. Korea
|
NA
|
prospective
|
IVF-ET
|
148/ 148
|
P > 25.2ng/ml (ROC analysis)
|
Day 3
|
OPR ↑, MR ↓
|
Thomsen et al, 2018
|
Denmark
|
May’14 to Jun’17
|
prospective
|
IVF/ICSI-ET
|
602/ 602
Early luteal phase − 432
Mid-luteal phase − 170
|
Early luteal phase -
P < 18.9ng/ml; P = 18.9–31.4 ng/ml; P = 31.8-125.8 ng/ml; P > 125.8ng/ml
Mid-luteal phase -
P < 47.2ng/ml; P = 47.2–78.6 ng/ml;
P = 78.6-125.8 ng/ml; P > 125.8ng/ml
|
Day 2,3 or 5
|
Optimal chance of pregnancy P = 60–100 nmol/L (early luteal phase) and P = 150- 250nmol/L (mid-luteal phase)
|
Netter et al, 2019
|
France
|
Jul’17 and Jun’18
|
retrospective
|
IVF/ICSI-ET
|
242/ 242
|
P < 36.1ng/ml
P = 36.1–79.2 ng/ml
P > 79.2ng/ml
|
Day 2–3
|
LBR ↑ when P > 252nmol/L
|
FET cycle – day of trigger in modified FET cycle
|
Groenewoud et al, 2017
|
Netherlands
|
Part of “ANTARTICA” trial
|
Secondary analysis of RCT
|
Modified NC FET
|
271/271
|
P > 1.47ng/ml/ ROC analysis
|
Day 3 or 5
|
LBR NS
|
FET cycle – at day before ovulation
|
Lee VC et al, 2014
|
China
|
Jan'06 and Dec’11
|
retrospective
|
NC FET
|
610/ 610
|
P > 1.57ng/ml arbitrary
|
Day 3
|
OPR and CPR NS
|
Wu D et al, 2022
|
China
|
Jan’18 to Apr’20
|
retrospective
|
NC FET
|
1159/ 1159
|
P > 1.0ng/ml
|
Day 3 or 5
|
LBR NS, CPR ↑, MR NS in day 3
LBR, CPR and MR NS in day 5
|
FET cycle – luteal phase
|
Akaeda et al, 2019
|
Japan
|
Sep’10 to Sep’15
|
retrospective
|
HRT FET
|
123/ 123
|
P < 5ng/ml; P = 5-9.9ng/ml
P = 10-14.9 ng/ml; P ≥ 15ng/ml
|
Day 2,3 or 5
|
Optimal chance of pregnancy P = 5-15ng/ml
|
Boynukalin et al, 2019
|
Turkey
|
Mar’18 to Aug’18
|
prospective
|
HRT FET
|
168/ 168
|
P < 13.6ng/ml; P = 13.6–24.3 ng/ml
P = 24.4–53.2 ng/ml; P > 53.2ng/ml
|
Day 5
|
OPR ↑, MR ↓when P > 13.6ng/ml
|
Alsbjerg et al, 2020
|
Denmark
|
Mar’18 and Apr’19
|
prospective
|
HRT FET
|
239/ 239
|
P < 8.8ng/ml; P = 8.8-14.2ng/ml
P > 14.2ng/ml
|
Day 5–6
|
OPR, MR NS
|
Liu & Wu, 2020
|
China
|
Jan’15 to Dec’18
|
retrospective
|
HRT FET
|
856/ 262 (only IM group)
|
P > 13.15ng/ml (arbitrary)
|
Day 2–3
|
LBR NS
|
Polat et al, 2020
|
Turkey
|
Oct’17 to Oct’19
|
retrospective
|
HRT FET
|
475/ 475
|
PV only:
P < 8.75ng/ml; P = 8.76–12.94 ng/ml; P = 12.95–20.42 ng/ml; P > 20.42ng/ml.
