The predictive role of serum uric acid levels before pregnancy in the development of gestational diabetes mellitus

Abstract

Objective

With the prevalence of obesity, growing age of first pregnancy, highly processed high-calorie diet, consumption of saturated fats as well as sedentary and stressful life, the prevalence of gestational diabetes mellitus (GDM) is increasing. We aimed to determine the predictive role of pre-pregnancy serum uric acid levels and the occurrence of GDM during pregnancy.

Methods

This study was a descriptive-analytical study that was performed retrospectively through case–control. The Subjects of this study were women over 18 years of age who were 24–28 weeks pregnant. All subjects of this study were evaluated for GDM based on the ADA guideline and were divided into case and control groups. All data relating to the period of 6 months before pregnancy of the study participants including blood uric acid level, blood pressure, etc. were collected and analyzed.

Results

In this study, 454 normal individuals without GDM and 478 others with GDM were examined. The mean serum uric acid showed to be 4.47 ± 1.33 mg/dl in patients with GDM but 3.7 ± 1.25 mg/dl in patients without GDM (p value = 0.001). The results of the regression test showed that the incidence of GDM in people with blood uric acid levels of 4.1–5, 5.1–6, and more than 6 mg/dl is 2.46, 3.42, and 4.9 times higher in people with uric acid levels less than 3 mg/dl, respectively.

Conclusion

The present study identified that serum uric acid levels in the period of 6 months before pregnancy can be used as an independent predictor marker for future GDM.

Introduction

Gestational diabetes mellitus (GDM) is a common pregnancy complication defined by glucose intolerance, which is first initiated or diagnosed in pregnancy [1]. Patients with GDM are higher than seven times more likely to develop type 2 diabetes within 5 to 10 years after delivery [2]. In addition, the children of these patients are more likely to develop type 2 diabetes [3].

The prevalence of GDM is reported to be between 4 and 14% in different parts of the world [4]. Approximately 7% of pregnant women in the United States, equivalent to 200,000 people, suffer from this type of diabetes [5].

The prevalence of this diabetes in the United Kingdom and Europe is reported to be 5% and 2–6%, respectively [6]. A higher prevalence has been reported in African, Asian, and Indian races. Its rate is, however, far varying in Iran and shows to be between 0.7 and 18.2%, which is reported to be 9.4% on average [7, 8]. Patients with this disease and their babies are also exposed to other risks, such that 40% of women with GDM develop type 2 diabetes within 30 years [9]. Between 30 and 70% of them will develop GDM in the next pregnancy and 6% of their babies will develop diabetes during their lifetime [10, 11].

Among the most important complications that GDM can have on the mother, we can mention complications such as premature delivery, infectious complications, and high blood pressure [12, 13]. It can also have serious consequences for the baby, such as fetal death in utero, congenital anomalies, fetal developmental disorders, childhood obesity, polycythemia, type 2 diabetes, as well as cardiovascular disease, and metabolic syndrome [14].

Risk factors for GDM include maternal age, high pregnancy rate, family history of diabetes, race, sleep-disordered breathing, and pre-pregnancy obesity [15–18]. One of the most important issues for doctors when dealing with women who are planning to become pregnant is how a woman's current health can affect the pregnancy process [19]. The importance of improving maternal health, as well as reducing complications and adverse pregnancy outcomes due to birth, has led to shifts in attention from prenatal care and assessments to the time before pregnancy [10]. Undoubtedly, prenatal care is necessary and important to find fetal complications and abnormalities, but this prenatal care alone is not enough to reduce the risk factors that affect the outcome of pregnancy [20]. The CDC (Centers for Disease Control and Prevention) considers pre-pregnancy health as a set of interventions aimed at identifying and controlling biomedical, behavioral, and social risks for women's health or pregnancy outcomes through management and prevention [10]. Studies have shown that lifestyle changes in the early stages of pregnancy to prevent GDM have been associated with little success, as pathogenic processes leading to GDM begin before pregnancy [21, 22].

