Prediction of Symptomatic Venous Thromboembolism in Critically Ill Patients: The ICU-Venous Thromboembolism Score

Development and Validation of an ICU-Venous Thromboembolism Prediction Model Using Machine Learning Approaches: A Multicenter Study

Purpose

The purpose of this study was to establish and validate machine learning-based models for predicting the risk of venous thromboembolism (VTE) in intensive care unit (ICU) patients.

Patients and Methods

The clinical data of 1494 ICU patients who underwent Doppler ultrasonography or venography between December 2020 and March 2023 were extracted from three tertiary hospitals. The Boruta algorithm was used to screen the essential variables associated with VTE. Five machine learning algorithms were employed: Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Gradient Boosting Decision Tree (GBDT), and Logistic Regression (LR). Hyperparameter optimization was conducted on the predictive model of the training dataset. The performance in the validation dataset was measured using indicators, including the area under curve (AUC) of the receiver operating characteristic (ROC) curve, specificity, and F1 score. Finally, the optimal model was interpreted using the SHapley Additive exPlanation (SHAP) package.

Results

The incidence of VTE among the ICU patients in this study was 26.04%. We screened 19 crucial features for the risk prediction model development. Among the five models, the RF model performed best, with an AUC of 0.788 (95% CI: 0.738-0.838), an accuracy of 0.759 (95% CI: 0.709-0.809), a sensitivity of 0.633, and a Brier score of 0.166.

Conclusion

A machine learning-based model for prediction of VTE in ICU patients were successfully developed, which could assist clinical medical staff in identifying high-risk populations for VTE in the early stages so that prevention measures can be implemented to reduce the burden on the ICU patients.

External Validation of the ICU-Venous Thromboembolism Risk Assessment Model in Adult Critically Ill Patients

BACKGROUND

Currently, no universally accepted standardized VTE risk assessment model (RAM) is specifically designed for critically ill patients. Although the ICU-venous thromboembolism (ICU-VTE) RAM was initially developed in 2020, it lacks prospective external validation.

OBJECTIVES

To evaluate the predictive performance of the ICU-VTE RAM in terms of VTE occurrence in mixed medical-surgical ICU patients.

METHODS

We prospectively enrolled adult patients in the ICU. The ICU-VTE score and Caprini or Padua score were calculated at admission, and the incidence of in-hospital VTE was investigated. The performance of the ICU-VTE RAM was evaluated and compared with that of Caprini or Padua RAM using the receiver operating curve.

RESULTS

We included 269 patients (median age: 70 years; 62.5% male). Eighty-three (30.9%) patients experienced inpatient VTE. The AUC of the ICU-VTE RAM was 0.743 (95% CI, 0.682-0.804, P < 0.001) for mixed medical-surgical ICU patients. Comparatively, the performance of the ICU-VTE RAM was superior to that of the Pauda RAM (AUC: 0.727 vs 0.583, P < 0.001) in critically ill medical patients and the Caprini RAM (AUC: 0.774 vs 0.617, P = 0.128) in critically ill surgical patients, although the latter comparison was not statistically significant.

CONCLUSIONS

The ICU-VTE RAM may be a practical and valuable tool for identifying and stratifying VTE risk in mixed medical-surgical critically ill patients, aiding in managing and preventing VTE complications.

Development of a Risk Assessment Tool for Venous Thromboembolism among Hospitalized Patients in the ICU

BACKGROUND

ICU patients have a high incidence of VTE. The American College of Chest Physicians antithrombotic practice guidelines recommend assessing the risk of VTE in all ICU patients. Although several VTE risk assessment tools exist to evaluate the risk factors among hospitalized patients, there is no validated tool specifically for assessing the risk of VTE in ICU patients.

METHODS

A retrospective corhort study was conducted between June 2018 and October 2022. We obtained data from the electronic medical records of patients with a variety of diagnoses admitted to a mixed ICU. Multivariable logistic regression analysis was used to evaluate the independent risk factors of VTE. Receiver operating characteristic (ROC) curves were used to analyse the predictive accuracy of different tools.