PV + IM:
P < 11.75 ng/ml; P = 11.76–19.86 ng/ml; P = 19.87–31.79 ng/ml; P > 31.79 ng/ml
|
Day 5–6
|
No correlation between serum P level and OPR, CPR or MR
|
Shiba et al, 2021
|
Japan
|
Dec’16 to Dec’17
|
secondary analysis of RCT
|
HRT FET
|
235/235
|
P < 7.8ng/ml; P = 7.8-10.8ng/ml
P = 10.8-13.7ng/ml; P > 13.7ng/ml
|
Day 3 or 5
|
LBR, CPR and MR NS
|
Alyasin et al, 2021
|
Iran
|
Feb’19 and Feb’20
|
prospective
|
HRT FET
|
258/ 258
|
P < 19 ng/ml; P = 19–29 ng/ml
P = 29–49 ng/ml; P > 49 ng/ml
|
Day 5
|
LBR and CPR significantly lower in 4th quartile, MR NS
|
Maignien et al, 2022
|
France
|
Jan’19 and Mar’20
|
retrospective
|
HRT FET
|
915/915
|
< 9.8ng/ml (previous study)
|
Day 5
|
LBR ↓, CPR NS and MR ↑
|
Melo et al, 2022
|
UK
|
January 2020
|
prospective
|
NC FET/ HRT FET
|
402/402
|
< 7.8ng/ml (10th centile)
|
Day 5
|
LBR↑, CPR ↑ and MR ↓ when P4 increasing trend
|
Table showing characteristics of included studies with their progesterone threshold/ range and the summary of pregnancy outcomes reported in each studies. |
Abbreviations: COS, controlled ovarian stimulation; CPR, clinical pregnancy rate; ET, embryo transfer; EP, elevated progesterone; FET, frozen embryo transfer; HRT, hormone replacement therapy; ICSI, intracytoplasmic sperm injection; IM, intramuscular; IVF, in vitro fertilization, LBR, live birth rate; MR, miscarriage rate; NC, natural cycle; NEP, non-elevated progesterone; NS, non-significant; OPR, ongoing pregnancy rate; P/P4, progesterone; RCT, randomized-controlled trial; TVOR, transvaginal oocyte retrieval |
Study characteristics
Fresh ovarian stimulation cycle with ET
i. At basal follicular phase
Three studies3,12,13 reported progesterone monitoring in this category. Serum progesterone was measured on day 2 of the menstrual cycle. Two thresholds were identified, P>0.65ng/ml and P>1.5ng/ml. Two studies reported using D3 embryos3,12 and one study reported both D3 and D5 embryos13 (Table 1).
ii. At day of ovulation trigger
Forty-three studies had progesterone monitoring in this category14-56. The trigger used were HCG or agonist trigger. The data for EP and NEP groups in the RCTs were pooled together and analysed as cohort studies. The progesterone threshold ranged from 0.9ng/ml to 2.0ng/ml. Twenty-one studies reported using D3 embryos14-20,22,23,27, 30,32,34,36,40,45-48,50,56, four studies reported using D5 embryos31,33,37,54, eighteen studies reported using both D3 and D5 embryos21,24,25,29,38,39,41-44,49,51-53,55 and three studies did not specify the stages of embryo used26,28,35 (Table 1).
iii. At egg collection
Three studies reported progesterone monitoring in this category57-59. The progesterone threshold level used ranged from 2ng/ml to 12ng/ml. All three studies reported using D3 embryos (Table 1).
iv. At luteal phase
Three studies reported progesterone monitoring in this category60-62. The timing of serum progesterone measurements varied widely from the day of ET (two studies)61,62 and after ET (one study)60. Two studies used vaginal suppositories60,61 and one study used oral progesterone62. One study60 reported a single progesterone threshold level (<25.2ng/ml) and the other two studies61,62 reported progesterone level in ranges (<115nmol/L, 115-252nmol/L and >252nmol/L (61); 10th/50th/90th percentile for early luteal phase and 25th/50th/75th percentile in mid luteal phase). Two studies reported the use of D3 embryos60,62 and one study reported using both D3 and D5 embryos61 (Table 1).
