The mean age of the 1870 lactating women was 29.86 years with a pre-pregnancy BMI of 22.36 ± 3.33 kg/m2, a pre-partum BMI of 28.10 ± 3.79 kg/m2, 42 days postpartum BMI of 24.81 ± 3.58 kg/m2 and a current lactation BMI of 24.27 ± 3.89 kg/m2. The changes in BMI of the MetS and non-MetS groups during pregnancy and lactation are shown in Fig. 2(A), and there was a statistical difference in BMI at all stages between the two groups (P < 0.0001) (Table 1). The mean age of lactating children was 7.92 ± 5.01 months. There were 660 (35.29%) participants with 1 parity and 1124 (60.11%) with 2 parities. Light manual workers accounted for 1822 (81.39%), medium manual laborers 340 (18.18%). There were 251 subjects with MetS, while 983 subjects with abdominal obesity, 292 with high BP, 243 with high TG, 555 with low HDL, and 188 with elevated FG.
Table 1
Demographic and clinical characteristics of MetS and Non-MetS groups
| MetS (n = 251) | Non-MetS (n = 1619) | t/χ2 | P-value |
Age (years) | 31.75(5.72) | 29.57(5.14) | 5.70 | < 0.0001 |
BMI at prepregnancy | 24.86(3.77) | 21.97(3.08) | 11.54 | < 0.0001 |
BMI before labor | 30.58(4.51) | 27.71(3.51) | 9.61 | < 0.0001 |
BMI at 42 days postpartum | 27.33(3.78) | 24.42(3.39) | 11.54 | < 0.0001 |
BMI at current | 28.07(3.85) | 23.68(3.54) | 18.07 | < 0.0001 |
Parity numbers | 1.81(0.65) | 1.68(0.56) | 2.90 | 0.0040 |
Age of children (mouths) | 8.36(5.24) | 7.85(4.99) | 1.50 | 0.1332 |
Drinker (%) | 5 (1.99) | 33 (2.04) | 0.0023 | 0.9614 |
Smoker (%) | 2(0.80) | 10(0.62) | 0.1094 | 0.7408 |
Physical activity level (%) | | | | |
Light | 206(10.97) | 1316(70.07) | 1.2671 | 0.5307 |
Moderate | 45(2.40) | 295(15.71) | | |
Vigorous | 0 | 8(0.43) | | |
History of gestational diabetes (%) | 19(7.57) | 63(3.89) | 7.0130 | 0.0081 |
History of gestational hypertension (%) | 18(7.17) | 37(2.29) | 18.1728 | < 0.0001 |
TC (mmol/L) | 4.65(0.97) | 4.28(0.84) | 5.63 | < 0.0001 |
TG (mmol/L) | 2.07(1.14) | 0.91(0.48) | 15.80 | < 0.0001 |
HDL (mmol/L) | 1.15(0.23) | 1.56(0.34) | -24.25 | < 0.0001 |
LDL (mmol/L) | 2.76(0.89) | 2.45(0.75) | 5.19 | < 0.0001 |
Serum zinc (µg/dL) | 3.93(14.55) | 2.91(12.12) | 1.05 | 0.2939 |
Blood glucose (mmol/L) | 5.38(1.29) | 4.81(0.59) | 6.85 | < 0.0001 |
Ferritin (ng/mL) | 63.75(49.55) | 48.42(38.99) | 4.68 | < 0.0001 |
Transferrin receptor (mg/L) | 3.43(1.34) | 3.3(1.5) | 1.42 | 0.1551 |
Hemoglobin (g/L) | 135.39(11.83) | 131.47(13.39) | 4.80 | < 0.0001 |
Serum Vitamin B12 (pg/mL) | 454.72(200.44) | 462.3(235.21) | -0.54 | 0.5867 |
Serum folic acid (ng/mL) | 7.75(23.76) | 24.92(106.02) | -5.66 | < 0.0001 |
Hs-CRP (mg/L) | 2.49(3.03) | 1.65(4.36) | 3.83 | < 0.0001 |
Albumin (g/L) | 46.97(3.16) | 47.64(3.41) | -2.92 | 0.0036 |
Total protein (g/L) | 75.64(5.26) | 76.06(5.14) | -1.22 | 0.2243 |
Serum Vitamin A (µmol/L)) | 0.52(0.16) | 0.48(0.27) | 3.80 | 0.0002 |
Serum Vitamin D (ng/mL) | 17.28(5.52) | 17.25(6.28) | 0.08 | 0.9336 |