RESULTS

A total of 566 patients were included, and VTE occurred in 89 patients (15.7%), 62.9% was asymptomatic VTE. A prediction model (the ICU-VTE prediction model) was derived from the independent risk factors identified using multivariate analysis. The ICU-VTE prediction model included eight independent risk factors: history of VTE (3 points), immobilization ≥4 days (3 points), multiple trauma (3 points), age ≥70 years (2 points), platelet count >250 × 103/μL (2 points), central venous catheterization (1 point), invasive mechanical ventilation (1 point), and respiratory failure or heart failure (1 point). Patients with a score of 0-4 points had a low (1.81%) risk of VTE. Patients were at intermediate risk, scoring 5-6 points, and the overall incidence of VTE in the intermediate-risk category was 17.1% (odds ratio [OR], 11.1; 95% confidence interval [CI], 4.2-29.4). Those with a score ≥7 points had a high (44.1%) risk of VTE (OR, 42.6; 95% CI, 16.4-110.3). The area under the curve (AUC) of the ICU-VTE prediction model was 0.838, and the differences in the AUCs were statistically significant between the ICU-VTE prediction model and the other three tools (ICU-VTE score, Z = 3.723, P < 0.001; Caprini risk assessment model, Z = 6.212, P < 0.001; Padua prediction score, Z = 7.120, P < 0.001).

CONCLUSIONS

We identified eight independent risk factors for acquired VTE among hospitalized patients in the ICU, deriving a new ICU-VTE risk assessment model. The model aims to predict asymptomatic VTE in ICU patients. The new model has higher predictive accuracy than the current tools. A prospective study is required for external validation of the tool and risk stratification in ICU patients.

Association of early mobility with the incidence of deep-vein thrombosis and mortality among critically ill patients: a post hoc analysis of PREVENT trial

BACKGROUND

This study assessed the mobility levels among critically ill patients and the association of early mobility with incident proximal lower-limb deep-vein thrombosis and 90-day mortality.

METHODS

This was a post hoc analysis of the multicenter PREVENT trial, which evaluated adjunctive intermittent pneumatic compression in critically ill patients receiving pharmacologic thromboprophylaxis with an expected ICU stay ≥ 72 h and found no effect on the primary outcome of incident proximal lower-limb deep-vein thrombosis. Mobility levels were documented daily up to day 28 in the ICU using a tool with an 8-point ordinal scale. We categorized patients according to mobility levels within the first 3 ICU days into three groups: early mobility level 4-7 (at least active standing), 1-3 (passive transfer from bed to chair or active sitting), and 0 (passive range of motion). We evaluated the association of early mobility and incident lower-limb deep-vein thrombosis and 90-day mortality by Cox proportional models adjusting for randomization and other co-variables.

RESULTS

Of 1708 patients, only 85 (5.0%) had early mobility level 4-7 and 356 (20.8%) level 1-3, while 1267 (74.2%) had early mobility level 0. Patients with early mobility levels 4-7 and 1-3 had less illness severity, femoral central venous catheters, and organ support compared to patients with mobility level 0. Incident proximal lower-limb deep-vein thrombosis occurred in 1/85 (1.3%) patients in the early mobility 4-7 group, 7/348 (2.0%) patients in mobility 1-3 group, and 50/1230 (4.1%) patients in mobility 0 group. Compared with early mobility group 0, mobility groups 4-7 and 1-3 were not associated with differences in incident proximal lower-limb deep-vein thrombosis (adjusted hazard ratio [aHR] 1.19, 95% confidence interval [CI] 0.16, 8.90; p = 0.87 and 0.91, 95% CI 0.39, 2.12; p = 0.83, respectively). However, early mobility groups 4-7 and 1-3 had lower 90-day mortality (aHR 0.47, 95% CI 0.22, 1.01; p = 0.052, and 0.43, 95% CI 0.30, 0.62; p < 0.0001, respectively).

CONCLUSIONS

Only a small proportion of critically ill patients with an expected ICU stay ≥ 72 h were mobilized early. Early mobility was associated with reduced mortality, but not with different incidence of deep-vein thrombosis. This association does not establish causality, and randomized controlled trials are required to assess whether and to what extent this association is modifiable.

TRIAL REGISTRATION

The PREVENT trial is registered at ClinicalTrials.gov, ID: NCT02040103 (registered on 3 November 2013) and Current controlled trials, ID: ISRCTN44653506 (registered on 30 October 2013).