Modifiable risk factors such as pre-pregnancy or early pregnancy diet can be employed as a goal in preventing GDM. Diabetes and hypertension during pregnancy predict the future progression of cardiovascular disease and metabolic syndrome [23]. Hyperuricemia is also associated with markers of metabolic syndrome, including obesity, dyslipidemia, and hypertension [24–28]. Uric acid in non-pregnant women is an independent risk factor for progression toward type 2 diabetes over 10 years, and this association is stronger in women than men [29]. In non-pregnant women, uric acid is associated with insulin resistance [30]. In pregnancy, uric acid is also related to insulin resistance in women with gestational hypertension [31]. Studies have shown that serum uric acid levels are higher in women with GDM at 24–28 weeks of gestation [32]. Unrelated to BMI, uric acid has also been proven to be higher in non-pregnant women with a history of GDM [33].

Regarding the fact that uric acid is associated with insulin resistance and the incidence of GDM [34], this study aimed to determine and compare the serum levels of pre-pregnancy uric acid in women with GDM and the control group such that in the case of the existence of any positive correlation, it can provide an opportunity to prevent GDM.

Materials and methods

This is a retrospective analytical-descriptive study with case–control groups. In this research, pregnant women over 18 years of age who were diagnosed with GDM during 24–28 weeks of gestational screening tests and whose pre-pregnancy tests and information were available, including age, family history of diabetes, BMI, and blood pressure, were evaluated (All of the participated in this study were examined 6 mounts prior to conception in order to receive per-pregnancy care and counseling. The results of their prior assessments were utilized for this study when these individuals returned to the clinic between weeks 24 and 28 of pregnancy.). According to the American Diabetes Association (ADA) guidelines, if a patient’s fasting plasma glucose (FPG) was greater than or equal to 92 mg/dl, or one-hour glucose after consumption of 75 mg glucose was higher than or equal to 180 mg/dl, or two-hour glucose appeared to be greater than or equal to 153 mg/dl, GDM was diagnosed and the patient was entered into the study with her own consent. Given the significant level of 5%, test power of 80%, and according to previous study results and the prevalence of 10% high uric acid in normal people and 50% in GDM, as well as deploying the statistical formula, the number of samples in the group was calculated to be 100. The patients were entered into the study until the sample size was completed and then for the control group, pregnant women who did not show GDM via gestational diabetes mellitus screening were also included in the study as the control group if, of course, they were consented.

Patients with type 2 diabetes or any type of pre-pregnancy diabetes treated with antidiabetic drugs, patients with renal failure (GFR < 90), and those with a history of gout or the subjects treated with uric acid-lowering drugs were excluded from the study.

By examining the patient’s information, the data needed for the study comprising uric acid level during 6 months before pregnancy, blood pressure, GDM history, GDM family history, TSH (thyroid-stimulating hormone), FPG, and BMI (Body Mass Index) were collected and recorded. Finally, the data were analyzed by SPSS software version 21 using Chi-Square tests, logistic regression, and odds ratio (OR).

Results

This is a case–control probe attempting to investigate the relationship between serum uric acid levels and the incidence of GDM. In this study, examinations were carried out for 454 normal individuals in terms of GDM and 478 others affected with GDM. The demographic information and the laboratory data of the two groups are set out in Table 1.

Table 1.