Systolic pressure (mmHg) | 127.07(14.14) | 115.51(10.72) | 12.41 | < 0.0001 |
Diastolic pressure (mmHg) | 78.79(11) | 71.05(8.72) | 10.64 | < 0.0001 |
Note: Data were presented as mean (SD) or n (%). |
Table 1 listed the demographic and clinical characteristics of the MetS and non-MetS groups. Compared to the non-MetS group, the MetS group had a higher age, BMI by stage, parity, history of gestational diabetes (%), history of gestational hypertension (%), TC, TG, LDL, blood glucose, serum ferritin, hemoglobin, Hs-CRP, vitamin A, systolic pressure and diastolic pressure (P < 0.05). Also, HDL, serum folic acid and albumin were lower in the MetS group than in the non-MetS group (P < 0.05).
Table 2 presented the logistic regression results for the association between BMI at different stages and the risk of current MetS lactation. The model found a significant positive association between increasing BMI at each stage and the ORs of both MetS and its components after several adjustments for confounders (P < 0.05). The same positive correlation was also found for the difference value between BMI at current and BMI at 42 days postpartum as independent variables (P < 0.05). The sensitivity analysis of BMI at each stage on the current risk of MetS prevalence in lactating women was concentrated in Table 3. The results showed that the positive association between BMI at all stages and MetS risk in lactation was almost consistent in all stratified analyses. And there was an interaction of maternal age on the relationship between BMI at 42 days postpartum and MetS risk in lactating women (P = 0.0230).
Table 2
Association of BMI stage and the prevalence of MetS and its components
| | Abdominal obesity | High BP | High TG | Low HDL | Elevated FG | MetS |
BMI at prepregnancy | | | | | | | |
| Crude OR (95% CI) | 1.459 (1.399, 1.520) | 1.171 (1.130, 1.214) | 1.149 (1.106, 1.193) | 1.133 (1.099, 1.167) | 1.141 (1.094, 1.189) | 1.270 (1.221, 1.321) |
| Adjusted OR* (95% CI) | 1.449 (1.389,1.511) | 1.161 (1.119, 1.205) | 1.133 (1.090, 1.178) | 1.139 (1.104, 1.175) | 1.125 (1.078, 1.175) | 1.256 (1.206, 1.308) |
| Adjusted OR† (95% CI) | 1.452 (1.391, 1.515) | 1.141 (1.098, 1.186) | 1.132 (1.087, 1.178) | 1.138 (1.102, 1.174) | 1.109 (1.061, 1.160) | 1.243 (1.192, 1.296) |
BMI before labor | | | | | | | |
| Crude OR (95% CI) | 1.324 (1.280, 1.369) | 1.118 (1.081, 1.155) | 1.108 (1.071, 1.148) | 1.101 (1.071, 1.131) | 1.090 (1.050, 1.131) | 1.215 (1.171, 1.262) |
| Adjusted OR* (95% CI) | 1.318 (1.274, 1.364) | 1.111 (1.074, 1.149) | 1.099 (1.061, 1.139) | 1.102 (1.072, 1.133) | 1.082 (1.042, 1.124) | 1.208 (1.162, 1.255) |
| Adjusted OR† (95% CI) | 1.321 (1.276, 1.368) | 1.092 (1.055, 1.131) | 1.098 (1.059, 1.139) | 1.101 (1.071, 1.133) | 1.071 (1.030, 1.114) | 1.201 (1.155, 1.249) |
BMI at 42 days postpartum | | | | | | | |