Implementation of neurocritical care in Thailand

Dedicated neurointensive care units and neurointensivists are rarely available in Thailand, a developing country, despite the high burden of life-threatening neurologic illness, including strokes, post-cardiac arrest brain injury, status epilepticus, and cerebral edema from various etiologies. Therefore, the implementation of neurocritical care is essential to improve patient outcomes. With the resource-limited circumstances, the integration of neurocritical care service by collaboration between intensivists, neurologists, neurosurgeons, and other multidisciplinary care teams into the current institutional practice to take care of critically-ill neurologic patients is more suitable than building a new neurointensive care unit since this approach can promptly be made without reorganization of the hospital system. Providing neurocritical care knowledge to internal medicine and neurology residents and critical care fellows and developing a research system will lead to sustainable quality improvement in patient care. This review article will describe our current situation and strategies to implement neurocritical care in Thailand.

Diagnostic accuracy of multiorgan point-of-care ultrasound compared with pulmonary computed tomographic angiogram in critically ill patients with suspected pulmonary embolism

BACKGROUND

Critically ill patients have a higher incidence of pulmonary embolism (PE) than non-critically ill patients, yet no diagnostic algorithm has been validated in this population, leading to the overuse of pulmonary artery computed tomographic angiogram (CTA). This study aimed to comparatively evaluate the diagnostic accuracy of point-of-care ultrasound (POCUS) combined with laboratory data versus CTA in predicting PE in critically ill patients.

METHODS

A prospective diagnostic accuracy study. Critically ill patients with suspected acute PE undergoing CTA were prospectively enrolled. Demographic and clinical data were collected from electronic medical records. Blood samples were collected, and the Wells and revised Geneva scores were calculated. Standardized multiorgan POCUS and CTA were performed. The discriminatory power of multiorgan POCUS combined with biochemical markers was tested using ROC curves, and multivariate analysis was performed.

RESULTS

A total of 88 patients were included, and 37 (42%) had PE. Multivariate analysis showed a relative risk (RR) of PE of 2.79 (95% CI, 1.61-4.84) for the presence of right ventricular (RV) dysfunction, of 2.54 (95% CI, 0.89-7.20) for D-dimer levels >1000 ng/mL, and of 1.69 (95% CI, 1.12-2.63) for the absence of an alternative diagnosis to PE on lung POCUS or chest radiograph. The combination with the highest diagnostic accuracy for PE included the following variables: 1- POCUS transthoracic echocardiography with evidence of RV dysfunction; 2- lung POCUS or chest radiograph without an alternative diagnosis to PE; and 3- plasma D-dimer levels >1000 ng/mL. Combining these three findings resulted in an area under the curve of 0.85 (95% CI, 0.77-0.94), with 50% sensitivity and 96% specificity.

CONCLUSIONS

Multiorgan POCUS combined with laboratory data has acceptable diagnostic accuracy for PE compared with CTA. The combined use of these methods might reduce CTA overuse in critically ill patients.

Indian Society of Critical Care Medicine Consensus Statement for Prevention of Venous Thromboembolism in the Critical Care Unit

Deep vein thrombosis (DVT) is a preventable complication of critical illness, and this guideline aims to convey a pragmatic approach to the problem. Guidelines have multiplied over the last decade, and their utility has become increasingly conflicted as the reader interprets all suggestions or recommendations as something that must be followed. The nuances of grade of recommendation vs level of evidence are often ignored, and the difference between a "we suggest" vs a "we recommend" is overlooked. There is a general unease among clinicians that failure to follow the guidelines translates to poor medical practice and legal culpability. We attempt to overcome these limitations by highlighting ambiguity when it occurs and refraining from dogmatic recommendations in the absence of robust evidence. Readers and practitioners may find the lack of specific recommendations unsatisfactory, but we believe that true ambiguity is better than inaccurate certainty. We have attempted to comply with the guidelines on how to create guidelines.1 And to overcome the poor compliance with these guidelines.2 Some observers have expressed concern that DVT prophylaxis guidelines may cause more harm than good.3 We have placed greater emphasis on large randomized controlled trials (RCTs) with clinical end point and de-emphasized RCTs with surrogate end points and also de-emphasized hypothesis generating studies (observational studies, small RCTs, and meta-analysis of these studies). We have de-emphasized RCTs in non-intensive care unit populations like postoperative patients or those with cancer and stroke. We have also considered resource limitation settings and have avoided recommending costly and poorly proven therapeutic options.

