Deep learning model to discriminate diverse infection types based on pairwise analysis of host gene expression.

Accurate detection of pathogens, particularly distinguishing between Gram-positive and Gram-negative bacteria, could improve disease treatment. Host gene expression can capture the immune system's response to infections caused by various pathogens. Here, we present a deep neural network model, bvnGPS2, which incorporates the attention mechanism based on a large-scale integrated host transcriptome dataset to precisely identify Gram-positive and Gram-negative bacterial infections as well as viral infections. We performed analysis of 4,949 blood samples across 40 cohorts from 10 countries using our previously designed omics data integration method, iPAGE, to select discriminant gene pairs and train the bvnGPS2. The performance of the model was evaluated on six independent cohorts comprising 374 samples. Overall, our deep neural network model shows robust capability to accurately identify specific infections, paving the way for precise medicine strategies in infection treatment and potentially also for identifying subtypes of other diseases.

The relationship between sarcopenia and mortality in Chinese community-dwelling adults: a 7-year cohort study with propensity score matching and Mendelian randomization.

Background: Sarcopenia has been linked to adverse health outcomes, including an increased risk of mortality. This study aimed to assess the 7-year mortality risk of sarcopenia in a community-based population in China and explore the causal relationship between components of sarcopenia and any death. Methods: Data were sourced from the China Health and Retirement Longitudinal Study (CHARLS) conducted between 2011 and 2018. Sarcopenia was diagnosed using the Asian Working Group for Sarcopenia (AWGS) 2019 criteria. Logistic regression, Kaplan-Meier (KM) survival analysis, and propensity score matching with inverse probability of treatment weighting were used. Mendelian randomization (MR) analyses, conducted using European population data, were utilized to assess causality between sarcopenia and any death. Results: The study included 9,006 participants: 3,892 had no sarcopenia, 3,570 had possible sarcopenia, 1,125 had sarcopenia, and 419 had severe sarcopenia. Over 7 years of follow-up, there were 871 deaths, including 196 with sarcopenia and 133 with severe sarcopenia. The KM curves showed that sarcopenia had a higher risk of mortality. Compared to those of no sarcopenia, the odds ratios (ORs) of sarcopenia for 7-year mortality were 1.41 (95% CI, 1.06-1.87) after adjusting for confounding variables (p < 0.05). The ORs of severe sarcopenia were 2.11 (95% CI, 1.51-2.95). Propensity score matching analysis and inverse probability of treatment weighting analysis confirmed these findings. The adjusted ORs of sarcopenia and 7-year mortality were 2.94 (95% CI, 1.6-5.39) in the 45-60 age group, 1.72 (95% CI, 1.11-2.68) in the 60-80 age group, and 5.03 (95% CI, 0.48-52.65) in the ≥80 age group. The ORs of severe sarcopenia and 7-year mortality were 6.92 (95% CI, 1.95-24.5) in the 45-60 age group, 2.59 (95% CI, 1.61-4.17) in the 60-80 age group, and 12.52 (95% CI, 1.18-133.18) in the ≥80 age group. The MR analyses, leveraging the inverse variance weighted (IVW) method, unveiled substantial causal links between low hand grip strength in individuals aged 60 and older, the usual walking pace, and mortality risk. Conclusion: This study underscores the significant impact of sarcopenia and its components on mortality risk within the Chinese population. Particularly, low hand grip strength and usual walking pace emerged as noteworthy contributors to mortality risk.

The association of handgrip strength with all-cause and cardiovascular mortality: results from the National Health and Nutrition Examination Survey database prospective cohort study with propensity score matching.

