Preplanned Studies: Molybdenum Concentration and the Risk of Spontaneous Preterm Birth: A Nested Case-Control Study — Beijing Municipality, China, 2018–2020

Spontaneous preterm births (SPB), with no identifiable causes, account for a significant 60%–70% of all preterm births (1). These are responsible for about 35% of neonatal fatalities linked to premature births (2). Given its high incidence and consequences, the prevention of preterm births is increasingly gaining traction as a critical public health concern. An essential ultratrace element, molybdenum (Mo), plays a role in several human metalloflavoproteins. Studies have revealed that the maternal urinary Mo concentration during pregnancy bears an inverse association with fetal abdominal circumference and estimated fetal weight while also positively correlating with childhood blood pressure (3-4). This study aimed to monitor Mo levels in pregnant women’s plasma and analyze any potential link to SPB risk. We found the maternal plasma Mo concentration in the SPB group was significantly greater than that in the control group during the first trimester [2.12 (1.56–2.69) vs. 2.02 (1.46–2.50) ng/mL, P=0.027]. Higher Mo levels (≥2.584 ng/mL) recorded in the first trimester were linked with an elevated SPB risk after adjusting for maternal age, ethnicity, pre-pregnancy body mass index (PPBMI), parity, mode of delivery, education, and income levels [odds ratio (OR)=2.074 (1.212, 3.550), P=0.008]. However, when the Mo concentration stood at ≥3.554 ng/mL during the third trimester, the OR for SPB reached 1.732 (1.016, 2.953), P=0.044, again adjusted for the aforementioned factors. Furthermore, it is crucial to note that the current reference range for blood Mo concentrations in pregnant women across various Chinese regions remains undefined. Hence, clinicians must be vigilant against illnesses potentially arising from high Mo concentrations.

We conducted a nested case-control study with subjects drawn from the Beijing birth cohort. In estimating our sample size, we predicated on a high-level Mo exposure rate of approximately 50% in the term delivery group, a common odds ratio of 2 (5), a two-sided importance level of α=0.05 and β=0.10, and a case-control proportion of 1∶1. Calculations performed using the PASS (version 11.0; NCSS statistical software, Kaysville, USA), indicated a minimum sample size of 179 patients required for the SPB group. Eligibility criteria included women aged between 18 and 44 years willing to undergo routine prenatal examinations and deliveries at the Beijing Obstetrics and Gynecology Hospital, and who provided signed consent. Following an overnight fast (≥8 hours), we obtained 2.0 mL venous blood samples from each participant during the first (7 to 9 weeks of gestation) and third trimesters (33 to 34 weeks). SPB was defined as preterm birth with intact fetal membranes and preterm premature rupture of membranes before the 37th week of gestation, excluding iatrogenic preterm birth caused by medical interventions such as maternal fetal induction or cesarean section. After excluding women with twin pregnancies, cervical inadequacies, a history of preterm birth, placental abruption, iatrogenic preterm birth due to hypertension, diabetes, central placental previa, abnormal liver function, acute fatty liver, cholestasis, appendicitis, and significant cardiac insufficiency, liver, kidney, and brain diseases, as well as those on bed-rest; we selected a cohort of 29,303 live singletons, 617 of whom were SPB cases. We further narrowed it down to 236 SPB patients who had blood samples collected during both the first and third trimesters. A control group of 236 subjects composed of healthy pregnant women with gestational weeks of <39 or >41 was also selected (Supplementary Figure S1). The study was granted approval by the Ethics Committee of the Beijing Obstetrics and Gynecology Hospital, Capital Medical University.

The Mo concentrations were determined using inductively coupled plasma mass spectrometry (ICP-MS, ELAN DRC II, PerkinElmer, USA). Standardized plasma (ClinChek® - Plasma Control, Level II, Germany) samples served as controls for quality assurance. The detection limit (LOD) for Mo in the plasma was set at 0.04 ng/mL and our analysis found that the concentrations of Mo in all samples exceeded this threshold. In the standard plasma samples, the median reference value for Mo was 5.96 ng/mL, ranging from a minimum of 4.77 ng/mL to a maximum of 7.15 ng/mL. To investigate the correlation between Mo levels and the risk of SPB anomalies, we executed an unconditional logistic regression analysis utilizing the IBM SPSS Statistics software (version 26.0; IBM Corporation, Armonk, NY, USA). We deemed a two-tail P-value of less than 0.05 as statistically significant in all statistical analyses.

