A Machine Learning Model for the Accurate Prediction of 1-Year Survival in TAVI Patients: A Retrospective Observational Cohort Study.

BACKGROUND: predicting the 1-year survival of patients undergoing transcatheter aortic valve implantation (TAVI) is indispensable for managing safe early discharge strategies and resource optimization. METHODS: Routinely acquired data (134 variables) were used from 629 patients, who underwent transfemoral TAVI from 2012 up to 2018. Support vector machines, neuronal networks, random forests, nearest neighbour and Bayes models were used with new, previously unseen patients to predict 1-year mortality in TAVI patients. A genetic variable selection algorithm identified a set of predictor variables with high predictive power. RESULTS: Univariate analyses revealed 19 variables (clinical, laboratory, echocardiographic, computed tomographic and ECG) that significantly influence 1-year survival. Before applying the reject option, the model performances in terms of negative predictive value (NPV) and positive predictive value (PPV) were similar between all models. After applying the reject option, the random forest model identified a subcohort showing a negative predictive value of 96% (positive predictive value = 92%, accuracy = 96%). CONCLUSIONS: Our model can predict the 1-year survival with very high negative and sufficiently high positive predictive value, with very high accuracy. The "reject option" allows a high performance and harmonic integration of machine learning in the clinical decision process.

Management of closed sternal incision after bilateral internal thoracic artery grafting with a single-use negative pressure system.

Single-use, closed incision management (CIM) systems offer a practical means of delivering negative pressure wound therapy to patients. This prospective study evaluates the Prevena™ Therapy system in a cohort of coronary patients at high risk of deep sternal wound infection (DSWI). Fifty-three consecutive patients undergoing bilateral internal thoracic artery (BITA) grafting were preoperatively elected for CIM with the Prevena™ Therapy system, which was applied immediately after surgery. The actual rate of DSWI in these patients was compared with the expected risk of DSWI according to two scoring systems specifically created to predict either DSWI after BITA grafting (Gatti score) or major infections after cardiac surgery (Fowler score). The actual rate of DSWI was lower than the expected risk of DSWI by the Gatti score (3.8 vs. 5.8%, p = 0.047) but higher than by the Fowler score (2.3%, p = 0.069). However, while the Gatti score showed very good calibration (χ2 = 4.8, p = 0.69) and discriminatory power (area under the receiver-operating characteristic curve 0.838), the Fowler score showed discrete calibration (χ2 = 10.5, p = 0.23) and low discriminatory power (area under the receiver-operating characteristic curve 0.608). Single-use CIM systems appear to be useful to reduce the risk of DSWI after BITA grafting. More studies have to be performed to make stronger this finding.

Active clearance of chest drainage catheters reduces retained blood.

OBJECTIVE: Chest tubes are used to clear blood from around the heart and lungs after heart surgery, but they can be obstructed by a blood clot, leading to retained blood syndrome (RBS). We sought to examine the frequency of RBS and associated morbidity, and to determine the influence of a preventative active chest tube clearance (ATC) protocol on these outcomes. METHODS: A multidisciplinary team developed a simple protocol to institute ATC to preventatively clear chest tubes of clot during the first 24 hours after heart surgery. An extensive educational in-service was performed before universal implementation (phase 1). We retrospectively compared data collected prospectively from 1849 patients before universal implementation (phase 0) with data from 256 patients collected prospectively after universal implementation (phase 2), and then used propensity matching for outcomes assessment. RESULTS: In propensity-matched patients, 19.9% of patients had interventions for RBS (phase 0). After the implementation of ATC (phase 2), the percent of patients with interventions for RBS was reduced to 11.3%, representing a 43% reduction in RBS (P = .0087). These patients had a 33% reduced incidence of postoperative atrial fibrillation from 30% (78 out of 256) in phase 0 to 20% (52 out of 256) in phase 2. (P = .013). CONCLUSIONS: ATC is associated with a reduced need for interventions for RBS and postoperative atrial fibrillation. Our findings underscore the importance of maintaining chest tube patency in the early hours after cardiac surgery.

A Retrospective Study to Evaluate Use of Negative Pressure Wound Therapy in Patients Undergoing Bilateral Internal Thoracic Artery Grafting.

Bilateral internal thoracic artery (BITA) grafting may be associated with a higher risk of postoperative deep sternal wound infection than monolateral internal thoracic artery grafting due to a limited blood supply to the thoracic chest wall. Because preliminary studies suggest negative pressure wound therapy (NPWT) may reduce the risk of infection, a retrospective chart review of 129 patients who underwent BITA between February 2003 and October 2014 was conducted. Of those, 21 patients received NPWT for 5 days immediately following surgery and the incisions of 108 patients were covered with a conventional gauze dressing. Patient demographic and history variables as well as surgical procedure and outcome variables were abstracted. Outcome variables assessed included infection, need for transfusion, and length of hospital stay. The NPWT group was significantly younger (average age 55.9 ± 7.6 versus 60 ± 10.5 years, P = 0.049), had fewer urgent/emergent surgeries (4 [19%] versus 36 [33.3%], P = 0.247), and had significantly lower surgical risk scores (2.0 ± 2.3 versus 3.8 ± 2.8, P = 0.010). The rate of deep sternal wound infections was lower in the NPWT than in the control group, but the difference was not statistically significant (0% versus 5.6%, P = 0.336). Sternal instability was noted in 4 control patients, requiring wound re-exploration versus 0 in the NPWT group (3.7% versus 0%, P = 0.487). One (1) patient in the NPWT group had postoperative bleeding that required removal of the device. The rates of re-thoracotomy due to bleeding were 9.3% in the control compared to 4.8% in the NPWT group (P = 0.435), which translated into a greater need for blood transfusions (1.77 ± 3.4 units versus 0.3 3± 0.7 units, P = 0.056) and larger chest drainage volume (997.8 ± 710 mL versus 591.2 ± 346 mL, P = 0.012) in the control group. Hospital stay was longer in the control group, but the difference was not statistically significant (12 ± 8.8 days versus 9.4 ± 4.2 days, P = 0.184). These preliminary results are encouraging, and prospective, randomized, controlled clinical studies to compare the efficacy, effectiveness, and costeffectiveness of NPWT to other wound management modalities following cardiac surgery are warranted.