Comparison of demographic and laboratory variables in the two groups (the variables were obtained in the pre-conception period)

Variables	Participants		p value
	GDM group (n = 478)	Normal group (n = 454)
Age	3.92 ± 29.17	3.77 ± 27.38	0.001
BMI	2.19 ± 27.7	1.8 ± 26.9	0.001
History of previous GDM	170 (35.6%)	113 (24.9%)	0.001
Family history	184 (38.5%)	112 (24.7%)	0.001
Uric acid (mg/dl)	1.33 ± 4.47	1.25 ± 3.7	0.001
FPG (mg/dl)	6.34 ± 99.13	5.38 ± 82.8	0.001
TSH (mIU/L)	0.5 ± 1.23	0.5 ± 1.39	0.001
Systolic blood pressure (mm Hg)	14.7 ± 124.16	13.6 ± 121.5	0.005
Diastolic blood pressure (mm Hg)	11.63 ± 71.5	12.72 ± 69.9	0.045

The mean serum uric acid in patients with GDM was 4.47 ± 1.33 mg/dl and in normal individuals was 3.7 ± 1.25 mg/dl, which was significant concerning p value = 0.001; therefore, the mean serum uric acid level 6 months before pregnancy in patients with GDM turned out to be higher.

After univariate analysis, the results projected the incidence of GDM in patients over 30 years to be 4.37 times higher than in those under 20. But the risk of developing GDM in people aged 20 to 24 and 25 to 30 failed to be different from those under 20. Also, the incidence of GDM in people with a BMI of 25 to 29.9 and above 30 appeared to be, respectively, 1.69 and 5.22 times higher compared with people carrying a BMI of lower than 25.

With the uric acid level lower than 4 mg/dl, the majority of the patients were normal while with uric acid 5.1–6, mg/dl about 68.7%, and with uric acid higher than 6 mg/dl, 75.9% of them showed GDM. Chi-Square test results demonstrated an increment in the incidence of GDM as uric acid levels increased (Table 2).

Table 2.

Comparison of serum uric acid levels in patients with and without gestational diabetes (the variables were obtained in the pre-conception period)

Uric acid level (mg/dl)	Participants			OR (95% CI)	p value
	GDM group	Normal group	Total
Lower than 3	84 (39.1%)	131 (60.9%)	215 (100%)	Ref
3.1–4	77 (31.4%)	168 (68.6%)	245 (100%)	0.71 (0.48–1.05)	0.08
4.1–5	95 (61.3%)	60 (38.7%)	155 (100%)	2.46 (1.61–3.77)	0.001
5.1–6	178 (68.7%)	81 (31.3%)	259 (100%)	3.42 (2.34–5)	0.001
Higher than 6	44 (75.9%)	14 (24.1%)	58 (100%)	4.9 (2.53–9.49)	0.001
Total	478 (51.3%)	454 (48.7%)	932 (100%)

In univariate analysis, the incidence of GDM was significantly associated with a history of GDM, positive family history. The likelihood of developing gestational diabetes mellitus in people with a positive history of GDM proved 1.66 (1.23–2.18 CI 95%) times higher compared to those without a history of GDM, in those with a positive family history of 1.91 (1.42–2.5 CI 95%) times greater compared to people without a family history. Also, the incidence of GDM in people with stage one and two blood pressure (in this study, the blood pressure classification is based on AHA (American Heart Association) guidelines) turned out to be 1.4 and 1.53 times higher than in people with normal blood pressure, respectively, which proved to be significant. But there was no difference between pre-hypertension and normal individuals and no significant relationship was found.

In evaluating all variables, the results of the logistic regression test revealed that among the studied variables, uric acid along with other factors such as age, BMI, family history, and GDM history can act as predictors of GDM in pregnant women. According to OR, after age and BMI, which are important factors in the incidence of GDM, a uric acid level greater than 4.6 mg/dl is also a positive and significant factor. Family history and history of GDM are, moreover, effective factors in the development of GDM (Table 3).

Table 3.

