First-trimester blood urea nitrogen and risk of gestational diabetes mellitus

A stepwise prediction and interpretation of gestational diabetes mellitus: Foster the practical application of machine learning in clinical decision

Background

Machine learning has shown to be an effective method for early prediction and intervention of Gestational diabetes mellitus (GDM), which greatly decreases GDM incidence, reduces maternal and infant complications and improves the prognosis. However, there is still much room for improvement in data quality, feature dimension, and accuracy. The contributions and mechanism explanations of clinical data at different pregnancy stages to the prediction accuracy are still lacking. More importantly, current models still face notable obstacles in practical applications due to the complex and diverse input features and difficulties in redeployment. As a result, a simple, practical but accurate enough model is urgently needed.

Design and methods

In this study, 2309 samples from two public hospitals in Shenzhen, China were collected for analysis. Different algorithms were systematically compared to build a robust and stepwise prediction system (level A to C) based on advanced machine learning, and models under different levels were interpreted.

Results

XGBoost reported the best performance with ACC of 0.922, 0.859 and 0.850, AUC of 0.974, 0.924 and 0.913 for the selected level A to C models in the test set, respectively. Tree-based feature importance and SHAP method successfully identified the commonly recognized risk factors, while indicated new inconsistent impact trends for GDM in different stages of pregnancy.

Conclusion

A stepwise prediction system was successfully established. A practical tool that enables a quick prediction of GDM was released at https://github.com/ifyoungnet/MedGDM.This study is expected to provide a more detailed profiling of GDM risk and lay the foundation for the application of the model in practice.

Association between blood urea nitrogen levels and diabetic retinopathy in diabetic adults in the United States (NHANES 2005-2018)

Objective

To investigate the association between blood urea nitrogen (BUN) levels and diabetic retinopathy (DR) in adults with diabetes mellitus (DM).

Methods

Seven cycles of cross-sectional population information acquired from NHANES(national health and nutrition examination surveys) 2005-2018 were collected, from which a sample of diabetic adults was screened and separated into two groups based on whether or not they had DR, followed by weighted multivariate regression analysis. This study collected a complete set of demographic, biological, and sociological risk factor indicators for DR. Demographic risk factors comprised age, gender, and ethnicity, while biological risk factors included blood count, blood pressure, BMI, waist circumference, and glycated hemoglobin. Sociological risk factors included education level, deprivation index, smoking status, and alcohol consumption.

Results

The multiple regression model revealed a significant connection between BUN levels and DR [odds ratio =1.04, 95% confidence interval (1.03-1.05), p-value <0.0001],accounting for numerous variables. After equating BUN levels into four groups, multiple regression modeling showed the highest quartile (BUN>20 mg/dl) was 2.22 times more likely to develop DR than the lowest quartile [odds ratio =2.22, 95% confidence interval (1.69-2.93), p- value <0.0001]. Subgroup analyses revealed that gender, race, diabetes subtype, and duration of diabetes had a regulating effect on the relationship between BUN and DR.

Conclusion

BUN levels were related with an increased prevalence of DR, particularly in individuals with BUN >20 mg/dl. These findings highlight the significance of BUN level in assessing the risk of DR.

Association between blood urea nitrogen levels and the risk of diabetes mellitus in Chinese adults: secondary analysis based on a multicenter, retrospective cohort study

Background

As one of the recognized indicators of kidney function, blood urea nitrogen (BUN) is a key marker of metabolic diseases and other diseases. Currently, data on the relationship of BUN levels with the risk of diabetes mellitus (DM) in Chinese adults are sparse. This study aimed to investigate the correlation between BUN levels and DM risk in Chinese adults.

Data and methods

This study is a secondary analysis of a multicenter, retrospective cohort study with data from the Chinese health screening program in the DATADRYAD database. From 2010 to 2016, health screening was conducted on 211833 Chinese adults over the age of 20 in 32 locations and 11 cities in China, and there was no DM at baseline. Cox proportional hazards regression analysis assessed an independent correlation between baseline BUN levels and the risk of developing DM. The Generalized Sum Model (GAM) and smoothed curve fitting methods were used to explore the nonlinear relationship. In addition, subgroup analyses were performed to assess the consistency of correlations between different subgroups and further validate the reliability of the results.

Results

After adjusting for potential confounding factors (age, sex, etc.), BUN levels were positively correlated with the occurrence of DM (HR=1.11, 95% CI (1.00~1.23)). BUN level had a nonlinear relationship with DM risk, and its inflection point was 4.2mmol/L. When BUN was greater than 4.2mmol/L, BUN was positively correlated with DM, and the risk of DM increased by 7% for every 1 mmol/L increase in BUN (P<0.05). Subgroup analysis showed that a more significant correlation between BUN levels and DM was observed in terms of sex, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL), alaninetransaminase (ALT), aspartate transaminase (AST), creatinine (Cr) and smoking status (interaction P<0.05).

