Totally, 6890 cycles were assessed for eligibility during February 2019 and December 2020 (Fig. 1). We dropped those with age over 40 (n = 180) and missing data on follow-up (n = 4319), as well as those with chromosomal or genetic abnormalities (n = 30), endocrine disease or abnormal liver and kidney function (n = 258), immune disease (n = 16), PCOS (n = 118) and unexplained infertility (n = 6). Finally, 1963 cycles were remained for analyses and divided into two groups, including endometriosis group (n = 412) and control group (n = 1551).
Baseline Characteristics In Endometriosis Patients And Controls
Baseline characteristics of two study groups were summarized in Table 1. Significant differences were found for types of infertility (P < 0.001), history of miscarriage [116(28.2%) vs 928(59.8%), P < 0.001], number of previous pregnancies [0.0(0.0–1.0) vs 0.0(0.0–2.0), P < 0.001] and serum CA125 levels [24.00(15.10–41.20) vs 15.10(11.00-21.60) U/mL, P < 0.001]. Nevertheless, there were no differences in age, BMI, smoking, drinking, duration of infertility, type of ART and history of preterm delivery between two groups (P > 0.05, respectively).
Table 1
Baseline characteristics of two study groups
Characteristics | Endometriosis (n = 412) | Controls (n = 1551) | P value |
Maternal age (years) | 32 (30–35) | 33 (30–36) | 0.129 |
BMI (kg/m2) | 21.0 (19.5–22.9) | 21.3 (19.6–23.3) | 0.052 |
Smoking | | | 0.060 |
No [n (%)] | 409 (99.3) | 1518 (97.9) | |
Yes [n (%)] | 3 (0.7) | 33 (2.1) | |
Drinking | | | NA |
No [n (%)] | 412 (100.0) | 1550 (99.9) | |
Yes [n (%)] | 0 (0.0) | 1 (0.1) | |
Duration of infertility (years) | 3.0 (1.5-4.0) | 2.8 (1.0–4.0) | 0.248 |
Type of infertility | | | < 0.001 |
Primary infertility [n (%)] | 258 (62.6) | 537 (34.6) | |
Secondary infertility [n (%)] | 120 (29.1) | 838 (54.0) | |
Less than one year [n (%)] | 34 (8.3) | 176 (11.3) | |
Type of ART | | | 0.437 |
IVF [n (%)] | 337 (79.5) | 1195 (77.6) | |
Half ICSI [n (%)] | 8 (1.9) | 21 (1.4) | |
ICSI [n (%)] | 79 (18.6) | 323 (21.0) | |
History of preterm delivery [n (%)] | 3 (0.7) | 21 (1.4) | 0.438 |
History of miscarriage [n (%)] | 116 (28.2) | 928 (59.8) | < 0.001 |
Number of previous pregnancies (n) | 0.0 (0.0–1.0) | 0.0 (1.0–2.0) | < 0.001 |
CA125 (U/mL) | 24.00 (15.10–41.20) | 15.10 (11.00-21.60) | < 0.001 |
Values are expressed as mean ± standard deviation, median (interquartile range) or number (%). |
BMI body mass index, ART assisted reproductive technology, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection |
Metabolic Indexes In Endometriosis Patients And Controls
As shown in Table 2, there were no statistically differences in basal serum levels of estradiol (E2) and progesterone (P), Glu, triglycerides (TG), total protein (TP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine, urea nitrogen, uric acid and homocysteine (HCY) between the two groups (P > 0.05, respectively). While we found significantly lower serum basal testosterone (T) [0.50(0.00-0.80) vs 0.60(0.00-0.90) nmol/L, P = 0.005], higher serum INS [6.90(5.10–9.50) vs 6.50(4.80–8.90) µU/mL, P = 0.005], TC [4.35(3.92–4.80) vs 4.27(3.81–4.77) mmol/L, P = 0.036], HDL-C [1.36(1.19–1.57) vs 1.32(1.14–1.52) mmol/L, P = 0.005] and LDL-C [2.63(2.17–3.01) vs 2.54(2.12–2.94) mmol/L, P = 0.043] in endometriosis group compared with controls.
