Prediction of red blood cell transfusion after orthopedic surgery using an interpretable machine learning framework.

Objective: Postoperative red blood cell (RBC) transfusion is widely used during the perioperative period but is often associated with a high risk of infection and complications. However, prediction models for RBC transfusion in patients with orthopedic surgery have not yet been developed. We aimed to identify predictors and constructed prediction models for RBC transfusion after orthopedic surgery using interpretable machine learning algorithms. Methods: This retrospective cohort study reviewed a total of 59,605 patients undergoing orthopedic surgery from June 2013 to January 2019 across 7 tertiary hospitals in China. Patients were randomly split into training (80%) and test subsets (20%). The feature selection method of recursive feature elimination (RFE) was used to identify an optimal feature subset from thirty preoperative variables, and six machine learning algorithms were applied to develop prediction models. The Shapley Additive exPlanations (SHAP) value was employed to evaluate the contribution of each predictor towards the prediction of postoperative RBC transfusion. For simplicity of the clinical utility, a risk score system was further established using the top risk factors identified by machine learning models. Results: Of the 59,605 patients with orthopedic surgery, 19,921 (33.40%) underwent postoperative RBC transfusion. The CatBoost model exhibited an AUC of 0.831 (95% CI: 0.824-0.836) on the test subset, which significantly outperformed five other prediction models. The risk of RBC transfusion was associated with old age (>60 years) and low RBC count (<4.0 × 1012/L) with clear threshold effects. Extremes of BMI, low albumin, prolonged activated partial thromboplastin time, repair and plastic operations on joint structures were additional top predictors for RBC transfusion. The risk score system derived from six risk factors performed well with an AUC of 0.801 (95% CI: 0.794-0.807) on the test subset. Conclusion: By applying an interpretable machine learning framework in a large-scale multicenter retrospective cohort, we identified novel modifiable risk factors and developed prediction models with good performance for postoperative RBC transfusion in patients undergoing orthopedic surgery. Our findings may allow more precise identification of high-risk patients for optimal control of risk factors and achieve personalized RBC transfusion for orthopedic patients.

Infant exposure to parabens, triclosan, and triclocarban via breastfeeding and formula supplementing in southern China.

Parabens, triclosan (TCS), and triclocarban (TCC) are antimicrobial additives that are widely used in personal care products (PCPs) and may dysregulate infant gut microbiota and induce a series of chronic diseases. Dietary intake may be an underestimated exposure route of such antimicrobial additives in infants, but relevant data remain scarce. Therefore, this study determined five common preservatives, including methyl- (MeP), ethyl- (EtP), propyl- (PrP), butyl- (BuP), and benzyl-paraben (BeP), and two antimicrobials TCS and TCC, in major infant food sources (breastmilk, milk-based infant formula [MIF], and cereal-based complementary food [CCF]) in southern China. The health risks associated with dietary exposure among infants across different months of age were also evaluated. The results demonstrated a high incidence of MeP, EtP, PrP, and BeP in processed infant food products, while TCS and TCC were mainly detected in maternal breastmilk. Notably, MeP and EtP were found in all of the MIFs tested, while MeP, EtP, and BeP were detected in 85.6 %-100 % of the CCFs. By incorporating the human equivalent dose and an additional 10-fold margin of safety for infants into the health risk assessment, the 95th percentile hazard quotient of PrP via the ingestion of breastmilk among neonates exceeded 1. For the first time, the results showed that exposure to PrP via breastmilk intake may pose a considerable health risk to urban neonates in southern China. The health risks caused by antimicrobial exposure via ingesting MIF and CCF among infants were negligible. Thus, we recommend breastfeeding women reduce their consumption of PCPs and processed food, especially during the first month after delivery.

Association between organophosphorus flame retardants exposure and cognitive impairment among elderly population in southern China.

The wide application of organophosphorus flame retardants (OPFRs) in consumer products leads to their ubiquitous occurrence. The neurotoxicity of OPFRs has been raised, whereas evidence from the elderly population were rather scarce. Hence, a case-control study was conducted based on the Shenzhen Aging-related Disorder Cohort. A total of 184 cases [Mini-mental State Examination (MMSE) < 24] and 795 participants as controls (MMSE ≥24) were recruited. Eight metabolites of OPFRs (m-OPFRs) in urine samples were measured, including bis(2-butoxyethyl) phosphate (BBOEP), bis(2-chloroethyl) phosphate (BCEP), bis(1-chloro-2-propyl) phosphate (BCIPP), bis(1,3-dichloro-2-propyl) phosphate (BDCIPP), dibutyl phosphate (DBP), di-o-cresyl phosphate (DOCP), di-p-cresyl phosphate (DPCP) and diphenyl phosphate (DPHP). The detection frequencies of m-OPFRs ranged from 88.8 % to 95.4 %. BCEP had the highest median concentration (0.93 µg/L), followed by BCIPP (0.32 µg/L), DPHP (0.27 µg/L) and DBP (0.20 µg/L). Significant correlations were found between all pairs of urinary m-OPFRs with correlation coefficients ranging from 0.22 to 0.71 (p< 0.05). Logistic regression models showed that urinary concentrations of BDCIPP (adjusted odds ratio [OR]: 1.25, 95 % confidential interval [CI]: 1.04-1.50) and DBP (adjusted OR: 1.10, 95 % CI: 1.01-1.20) were positively associated with lower cognitive functions. Furthermore, a nonlinear dose-response relationship was found between urinary BDCIPP concentration and cognitive decline. To our knowledge, this is the first report on OPFR exposure and cognitive impairment among elderly population. Further toxicological tests of BDCIPP and DBP are needed to illustrate these results.

Antibiotics in urine from general adults in Shenzhen, China: Demographic-related difference in exposure levels.

Misuse or overuse of antibiotics can have a variety of detrimental microbial effects. However, the body burden of antibiotics in the general population is currently unclear. In this cross-sectional study, we determined four classes of widely-applied antibiotics (3 imidazoles, 2 sulfonamides, 5 quinolones, and 2 chloramphenicols) in urine samples from 1170 adult residents in Shenzhen, China. Antibiotics were detected in 30.8 % of all urine samples with concentrations ranging from

Impact of population movement on the spread of 2019-nCoV in China.

Since Dec 2019, China has experienced an outbreak caused by a novel coronavirus, 2019-nCoV. A travel ban was implemented for Wuhan, Hubei on Jan 23 to slow down the outbreak. We found a significant positive correlation between population influx from Wuhan and confirmed cases in other cities across China (R2 = 0.85, P < 0.001), especially cities in Hubei (R2 = 0.88, P < 0.001). Removing the travel restriction would have increased 118% (91%-172%) of the overall cases for the coming week, and a travel ban taken three days or a week earlier would have reduced 47% (26%-58%) and 83% (78%-89%) of the early cases. We would expect a 61% (48%-92%) increase of overall cumulative cases without any restrictions on returning residents, and 11% (8%-16%) increase if the travel ban stays in place for Hubei. Cities from Yangtze River Delta, Pearl River Delta, and Capital Economic Circle regions are at higher risk.