FET cycle
i. At ovulation trigger in modified NC-FET cycle
One study reported EP at ovulation trigger63. The progesterone threshold level was >1.47ng/ml (Table 1).
ii. Before ovulation in NC-FET cycle
Two studies reported EP in this category64,65, and ovulation was determined by either monitoring of LH surge or when the collapse of the dominant follicle was observed during transvaginal scan. The progesterone threshold levels were >1.0ng/ml and >1.57ng/ml. One study reported the use of D3 embryos64 and the other study reported using both D3 and D5 embryos65 (Table 1).
iii. At luteal phase
No studies reported progesterone monitoring in natural FET cycle with or without progesterone supplementation. Nine studies reported progesterone monitoring in this category66-74. All the studies apart from one74 were medicated FET cycles. Melo et al.74 included women from natural, and medicated FET cycles. The timing of serum progesterone measurements varied widely from the day of ET (seven studies)66,67,70-74 and after ET (two studies)68,69. Three studies used vaginal suppositories70,71,73, two studies used intramuscular injections67,69 and four studies used a combination of progesterone support68,70,72,74. Three studies69,73,74 reported single progesterone threshold level (<7.8ng/ml, <9.8ng/ml and <13.15ng/ml) and the remaining six studies66-68,70-72 reported progesterone value according to quartiles or percentiles. One study reported the use of D3 embryos69, six studies reported using D5 embryos67,68,70,72-74 and two studies reported using both D3 and D5 embryos66,71 (Table 1).
Outcomes: Fresh ovarian stimulation cycle with ET
A. At basal follicular phase
There was no difference in LBR in the EP compared to the NEP at threshold level > 1.5ng/ml, (OR 0.76, 95% CI 0.39–1.49, I2 = 0%, 2 studies, N = 309) (Fig. 3).
Three studies3,12,13 reported CPR over two different threshold levels (> 0.65ng/ml and > 1.5ng/ml). There was no difference in CPR in the EP compared to the NEP (P > 0.65ng/ml, OR 1.41, 95% CI 0.93–2.13, 1 study, N = 464; P > 1.5ng/ml, OR 0.81, 95% CI 0.38–1.71, I2 = 24%, 2 studies, N = 309) (Fig. 3).
We were unable to meta-analyse OPR and MR in a meaningful way as they are single studies. Data from single studies were summarised in Supplementary Table S3.
B. At day of ovulation trigger
Seventeen studies20,25,27,29,31,32,37,38,42,44,47–50,53−55 reported LBR. The threshold levels ranged between > 0.9ng/ml to > 2.0ng/ml. EP on the day of trigger was associated with decreased LBR across 3 threshold levels (P > 1.0ng/ml: OR 0.40, 95% CI 0.23–0.69, I2 = 48%, 2 studies, N = 2805; P > 1.1ng/ml: OR 0.70, 95% CI 0.53–0.93, I2 = 42%, 2 studies, N = 3186; P > 2.0ng/ml: OR 0.37, 95% CI 0.24–0.58, I2 = 0%, 2 studies, N = 2257) and no difference in LBR at 2 thresholds (P > 1.3ng/ml: OR 0.89, 95% CI 0.56–1.41, I2 = 0%, 2 studies, N = 429; P > 1.5ng/ml: OR 0.83, 95% CI 0.66–1.05, I2 = 52%, 6 studies, N = 8170) (Fig. 4a).
Five studies16,20,33,34,42 reported OPR. The threshold levels ranged between > 1.1ng/ml to > 1.9ng/ml. Elevated progesterone level on the day of trigger was associated with decreased OPR in P > 1.5ng/ml compared to those with NEP (OR 0.61, 95% CI 0.44–0.84, I2 = 0%, 2 studies, N = 2070) (Fig. 4b).
Forty studies14,15,17–26,28−33,35–43,45−56 reported CPR. The threshold levels ranged between > 0.9ng/ml to > 2.0ng/ml. EP on the day of trigger was associated with decreased CPR across 4 threshold levels: P > 1.0ng/ml; P > 1.1ng/ml; P > 1.5ng/ml and P > 2.0ng/ml (Fig. 4c) and no difference in CPR over 4 thresholds: P > 0.9ng/ml; P > 1.2ng/ml; P > 1.3ng/ml; P > 1.7ng/ml.