| Crude OR (95% CI) | 1.435 (1.380, 1.492) | 1.129 (1.091, 1.168) | 1.133 (1.092, 1.174) | 1.111 (1.080, 1.144) | 1.110 (1.068, 1.155) | 1.245 (1.197, 1.294) |
| Adjusted OR* (95% CI) | 1.428 (1.373, 1.486) | 1.119 (1.080, 1.158) | 1.120 (1.079, 1.162) | 1.115 (1.083, 1.147) | 1.101 (1.057, 1.146) | 1.233 (1.185, 1.283) |
| Adjusted OR† (95% CI) | 1.432 (1.375, 1.491) | 1.100 (1.061, 1.140) | 1.117 (1.076, 1.160) | 1.112 (1.080, 1.146) | 1.088 (1.043, 1.134) | 1.222 (1.173, 1.272) |
BMI at current | | | | | | | |
| Crude OR (95% CI) | 1.818 (1.720, 1.921) | 1.178 (1.141, 1.216) | 1.179 (1.140, 1.219) | 1.162 (1.131, 1.193) | 1.150 (1.109, 1.192) | 1.331 (1.281, 1.383) |
| Adjusted OR* (95% CI) | 1.810 (1.713, 1.913) | 1.173 (1.136, 1.211) | 1.172 (1.133, 1.213) | 1.167 (1.135, 1.199) | 1.143 (1.101, 1.186) | 1.330 (1.279, 1.383) |
| Adjusted OR† (95% CI) | 1.813 (1.715, 1.916) | 1.165 (1.127, 1.204) | 1.172 (1.132, 1.213) | 1.166 (1.134, 1.198) | 1.133 (1.091, 1.176) | 1.323 (1.271, 1.376) |
△BMI before labor - BMI at prepregnancy | | | | | | | |
| Crude OR (95% CI) | 0.993 (0.960, 1.028) | 0.969 (0.921, 1.021) | 0.990 (0.939, 1.044) | 0.993 (0.956, 1.032) | 0.966 (0.908, 1.029) | 0.997 (0.947, 1.049) |
| Adjusted OR* (95% CI) | 1.002 (0.967, 1.038) | 0.979 (0.931, 1.030) | 1.001 (0.951, 1.053) | 0.994 (0.956, 1.032) | 0.978 (0.919, 1.040) | 1.010 (0.962, 1.060) |
| Adjusted OR† (95% CI) | 1.006 (0.971, 1.043) | 0.972 (0.921, 1.027) | 1.002 (0.952, 1.054) | 0.995 (0.958, 1.034) | 0.984 (0.927, 1.044) | 1.012 (0.965, 1.063) |
△BMI at 42 days postpartum - BMI before labor | | | | | | | |
| Crude OR (95% CI) | 1.097 (1.042, 1.154) | 0.988 (0.923, 1.058) | 1.035 (0.961, 1.116) | 1.000 (0.947, 1.056) | 1.025 (0.943, 1.114) | 1.017 (0.945, 1.094) |
| Adjusted OR* (95% CI) | 1.083 (1.029, 1.141) | 0.974 (0.910, 1.043) | 1.022 (0.947, 1.102) | 0.999 (0.946, 1.055) | 1.013 (0.930, 1.103) | 1.000 (0.928, 1.077) |
| Adjusted OR† (95% CI) | 1.082 (1.028, 1.140) | 0.980 (0.912, 1.052) | 1.017 (0.942, 1.098) | 0.996 (0.943, 1.052) | 1.010 (0.928, 1.099) | 0.995 (0.923, 1.072) |
△BMI at current - BMI at 42 days postpartum | | | | | | | |
| Crude OR (95% CI) | 1.273 (1.219, 1.330) | 1.133 (1.083, 1.185) | 1.129 (1.076, 1.184) | 1.127 (1.084, 1.171) | 1.110 (1.053, 1.169) | 1.208 (1.150, 1.268) |
| Adjusted OR* (95% CI) | 1.276 (1.220, 1.334) | 1.140 (1.088, 1.193) | 1.136 (1.082, 1.193) | 1.129 (1.086, 1.174) | 1.117 (1.060, 1.178) | 1.222 (1.163, 1.284) |
| Adjusted OR† (95% CI) | 1.279 (1.223,1.337) | 1.160 (1.106, 1.216) | 1.139 (1.084, 1.197) | 1.131 (1.088, 1.176) | 1.122 (1.064, 1.184) | 1.236 (1.174, 1.300) |
*Model 1, adjusted for age, age of children (month), physical activity level, current smoking status, current alcohol drinking status. †Model 2, adjusted for Model 1, parity numbers, history of gestational diabetes, history of gestational hypertension, hemoglobin. |
Table 3