Table 2
Metabolic indexes, ovarian function and ART outcomes of two study groups
Characteristics | Endometriosis (n = 412) | Controls (n = 1551) | P value |
Basal E2 (pmol/L) | 114.65 (67.97-156.58) | 110.70 (67.70-153.60) | 0.405 |
Basal P (nmol/L) | 1.14 (0.75–1.55) | 1.12 (0.76–1.50) | 0.644 |
Basal T (nmol/L) | 0.50 (0.00-0.80) | 0.60 (0.00-0.90) | 0.005 |
Glu (mmol/L) | 5.01 (4.79–5.27) | 5.04 (4.80–5.31) | 0.134 |
INS (µU/mL) | 6.90 (5.10–9.50) | 6.50 (4.80–8.90) | 0.005 |
TG (mmol/L) | 0.92 (0.71–1.19) | 0.92 (0.69–1.27) | 0.435 |
TC (mmol/L) | 4.35 (3.92–4.80) | 4.27 (3.81–4.77) | 0.036 |
HDL-C (mmol/L) | 1.36 (1.19–1.57) | 1.32 (1.14–1.52) | 0.005 |
LDL-C (mmol/L) | 2.63 (2.17–3.01) | 2.54 (2.12–2.94) | 0.043 |
TP (g/L) | 71.97 ± 4.49 | 72.20 ± 4.77 | 0.318 |
ALT (U/L) | 13.00 (10.00–17.00) | 13.00 (10.00–18.00) | 0.178 |
AST (U/L) | 17.00 (15.00–20.00) | 18.00 (15.00–20.00) | 0.068 |
Creatinine (µmoI/L) | 56.00 (50.00-62.60) | 55.10 (49.00–62.00) | 0.202 |
Urea nitrogen (mmol/L) | 3.78 (3.21–4.48) | 3.76 (3.18–4.51) | 0.621 |
Uric acid (µmoI/L) | 266.00 (234.25–303.00) | 270.00 (231.00-312.00) | 0.277 |
HCY (nmol/L) | 9.70 (8.50–10.80) | 9.80 (8.50–11.10) | 0.281 |
AMH (ng/mL) | 2.02 (1.06–3.49) | 2.53 (1.48–4.07) | < 0.001 |
AFC (n) | 8.00 (5.00–11.00) | 10.00 (7.00–12.00) | < 0.001 |
Basal FSH (IU/L) | 6.74 (5.26–8.35) | 6.32 (4.81–7.87) | < 0.001 |
Basal LH (IU/L) | 4.19 ± 2.40 | 4.60 ± 2.66 | 0.031 |
Gn dosage (IU) | 2025.00 (1575.00-2700.00) | 2025.00 (1575.00-2475.00) | 0.664 |
Gn days (day) | 9.00 (8.00–12.00) | 9.00 (8.00–11.00) | 0.058 |
Number of retrieved oocytes (n) | 7.00 (4.00–11.00) | 9.00 (5.00–14.00) | < 0.001 |
Fertilization rate [%] | 64.3 (2158/3357) | 64.0 (9714/15172) | 0.778 |
Cleavage rate [%] | 23.7 (512/2158) | 24.1 (2338/9714) | 0.736 |
Number of transferable embryos (n) | 0.76 ± 0.93 | 0.77 ± 0.93 | 0.776 |
Number of high-quality embryos (n) | 0.57 ± 0.82 | 0.60 ± 0.83 | 0.402 |
High-quality embryos rate [%] | 12.1 (232/1912) | 11.1 (936/8435) | 0.196 |
Number of embryos transferred (n) | 1.81 ± 0.39 | 1.78 ± 0.41 | 0.389 |
Implantation rate [%] | 38.7 (122/315) | 36.0 (429/1191) | 0.375 |
Ectopic pregnancy rate [%] | 2.3 (4/174) | 1.5 (10/669) | 0.684 |
Clinical pregnancy rate [%] | 51.7 (90/174) | 49.5 (331/669) | 0.597 |
Miscarriage rate [%] | 6.9 (12/174) | 5.7 (38/669) | 0.606 |
Delivery rate [%] | 45.4 (79/174) | 43.8 (293/669) | 0.606 |
Live birth rate [%] | 54.0 (94/174) | 54.0 (361/669) | 0.988 |
Values are expressed as mean ± standard deviation, median (interquartile range) or number (%). |
E2 estradiol, P progesterone, T testosterone, Glu glucose, INS insulin, TG triglycerides, TC total cholesterol, HDL-C high density lipoprotein cholesterol, LDL-C low density lipoprotein cholesterol, TP total protein, ALT alanine aminotransferase, AST aspartate aminotransferase, HCY homocysteine, AMH Anti-Müllerian hormone, AFC antral follicle counting, FSH follicle stimulating hormone, LH luteinizing hormone, Gn gonadotropin |
Prediction Of Endometriosis By Metabolic Indexes
After adjusting for potential confounders, number of previous pregnancies [adjusted odds ratio (aOR) 0.51, 95% confidence interval (CI) 0.43–0.62; P < 0.001], serum CA125 levels (aOR 1.02, 95% CI 1.01–1.03; P < 0.001), serum Glu (aOR 0.74, 95% CI 0.56–0.97; P = 0.027) and serum INS (aOR 1.03, 95% CI 1.01–1.04; P = 0.002) were found to be significantly associated with endometriosis (Table 3). Besides, compared with subjects with primary infertility, those with secondary infertility suffered from decreased incidence of endometriosis (aOR 0.70, 95% CI 0.50–0.97; P = 0.030). The aORs and their 95% confidence intervals (CI) were extracted and a forest plot graphic was built [13].