Nine studies15,25,32,37,48,49,52,53,55 reported MR. The threshold levels ranged between > 0.9ng/ml to > 1.7ng/ml. There was no difference in MR in EP compared to NEP across all threshold levels: P > 0.9ng/ml and P > 1.5ng/ml (Fig. 4d). Data from single studies were summarised in Supplementary Table S3.
Subgroup analysis on Day 3 embryo at ovulation trigger
When we analysed studies which reported on only D3 embryos, there was a decreased LBR at threshold level > 1.0ng/ml (OR 0.46, 95% CI 0.38–0.55, I2 = 48%, 2 studies, N = 2805) and > 1.5ng/ml (OR 0.68, 95% CI 0.47–0.98, I2 = 19%, 2 studies, N = 867) (Fig. 4e) and a decreased CPR at all threshold levels (P > 0.9ng/ml; OR 0.74, 95% CI 0.52–1.04, I2 = 58%, 2 studies, N = 577; P > 1.0ng/ml; OR 0.49, 95% CI 0.42–0.57, I2 = 3%, 3 studies, N = 3323; P > 1.1ng/ml; OR 0.67, 95% CI 0.53–0.83, I2 = 0%, 2 studies, N = 2444; P > 1.2ng/ml; OR 0.59, 95% CI 0.44–0.79, I2 = 49%, 6 studies, N = 844; P > 1.3ng/ml; OR 0.62, 95% CI 0.40–0.96, I2 = 15%, 2 studies, N = 631; P > 1.5ng/ml; OR 0.18, 95% CI 0.16–0.20, I2 = 93%, 6 studies, N = 13870; P > 2.0ng/ml; OR 0.43, 95% CI 0.31–0.59, I2 = 0%, 3 studies, N = 1949) (Fig. 4f).
Subgroup analysis on Day 5 embryo at ovulation trigger
When we analysed studies which reported on only D5 embryos, there was no difference in LBR (P > 1.5ng/ml; OR 0.96, 95% CI 0.81–1.14, I2 = 55%, 3 studies, N = 5174) and CPR (P > 1.5ng/ml; OR 0.90, 95% CI 0.78–1.04, I2 = 50%, 6 studies, N = 5705) between EP and NEP groups (Fig. 4g).
C. At egg collection
One study59 reported LBR at threshold level > 2ng/ml; one study57 reported OPR at threshold level > 11.7ng/ml; three studies57–59 reported CPR at different threshold levels > 2ng/ml, > 11.7ng/ml and > 12ng/ml; two studies58,59 reported MR at threshold levels > 2ng/ml and > 12ng/ml. Data from single studies were summarised in Supplementary Table S3.
D. Luteal phase
Two studies61,62 reported LBR at threshold value < 18.9ng/ml, < 31.4ng/ml, < 125.8ng/ml, < 47.2ng/, < 78.6ng/ml, < 125.8ng/ml, < 36.1ng/ml and < 79.2ng/ml. No studies reported on OPR, three studies60–62 reported CPR and MR. The threshold value used were < 18.9ng/ml, < 31.4ng/ml, < 125.8ng/ml, < 25.2ng/ml, < 47.2ng/ml, < 78.6ng/ml, < 125.8ng/ml, < 36.1ng/ml and < 79.2ng/ml (Fig. 5). Data from various threshold values were summarised in Supplementary Table S4.
Outcomes: FET cycle
A. Before ovulation in a natural FET cycle
One study65 reported LBR at threshold level > 1.0ng/ml; one study64 reported OPR at threshold level > 1.57ng/ml; two studies64,65 reported CPR and MR at threshold levels > 1.0ng/ml and > 1.57ng/ml. (Supplementary Table S3)
B. Luteal phase
In medicated FET cycles, four studies69,71–73 reported on LBR, three studies67,68,70 reported on OPR, seven studies66–69,71−73 reported on CPR and six studies67,68,70–73 reported on MR at various threshold values (Fig. 6). There were no similarities between the threshold values used and wide variation of the timing of progesterone measurement. Data from various threshold values were summarised in Supplementary Table S5. In both natural cycle and medicated FET cycle, Melo et al.74 reported LBR, CPR and MR as summarised in Supplementary Table S6.