Objective: To investigate the association between handgrip strength (HGS) with all-cause and cardiovascular disease (CVD) mortality in US adults. Method: We analyzed data from the National Health and Nutrition Examination Survey (NHANES) prospective cohort study (2011-2014) with 10,470 participants. The cox regression analysis, Kaplan-Meier survival curves, fitted curves, ROC curves, and propensity score-matched analysis (PSM) with inverse probability of treatment weighting (IPTW), SMRW (PSM with repeated weights), PA (pairwise algorithm), and OW (overlap weighting) regression analysis were performed to assess the relationship between HGS and all-cause and CVD mortality. Results: The low HGSs (men <37.4 kg, women <24 kg), was found to be associated with higher all-cause and CVD mortality in a reverse J-shaped curve (p < 0.05). Adjusting for multiple covariates including age, BMI, race, education level, marriage status, smoking and alcohol use, and various comorbidities, the hazard ratio (HR) for all-cause mortality in the lowest HGS quintile 1 (Q1) was 3.45 (2.14-5.58) for men and 3.3 (1.88-5.79) for women. For CVD mortality, the HR was 2.99 (1.07-8.37) for men and 10.35 (2.29-46.78) for women. The area under the curve (AUC) for HGS alone as a predictor of all-cause mortality was 0.791 (0.768-0.814) for men and 0.780 (0.752-0.807) for women (p < 0.05), while the AUC for HGS and age was 0.851 (0.830-0.871) for men and 0.848 (0.826-0.869) for women (p < 0.05). For CVD mortality, the AUC for HGS alone was 0.785 (95% CI 0.738-0.833) for men and 0.821 (95% CI 0.777-0.865) for women (p < 0.05), while the AUC for HGS and age as predictors of all-cause mortality was 0.853 (0.861-0.891) for men and 0.859 (0.821-0.896) for women (p < 0.05). The HGS Q1 (men <37.4 kg and women <24 kg) was matched separately for PSM. After univariate, multivariate Cox regression models, PSM, IPTW, SMRW, PA, and OW analyses, women had 2.37-3.12 and 2.92-5.12 HRs with low HGS for all-cause and CVD mortality, while men had 2.21-2.82 and 2.33-2.85 for all-cause and CVD mortality, respectively (p < 0.05). Conclusion: Adults with low HGS exhibited a significantly increased risk of both all-cause and CVD mortality, regardless of gender. Additionally, low HGS served as an independent risk factor and predictor for both all-cause and CVD mortality.

Using machine learning algorithms for predicting cognitive impairment and identifying modifiable factors among Chinese elderly people.

Objectives: This study firstly aimed to explore predicting cognitive impairment at an early stage using a large population-based longitudinal survey of elderly Chinese people. The second aim was to identify reversible factors which may help slow the rate of decline in cognitive function over 3 years in the community. Methods: We included 12,280 elderly people from four waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS), followed from 2002 to 2014. The Chinese version of the Mini-Mental State Examination (MMSE) was used to examine cognitive function. Six machine learning algorithms (including a neural network model) and an ensemble method were trained on data split 2/3 for training and 1/3 testing. Parameters were explored in training data using 3-fold cross-validation and models were evaluated in test data. The model performance was measured by area-under-curve (AUC), sensitivity, and specificity. In addition, due to its better interpretability, logistic regression (LR) was used to assess the association of life behavior and its change with cognitive impairment after 3 years. Results: Support vector machine and multi-layer perceptron were found to be the best performing algorithms with AUC of 0.8267 and 0.8256, respectively. Fusing the results of all six single models further improves the AUC to 0.8269. Playing more Mahjong or cards (OR = 0.49,95% CI: 0.38-0.64), doing more garden works (OR = 0.54,95% CI: 0.43-0.68), watching TV or listening to the radio more (OR = 0.67,95% CI: 0.59-0.77) were associated with decreased risk of cognitive impairment after 3 years. Conclusions: Machine learning algorithms especially the SVM, and the ensemble model can be leveraged to identify the elderly at risk of cognitive impairment. Doing more leisure activities, doing more gardening work, and engaging in more activities combined were associated with decreased risk of cognitive impairment.

Prototypical Graph Contrastive Learning.

Graph-level representations are critical in various real-world applications, such as predicting the properties of molecules. However, in practice, precise graph annotations are generally very expensive and time-consuming. To address this issue, graph contrastive learning constructs an instance discrimination task, which pulls together positive pairs (augmentation pairs of the same graph) and pushes away negative pairs (augmentation pairs of different graphs) for unsupervised representation learning. However, since for a query, its negatives are uniformly sampled from all graphs, existing methods suffer from the critical sampling bias issue, i.e., the negatives likely having the same semantic structure with the query, leading to performance degradation. To mitigate this sampling bias issue, in this article, we propose a prototypical graph contrastive learning (PGCL) approach. Specifically, PGCL models the underlying semantic structure of the graph data via clustering semantically similar graphs into the same group and simultaneously encourages the clustering consistency for different augmentations of the same graph. Then, given a query, it performs negative sampling via drawing the graphs from those clusters that differ from the cluster of query, which ensures the semantic difference between query and its negative samples. Moreover, for a query, PGCL further reweights its negative samples based on the distance between their prototypes (cluster centroids) and the query prototype such that those negatives having moderate prototype distance enjoy relatively large weights. This reweighting strategy is proven to be more effective than uniform sampling. Experimental results on various graph benchmarks testify the advantages of our PGCL over state-of-the-art methods. The code is publicly available at https://github.com/ha-lins/PGCL.