How to cite this article

Jagiasi BG, Chhallani AA, Dixit SB, Kumar R, Pandit RA, Govil D, et al. Indian Society of Critical Care Medicine Consensus Statement for Prevention of Venous Thromboembolism in the Critical Care Unit. Indian J Crit Care Med 2022;26(S2):S51-S65.

The ICU-venous thromboembolism score and tumor grade can predict inhospital venous thromboembolism occurrence in critical patients with tumors

Background

Venous thromboembolism (VTE) is a threat to the prognosis of tumor patients, especially for critically ill patients. No uniform standard model of VTE risk for critically ill patients with tumors was formatted by now. We thus analyzed risk factors of VTE from the perspectives of patient, tumor, and treatment and assessed the predictive value of the ICU-VTE score, which consisted of six independent risk factors (central venous catheterization, 5 points; immobilization ≥ 4 days, 4 points; prior VTE, 4 points; mechanical ventilation, 2 points; lowest hemoglobin during hospitalization ≥ 90 g/L, 2 points; and baseline platelet count > 250,000/μL, 1 points).

Methods

We evaluated the data of tumor patients admitted to the intensive care unit of the Peking University Cancer Hospital between November 2011 and January 2022; 560 cases who received VTE-related screening during hospitalization were chosen for this retrospective study.

Results

The inhospital VTE occurrence rate in our cohort was 55.7% (312/560), with a median interval from ICU admission to VTE diagnosis of 8.0 days. After the multivariate logistic regression analysis, several factors were proved to be significantly associated with inhospital VTE: age ≥ 65 years, high tumor grade (G3–4), medical diseases, fresh frozen plasma transfusion, and anticoagulant prophylaxis. The medium-high risk group according to the ICU-VTE score was positively correlated with VTE when compared with the low-risk group (9–18 points vs. 0–8 points; OR, 3.13; 95% CI, 2.01–4.85, P < 0.001). The AUC of the ICU-VTE scores according to the ROC curve was 0.714 (95% CI, 0.67–0.75, P < 0.001).

Conclusions

The ICU-VTE score, as well as tumor grade, might assist in the assessment of inhospital VTE risk for critically ill patients with tumors. The predictive accuracy might be improved when combining two of them; further follow-up researches are needed to confirm it.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-022-02705-z.

Development and Validation of a Risk Assessment Model for Venous Thromboembolism in Patients With Invasive Mechanical Ventilation

Background

Patients with invasive mechanical ventilation may be at high risk of acquiring venous thromboembolism (VTE). We aim to develop risk assessment models for predicting the improvement of VTE in invasively ventilated patients.

Methodology

A total of 6,734 invasively ventilated patients enrolled from the Medical Information Mart for Intensive Care-III (MIMIC-III) database were used as input for model development and internal validation, while data from 168 patients from Nanfang Hospital were used for external validation. Logistic regression was performed based on predictive factors derived from least absolute shrinkage and selection operator (LASSO) regression analysis and logistic regression with backward selection to develop two Risk Assessment Models (RAM), namely, I and II, for the prediction of VTE, respectively. Model selection was performed by evaluation of the area under the receiver operating characteristic curve (AUC), the goodness of fit with calibration curves, and decision curve analyses (DCA).

Results

RAM-I included prior history of VTE, in-hospital immobilization, infection, glucose, the use of antiplatelet, and activated partial thromboplastin time (APTT) as variables, while RAM-II included prior history of VTE, in-hospital immobilization, infection, ischemic stroke, glucose, the use of antiplatelet and APTT as variables. Compared with RAM-I and ICU-Venous Thromboembolism Score, RAM-II exhibited better discrimination in the training dataset (AUC = 0.826), internal validation dataset (AUC = 0.771), and external validation dataset (AUC = 0.770). Additionally, DCA demonstrated that RAM-II was clinically beneficial. Inspection of the calibration curves revealed good agreement between the predictions and observations.

