Interpretation of the PREVENT study findings on the adjunctive role of intermittent pneumatic compression to prevent venous thromboembolism

Venous thromboembolic prophylaxis: current practice of surgeons in Australia and New Zealand for major abdominal surgery

BACKGROUND

Surgical prophylaxis for venous thrombo-embolic disease (VTE) includes risk assessment, chemical prophylaxis and mechanical prophylaxis (graduated compression stockings [GCS] and/or intermittent pneumatic compression devices [IPCD]). Although there is overwhelming evidence for the need and efficacy of VTE prophylaxis in patients at risk, only about a third of those who are at risk of VTE receive appropriate prophylaxis.

OBJECTIVE

There is debate as to the best combination of VTE prophylaxis following abdominal surgery due to lack of evidence. The aim of this survey was to understand this gap between knowledge and practice.

METHODS

In 2019 and 2020, a survey was conducted to investigate the current practice of venous thromboembolism (VTE) prophylaxis for major abdominal surgery, with a focus on colorectal resections. The study received ethics approval and involved distributing an 11-item questionnaire to members of two professional surgical societies: the Colorectal Surgical Society of Australia and New Zealand (CSSANZ) and the General Surgeons Australia (GSA).

RESULTS

From 214 surgeons: 100% use chemical prophylaxis, 68% do not use a risk assessment tool, 27% do not vary practice according to patient risk factors while > 90% use all three forms of VTE prophylaxis at some stage of treatment. Most surgeons do not vary practice between laparoscopic and open colectomy/major abdominal surgery and only 33% prescribe post-discharge chemical prophylaxis. 42% of surgeons surveyed had equipoise for a clinical trial on the use of IPCDs and the vast majority (> 95%) feel that IPCDs should provide at least a 2% improvement in VTE event rate in order to justify their routine use.

CONCLUSION

Most surgeons in Australia and New Zealand do not use risk assessment tools and use all three forms of prophylaxis regardless. Therfore there is a gap between practice and VTE prophylaxis for the use of mechanical prophylaxis options. Further research is required to determine whether dual modality mechanical prophylaxis is incrementally efficacious. Trial Registration- Not Applicable.

Machine Learning to Dynamically Predict In-Hospital Venous Thromboembolism After Inguinal Hernia Surgery: Results From the CHAT-1 Study

BACKGROUND

The accuracy of current prediction tools for venous thromboembolism (VTE) events following hernia surgery remains insufficient for individualized patient management strategies. To address this issue, we have developed a machine learning (ML)-based model to dynamically predict in-hospital VTE in Chinese patients after hernia surgery.

METHODS

ML models for the prediction of postoperative VTE were trained on a cohort of 11 305 adult patients with hernia from the CHAT-1 trial, which included patients across 58 institutions in China. In data processing, data imputation was conducted using random forest (RF) algorithm, and balanced sampling was done by adaptive synthetic sampling algorithm. Data were split into a training cohort (80%) and internal validation cohort (20%) prior to oversampling. Clinical features available pre-operatively and postoperatively were separately selected using the Sequence Forward Selection algorithm. Nine-candidate ML models were applied to the pre-operative and combined datasets, and their performance was evaluated using various metrics, including area under the receiver operating characteristic curve (AUROC). Model interpretations were generated using importance scores, which were calculated by transforming model features into scaled variables and representing them in radar plots.

RESULTS

The modeling cohort included 2856 patients, divided into 2536 cases for derivation and 320 cases for validation. Eleven pre-operative variables and 15 combined variables were explored as predictors related to in-hospital VTE. Acceptable-performing models for pre-operative data had an AUROC ≥ 0.60, including logistic regression, support vector machine with linear kernel (SVM_Linear), attentive interpretable Tabular learning (TabNet), and RF. For combined data, logistic regression, SVM_Linear, and TabNet had better performance, with an AUROC ≥ 0.65 for each model. Based on these models, 7 pre-operative predictors and 10 combined predictors were depicted in radar plots.

CONCLUSIONS

A ML-based approach for the identification of in-hospital VTE events after hernia surgery is feasible. TabNet showed acceptable performance, and might be useful to guide clinical decision making and VTE prevention. Further validated study will strengthen this finding.

Prediction Model of in-Hospital Venous Thromboembolism in Chinese Adult Patients after Hernia Surgery: The CHAT Score

BACKGROUND

Venous thromboembolism (VTE) events after hernia surgery influence prognosis and life quality and may be preventable. This study aimed to develop a useful model for predicting in-hospital VTE in Chinese patients after hernia surgery.

METHODS

Patients after hernia surgery were retrospectively recruited from 58 institutions (n = 14 322). Totally, 36 potential predictors were involved in the regression analysis. Weighted points were assigned to the predictors of in-hospital VTE identified in the multivariate logistic regression analysis and a prediction model was established. Decision curve analysis was performed to evaluate the net clinical benefit between the established and Caprini models.

RESULTS

A total of 11 707 patients were included and five variables were explored as predictors related to in-hospital VTE: varicose veins of lower extremity, history of VTE, family history of thrombosis, interruption of antithrombotic agents, and reducible hernia. The prediction model (the CHAT score) revealed a good performance metrics (c-statistic, 0.81 [95% CI, 0.80 to 0.81]; Nagelkerke R², 0.27 [95% CI, 0.26 to 0.30]; Brier score, 0.16 [95% CI, 0.13 to 0.23]). The rate of in-hospital VTE after hernia surgery at low-risk (-4 points), intermediate-risk (0-1 points), high-risk (4 points) and very high-risk (≥5 points) were 0.05%, 0.39%, 0.73% and 8.62%, respectively. The CHAT score identified a considerable variability (from 0.05% to 8.62%) for in-hospital VTE among the overall population after hernia surgery. Decision curve analysis found a superior net benefit of the established model than the Caprini score.

CONCLUSIONS

The CHAT score is likely to be a practical 5-item supporting tool to identify patients at high risk of in-hospital VTE after hernia surgery that might assist in decision making and VTE prevention. Further validated study will strengthen this finding.