Results of multivariate logistic regression analysis in predicting gestational diabetes (the variables were obtained in the pre-conception period)

Variable	OR	95% CI	p value
Uric acid (mg/dl)
Lower than 3	Ref	–	–
3.1–4	0.74	1.13–0.49	0.16
4.1–5	2.6	4.27–1.7	0.001
5.1–6	4.12	6.24–2.72	0.001
Higher than 6	4.46	8.99–2.21	0.001
Age
≤20	Ref	–	–
24–20	1.17	0.45–3.04	0.74
30–25	1.85	0.76–4.47	0.17
>30	4.7	1.87–11.79	0.001
BMI
<24.9	Ref	–	–
29.9–25	1.63	0.94–2.84	0.08
≥30	5.89	2.77–12.52	0.001
Blood pressure* (BP) (mm Hg)
Normal	Ref	–	–
Pre-hypertension	1.09	0.76–1.56	0.061
Stage I	1.48	1.04–2.09	0.02
Stage II	1.59	1.03–2.44	0.03
History of previous GDM	1.4	1.01–1.94	0.039
Family history	2.12	1.53–2.93	0.001

^*According to AHA guideline. Normal systolic BP less than 120 mm Hg and diastolic BP less than 80 mm Hg, Pre-hypertension (elevated) systolic BP between 120 and 129 mm Hg and diastolic BP less than 80 mm Hg, Stage I systolic BP between 130 and 139 mm Hg or diastolic BP between 80 and 89 mm Hg, Stage II systolic BP of 140 mm Hg or higher or diastolic BP of 90 mm Hg or higher

Discussion

Due to the limited number of studies in this area, the present study was performed to research the relationship between serum uric acid levels before pregnancy and the incidence of GDM that the results of which indicated the mean serum uric acid level in the patients affected with GDM is 4.47 mg/dl but in patients without GDM, it is 3.7 mg/d; this indicates the prevalence of a higher mean serum uric acid level before pregnancy in GDM patients. Moreover, regression test results confirmed that variables such as age (over 30 years), BMI, blood pressure, GDM history, and GDM family history alone are associated with an increased incidence of GDM. The results on uric acid levels also revealed that the incidence of GDM increases in people with uric acid above 4 mg/dl, i.e., in uric acid above 4, 5, and 6 mg/dl, 61.3%, 68.7%, and 75.9% of people can be affected by GDM, respectively. Since different factors impact the incidence of GDM, the results of multivariate regression identified that the most important parameters influencing the incidence of GDM are age over 30 years and BMI above 30 after these two factors, serum uric acid level was one of the most important predictors of GDM. The risk of GDM in people with uric acid levels of 4–5, 5–6, and higher than 6 mg/dl shows to be, respectively, 2.69, 4.12, and 4.46 times higher than people with uric acid less than 3. Multivariate analysis also showed that other influential variables in the incidence of GDM are blood pressure (stages 1 and 2), family history as well as an individual history of GDM.

In a study carried out by Dr. Khosro Beigi on 30 women with GDM and 30 healthy pregnant women, the mean serum level of uric acid in healthy pregnant women was projected to be 4.76 mg/dl, and in diabetic women amounted to 5.06 mg/dl, and failed to make a significant difference. However, serum uric acid levels were checked in their study at 24–28 weeks of gestation; while in the present study, uric acid levels were assessed during 6 months prior to pregnancy [37]. The results of Khosro Beigi et al. study suggested that serum uric acid levels bear a significant direct correlation with the HOMA-IR index (Homeostatic Model Assessment of Insulin Resistance) in the diabetic group (r = 0.38) but a slight correlation with the QUICKI index (Quantitative insulin sensitivity check index) [37]. In that study, it was stated that a slight increase in uric acid could be a risk factor for insulin resistance in GDM.