Association of hypertensive disorders of pregnancy with infant growth in the first 36 months of life.

To investigate the association of chronic hypertension, gestational hypertension, and preeclampsia diseases with infant growth in the first 36 months of life, we conducted a retrospective birth cohort of 31,734 children born in Zhoushan Maternal and Child Care Hospital between January 2001 and May 2018. Birthweight, gestational age, and infant growth (weight, height, weight/height-for-age Z score, the weight gain during childhood) were the main outcomes. The associations of chronic hypertension, gestational hypertension, and preeclampsia diseases with birth outcomes and infant growth at children's age of 3, 6, 12, 18, and 24 months were analyzed by multivariable regression models. Gestational hypertension, preeclampsia diseases, and chronic hypertension were significantly associated with lower birthweight and shorter gestational age. Both gestational hypertension and preeclampsia diseases were respectively inversely associated with weight, weight-for-age Z score, height, and height-for-age Z score of children in the whole sample and sub-sample data analysis from birth to the age of 36 months, although correction for birthweight rendered the associations nonsignificant. No significant association of gestational hypertension, preeclampsia diseases, and chronic hypertension with weight gain was found. Conclusion: The inverse associations of gestational hypertension and preeclampsia diseases with infant growth in early childhood were mainly mediated by the effect of gestational hypertension and preeclampsia diseases on lower birthweight. What is Known: • Hypertensive disorders of pregnancy are associated with increased risk of adverse birth outcomes. What is New: • Both gestational hypertension and preeclampsia were respectively inversely associated with physical development of offspring from birth to the age of 36 months. • Lower birthweight might be the mediator of the inverse associations of gestational hypertension and preeclampsia diseases with infant growth in early childhood.

The Association of Gene Variants in the Vitamin D Metabolic Pathway and Its Interaction with Vitamin D on Gestational Diabetes Mellitus: A Prospective Cohort Study.

The present prospective study included 2156 women and investigated the effect of gene variants in the vitamin D (VitD) metabolic and glucose pathways and their interaction with VitD levels during pregnancy on gestational diabetes mellitus (GDM). Plasma 25(OH)D concentrations were measured at the first and second trimesters. GDM subtype 1 was defined as those with isolated elevated fasting plasma glucose; GDM subtype 2 were those with isolated elevated postprandial glucose at 1 h and/or 2 h; and GDM subtype 3 were those with both elevated fasting plasma glucose and postprandial glucose. Six Gc isoforms were categorized based on two GC gene variants rs4588 and rs7041, including 1s/1s, 1s/2, 1s/1f, 2/2, 1f/2 and 1f/1f. VDR-rs10783219 and MTNR1B-rs10830962 were associated with increased risks of GDM and GDM subtype 2; interactions between each other as well as with CDKAL1-rs7754840 were observed (Pinteraction < 0.05). Compared with the 1f/1f isoform, the risk of GDM subtype 2 among women with 1f/2, 2/2, 1s/1f, 1s/2 and 1s/1s isoforms and with prepregnancy body mass index ≥24 kg/m2 increased by 5.11, 10.01, 10, 14.23, 19.45 times, respectively. Gene variants in VitD pathway interacts with VitD deficiency at the first trimester on the risk of GDM and GDM subtype 2.

Identifying urban built environment factors in pregnancy care and maternal mental health outcomes.

