The protective role of serum uric acid against premature membrane rupture in gestational diabetes: a cross-sectional study

In this cross-sectional study, subjects were recruited from the maternity clinic of the Central Hospital of Shanghai Fengxian District from November 2016 to January 2019. The patients were included according to the following criteria: 1) 20 to 40 years old, 2) had a pregnancy examination in our hospital, 3) were prepared to complete a 75-g oral glucose tolerance test (OGTT) screening at 24-28th weeks of gestation and 4) were to give birth in our hospital to ensure the integrity of the data. All subjects underwent a medical history evaluation, and those with the following conditions were excluded: 1) a history of smoking (n = 1), 2) diabetes diagnosis before pregnancy (n = 15), 3) polycystic ovary syndrome (n = 15), 4) hypertension (n = 8), 5) liver disease (n = 3), or 6) drugs that may affect glucose metabolism (n = 5). Finally, a total of 152 women with GDM and 304 individuals without GDM were included in this study.

Pre-pregnancy BMI (kg/m²) was calculated on the basis of self-reported weight measurements before the pregnancy.

Venous blood samples were drawn after a minimum of a 10-h overnight fast during 24-28th gestational weeks. Serum biochemical parameters were measured using a Beckman DXC800 automatic biochemical analyser. Haemoglobin A1C (HbA1c) levels were measured using high-performance liquid chromatography (TOSOH HLC-723 G7, Japan), and insulin levels were determined using an insulin detection kit and its reagent (Roche Diagnostics, Germany).

One-step 75 g OGTTs were performed during 24-28th gestational weeks. Fasting and postload blood glucose and insulin levels were measured at 0, 60, and 120 min. GDM was diagnosed according to the 2013 World Health Organization diagnostic criteria: fasting blood glucose ≥5.1 mmol/L,1-h OGTT blood glucose ≥10.0 mmol/L, or 2-h OGTT blood glucose ≥8.5 mmol/L.

All participants were inpatients before delivery. Weight gain during pregnancy was collected, and primary adverse outcomes were confirmed by experienced obstetricians. The natural rupture of foetal membranes before the first stage of labour was diagnosed as PROM. All subjects had delivered and finished follow-up by December 2019.

Table 1 lists the clinical characteristics of the patients involved in this study. There were differences in the levels of HBA1c, fasting plasma glucose (FPG), 1-h postprandial glucose (PG), 2-h PG, 2-h postprandial insulin (2-h INS), HOMA-IR, HOMA-β, white blood cells, BMI, fasting insulin (FIN), 1-h postprandial insulin (1-h INS) and triglycerides between the non-GDM and GDM groups (all P < 0.05). Moreover, there were significant differences in weight gain during pregnancy (P = 0.014), foetal birth weight (P = 0.003), rate of PROM (P = 0.033) and macrosomia (P = 0.004) between the two groups. There was a similar level of systolic blood pressure before delivery in the two groups.

The patients with GDM were divided into PROM and non-PROM groups, and the baseline data for the two subgroups are shown in Table 2. There were significant differences in age (P = 0.030) and uric acid (P = 0.030) between the two groups. There were no significant differences in uric acid between the GDM and non-GDM groups when PROM status was not considered (P = 0.130) (Table 1).

Binary logistic regression was conducted to assess the association of PROM with clinical variables with PROM as the dependent variable, as shown in Table 3. When the clinical variables (age, pre-pregnancy BMI, 1-h PG, parity, weight gain) that might affect outcomes were adjusted, uric acid (P = 0.043) was related to PROM in the GDM group in multivariate analysis. Furthermore, the uric acid of the whole population was stratified by three quantiles, and the results showed that the incidence of PROM decreased with the increase in blood uric acid tertiles. The corresponding ORs (95% CIs) were 0.13 [0.02–0.95] in the second tertile and 0.07 [0.01–0.93] in the upper tertile in comparison to the lower tertile (all P < 0.05).

Furthermore, the predictive value of different models for PROM was analysed using receiver operating characteristic (ROC) analysis. As shown in Fig. 1, ROC curve analysis showed that the area under the curve of the general clinical characteristic model 1 was 0.71 [0.54–0.87], P < 0.05. When uric acid was added as a variable, the area under the curve in model 2 reached 0.86 [0.75–0.98], P < 0.001.