Conclusion

High levels of BUN are associated with an increased risk of DM in Chinese adults, suggesting that active control of BUN levels may play an important role in reducing the risk of DM in Chinese adults.

Association between systemic immune-inflammation index and diabetes: a population-based study from the NHANES

Background

Systemic Immune-Inflammation Index (SII) has been reported to be associated with diabetes. We aimed to assess possible links between SII and diabetes.

Methods

Data were obtained from the 2017-2020 National Health and Nutrition Examination Survey (NHANES) database. After removing missing data for SII and diabetes, we examined patients older than 20 years. Simultaneously, the relationship between SII and diabetes was examined using weighted multivariate regression analysis, subgroup analysis, and smooth curve fitting.

Results

There were 7877 subjects in this study, the average SII was 524.91 ± 358.90, and the prevalence of diabetes was 16.07%. Weighted multivariate regression analysis found that SII was positively associated with diabetes, and in model 3, this positive association remained stable (OR = 1.04; 95% CI: 1.02-1.06; p = 0.0006), indicating that each additional unit of SII, the possibility of having diabetes increased by 4%. Gender, age, BMI, regular exercise, high blood pressure, and smoking did not significantly affect this positive link, according to the interaction test (p for trend>0.05).

Discussion

Additional prospective studies are required to examine the precise connection between higher SII levels and diabetes, which may be associated with higher SII levels.

MIDO GDM: an innovative artificial intelligence-based prediction model for the development of gestational diabetes in Mexican women

Given the barriers to early detection of gestational diabetes mellitus (GDM), this study aimed to develop an artificial intelligence (AI)-based prediction model for GDM in pregnant Mexican women. Data were retrieved from 1709 pregnant women who participated in the multicenter prospective cohort study 'Cuido mi embarazo'. A machine-learning-driven method was used to select the best predictive variables for GDM risk: age, family history of type 2 diabetes, previous diagnosis of hypertension, pregestational body mass index, gestational week, parity, birth weight of last child, and random capillary glucose. An artificial neural network approach was then used to build the model, which achieved a high level of accuracy (70.3%) and sensitivity (83.3%) for identifying women at high risk of developing GDM. This AI-based model will be applied throughout Mexico to improve the timing and quality of GDM interventions. Given the ease of obtaining the model variables, this model is expected to be clinically strategic, allowing prioritization of preventative treatment and promising a paradigm shift in prevention and primary healthcare during pregnancy. This AI model uses variables that are easily collected to identify pregnant women at risk of developing GDM with a high level of accuracy and precision.

Establishment and validation of a predictive nomogram for gestational diabetes mellitus during early pregnancy term: A retrospective study

Objective

This study aims to develop and evaluate a predictive nomogram for early assessment risk factors of gestational diabetes mellitus (GDM) during early pregnancy term, so as to help early clinical management and intervention.

Methods

A total of 824 pregnant women at Zhongnan Hospital of Wuhan University and Maternal and Child Health Hospital of Hubei Province from 1 February 2020 to 30 April 2020 were enrolled in a retrospective observational study and comprised the training dataset. Routine clinical and laboratory information was collected; we applied least absolute shrinkage and selection operator (LASSO) logistic regression and multivariate ROC risk analysis to determine significant predictors and establish the nomogram, and the early pregnancy files (gestational weeks 12-16, n = 392) at the same hospital were collected as a validation dataset. We evaluated the nomogram via the receiver operating characteristic (ROC) curve, C-index, calibration curve, and decision curve analysis (DCA).

Results

We conducted LASSO analysis and multivariate regression to establish a GDM nomogram during the early pregnancy term; the five selected risk predictors are as follows: age, blood urea nitrogen (BUN), fibrinogen-to-albumin ratio (FAR), blood urea nitrogen-to-creatinine ratio (BUN/Cr), and blood urea nitrogen-to-albumin ratio (BUN/ALB). The calibration curve and DCA present optimal predictive power. DCA demonstrates that the nomogram could be applied clinically.

Conclusion

An effective nomogram that predicts GDM should be established in order to help clinical management and intervention at the early gestational stage.

Associations between urea nitrogen and risk of depression among subjects with and without type 2 diabetes: A nationwide population-based study

Background

Depression and type 2 diabetes (T2D) are serious public health problems with irreversible health consequences and a significant economic burden on the healthcare system. Previous studies have suggested that blood urea nitrogen (BUN) was inversely longitudinally associated with incidence of diabetes and depression in adults, but few well-designed studies have examined the effects of status of T2D on the full range of relationship between BUN and depression.