BACKGROUNDS: Risk factors related to the built environment have been associated with women's mental health and preventive care. This study sought to identify built environment factors that are associated with variations in prenatal care and subsequent pregnancy-related outcomes in an urban setting. METHODS: In a retrospective observational study, we characterized the types and frequency of prenatal care events that are associated with the various built environment factors of the patients' residing neighborhoods. In comparison to women living in higher-quality built environments, we hypothesize that women who reside in lower-quality built environments experience different patterns of clinical events that may increase the risk for adverse outcomes. Using machine learning, we performed pattern detection to characterize the variability in prenatal care concerning encounter types, clinical problems, and medication prescriptions. Structural equation modeling was used to test the associations among built environment, prenatal care variation, and pregnancy outcome. The main outcome is postpartum depression (PPD) diagnosis within 1 year following childbirth. The exposures were the quality of the built environment in the patients' residing neighborhoods. Electronic health records (EHR) data of pregnant women (n = 8,949) who had live delivery at an urban academic medical center from 2015 to 2017 were included in the study. RESULTS: We discovered prenatal care patterns that were summarized into three common types. Women who experienced the prenatal care pattern with the highest rates of PPD were more likely to reside in neighborhoods with homogeneous land use, lower walkability, lower air pollutant concentration, and lower retail floor ratios after adjusting for age, neighborhood average education level, marital status, and income inequality. CONCLUSIONS: In an urban setting, multi-purpose and walkable communities were found to be associated with a lower risk of PPD. Findings may inform urban design policies and provide awareness for care providers on the association of patients' residing neighborhoods and healthy pregnancy.

Association of vitamin D and gene variants in the vitamin D metabolic pathway with preterm birth.

OBJECTIVES: The aim of this study was to explore the association of vitamin D (VitD) levels during pregnancy and its metabolic pathway genes with the risk for preterm birth (PTB) among pregnant women in southeast China. METHODS: This study was conducted in Zhoushan Maternal and Child Health Hospital, Zhejiang, from August 2011 to May 2018. Plasma 25-hydroxyvitamin vitamin D [25(OH)D] levels in three trimesters and single-nucleotide morphisms in the VitD metabolic pathway were measured. Relevant information was collected using questionnaires and an electronic medical recorder system. Multiple statistical methods including linear regression, logistic regression, and crossover analysis were applied. RESULTS: The prospective cohort study included 3465 pregnant women, of which 202 were PTB (week of gestation at delivery: 33.38 ± 4.05), accounting for 5.8%. After adjusting for potential confounders, VitD sufficiency (≥30 ng/mL) in the second and third trimesters was associated with longer gestational age at delivery compared with VitD deficiency (<20 ng/mL). However, no significant association was found between VitD with the risk for PTB. rs7041, rs10210408, and rs2228171 were associated with gestational week and the risk for PTB. Significant associations were found of rs10210408, rs2209314, rs1155563, rs2544381 and the status of VitD in the second and third trimester with the gestational week. We also found that rs7041 and VitD in the second trimester might exert interaction on gestational week and the risk for PTB (Pinter = 0.038; Pinter = 0.019); rs16846876 and VitD in the second trimester might exert interaction on gestational week (Pinter = 0.024); rs4334089 and VitD in the third trimester might exert interaction on gestational week (Pinter = 0.024). Similar results were found when we tested pregnant women's plasma 25(OH)D in the first and second trimesters. CONCLUSIONS: Women with VitD deficiency were associated with shorter gestational weeks. Single-nucleotide morphisms in VitD metabolic pathway genes were significantly associated with gestation week and the risk for PTB, mainly in vitamin D-binding protein (GC) and low-density lipoprotein-related protein 2 (LRP2)genes. Additionally, maternal VitD with GC gene and maternal VitD with vitamin D receptor (VDR) gene might exert interactions on the risk for PTB.

Association of vitamin D pathway gene polymorphisms with vitamin D level during pregnancy was modified by season and vitamin D supplement.

BACKGROUND & AIMS: This study aims to explore the associations of vitamin D (VD) metabolic pathway gene with 25(OH)D level in pregnant women and the interactions of SNP with season and VD supplement. METHODS: A total of 2658 pregnant women were selected from Zhoushan Pregnant Women Cohort study. Gestational 25(OH)D level and single nucleotide polymorphism (SNP) of VD metabolic pathway gene were detected. Multilinear regression models were used to estimate associations of SNPs with gestational 25(OH)D levels. Stratified analyses were performed to test the interactions of SNP with season and VD supplements. RESULTS: The mutations of rs2298849 and rs7041 on the GC gene were respectively associated with higher 25(OH)D in the first and third trimester; the mutations of seven SNPs (rs1155563, rs16846876, rs17467825, rs2282679, rs2298850, rs3755967, and rs4588) on the GC gene were respectively associated with lower 25(OH)D both in the first and third trimester, and lower changes in 25(OH)D during late pregnancy. The mutations of above seven SNPs, except for rs1155563, were also respectively associated with lower 25(OH)D in the second trimester, but to a lesser extent; Besides, pregnant women with mutation on CYP24A1-rs2209314 had a higher increment in 25(OH)D than their counterparts in the second trimester. The increasing dose effect of Gc isoform on 25(OH)D was observed. The associations of GC and LRP2 genes with 25(OH)D modified by season and VD supplements. CONCLUSIONS: The polymorphisms of VD metabolic pathway gene were associated with gestational 25(OH)D, and the associations differ by seasons and VD supplements. Gc isoform exerted a profound influence on gestational 25(OH)D.