Conclusions

A RAM for VTE in invasively ventilated patients was developed with reasonable performance.

Anti-factor Xa Activity Is Not Associated With Venous Thromboembolism in Critically Ill Patients Receiving Enoxaparin for Thromboprophylaxis: A Retrospective Observational Study

Background

Anti-factor Xa activity has been suggested as a surrogate parameter for judging the effectiveness of pharmacological thromboprophylaxis with low molecular weight heparins in critically ill patients. However, this practice is not supported by evidence associating low anti-factor Xa activity with venous thromboembolism.

Methods

We performed a retrospective observational study including 1,352 critically ill patients admitted to 6 intensive care units of the Medical University of Vienna, Austria between 01/2015 and 12/2018. Included patients received prophylactically dosed enoxaparin (≤100 IU/kg body weight per day). We analyzed median peak, 12-h trough and 24-h trough anti-factor Xa activity per patient and compared anti-factor Xa activity between patients without vs. with venous thromboembolic events.

Results

19 patients (1.4%) developed a total of 22 venous thromboembolic events. We did not observe a difference of median (IQR) anti-factor Xa activity between patients without venous thromboembolism [peak 0.22 IU/mL (0.14–0.32); 12-h trough 0.1 IU/mL (<0.1–0.17), 24-h trough < 0.1 IU/mL (<0.1– <0.1)] vs. patients with venous thromboembolism [peak 0.33 IU/mL (0.14–0.34); 12-h trough 0.12 IU/mL (<0.1–0.26); 24-h trough < 0.1 IU/mL (<0.1–<0.1)].

Conclusion

Patients who developed venous thromboembolism had anti-factor Xa activities comparable to those who did not suffer from venous thromboembolism.

A New Risk Assessment Model for Hospital-Acquired Venous Thromboembolism in Critically Ill Children: A Report From the Children's Hospital-Acquired Thrombosis Consortium

OBJECTIVES

To create a risk model for hospital-acquired venous thromboembolism in critically ill children upon admission to an ICU.

DESIGN

Case-control study.

SETTING

ICUs from eight children's hospitals throughout the United States.

SUBJECTS

Critically ill children with hospital-acquired venous thromboembolism (cases) 0-21 years old and similar children without hospital-acquired venous thromboembolism (controls) from January 2012 to December 2016. Children with a recent cardiac surgery, asymptomatic venous thromboembolism, or a venous thromboembolism diagnosed before ICU admission were excluded.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The multi-institutional Children's Hospital-Acquired Thrombosis registry was used to identify cases and controls. Multivariable logistic regression was used to determine the association between hospital-acquired venous thromboembolism and putative risk factors present at or within 24 hours of ICU admission to develop the final model. A total of 548 hospital-acquired venous thromboembolism cases (median age, 0.8 yr; interquartile range, 0.1-10.2) and 187 controls (median age, 2.4 yr; interquartile range, 0.2-8.3) were analyzed. In the multivariable model, recent central venous catheter placement (odds ratio, 4.4; 95% CI, 2.7-7.1), immobility (odds ratio 3.6, 95% CI, 2.1-6.2), congenital heart disease (odds ratio 2.9, 95% CI, 1.7-4.7), length of hospital stay prior to ICU admission greater than or equal to 3 days (odds ratio, 2.5; 95% CI, 1.1-5.6), and history of autoimmune/inflammatory condition or current infection (odds ratio, 2.1; 95% CI, 1.2-3.4) were each independently associated with hospital-acquired venous thromboembolism. The risk model had an area under the receiver operating characteristic curve of 0.79 (95% CI, 0.73-0.84).

CONCLUSIONS

Using the multicenter Children's Hospital-Acquired Thrombosis registry, we identified five independent risk factors for hospital-acquired venous thromboembolism in critically ill children, deriving a new hospital-acquired venous thromboembolism risk assessment model. A prospective validation study is underway to define a high-risk group for risk-stratified interventional trials investigating the efficacy and safety of prophylactic anticoagulation in critically ill children.