Table 3
Odd ratios for endometriosis in these patients
| Crude OR (95% CI) | P value | Adjusted OR (95% CI) | P value |
Maternal age (years) | 0.99 (0.96–1.01) | 0.298 | Removed | |
Smoking | | | | |
No [n (%)] | Reference | | Reference | |
Yes [n (%)] | 2.96 (0.90–9.71) | 0.073 | 1.93 (0.48–7.74) | 0.352 |
Number of previous pregnancies (n) | 0.46 (0.40–0.53) | < 0.001 | 0.51 (0.43–0.62) | < 0.001 |
Type of infertility | | | | |
Primary infertility [n (%)] | Reference | | Reference | |
Secondary infertility [n (%)] | 0.40 (0.27–0.60) | < 0.001 | 0.70 (0.50–0.97) | 0.030 |
Less than one year [n (%)] | 0.30 (0.23–0.38) | < 0.001 | 1.07 (0.66–1.72) | 0.793 |
CA125 (U/mL) | 1.02 (1.02–1.03) | < 0.001 | 1.02 (1.01–1.03) | < 0.001 |
AST (U/L) | 1.00 (0.98–1.01) | 0.589 | Removed | |
BMI (kg/m2) | 0.96 (0.92-1.00) | 0.058 | 0.99 (0.94–1.04) | 0.676 |
Glu (mmol/L) | 0.74 (0.58–0.94) | 0.014 | 0.74 (0.56–0.97) | 0.027 |
INS (µU/mL) | 1.02 (1.00-1.03) | 0.010 | 1.03 (1.01–1.04) | 0.002 |
TG (mmol/L) | 0.88 (0.72–1.08) | 0.213 | Removed | |
TC (mmol/L) | 1.16 (1.01–1.33) | 0.031 | 0.95 (0.72–1.25) | 0.706 |
HDL-C (mmol/L) | 1.63 (1.14–2.35) | 0.008 | 1.54 (0.94–2.54) | 0.088 |
LDL-C (mmol/L) | 1.16 (0.99–1.35) | 0.065 | 1.26 (0.96–1.72) | 0.137 |
Basal E2 (pmol/L) | 1.00 (1.00–1.00) | 0.894 | Removed | |
Basal P (nmol/L) | 1.03 (0.97–1.09) | 0.356 | Removed | |
Basal T (nmol/L) | 0.98 (0.90–1.06) | 0.595 | Removed | |
OR odd ratio, AST aspartate aminotransferase, BMI body mass index, Glu glucose, INS insulin, TG triglycerides, TC total cholesterol, HDL-C high density lipoprotein cholesterol, LDL-C low density lipoprotein cholesterol, E2 estradiol, P progesterone, T testosterone |
Furthermore, we performed AUC and receiver operator control curves (ROC analysis) to assess whether the statistically different factors found in Table 1 could be used as indicators to predict the occurrence of endometriosis [13](Fig. 2). Results showed Glu and INS had a sensitivity of 39.9% and 41.3%, specificity of 66.5% and 67.5%, AUC of 0.52 and 0.55, respectively. When combining previous pregnancies, serum CA125, serum Glu and INS, the mode had a sensitivity of 73.9%, specificity of 67.8% and AUC of 0.77.
The Effect Of Altered Metabolic Indexes On The Number Of Retrieved Oocytes In Endometriosis
As shown in Table 2, there were statistically significant differences in AMH [2.02(1.06–3.49) vs 2.53(1.48–4.07) ng/mL, P < 0.001], AFC [8.00(5.00–11.00) vs 10.00(7.00–12.00), P < 0.001], FSH [6.74(5.26–8.35) vs 6.32(4.81–7.87) IU/L, P < 0.001] and LH [(4.19 ± 2.40 vs 4.60 ± 2.66 IU/L, P = 0.031)] between endometriosis group and control group. Furthermore, the number of retrieved oocytes in endometriosis patients was significantly lower than that in control group [7.00(4.00–11.00) vs 9.00(5.00–14.00), P < 0.001], without differences in gonadotropin (Gn) dosage and Gn days.
Given that the number of retrieved oocytes was closely related to ART outcomes of endometriosis patients, we explored whether alterations in serum Glu and INS played a role in the numbers of retrieved oocytes in endometriosis, we conducted multi-factor linear regression analysis. As shown in Table 4, the number of retrieved oocytes was positively correlated with INS [0.07(0.00-0.14), P = 0.048] in endometriosis group. In control group, the number of retrieved oocytes was negatively correlated with Glu [-0.80(-1.48 - -0.12), P = 0.021].