Stratified analyses of MetS risk and BMI stage by age, age of children (mouths), parity, physical activity level, hemoglobin
| BMI at prepregnancy | | BMI before labor | | BMI at 42 days postpartum | | BMI at current | | △BMI before labor - BMI at prepregnancy | | △BMI at 42 days postpartum - BMI before labor | | △BMI at current - BMI at 42 days postpartum | |
| MetS (OR, 95%CI) | P Value for Interaction | MetS (OR, 95%CI) | P Value for Interaction | MetS (OR, 95%CI) | P Value for Interaction | MetS (OR, 95%CI) | P Value for Interaction | MetS (OR, 95%CI) | P Value for Interaction | MetS(OR, 95%CI) | P Value for Interaction | MetS (OR, 95%CI) | P Value for Interaction |
Age ( years) | | 0.1585 | | 0.1872 | | 0.0230 | | 0.1891 | | 0.3572 | | 0.2554 | | 0.2479 |
< 30 | 1.230 (1.155, 1.310) | | 1.177 (1.111, 1.247) | | 1.196 (1.127, 1.269) | | 1.306 (1.233, 1.383) | | 0.986 (0.900, 1.079) | | 1.031 (0.914, 1.162) | | 1.258 (1.171, 1.350) | |
≥ 30 | 1.254 (1.185, 1.327) | | 1.217 (1.152, 1.286) | | 1.251 (1.181,1.326) | | 1.347 (1.272, 1.426) | | 1.025 (0.967, 1.085) | | 0.981 (0.889, 1.083) | | 1.207 (1.121, 1.124) | |
Age of children (mouths) | | 0.9235 | | 0.5235 | | 0.7624 | | 0.5239 | | 0.0438 | | 0.0408 | | 0.0358 |
< 12 months | 1.240 (1.183, 1.301) | | 1.218 (1.164, 1.275) | | 1.230 (1.173, 1.289) | | 1.320 (1.260, 1.382) | | 1.046 (0.993, 1.101) | | 0.957 (0.879, 1.041) | | 1.188 (1.119, 1.262) | |
12–23 months | 1.257 (1.150, 1.373) | | 1.159 (1.066, 1.260) | | 1.217 (1.115, 1.328) | | 1.365 (1.253, 1.487) | | 0.883 (0.787, 0.991) | | 1.128 (0.968, 1.313) | | 1.364 (1.229, 1.514) | |
Parity | | 0.4000 | | 0.1903 | | 0.1234 | | 0.1383 | | 0.0835 | | 0.9756 | | 0.149 |
Primiparous | 1.254 (1.165, 1.350) | | 1.241 (1.158, 1.331) | | 1.257 (1.171, 1.348) | | 1.363 (1.269, 1.465) | | 1.083 (0.991, 1.183) | | 0.966 (0.853, 1.095) | | 1.183 (1.092, 1.282) | |
Multiparous | 1.247 (1.185, 1.313) | | 1.186 (1.130, 1.244) | | 1.213 (1.153, 1.276) | | 1.312 (1.249, 1.378) | | 0.947 (0.880, 1.018) | | 1.028 (0.936, 1.130) | | 1.269 (1.187, 1.357) | |
Physical activity level | | 0.3631 | | 0.5838 | | 0.6447 | | 0.8915 | | 0.8935 | | 0.7011 | | 0.6538 |
Light | 1.252 (1.195, 1.311) | | 1.207 (1.155, 1.260) | | 1.226 (1.172, 1.282) | | 1.327 (1.270, 1.387) | | 1.014 (0.963, 1.067) | | 0.986 (0.908, 1.070) | | 1.232 (1.163, 1.305) | |
Moderate & Vigorous | 1.207 (1.089, 1.338) | | 1.167 (1.060, 1.284) | | 1.200 (1.085, 1.328) | | 1.310 (1.193, 1.439) | | 0.991 (0.852, 1.153) | | 1.038 (0.864, 1.248) | | 1.250 (1.118, 1.397) | |
Hemoglobin (g/L) | | 0.3270 | | 0.2869 | | 0.5624 | | 0.7220 | | 0.1134 | | 0.9527 | | 0.1441 |
< 133(median) | 1.290 (1.206, 1.381) | | 1.184 (1.116, 1.257) | | 1.207 (1.134, 1.284) | | 1.334 (1.255, 1.418) | | 0.921 (0.834, 1.018) | | 1.014 (0.894, 1.150) | | 1.306 (1.207, 1.414) | |
