Search

Introduction: With widespread adoption of enhanced recovery protocols and a push toward shorter length of stay (LOS) following colon surgery, the extent to which complications have shifted to the post-discharge setting is unknown. The objectives of this study were to (1) characterize changes in LOS and post-discharge complications over time and (2) evaluate risk factors associated with post-discharge complications.Methods: Patients who underwent elective colon resection from 2012 to 2018 were identified from the ACS NSQIP Colectomy-Targeted Dataset. Changes in LOS and the proportion of post-discharge complications were evaluated over time, and predictors of post-discharge complications were assessed using multivariable logistic regression.Results: Of the 98,136 patients who underwent colon resection, median LOS decreased from 5 days in 2012 to 4 days in 2018. Overall, 30-day complication rate was 21.5%, which decreased during the study period (25.8 to 19.1%, p < 0.001). Of the 13 individual complications evaluated, 4 demonstrated a significant increase in the proportion of post-discharge events including overall SSI (55.8 to 63.3%, p = 0.002), superficial SSI (57.3 to 75.7%, p < 0.001), wound disruption (46.0 to 62.1%, p = 0.047), and UTI (41.5 to 62.7%, p < 0.001). Factors associated with the development of any post-discharge complication included female sex, ASA III/IV/V, dependent functional status, and higher BMI. Intraoperative factors included wound class, operation time, and approach.Conclusions: Although LOS and 30-day complications decreased over time, the proportion of events occurring post-discharge increased for several complications. We identified specific factors associated with post-discharge complications which emphasize the importance of a patient monitoring program to early identify and manage post-discharge complications. [ABSTRACT FROM AUTHOR]

Introduction: Carotid bifurcation stenosis may co-exist simultaneously with more proximal common carotid artery (CCA) atherosclerotic plaquing, primarily at the vessel origin in the aortic arch. This scenario is relatively infrequent and its' management does not have quality randomized data to support medical vs surgical treatment. It is logical to treat any high grade common carotid lesions proximal to a carotid bifurcation endarterectomy (CEA) site both to prevent perioperative emboli or thrombosis as well as future embolization. Prior long-term investigations of the combined treatment paradigm have been low volume analysis. Further, prior studies focus on perioperative outcomes with respect to stroke prevention. The only prior VQI study investigating mid-term outcomes following simultaneous CEA with proximal CCA endovascular therapy provided data on less than 10 patients beyond 1.5 years. The long-term follow-up (LFTU) component initiative within VQI has been emphasized in recent years, now allowing for much more robust LTFU analysis. Methods: Four cohorts were created for perioperative outcome analysis and Kaplan Meier freedom from event analysis: CEA in isolation for asymptomatic disease; CEA in isolation for symptomatic patients; CEA with proximal CCA endovascular intervention for asymptomatic; and, CEA with proximal CCA intervention for symptomatic patients. Binary logistic multivariable regression was performed for perioperative neurological event and 90-day mortality risk determination and Cox multivariable regression analysis was performed for long term freedom from cumulative ischemic neurological event and long-term mortality analysis. Symptomatology and type of surgery (CEA with or without CCA intervention) were individual variables in the multivariable analysis. Neurological ischemic event in this study encompassed transient ischemic attack (TIA) and stroke combined. Results: We noted a statistically significant (P <.001) escalation in rates of perioperative neurological event, myocardial infarction (MI), carotid re-exploration, 90 day mortality and combined neurological event and 90 day mortality moving from: A) asymptomatic CEA in isolation to B) symptomatic CEA in isolation to C) asymptomatic CEA combined with proximal CCA intervention to D) symptomatic CEA in combination with proximal CCA intervention. The positivity rate for the combined outcome of perioperative ischemic neurological event and 90 day mortality was 2.2% amongst asymptomatic CEA in isolation, 4.1% amongst symptomatic CEA in isolation, 4.4% amongst asymptomatic CEA in combination with proximal CCA intervention; and 8.8% in patients with symptomatic lesions undergoing combined CEA with proximal CCA intervention. On multivariable analysis patients undergoing CEA with proximal CCA endovascular intervention experienced greater risk for perioperative neurological ischemic event (aOR 2.03, 1.43-2.90, P <.001), combined perioperative neurological ischemic event and 90 day mortality (aOR 2.13, 1.62-2.80, P <.001), long term mortality (HR 1.62, 1.12-2.29, P <.001), and cumulative neurological ischemic event in long term follow up (HR 1.62, 1.12-2.29, P =.007). Amongst 4395 cumulative ischemic neurological events in all study patients, 34% were TIA. Conclusions: Carotid bifurcation endarterectomy in combination with proximal endovascular common carotid artery intervention caries an over two fold higher perioperative risk of neurologic ischemic event and 90 day mortality relative to CEA in isolation for asymptomatic and symptomatic cohorts respectively. After surgery, freedom from cerebral ischemia and mortality for patients undergoing dual intervention is closely aligned with patients undergoing CEA in isolation. Despite high adverse perioperative event rates for the combined CEA and CCA treatment, there is likely long term stroke reduction and mortality benefit to this approach in symptomatic patients based on the event free rates seen herein after initial hospital discharge. The benefit of treating asymptomatic tandem ICA and CCA lesions remains vague but the 4.4% perioperative neurologic event and death rate suggests that these patients would be better managed with medical therapy. [ABSTRACT FROM AUTHOR]

The accurate effluent prediction plays a crucial role in providing early warning for abnormal effluent and achieving the adjustment of feedforward control parameters during wastewater treatment. This study applied a dual-staged attention mechanism based on long short-term memory network (DA-LSTM) to improve the accuracy of effluent quality prediction. The results showed that input attention (IA) and temporal attention (TA) significantly enhanced the prediction performance of LSTM. Specially, IA could adaptively adjust feature weights to enhance the robustness against input noise, with R2 increased by 13.18%. To promote its long-term memory ability, TA was used to increase the memory span from 96 h to 168 h. Compared to a single LSTM model, the DA-LSTM model showed an improvement in prediction accuracy by 5.10%, 2.11%, 14.47% for COD, TP, and TN. Additionally, DA-LSTM demonstrated excellent generalization performance in new scenarios, with the R2 values for COD, TP, and TN increasing by 22.67%, 20.06%, and 17.14% respectively, while the MAPE values decreased by 56.46%, 63.08%, and 42.79%. In conclusion, the DA-LSTM model demonstrated excellent prediction performance and generalization ability due to its advantages of feature-adaptive weighting and long-term memory focusing. This has forward-looking significance for achieving efficient early warning of abnormal operating conditions and timely management of control parameters. [Display omitted] • Improving prediction accuracy of effluent quality in multivariate time series. • Dynamically adjusting feature weights and selecting critical features. • Adaptively focusing on historical information and enhancing long-term memory. • Outstanding generalization performance in novel scenarios. [ABSTRACT FROM AUTHOR]