O02 THE ROLE OF DEEP LEARNING IN PREDICTING COMPLEXITY AND COMPLICATIONS IN ABDOMINAL WALL RECONSTRUCTION

Aim The aim of our study was to evaluate the utility of image-based deep learning models (DLMs) to predict surgical complexity and postoperative outcomes in patients undergoing AWR. Material and Methods A prospective, tertiary center, hernia database was queried for open AWR patients with adequate pre-operative CT-scans. An 8-layer convolutional neural network (CNN) analyzed image characteristics in Python utilizing the open source Tensorflow© and OpenCV frameworks. Images were analyzed and batched into a training set (80%) and validation set (20%) used to analyze the model output, which was blinded to the CNN until testing. DLMs were run to assess surgical complexity based on need for component separation, surgical site infection (SSI), and pulmonary failure. The surgical complexity DLM was validated by comparison to 6 expert AWR surgeons. Results In total, 369 patient CT scans were utilized. The surgical complexity DLM performed well (ROC=0.744;p<0.0001), and when compared to surgeon prediction on the validation set, performed better with an accuracy of 81.3% compared to 65.0% (p < 0.0001). The SSI DLM was successful with an ROC of 0.898 (p < 0.0001). The DLM for predicting pulmonary failure was less effective with an ROC of 0.545 (p = 0.03). Conclusions DLMs were able to successfully predict surgical complexity and were more accurate than expert surgeons using objective, pre-operative imaging. DLMs were also successful in predicting SSI. This breakthrough may allow for enhanced pre-operative planning, including resource utilization and possible need for tertiary center referral. AI appears to be an exciting new management tool in complex AWR.

COVID-19: From viral infection to pulmonary failure

SARS-CoV-2 is a virus which promoted a worldwide pandemic outbreak in 2020. The virus is highly infectious and is able to contaminate a lot of people in a short time period. The disease promoted by the virus, named COVID-19, can cause different symptoms such as fever, cough, muscle pain, headache, prostration, diarrhea, neurological complications, dermic manifestations, pulmonary impairment, dyspnea, coagulopathies, organ failure, and death. Here, we show how the infection occurs and the major characteristics observed in the lungs of patients with COVID-19. In addition, we explored the immunological activation in this environment by the virus and some treatments used in the severe phase of the disease.

An innovative ovine model of severe cardiopulmonary failure supported by veno-arterial extracorporeal membrane oxygenation

Background Refractory cardiogenic shock (CS), frequently complicated by pulmonary failure, often requires veno-arterial extracorporeal membrane oxygenation (VA-ECMO) to sustain end-organ perfusion. Currently available animal models, such as the coronary ligation model, result in highly variable injury profiles and unacceptably high levels of subsequent ventricular fibrillation, cardiac arrest, and death. As the use of ECMO increases, there is a growing need for a clinically relevant, robust, and titratable model of severe cardiopulmonary failure supported by VA-ECMO. Methods Six sheep (60 ± 6 kg) were anaesthetized, intubated and mechanically ventilated. VA-ECMO was initially carried out at a flow rate of 1 L/min. CS was induced through 1-mL left ventricle myocardial injections of 96% ethanol and confirmed when systolic blood pressure (SBP) was < 90 mmHg and lactate > 4 mmol/L. Then, pulmonary failure was confirmed when PaO2 was < 60 mmHg through substantial decrease in the ventilatory support. Thereafter, VA-ECMO support was increased to obtain a mean arterial pressure of 65 mmHg. Echocardiography and cardiac Troponin I (cTnI) analysis were performed at baseline, upon CS confirmation, establishment of pulmonary failure and hourly thereafter. After 5h, the animals were euthanised and the heart collected for histological and macroscopic assessment. Results Ethanol (58 ± 23 mL) rapidly induced CS in all animals. cTnI levels increased near 5000-fold. SBP decreased from 97 ± 18 mmHg at baseline to 67 ± 14 mmHg upon CS, and lactate from 1.4 ± 0.8 to 4.7 ± 0.9 mmol/L, respectively. Echocardiography studies demonstrated a decrease in the left ventricular fractional area change from 34 ± 9% upon baseline to 16 ± 7% after CS. Analysis of myocardial tissue samples corroborated extensive cellular necrosis and inflammatory infiltrates. Conclusions We present a novel titratable model of severe cardiopulmonary failure in animal on VA-ECMO, through intramyocardial ethanol injections and reduction in ventilatory support. This model could be essential to further characterize left-sided heart failure and develop future treatments.