Changing patterns of prenatal depression and anxiety status in different trimesters and modified form of Zung scales for pregnant women.

BACKGROUND: We described the changing patterns of depression and anxiety status in different trimesters among Chinese pregnant women, and identified the modified form of SDS/SAS for pregnant women and assessed its reliability and validity. METHODS: Changing patterns of depression/ anxiety status in different trimesters were described. The modified form of SDS/SAS was identified for pregnant women. Cohen's Kappa to measure agreement with SDS/SAS, and the ROC analysis was performed to assess its validity. RESULTS: The SDS score in 1st trimester was higher than 2nd and 3rd trimester; there was no significant difference between SDS score in 2nd and 3rd trimester. Modified form of SDS evaluated the depression; the areas under the curve (AUC) in testing group were up to 0.988, 0.989 and 0.992 for 1st, 2nd and 3rd trimester, respectively. Modified form of SAS evaluated the anxiety, the AUC in testing group were up to 0.987, 0.985, 0.987 for 1st, 2nd and 3rd trimester, respectively. CONCLUSION: Pregnant women had higher severity of depression and anxiety status in 1st trimester than that in 2nd and 3rd trimester. The modified form of SDS/SAS may be more brief and suitable to assess the depression and anxiety status in pregnant women.KEY POINTSPregnant women had a higher severity of depression and anxiety status in the 1st trimester than that in the 2nd and 3rd trimester.The present study suggests that prenatal depression and anxiety status are prevalent in Chinese pregnant women.Prevention or treatments focus on high-score items of SDS and SAS would be beneficial for rectifying prenatal depression and anxiety.

Development and validation of a machine learning algorithm for predicting the risk of postpartum depression among pregnant women.

OBJECTIVE: There is a scarcity in tools to predict postpartum depression (PPD). We propose a machine learning framework for PPD risk prediction using data extracted from electronic health records (EHRs). METHODS: Two EHR datasets containing data on 15,197 women from 2015 to 2018 at a single site, and 53,972 women from 2004 to 2017 at multiple sites were used as development and validation sets, respectively, to construct the PPD risk prediction model. The primary outcome was a diagnosis of PPD within 1 year following childbirth. A framework of data extraction, processing, and machine learning was implemented to select a minimal list of features from the EHR datasets to ensure model performance and to enable future point-of-care risk prediction. RESULTS: The best-performing model uses from clinical features related to mental health history, medical comorbidity, obstetric complications, medication prescription orders, and patient demographic characteristics. The model performances as measured by area under the receiver operating characteristic curve (AUC) are 0.937 (95% CI 0.912 - 0.962) and 0.886 (95% CI 0.879-0.893) in the development and validation datasets, respectively. The model performances were consistent when tested using data ending at multiple time periods during pregnancy and at childbirth. LIMITATIONS: The prevalence of PPD in the study data represented a treatment prevalence and is likely lower than the illness prevalence. CONCLUSIONS: EHRs and machine learning offer the ability to identify women at risk for PPD early in their pregnancy. This may facilitate scalable and timely prevention and intervention, reducing negative outcomes and the associated burden.

Hemoglobin concentration and iron supplement during pregnancy were associated with an increased risk of gestational diabetes mellitus.