Coagulopathies in Intensive Care Medicine: Balancing Act between Thrombosis and Bleeding

Patient Blood Management advocates an individualized treatment approach, tailored to each patient's needs, in order to reduce unnecessary exposure to allogeneic blood products. The optimization of hemostasis and minimization of blood loss is of high importance when it comes to critical care patients, as coagulopathies are a common phenomenon among them and may significantly impact morbidity and mortality. Treating coagulopathies is complex as thrombotic and hemorrhagic conditions may coexist and the medications at hand to modulate hemostasis can be powerful. The cornerstones of coagulation management are an appropriate patient evaluation, including the individual risk of bleeding weighed against the risk of thrombosis, a proper diagnostic work-up of the coagulopathy's etiology, treatment with targeted therapies, and transfusion of blood product components when clinically indicated in a goal-directed manner. In this article, we will outline various reasons for coagulopathy in critical care patients to highlight the aspects that need special consideration. The treatment options outlined in this article include anticoagulation, anticoagulant reversal, clotting factor concentrates, antifibrinolytic agents, desmopressin, fresh frozen plasma, and platelets. This article outlines concepts with the aim of the minimization of complications associated with coagulopathies in critically ill patients. Hereditary coagulopathies will be omitted in this review.

Basic ultrasound head-to-toe skills for intensivists in the general and neuro intensive care unit population: consensus and expert recommendations of the European Society of Intensive Care Medicine

Purpose

To provide consensus, and a list of experts’ recommendations regarding the basic skills for head-to-toe ultrasonography in the intensive care setting.

Methods

The Executive Committee of the European Society of Intensive Care (ESICM) commissioned the project and supervised the methodology and structure of the consensus. We selected an international panel of 19 expert clinicians–researchers in intensive care unit (ICU) with expertise in critical care ultrasonography (US), plus a non-voting methodologist. The panel was divided into five subgroups (brain, lung, heart, abdomen and vascular ultrasound) which identified the domains and generated a list of questions to be addressed by the panel. A Delphi process based on an iterative approach was used to obtain the final consensus statements. Statements were classified as a strong recommendation (84% of agreement), weak recommendation (74% of agreement), and no recommendation (less than 74%), in favor or against.

Results

This consensus produced a total of 74 statements (7 for brain, 20 for lung, 20 for heart, 20 for abdomen, 7 for vascular Ultrasound). We obtained strong agreement in favor for 49 statements (66.2%), 8 weak in favor (10.8%), 3 weak against (4.1%), and no consensus in 14 cases (19.9%). In most cases when consensus was not obtained, it was felt that the skills were considered as too advanced. A research agenda and discussion on training programs were implemented from the results of the consensus.

Conclusions

This consensus provides guidance for the basic use of critical care US and paves the way for the development of training and research projects.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00134-021-06486-z.

Thromboprophylaxis in critically ill patients: balancing on a tightrope

Venous thromboembolism (VTE), including deep vein thrombosis and pulmonary embolism, is a common and potentially fatal complication in the intensive care unit (ICU). Critically ill patients have some special characteristics that increase the risk for VTE and complicate risk stratification and diagnosis. Given the positive effect of thromboprophylaxis on main outcomes, its use is mandatory in these patients, which is documented by various studies and recommended by all published guidelines. However, anticoagulation management is not an easy issue in clinical practice, as the critical patient may be at high risk for thrombosis or, conversely, at increased risk of bleeding or may balance between thrombotic and bleeding risk. Thrombotic and bleeding risk scoring should be evaluated daily in order to select the appropriate form of thromboprophylaxis. The selection depends on the degree of bleeding risk and the subgroup of ICU patients involved, such as patients with sepsis, acute brain injury, major trauma or coronavirus disease-2019. If there is no bleeding risk or other contraindication, the patient should receive pharmacologic thromboprophylaxis with unfractionated heparin or low molecular weight heparins, weighing the advantages of each agent. If the patient is at high risk of bleeding or there is a contraindication to pharmacologic prophylaxis, he should receive mechanical thromboprophylaxis mainly with intermittent pneumatic compression or graduated compression stockings. Thromboprophylaxis compliance with the guidelines is a prerequisite for moving from theory to practice. Direct oral anticoagulants have been studied in ICU patients and have no place at present in VTE prophylaxis requiring further research.