High-Dose Chemotherapy with Bendamustin and Melphalan Improves the Rate of Complete Remission in Myeloma Patients in First Remission Compared to Standard Melphalan Alone

Introduction: Consolidation of first-line induction treatment in myeloma (MM) patients (pts) with 200 mg/m2 melphalan chemotherapy (HDCT) and autologous stem cell transplantation (ASCT) was established as standard of care three decades ago. However, definite cure in myeloma patients remains exceptional due to residual disease escaping intensive treatment, and almost all patients will ultimately relapse at earlier or later time points following ASCT. Thus, improving efficacy of HDCT in MM remains an unresolved issue. Methods: We performed a phase-II randomized trial comparing standard 200 mg/m2 Melphalan (Mel) HDCT to experimental HDCT treatment with 200 mg/m2 bendamustine, a bifunctional alkylating agent, given at days -4 and -3, combined with 200 mg/m2 melphalan split on days -2 and -1 at 100 mg/m2 (BenMel) before ASCT in MM pts. Patients had up to four cycles of first-line induction treatment with bortezomib, lenalidomide and dexamethasone. After ASCT, pts received lenalidomide maintenance treatment for two years. The primary endpoint was to show a 15% improvement of the rate of complete remission (sCR+CR) after HDCT with BenMel compared to Mel alone. MRD assessment from the bone marrow was performed by multiparameter flow cytometry after hematological engraftment following HDCT/ASCT. MRD negativity was defined as clonal plasma cells below 10(-5). Results: We randomized 120 myeloma pts (60 patients in each arm), with high-risk genetic abnormalities present in 21.3% of the patients. The median age was 63 years (range 35-74). The sCR/CR rate after ASCT before initiation of lenalidomide maintenance treatment was better in the BenMel arm compared to Mel alone (70.0% vs 51.7%; p=.039). The post-ASCT remission rates in detail were sCR 40.0% vs 31.7% (p=.341); CR 30.0% vs 20.0% (p=.205); VGPR 16.7% vs 33.3% (p=.035); and PR 13.3% vs 15.0% (p=.793). MRD negativity assessed in the bone marrow by flow cytometry was observed in 26 (45.6%) of the BenMel treated pts compared to 22 (37.9%) of the Mel pts. Median time until neutrophil engraftment was 11 days after BenMel vs 12 days after Mel (p=.096), and median time until platelet engraftment was 13 days in both arms (p=0.367); all pts had full engraftment of both cell lineages. Prolonged hospitalization duration was seen in BenMel pts (median 19 vs 18 days; p=.006) due to the longer BenMel treatment administration. Fully reversible acute renal insufficiency occurred in three (5%) BenMel pts compared to none of the Mel pts (p=.250). No treatment-related mortality was seen in both groups. ICU admissions were necessary in 3 pts (5%) in the BenMel group (ARDS, septic shock, pulmonary failure), and 2 Mel treated pts (3.3%; due to pulmonary failure and decompensated cardiomyopathy). The PFS rates at 12 months were 95% in BenMel pts and 91% in Mel treated pts (p=.551). OS at 12 months was 96% for both groups (p=.262), and median PFS and OS were not reached in both groups. Conclusions: Our data confirm that high-dose bendamustine combined with melphalan HDCT before ASCT in MM patients is safe and well tolerated. In particular, bendamustine-associated renal toxicity was manageable and reversible in all patients, and hematopoietic engraftment was comparable to standard melphalan HDCT. HDCT with BenMel improves the sCR/CR rate compared to standard melphalan alone. Thus, BenMel HDCT before ASCT warrants further investigation aiming to improve the long-term survival rates of MM patients, eventually combined with new maintenance strategies in the post-transplant period. Figure 1 Disclosures No relevant conflicts of interest to declare.