Table 4
Multivariate logistic regression predictors of the number of retrieved oocytes
| Endometriosis (n = 412) | | Controls (n = 1551) |
β(95% CI) | P value | | β(95% CI) | P value |
Maternal age (years) | -0.52 (-0.68 - -0.31) | < 0.001 | | -0.57 (-0.64 - -0.49) | < 0.001 |
CA125 (U/mL) | 0.00 (-0.01–0.00) | 0.201 | | -0.01 (-0.02–0.01) | 0.314 |
Number of previous pregnancies (n) | 0.95 (0.26–1.64) | 0.007 | | 0.25 (0.02–0.47) | 0.032 |
BMI (kg/m2) | 0.00 (-0.23–0.154) | 0.984 | | 0.01 (-0.11–0.14) | 0.833 |
Basal E2 (pmol/L) | 0.00 (0.00–0.00) | 0.212 | | -0.00 (0.00–0.00) | 0.120 |
Basal P (nmol/L) | -0.18 (-0.43–0.07) | 0.161 | | -0.16 (-0.20–0.17) | 0.864 |
Basal T (nmol/L) | 0.99 (0.37–1.60) | 0.002 | | 0.16 (-0.04–0.36) | 0.122 |
TG (mmol/L) | -0.29 (-1.24–0.67) | 0.552 | | -0.06 (-0.63–0.51) | 0.835 |
TC (mmol/L) | 1.70 (0.38–3.03) | 0.012 | | 0.88 (0.10–1.67) | 0.028 |
HDL-C (mmol/L) | -0.83 (-3.18–1.52) | 0.490 | | -0.78 (-2.14–0.58) | 0.260 |
LDL-C (mmol/L) | -1.75 (-3.19 - -0.31) | 0.018 | | -0.59 (-1.43 - -0.25) | 0.168 |
Glu (mmol/L) | 0.19 (-0.90–1.28) | 0.734 | | -0.80 (-1.48 - -0.12) | 0.021 |
INS (µU/mL) | 0.07 (0.00–0.14) | 0.048 | | -0.03 (-0.08–0.02) | 0.264 |
BMI body mass index, E2 estradiol, P progesterone, T testosterone, TG triglycerides, TC total cholesterol, HDL-C high density lipoprotein cholesterol, LDL-C low density lipoprotein cholesterol, Glu glucose, INS insulin |
Comparisons Of Art Outcomes In Endometriosis Patients And Controls
We further compared ART outcomes in two groups (Table 2). There were no statistically significant differences in fertilization rate, cleavage rate, number of transferable embryos, number of high-quality embryos, high-quality embryos rate, number of embryos transferred, implantation rate, clinical pregnancy rate, miscarriage rate, ectopic pregnancy rate, delivery rate and live birth rate between the study groups (P > 0.05, respectively).
Neonatal Outcomes And Complications In Two Groups
Neonate outcomes in two groups were illustrated in Supplementary Table 2[13]. There were no statistically significant differences in gestational week, manner of childbirth, rate of twins as well as gender and weight of both single and twin babies. As shown in Supplementary Table 3[13], no significant difference was found in incidences of pregnancy complications (GDM, gestational hypertension, intra-hepatic cholestasis of pregnancy, placenta previa, placental abruption, premature rupture of membranes, umbilical cord around neck, postpartum hemorrhage, infection and hypothyroidism) or neonatal complications (neonatal respiratory distress syndrome, hypoglycemia, jaundice and infection) between endometriosis group and control group.
We further explored effects of serum Glu and INS on incidence of GDM in both groups (Table 5), and found serum Glu [(aOR 12.95, 95% CI 1.69–99.42; P = 0.014) were significantly associated with incidence of GDM in both groups (aOR 4.15, 95% CI 1.50–11.53; P = 0.006).
Table 5
Effects of Glu and INS on GDM of two study groups
| N (-/+) | GDM (%) | Crude OR (95% CI) | P value | Adjusted OR (95% CI) | P value |
Glu (mmol/L) | | | | | | |
Endometriosis | 66/13 | 16.5 | 7.90 (1.23–50.80) | 0.027 | 12.95 (1.69–99.42) | 0.014 |
Controls | 264/29 | 9.9 | 3.27 (1.28–8.39) | 0.013 | 4.15 (1.50-11.53) | 0.006 |
INS (µU/ml) | | | | | | |
Endometriosis | 66/13 | 16.5 | 1.00 (0.93–1.07) | 0.973 | 1.01 (0.94–1.08) | 0.822 |
Controls | 264/29 | 9.9 | 0.98 (0.89–1.08) | 0.687 | 0.99 (0.90–1.09) | 0.819 |
GDM gestational diabetes mellitus, OR odd ratio, Glu glucose, INS insulin |