Genetic factors increase the identification efficiency of predictive models for dyslipidaemia: a prospective cohort study

Background

Few studies have developed risk models for dyslipidaemia, especially for rural populations. Furthermore, the performance of genetic factors in predicting dyslipidaemia has not been explored. The purpose of this study is to develop and evaluate prediction models with and without genetic factors for dyslipidaemia in rural populations.

Methods

A total of 3596 individuals from the Henan Rural Cohort Study were included in this study. According to the ratio of 7:3, all individuals were divided into a training set and a testing set. The conventional models and conventional+GRS (genetic risk score) models were developed with Cox regression, artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM) classifiers in the training set. The area under the receiver operating characteristic curve (AUC), net reclassification index (NRI), and integrated discrimination index (IDI) were used to assess the discrimination ability of the models, and the calibration curve was used to show calibration ability in the testing set.

Results

Compared to the lowest quartile of GRS, the hazard ratio (HR) (95% confidence interval (CI)) of individuals in the highest quartile of GRS was 1.23(1.07, 1.41) in the total population. Age, family history of diabetes, physical activity, body mass index (BMI), triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were used to develop the conventional models, and the AUCs of the Cox, ANN, RF, and GBM classifiers were 0.702(0.673, 0.729), 0.736(0.708, 0.762), 0.787 (0.762, 0.811), and 0.816(0.792, 0.839), respectively. After adding GRS, the AUCs increased by 0.005, 0.018, 0.023, and 0.015 with the Cox, ANN, RF, and GBM classifiers, respectively. The corresponding NRI and IDI were 25.6, 7.8, 14.1, and 18.1% and 2.3, 1.0, 2.5, and 1.8%, respectively.

Conclusion

Genetic factors could improve the predictive ability of the dyslipidaemia risk model, suggesting that genetic information could be provided as a potential predictor to screen for clinical dyslipidaemia.

Trial registration

The Henan Rural Cohort Study has been registered at the Chinese Clinical Trial Register. (Trial registration: ChiCTR-OOC-15006699. Registered 6 July 2015 - Retrospectively registered).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-021-01439-3.

Incidence, prognostic factors, and outcomes of venous thromboembolism in critically ill patients: data from two prospective cohort studies

BACKGROUND

The objective of this study was to describe the prevalence, incidence, prognostic factors, and outcomes of venous thromboembolism in critically ill patients receiving contemporary thrombosis prophylaxis.

METHODS

We conducted a pooled analysis of two prospective cohort studies. The outcomes of interest were in-hospital pulmonary embolism or lower extremity deep vein thrombosis (PE-LDVT), in-hospital nonleg deep vein thrombosis (NLDVT), and 90-day mortality. Multivariable logistic regression analysis was used to evaluate the association between predefined baseline prognostic factors and PE-LDVT or NLDVT. Cox regression analysis was used to evaluate the association between PE-LDVT or NLDVT and 90-day mortality.

RESULTS

A total of 2208 patients were included. The prevalence of any venous thromboembolism during 3 months before ICU admission was 3.6% (95% CI 2.8-4.4%). Out of 2166 patients, 47 (2.2%; 95% CI 1.6-2.9%) developed PE-LDVT and 38 patients (1.8%; 95% CI 1.2-2.4%) developed NLDVT. Renal replacement therapy (OR 3.5 95% CI 1.4-8.6), respiratory failure (OR 2.0; 95% CI 1.1-3.8), and previous VTE (OR 3.6; 95% CI 1.7-7.7) were associated with PE-LDVT. Central venous catheters (OR 5.4; 95% CI 1.7-17.8) and infection (OR 2.2; 95% CI 1.1-4.3) were associated with NLDVT. Occurrence of PE-LDVT but not NLDVT was associated with increased 90-day mortality (HR 2.7; 95% CI 1.6-4.6, respectively, 0.92; 95% CI 0.41-2.1).

CONCLUSION

Thrombotic events are common in critically ill patients, both before and after ICU admittance. Development of PE-LDVT but not NLDVT was associated with increased mortality. Prognostic factors for developing PE-LDVT or NLDVT despite prophylaxis can be identified at ICU admission and may be used to select patients at higher risk in future randomized clinical trials.

TRIAL REGISTRATION

NCT03773939.