BACKGROUND: Hemoglobin (Hb) measurement is a conventional test during perinatal visits. Hb concentration is related to iron supplement. However, studies focusing on Hb levels, iron supplement, and pregnancy outcomes are scarce. This study aimed to determine whether Hb levels and iron supplement were associated with the risk of gestational diabetes mellitus (GDM). METHODS: A running hospital-based cohort was conducted from August, 2011. The demographic data and medical information were collected individually through questionnaires and patient medical records. Multiple linear regression was applied for the association between Hb levels, iron supplement, and blood glucose. Multiple logistic regression was used for evaluating odds ratios between Hb levels, iron supplement, and GDM. RESULTS: Hb levels during first (T1) and second trimester (T2) of pregnancy were significantly and positively associated with blood glucose and GDM risk. After adjusting for age, prepregnancy body mass index, and other risk factors, pregnant women with Hb ≥ 11 g/dL and iron supplement had higher postprandial blood glucose at 1 hour (Hb ≥ 11 g/dL in T2 and iron supplement in T1: ß = 0.860，P = <0.001; Hb ≥ 11 g/dL in T2 and iron supplement in T2: ß = 0.960，P < 0.001; Hb ≥ 11 g/dL in T1 and iron supplement in T2: ß = 1.133, P = 0.033) and GDM risks (odds ratio [OR] = 1.53, 95% confidence interval [CI]: 1.05-2.24; OR = 1.92, 95% CI: 1.13-3.35; OR = 2.15, 95% CI: 1.07-4.34, respectively), compared with those with Hb < 11 g/dL and without iron supplement. CONCLUSION: High Hb concentration and iron supplements without anemia increased postprandial blood glucose and risks for GDM. It indicates that pregnant women with good Hb levels should not be advised to take iron supplements during pregnancy.

Forecasting the long-term trend of COVID-19 epidemic using a dynamic model.

The current outbreak of coronavirus disease 2019 (COVID-19) has recently been declared as a pandemic and spread over 200 countries and territories. Forecasting the long-term trend of the COVID-19 epidemic can help health authorities determine the transmission characteristics of the virus and take appropriate prevention and control strategies beforehand. Previous studies that solely applied traditional epidemic models or machine learning models were subject to underfitting or overfitting problems. We propose a new model named Dynamic-Susceptible-Exposed-Infective-Quarantined (D-SEIQ), by making appropriate modifications of the Susceptible-Exposed-Infective-Recovered (SEIR) model and integrating machine learning based parameter optimization under epidemiological rational constraints. We used the model to predict the long-term reported cumulative numbers of COVID-19 cases in China from January 27, 2020. We evaluated our model on officially reported confirmed cases from three different regions in China, and the results proved the effectiveness of our model in terms of simulating and predicting the trend of the COVID-19 outbreak. In China-Excluding-Hubei area within 7 days after the first public report, our model successfully and accurately predicted the long trend up to 40 days and the exact date of the outbreak peak. The predicted cumulative number (12,506) by March 10, 2020, was only 3·8% different from the actual number (13,005). The parameters obtained by our model proved the effectiveness of prevention and intervention strategies on epidemic control in China. The prediction results for five other countries suggested the external validity of our model. The integrated approach of epidemic and machine learning models could accurately forecast the long-term trend of the COVID-19 outbreak. The model parameters also provided insights into the analysis of COVID-19 transmission and the effectiveness of interventions in China.

Anemia and Diabetic Kidney Disease Had Joint Effect on Diabetic Retinopathy Among Patients With Type 2 Diabetes.

Purpose: Whether the association between diabetic kidney disease (DKD) and diabetic retinopathy (DR) in patients with type 2 diabetes mellitus (T2DM) is leveraged by anemia remains unclear. This study is to evaluate the joint effect of DKD and anemia on DR. Methods: Data were collected from electronic medical records of 1389 patients with T2DM in the Yiwu Central Hospital of Zhejiang Province from 2018 to 2019. Based on retinal examination findings, patients were classified as without diabetic retinopathy (non-DR), non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR). Odds ratio (OR) from multinomial logistic regression models adjusting for potential risk factors of DR were used to evaluate associations of DKD, renal function measures, and anemia with risk of NPDR and PDR. Path analysis was performed to help understand the association of DKD and hemoglobin (Hb) with DR. Results: The study included 901 patients with non-DR, 367 patients with NPDR and 121 patients with PDR. Both high DKD risk and abnormal renal function measures were significantly associated with PDR. Anemia was associated with increased risk of NPDR (OR = 1.75, 95% confidence interval [CI] = 1.18-2.58) and PDR (OR = 3.71, 95% CI = 2.23-6.18). DKD severity and anemia had joint effect on NPDR (OR = 2.29, 95% CI = 1.32-3.96) and PDR (OR = 11.31, 95% CI = 5.95-21.51). These associations were supported by path analysis. Conclusions: DKD severity, abnormal estimated glomerular filtration rate (eGFR), and urinary albumin/creatinine ratio (UACR) were associated with increased risk of DR in patients with T2DM, and anemia had joint effect on these associations. Improving Hb level may decrease the risk of DR in patients with T2DM.