≥ 133(median) | 1.223 (1.159, 1.290) | | 1.230 (1.164, 1.300) | | 1.243 (1.175, 1.314) | | 1.329 (1.259, 1.403) | | 1.042 (0.985, 1.102) | | 0.981 (0.893, 1.077) | | 1.185 (1.108, 1.267) | |
Logistic regression model with adjustment for age, age of children (month), physical activity level, current smoking status, current alcohol drinking status, parity numbers, history of gestational diabetes, history of gestational hypertension, and hemoglobin. |
Using ROC curve analysis, the cut-off values for differentiating MetS at each stage of pregnancy and lactation were obtained based on sensitivity, specificity and Youden Index standard (Fig. 2B). The best cut-off values for BMI at pre-pregnancy, BMI before labor, BMI at 42 days postpartum, BMI in lactation were 23.47, 30.49, 26.04 and 25.47kg/m2, according to the Youden Index standard. The heat map and hierarchical clustering of BMI and blood indicators for women at each stage of pregnancy and lactation are shown in Fig. 2C. Serum vitamin B12, folic acid, albumin, and vitamin D were grouped together with each stage of BMI. Transferrin receptor, hs-CRP, ferritin, vitamin A, serum zinc, hemoglobin, and total protein were grouped together. Transferrin receptor, hs-CRP, ferritin, vitamin A, serum zinc, hemoglobin, and total protein were grouped into one category. Fasting blood glucose was grouped into one category. The different stages of BMI, BMI at pre-pregnancy, BMI before labor, and BMI at 42 days postpartum were grouped into one category, and BMI at current was a separate category. In the logistic regression model-based RCS, ORs for MetS increased significantly with increasing BMI in women of reproductive age at all stages (Fig. 3). The non-linear spline test for BMI at 42 days postpartum, current BMI and MetS during lactation was statistically significant (P non-linear = 0.0223, 0.0003).
Figure 4 demonstrates the chain mediation effect and path coefficients of BMI at pre-pregnancy on the relationship between BMI at pre-pregnancy and MetS during pregnancy and lactation. The mediation effect arises through three chains of mediation: first, the indirect effect consisting of BMI at pre-pregnancy→BMI before labor→BMI at 42 days postpartum→BMI at current→MetS with Bootstrap 95% confidence interval [0.0501,0.0931], indicating that the chain mediating effect of BMI before labor, BMI at 42 days postpartum, and BMI at current was significant; second, the indirect effect consisting of BMI at pre-pregnancy→BMI at 42 days postpartum→BMI at current→MetS effect, Bootstrap 95% confidence interval [0.0223,0.0507], indicating that the chain mediated effect of BMI at 42 days postpartum, BMI at current was significant; third, the indirect effect composed of BMI at pre-pregnancy→BMI at current→MetS, Bootstrap 95% confidence interval [0.0845,0.1558], indicating a significant mediating effect of BMI at current. The total indirect effect accounted for 80.95% of the total effect.