To determine the association between uric acid (UA), impaired fasting glucose (IFG), impaired glucose tolerance (IGT), insulin resistance (IR), and low insulin sensitivity (IS) in youths with overweight/obesity (OW/OB), Di Bonito et al. undertook a cross-sectional study which involved 2248 youths, whose ages ranged from 5 to 17 on average [43]. In their study, the sample was stratified in sex-specific quintiles (Q1–Q5) of UA, and the associations with fasting glucose (FG), 2-h post-load glucose (2H-PG), IR, and low IS were investigated. IR and low IS were estimated by an assessment model of insulin resistance (HOMA-IR) and whole-body IS index (WBISI), respectively. IFG was defined as FG ≥ 100 < 126 mg/dL, IGT as 2H-PG ≥ 140 < 200 mg/dL, IR as HOMA-IR ≥ 75th percentile and low IS as WBISI ≤ 25th percentile by sex. Their findings demonstrated that whereas FG did not rise across sex-quintiles of UA, age, body mass index z-score, 2H-PG, HOMA-IR, and WBISI did. it was also found that rates of IGT and low IS increased progressively across quintiles of UA in youths with OW/OB, but that IFG and IR were exclusively related with the highest quintile of UA. According to their findings, UA could be used as a biomarker for impaired glucose metabolism [43].

In line with the study of Di Bonito et al., another study conducted by Hong-Qi Fan with the aim of examining whether serum uric acid is associated with 2-h post-load glucose (2-h PG) in people with impaired fasting plasma glucose (IFG) and/or HbA1c (IA1C). Their study included 1197 participants with IFG and/or IA1C (mean age 56.5 ± 10.3 years; 50.6% men and 49.4% women). The findings demonstrated that 2-h PG increased progressively and significantly in both men and women from the lower to the upper uric acid tertiles (P < 0.001). In addition, an increase in serum uric acid was substantially related with a 36% increased risk for 2-h newly diagnosed diabetes (2-h NDM) in multivariate logistic regression [44].

Among the parameters for which their effects on the incidence of gestational diabetes mellitus are known and carry the greatest impact on it, we can underline age and BMI. In this study, the results illustrated the effect of age over 30 years and BMI above 30 on escalating the risk of diabetes. The results are consistent with those of Emily et al.; they showed that BMI, and waist size before pregnancy can be associated with GDM. Moreover, in their study the OR level for BMI turned out to be 1.47; the reason for this discrepancy with what we found can be attributed to a lack of classification and quantitative evaluation of this variable [19].

In a study by Katherine Laughon et al., serum uric acid levels in the first trimester of pregnancy were assessed the results of which indicated that 46.6% of women affected with GDM showed uric acid over 3.57 mg/dl, And the risk of developing GDM at uric acid levels of 3.57–8.3 mg/dl was reported to be 3.25 times higher in this study. This effect persisted even after adjusting for age and BMI, which is consistent with the results of this study. In our study, with the uric acid level above 4.1 mg/dl, 61.3% of people were diabetic, and with that above 6 mg/dl, the risk was 4.46 times higher. However, Katherine's study identified that the effect of uric acid on increasing the risk of GDM is concentration dependent and with growing its concentration, the risk of GDM grows, which was consistent with the present study [35]. In addition, Katherine et al. Showed that at the cut-off point of 3.6 mg/dl for the uric acid level, the positive and negative predictive values as well as the level below the rock chart are 9%, 96.7%, and 0.7%, respectively. Although uric acid was strongly associated with BMI, the risk of developing GDM increased with increasing uric acid levels in the first trimester of pregnancy, independent of BMI, which is consistent with the present study. Unlike Katherine’s study, a research by Suriyamoorthi et al. suggested that uric acid in the first trimester of pregnancy is not associated with GDM. In their study, 33.3% of people with uric acid lower than 3.6 mg/dl and 17.5% of people with uric acid above 3.6 mg/dl had GDM, which was inconsistent with this and other studies possibly due to high BMI in the uric acid group lower than 3.6 mg/dl, because in this group the majority of the patients had a BMI above 25; while in the uric acid group above 3.6 mg/dl, the patients had a BMI mostly below 25 [36].