Uric acid fluctuation had no effect on renal function among gout patients.

BACKGROUND AND AIMS: To investigate the effects of serum uric acid (SUA) level and its fluctuation on renal dysfunction in gout patients. METHODS AND RESULTS: Data on gout patients was collected from Huzhou city electronic medical record system data sharing platform, and information about relevant diagnoses, prescriptions, biochemical indexes and imaging characteristics was extracted. The gout patients with baseline normal renal function were enrolled in this analysis, and the estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2 was defined as renal dysfunction. The generalized estimating equation and Cox regression analysis were used. A total of 1009 patients with gout were enrolled. Compared with the reference group (normal baseline SUA with endpoint SUA to be < 6 mg/dL), endpoint SUA ≥ 10 mg/dL was associated with an increased risk of renal dysfunction (baseline normal SUA group: HR [95% CI] = 3.28 [1.21, 8.91]; baseline high SUA group: HR [95% CI] = 3.01 [1.43, 6.35]). Subgroup analysis of 771 SUA stable gout patients demonstrated that SUA levels at 8-10 (excluding 10), and ≥10 mg/dL were significantly associated with an increased risk for renal dysfunction, with HR [95%CI] to be 1.99 [1.05, 3.77], and 2.98 [1.38, 6.43], respectively. CONCLUSION: Regardless of the baseline SUA level, SUA >10 mg/dL was a significant risk factor for renal dysfunction. SUA between 6 and 10 mg/dL was a potential risk factor for renal dysfunction. No significant correlation of SUA fluctuation and renal function was found.

Vitamin D nutrient status during pregnancy and its influencing factors.

BACKGROUND: Little is known about variation of vitamin D (VD) status during pregnancy among Chinese women. This study is to assess the change of VD status during pregnancy and its influencing factors among Chinese women. METHODS: A running cohort study has being conducted in southeast China. The pregnant women were interviewed and the peripheral blood samples were collected at the first (T1), second (T2) and third trimester (T3), respectively. 25(OH)D2 and 25(OH)D3 were measured by liquid chromatography tandem-mass spectrometry. Multiple linear and logistic regression models were applied to explore the associations of VD supplement with 25(OH)D concentration and VD deficiency, respectively. RESULTS: There were 4368 pregnant women enrolled in the current study. The 25(OH)D concentration increased notably with gestational week. The average plasma 25(OH)D concentration in T1, T2 and T3 was 18.94 ± 8.74, 23.05 ± 11.15, and 24.65 ± 11.59 ng/mL, respectively. Correspondingly, VD deficiency (25(OH)D < 20 ng/mL) rate was 65.26%, 33.56% and 32.12%. In T1 phase, higher pre-pregnancy BMI, more parity, sampling in summer/autumn were related to higher 25(OH)D level, and similar patterns were observed in T2 and T3 phase. There was positive dose-response effect between VD supplement frequency and 25(OH)D concentration during pregnancy, adjusting for potential confounders (T1: ß(SE) = 3.907 (0.319), P < 0.001; T2: ß(SE) = 2.780 (0.805), P < 0.001; T3: ß(SE) = 3.640 (1.057), P = 0.006). Not surprisingly, supplementing VD > 3 times/week reduced the risk of VD deficiency during pregnancy significantly, compared to without VD supplement (T1: OR = 0.30, 95% CI: 0.24-0.37; T2: 0.56, 0.38-0.82; T3: 0.67, 0.44-0.96). CONCLUSION: VD level increased with gestational week among Chinese pregnant women. High frequency of VD supplement during pregnancy is an effective way to reduce risk of VD deficiency, especially among the pregnant women with younger age, low prepregnancy BMI and primipara, and during winter and spring season.

Feature of trajectory of blood pressure among pregnant women with gestational hypertension.