What we already know about PROM focuses on infectious and external factors, such as genital infections by chlamydia and/or gonorrhoea, intrauterine infection and cervical abnormalities before pregnancy [13]. Studies that have looked for metabolic factors related to PROM in pregnant women are rare. One study from a hospital in Ethiopia showed that a history of abortion, history of PROM, history of caesarean section, and abnormal vaginal discharge were positively associated with PROM [1]. A Chinese study suggested that migration as a result of urbanization, high rates of induced abortion, and preterm birth are potential risk factors for preterm PROM [14]. A Swedish study of PROM at or near term found no obvious risk factors and concluded that PROM close to term can be difficult to manage [15]. GDM is associated with adverse pregnancy outcomes, including PROM [7]. A study from Pennsylvania found that cord blood uric acid was higher in preterm PROM vs. preterm birth, but maternal plasma uric acid and glucose tolerance state were not considered [16]. To the best of our knowledge, the relationship of uric acid levels with PROM in women with GDM has not been studied previously. No model for individual prediction of PROM risk has been validated or used in clinical practice. By distinguishing between GDM and women with normal glucose tolerance, we identified SUA as an independent factor related to PROM for women with GDM. Of utmost importance is that as the concentration of uric acid increases, the risk of PROM decreases in the patients with GDM, and elevated SUA levels suggest a lower risk incidence of PROM after adjustment for other confounders.

The concentration of serum uric acid has two sides in the pathophysiological process of many diseases. Hyperglycaemia accelerates purine nucleotide synthesis, which in turn stimulates nucleotide breakdown and increases the concentration of nucleotide degradation products, including superoxide molecules and uric acid [17]. These substances directly damage islet function [18] and worsen diabetes complications [19]. However, uric acid is an antioxidant that can protect cells from OS damage. The relationship between uric acid and cardiovascular event rates is J-shaped [20]. High SUA levels due to uric acid toxicity increase cardiovascular events, but when SUA levels are low, cardiovascular events are also increased; this analogous finding was observed in patients with vasospastic angina [21]. The effects of SUA are also the same in acute ischaemic stroke patients [22].

Reference values for pregnancy can be extremely important for making clinical decisions. Normal reference values are usually based on blood samples from nonpregnant women. However, uric acid levels are higher in women late in pregnancy than in nonpregnant women [23]. Additionally, hyperglycaemia increases oxidative stress products; hence SUA levels should also be increased in GDM patients. Women with GDM should have higher SUA levels in late pregnancy and in hyperglycaemic states. However, when the SUA levels do not increase correspondingly, women with GDM are more prone to PROM

A general health screening study of individuals revealed that SUA levels are positively associated with antioxidant potential; these results indicate that SUA can prevent oxidative damage [24]. GDM is characterized by increased levels of OS biomarkers. Placental histological examinations have indicated that OS plays an important role in inducing inflammation and impairing perfusion in the placenta [10]. Our study showed that SUA levels were inversely correlated with PROM in GDM subjects. OS damage is also an important factor in the initiation of cell apoptosis. GDM and type 2 diabetes mellitus (T2DM) subjects have increased expression levels of base excision repair proteins, and hyperglycaemia levels are associated with apoptosis [25]. Proteomics research confirmed by western blot and RT-PCR analyses revealed that placental tissues from PROM subjects had lower levels of the anti-apoptotic protein HSPA2 and higher expression levels of PRDX3 and annexin A1 (associated with apoptosis) than placental tissues from normal pregnancy subjects [26]. The increase in apoptosis caused by the decrease in antioxidant stress may be another important cause of PROM. We hypothesized that the antioxidant capacity of relatively low SUA levels cannot exert its usual protective effects; thus, women with low SUA are at risk of PROM. This study indicates that maternal age is another important factor affecting PROM, as it was significantly higher in GDM patients with PROM. This is consistent with the Chinese study of risk factors for PROM [14]. The percentage decrease in enzymatic antioxidants and antioxidant vitamins with age is significant [27]. In this study, when divided the subjects into two groups by age ≤ 29 years and age > 29 years old, the data showed that SUA was only associated with PROM in elderly pregnant women (age > 29 years old) in the GDM group (Supplemental Table 1). The protective effect of relatively high SUA levels on PROM may be more important in elderly women with GDM. After the implementation of the nationwide two-child policy in China, the number of elderly pregnant women increased, which may be another important factor in the association with adverse pregnancy outcomes.

There was some strength in this study. Blood biochemical tests (including SUA) were collected from all subjects at 24-28th weeks of gestation to exclude the effects of differences in gestational weeks on the results. All the subjects were women because of pregnancy, and none of the subjects consumed alcohol. Consequently, possible SUA-influencing factors, such as sex and alcohol consumption, were excluded. Except for the factors that may affect glucose metabolism, there were no other recruitment restrictions, thus the study population should be a good representation. There are some limitations to this study. This was a single centre study, so the number of subjects was limited. The follow-up period was short; therefore we could make no conclusions on the long-term pregnancy outcomes. The related uric acid metabolism pathways also need further study.