The study groups, consisting of 236 patients each in the SPB cohort and the control group, demonstrated comparable demographic characteristics with respect to maternal education, personal monthly income, maternal ethnicity, and fetal gender. Nevertheless, significant differences were observed concerning maternal delivery time (35.53±0.69 weeks for SPB vs. 39.70±0.74 weeks for controls, P<0.001), maternal age [under 35 years: 177 (75%) SPB vs. 196 (83.05%) controls; 35 years and older: 59 (25%) SPB vs. 40 (16.95%) controls, P=0.032], PPBMI [under 18.5 kg/m²: 29 (12.29%) SPB vs. 29 (12.29%) controls; 18.5–25 kg/m²: 149 (63.14%) SPB vs. 176 (74.58%) controls; 25 kg/m² and above: 58 (24.57%) SPB vs. 31 (13.13%) controls, P=0.021], parity [nulliparous: 154 (65.25%) SPB vs. 184 (77.97%) controls; multiparous: 82 (34.75%) SPB vs. 52 (22.03%) controls, P=0.002], mode of delivery [vaginal: 145 (61.44%) SPB vs. 176 (74.58%) controls; cesarean section: 91 (38.56%) SPB vs. 60 (25.42%) controls, P=0.002], fetal birth weight (2696.46±338.75 g SPB vs. 3429.94±299.46 g controls, P<0.001), and birth length (47.70±1.93 cm SPB vs. 50.25±0.90 cm controls, P<0.001) (Table 1).

During the first trimester, median maternal plasma Mo concentrations were measured at 2.07 (1.53–2.58) ng/mL for the entire cohort. Within this cohort, women in the SPB group had a slightly higher median Mo concentration of 2.12 (1.56–2.69) ng/mL compared to 2.02 (1.46–2.50) ng/mL in the control group, and this difference was statistically significant (P=0.027). By the third trimester, the overall median Mo concentration had increased to 2.59 (1.66–3.55) ng/mL. The SPB group’s median concentration rose to 2.66 (1.69–3.69) ng/mL, while the control group’s median was 2.54 (1.58–3.41) ng/mL, though this inter-group difference was not statistically significant (P=0.147). Notably, third trimester Mo concentrations were significantly elevated when compared to first trimester levels (P<0.001). Between the first and third trimesters, the SPB group experienced an increase in Mo concentration of 0.37 (range: −0.86 to 1.73) ng/mL and the control group an increase of 0.35 (range: −0.84 to 1.73) ng/mL, with no significant difference between the groups (P=0.775) (Table 2).

During the first trimester, when maternal plasma Mo concentration was at or above 2.584 ng/mL, the odds of SPB as determined by logistic regression analyses with various adjustments, were significantly increased with ORs of 1.878 [95% confidence interval (CI): 1.129, 3.124, P=0.015], 1.892 (95% CI: 1.118, 3.201, P=0.018), and 2.074 (95% CI: 1.212, 3.550, P=0.008). In the third trimester, the occurrence of SPB at a Mo concentration of 3.554 ng/mL or higher had an adjusted OR of 1.732 (95% CI: 1.016, 2.953, P=0.044), after factoring in variables such as maternal age, ethnicity, PPBMI, number of births (parity), mode of delivery, educational background, and income levels. In comparison to the lowest quartile of plasma Mo levels, the odds of experiencing SPB were 1.217 (95% CI: 0.715, 2.072), 1.084 (95% CI: 0.635, 1.848), and 2.074 (95% CI: 1.212, 3.550) for the second, third, and fourth quartiles, respectively, showing a significant trend (P_trend=0.017) in the third logistic regression model. In the first trimester, the P_trend values for the remaining two models were 0.036 and 0.039, respectively. For the third trimester, P_trend was found to be significant at 0.038 in logistic regression model 3 (Table 3).

The subgroup analysis results suggested a correlation between plasma Mo concentration and the risk of SPB in overweight, obese, and multiparous mothers during the first trimester (Supplementary Tables S1 and S2). Additionally, our sensitivity analysis results corroborate with our preliminary findings (Supplementary Table S3).