Methods

The analysis sample consisted of adults aged≥20 years from the 2007-2014 National Health and Nutrition Examination Survey (NHANES) who completed the Patient Health Questionnaire-9 (PHQ-9), involving 19,005 participants. By stratifying participants according to T2D status, we further assessed the difference between BUN and risk of depression in participants with and without T2D using multivariate logistic regression (interaction test).

Results

In this cross-sectional study, the association between BUN and depression prevalence appeared to differ between the T2D and non-T2D groups (OR: 1.00, 95% Cl: 0.95-1.05 vs. OR: 0.89, 95% Cl: 0.85-0.93). In addition, there was evidence of an interaction between BUN levels and T2D status in reducing the risk of depression (P value for interaction = 0.032.) The relationship between BUN and depressive symptoms was significant in non-T2D subjects (P < 0.001), but not in T2D (P = 0.940).

Conclusions

Our findings suggest that there is a significant relationship between BUN and depression, and T2D status may influence the association between BUN and the risk of depression. Such findings require further prospective studies to provide more evidence.

Serum uric acid in early pregnancy and risk of gestational diabetes mellitus: a cohort study of 85,609 pregnant women

AIMS

. - Higher serum uric acid (UA) has been associated with increased risk of type 2 diabetes mellitus. This cohort study examined whether there are any associations between serum UA in early pregnancy and the subsequent risk of gestational diabetes mellitus (GDM).

METHODS

. - This cohort study was conducted in Shanghai, China, and included 85,609 pregnant women. Generalised additive models were used to estimate the associations of serum UA with risk of GDM.

RESULTS

. - The prevalence of GDM was 14.0% (11,960/85,609). Non-linear associations between serum UA and GDM risk were observed and these associations varied by gestational ages. Only elevated serum UA levels at 13-18 weeks gestation was associated with substantially increased risk of GDM. Analysis by UA quintiles at 13-18 weeks gestation showed the odds ratios for GDM were 1.11 (95%CI, 1.03-1.20) for the second, 1.27 (95%CI, 1.17-1.37) for the third, 1.37 (95%CI, 1.27-1.48) for the fourth and 1.70 (95%CI, 1.58-1.84) for the fifth quintile of serum UA in comparison with the first quintile. Stratified analysis showed the associations of serum UA with GDM were stronger among pregnant women aged 35 years or older.

CONCLUSION

. - We found higher serum UA at 13-18 gestational weeks was a risk factor for GDM. Our findings provide new evidence for the role of serum UA in the prevention and early intervention of GDM, and highlighted the need for monitoring serum UA at 13-18 gestational weeks.

A Nomogram Model that Predicts the Risk of Diabetic Nephropathy in Type 2 Diabetes Mellitus Patients: A Retrospective Study

OBJECTIVE

To construct a novel nomogram model that predicts the risk of diabetic nephropathy (DN) incidence in Chinese patients with type 2 diabetes mellitus (T2DM).

METHODS

Questionnaire surveys, physical examinations, routine blood tests, and biochemical index evaluations were conducted on 1095 patients with T2DM from Guilin. A least absolute contraction selection operator (LASSO) regression and multivariable logistic regression analysis were used to screen out DN risk factors. A logistic regression analysis incorporating the screened risk factors was used to establish a predictive nomogram model. The performance of the nomogram model was evaluated using the C-index, an area under the receiver operating characteristic curve (AUC), calibration plots, and a decision curve analysis. Bootstrapping was applied for internal validation.

RESULTS

Independent predictors for DN incidence risk included gender, age, hypertension, medicine use, duration of diabetes, body mass index, blood urea nitrogen level, serum creatinine level, neutrophil to lymphocyte ratio, and red blood cell distribution width. The nomogram model exhibited moderate prediction ability with a C-index of 0.819 (95% confidence interval (CI): 0.783-0.853) and an AUC of 0.813 (95%CI: 0.778-0.848). The C-index from internal validation reached 0.796 (95%CI: 0.763-0.829). The decision curve analysis displayed that the DN risk nomogram was clinically applicable when the risk threshold was between 1 and 83%.

CONCLUSION

Our novel and simple nomogram containing 10 factors may be useful in predicting DN incidence risk in T2DM patients.

Circulating FABP4, nesfatin-1, and osteocalcin concentrations in women with gestational diabetes mellitus: a meta-analysis

OBJECTIVE

Recent studies have investigated the circulating adipocyte fatty acid binding protein (FABP4), nesfatin-1, and osteocalcin (OC) concentrations in women diagnosed with gestational diabetes mellitus (GDM), but the findings prove to be conflicting. The objective of this research was to systematically assess the relationship of circulating levels of above adipokines with GDM.