OBJECTIVE: We aimed to describe the feature and the trajectory of blood pressure (BP) among pregnant women with onset of gestational hypertension. METHODS: This epidemiology cohort study of pregnant women enrolled in Zhoushan, Zhejiang, included 4050 participants from the Zhoushan Pregnant Women Cohort. Each participant contributed up to eight serial perinatal visits of BP measurements from 2001 to 2018. Segmented mixed models were utilized to identify the dramatic change points in the relationship of BP elevation and gestational week among pregnant women with onset of gestational hypertension. RESULTS: Despite of in which gestational week gestational hypertension developed, the SBP and DBP levels of pregnant women in each gestational hypertension category maintained the stable and normal levels (SBP <119∼130 and DBP <76∼83 mmHg) before the accelerated point. However, after the accelerated point, BP dramatically developed to gestational hypertension in a very short time period. Meanwhile, the earlier gestational hypertension onset was, the higher the baseline BP were; meanwhile, the earlier the gestational hypertension onset was, the higher the difference in BP at the gestational week of gestational hypertension onset between women with and without gestational hypertension was. CONCLUSION: BP trajectories of gestational hypertension onset in different gestational week presented similar patterns. Meanwhile, the earlier gestational hypertension onset was, the higher the baseline BPs were. These findings show that BP monitoring during pregnancy is necessary, especially for women with high normal baseline BP.

The interaction between prepregnancy BMI and gestational vitamin D deficiency on the risk of gestational diabetes mellitus subtypes with elevated fasting blood glucose.

BACKGROUND & AIMS: To investigate the association of VitD with GDM, and examine the potential modifying effect of prepregnancy BMI in Chinese pregnant women. METHODS: 3318 pregnant women underwent oral glucose tolerance test (OGTT) were selected from Zhoushan Pregnant Women Cohort. Plasma VitD levels were measured in the first (T1) and/or second trimester (T2). Multiple linear and logistic regression models were used for evaluating the association of VitD with GDM. RESULTS: Prepregnancy BMI was positively associated with all three time-point glucose of OGTT. 25(OH)D level in T1 (ß = -0.003) and T2 (ß = -0.004), and its change from T1 to T2 (ß = -0.004) were significantly and inversely associated with fasting blood glucose (FBG) of OGTT, but not 1-h and 2-h postload blood glucose of OGTT, respectively. The negative associations of VitD and FBG were stronger among overweight/obese women. VitD deficiency (25(OH)D < 20 ng/ml) in T2 was associated with an increased risk of GDM with increased FBG, GDM subtype 1 (OR: 2.10) and subtype 3 (OR: 2.19). Moreover, prepregnancy BMI modified this effect on GDM subtype 1 (BMI < 24: OR = 1.42; BMI ≥ 24: OR = 9.61, P for interaction = 0.002). Lower VitD increment from T1 to T2 was associated with a higher risk for GDM among overweight/obese women. Additionally, GDM prevalence fluctuated with the season, i.e. lower in summer/fall and higher in winter/spring. CONCLUSIONS: Maternal VitD deficiency was associated with a higher risk of GDM subtype with increased FBG, and the risk is much greater among overweight/obesity women. The lower the VitD increment during pregnancy, the greater the risk of GDM, especially in overweight/obesity women. Furthermore, seasonal variation of GDM may be exhibited as a critical confounder in the association of VitD and GDM.

Using Electronic Health Records and Machine Learning to Predict Postpartum Depression.

Postpartum depression (PPD) is one of the most frequent maternal morbidities after delivery with serious implications. Currently, there is a lack of effective screening strategies and high-quality clinical trials. The ability to leverage a large amount of detailed patient data from electronic health records (EHRs) to predict PPD could enable the implementation of effective clinical decision support interventions. To develop a PPD prediction model, using EHRs from Weill Cornell Medicine and NewYork-Presbyterian Hospital between 2015-17, 9,980 episodes of pregnancy were identified. Six machine learning algorithms, including L2-regularized Logistic Regression, Support Vector Machine, Decision Tree, Naïve Bayes, XGBoost, and Random forest were constructed. Our model's best prediction performance achieved an AUC of 0.79. Race, obesity, anxiety, depression, different types of pain, antidepressants, and anti-inflammatory drugs during pregnancy were among the significant predictors. Our results suggest a potential for applying machine learning to EHR data to predict PPD and inform healthcare delivery.