Similar to these studies, Wolak et al. detected that uric acid levels in the first 20 weeks of pregnancy are associated with diabetes and preeclampsia. The prevalence of GDM in uric acid above 5.5 mg/dl was 10.5%; while in uric acid less than 2.4 mg/dl, its prevalence turned out to be 6.3%. Also, the mean uric acid level in people with GDM was significantly higher than in the non-diabetic group, which was consonant with the present study. Also, the OR level for uric acid in predicting GDM was 1.3, which means a 1.3 times escalation in the risk of GDM per each unit increment in uric acid [39]. In this regard, Zhou et al. showed that the incidence of GDM in nulliparous women with high uric acid at 20 weeks of gestation was 2.34 times higher. Their study also found that high triglyceride levels and low HDL levels were associated with an increased risk of 2.03- and 2.11-fold GDM, respectively, although these were quantitatively measured. Similar to the present study, in the classification of uric acid, their results showed that the incidence of GDM in uric acid above 243 µlmol/L was 1.84 times higher than in uric acid below 187 µlmol/L. Also, in their study, hyperuricemia was associated with an increased risk of preeclampsia [40]. Their study disclosed that evaluation of these factors alone bears little sensitivity and specificity in predicting the incidence of diabetes, but the association and evaluation of all these factors together, including age, BMI, and uric acid show a sensitivity of 80–92% and specificity of 50–53%. This finding was also observed in the present study concerning multivariate analysis. In the study by Zhou et al., at the cut-off point of 259.5 µlmol/L, uric acid had a sensitivity of 28% and a specificity of 85% [40]. In contrast to this study which reported low sensitivity to uric acid, the findings of Amuda et al. indicated that uric acid enhances with increasing age, and at a cut-off point of 3.6 mg/dl, it is suitable for uric acid with a surface area below the curve of 0.91 and a sensitivity and specificity of 92% and 99% for predicting GDM. In their study, people with uric acid above 3.6 mg/dl, up to 52%, and those with that below 3.6 mg/dl, up to 1.3% were affected with GDM [4]. Further, in another study by Kappaganthu et al. with a cut-off point of 3.4 mg/dl for uric acid, a sensitivity of 90%, as well as specificity of 95%, was reported with a negative predictive value of 99% in predicting diabetes [41].

In most investigations, elevated uric acid levels were identified as an independent variable to augment the risk of GDM, and the cut-off point in predicting GDM varied between 3.4 and 3.8 in different studies [35, 40–42]. However, the risk of diabetes incidence and its association with uric acid was concentration-dependent and heightened with increasing concentration [35, 40–42]. It seems that these discrepancies in different studies are ascribed to different periods of uric acid evaluation, which is mostly carried out in the first trimester, as well as other risk factors for diabetes and a critical factor that was not evaluated in these studies, i.e., the people’s diet. In this study, the role of diet on uric acid and GDM was not evaluated; this, however, can be the issue of intriguing investigations to come.

Among the strengths of the present study is the appropriate sample size and evaluation of an independent factor in predicting GDM through which it is possible to identify and manage the people at risk by pre-pregnancy screening tests. However, further extensive research can be far more beneficial due to the limited number of studies in this arena.

Conclusions

This study showed that the effect of uric acid in the blood is dependent on its concentration and with increasing its level, the risk of gestational diabetes mellitus increases, and this effect was present as an independent risk factor along with age, BMI, and other variables.

Data availability

The data that support the findings of this study are available on request from the corresponding author, [Golnaz Mohammadzade]. The data are not publicly available due to the fact that they’re containing information that could compromise the privacy of research participants.

Declarations

Conflict of interest

AG, GM, MGB, and AEM declare that they have no conflict of interest.

Ethical approval

We know of no conflicts of interest associated with this publication, and there has been no significant financial support for this work that could influence its outcome. This study was presented to the ethics committee of Shahid Sadoughi University of medical science of Yazd and was approved by this committee by its ethical standards on human experimentation Ethics verification code: IR.SSU.MEDICINE.REC.1398.020 Date of approval: 3.10.2021.