METHODS

Pubmed, Embase, Web of Science, Cochrane library, OVID, and Scopus were performed to locate articles published up to January 31, 2020. Pooled standard mean differences (SMDs) with 95% confidence intervals (CIs), and 95% predictive intervals (PIs) were calculated by random-effects models to compare levels of adipokines between GDM cases and control groups. Cumulative and single-arm meta-analyses were also performed.

RESULTS

Thirty-one studies comprising 4590 participants were included. No significant differences were found between GDM women and healthy controls in circulating nesfatin-1 levels (4.56 vs. 5.02 ng/mL; SMD = - 0.11, 95% CI -0.61-0.38, 95% PI -1.63-1.41). Nevertheless, circulating FABP4 and OC levels observed in GDM women outnumbered normal controls (FABP4, 23.68 vs. 16.04 ng/mL; SMD = 2.99, 95% CI 2.28-3.69, 95% PI 0.28-5.71; OC, 52.34 vs. 51.04 ng/mL; SMD = 0.68, 95% CI 0.31-1.05, 95% PI -0.48-1.84). The cumulative meta-analysis showed that the SMDs of circulating FABP4 and OC levels had stabilized between the two groups.

CONCLUSIONS

Elevated circulating FABP4 and OC levels were observed in GDM women, but nesfatin-1 levels did not change, the PI of OC crossed the no-effect threshold. The results suggested that FABP4 is more suitable as a biomarker of GDM compared to OC in a future study, which is useful in identifying pregnant women who are likely to develop GDM and providing prompt management strategies.

First-trimester blood urea nitrogen and risk of gestational diabetes mellitus

Prior studies indicated that urea increased insulin resistance and higher blood urea nitrogen (BUN) was associated with incident diabetes mellitus. However, it remains unclear whether BUN during the first trimester of pregnancy increases risk of gestational diabetes mellitus (GDM). We aimed to investigate the association between first‐trimester BUN and risk of incident GDM. We conducted a prospective, multicenter cohort study of pregnant women. A total of 13 448 eligible pregnant women with measured first‐trimester BUN levels were included in this analysis. Logistic regression analysis was used to estimate the relationship between BUN and GDM. Discrimination and reclassification for GDM by BUN were analysed. A total of 2973 (22.1%) women developed GDM. Compared with the lowest quartile of BUN, the third and fourth quartiles were associated with increased risk of GDM (adjusted odds ratios 1.21 [95% CI 1.07‐1.37] and 1.50 [95% CI 1.33‐1.69], respectively, P for trend <.001). The addition of BUN to conventional factor model improved discrimination (C statistic 0.2%, P = .003) and reclassification (net reclassification index 14.67%, P < .001; integrated discrimination improvement 0.12%, P < .001) for GDM. In conclusion, higher BUN concentrations during the first trimester of pregnancy were associated with increased risk of GDM, suggesting that BUN could be a potential predictor for GDM.

Association of Serum Uric Acid, Urea Nitrogen, and Urine Specific Gravity Levels at 16-18 Weeks of Gestation with the Risk of Gestational Diabetes Mellitus

OBJECTIVE

To evaluate the associations of serum uric acid (UA), urea nitrogen (UN), and urine specific gravity (USG) levels in the first trimester of pregnancy with the risk of gestational diabetes mellitus (GDM).

PATIENTS AND METHODS

A retrospective cohort study was conducted in 1,769 pregnant women aged 31.55 ± 3.91 years. UA, UN, and USG levels were measured during the 16-18th week of gestation. GDM was diagnosed by an oral 75 g glucose tolerance test during the 24-28th week of gestation.

RESULTS

A multivariate adjusted logistic regression analysis showed that UA levels in the highest quartile increased the risk of GDM by 55.7% (odds ratio [OR]: 1.557, 95% confidence interval [CI]: 1.055-2.298; p = 0.026) compared to those in the lowest quartile. USG levels in the second, third, and fourth quartiles increased the risk of GDM by 67.6% (95% CI: 1.090-2.421), 112.4% (95% CI: 1.446-3.119), and 94.5% (95% CI: 1.314-2.880), respectively, compared to those in the first quartile (p trend = 0.001). No significant association between UN levels and the GDM risk was observed. When the extreme composite biomarker score quartiles were compared, the adjusted OR (95% CI) for GDM was 1.909 (95% CI: 1.332-2.736). Age-stratified analyses revealed similar results in women aged ≤35 years only, but not in those aged >35 years.

CONCLUSION

Higher levels of UA and USG and a higher composite kidney function biomarker score during the 16-18th week of gestation were positively and independently associated with an increased risk of GDM.