Can machine learning provide preoperative predictions of biological hemostasis after extracorporeal circulation for cardiac surgery?

OBJECTIVES: The goal of this study was to improve decision making regarding the transfusion of patients at the end of extracorporeal circulation for cardiac surgery through machine learning predictions of the evolution of platelets counts, prothrombin ratio, and fibrinogen assay. METHODS: Prospective data with information about patient preoperative biology and surgery characteristics were collected at Institut Mutualiste Montsouris Hospital (Paris, France) for 10 months (n = 598). For each outcome of interest, instead of arbitrarily choosing 1 machine learning algorithm, we trained and tested a variety of algorithms together with the super learning algorithm, a state-of-the-art ensemble method that aggregates all the predictions and selects the best performing algorithm (total, 137 algorithms). We considered the top-performing algorithms and compared them to more standard and interpretable multivariable linear regression models. All algorithms were evaluated through their root mean squared error, a measure of the average difference between true and predicted values. RESULTS: The root mean squared error of the top algorithms for predicting the difference between pre- and postoperative platelet counts, prothrombin ratio, and fibrinogen assay were 38.27 × 10e9/L, 8.66%, and 0.44 g/L, respectively. The linear models had similar performances. CONCLUSIONS: Our machine learning algorithms accurately predicted prothrombin ratio and fibrinogen assay and less accurately platelet counts. As such, our models could provide an aid-decision tool for anesthetists in an operating room; future clinical trials addressing this hypothesis are warranted.

Mitroflow LXA structural deterioration following aortic valve replacement: a single-center experience.

BACKGROUND: Concerns have been previously raised regarding the potential early degeneration of the Mitroflow (Sorin Group Italia, Saluggia, Vercelli, Italy) bioprostheses. We aimed to evaluate our clinical experience with the Mitroflow LXA prosthesis for aortic valve replacement. METHODS: We prospectively analyzed data from 227 consecutive patients (133 males, mean age 73.9±9.2 years) implanted with the Mitroflow LXA between February 2007 and October 2011. Follow-up data were obtained by contacting the referring cardiologists. Kaplan-Meier curves were constructed for all-cause mortality, valve related mortality and structural valve degeneration (SVD). Multivariable analysis was conducted to identify SVD predictors. RESULTS: Median follow-up time was 54.2±37.9 months and completeness of follow-up was 95%. Overall mortality in the entire series was at 31% (N.=71) and mortality from cardiac or unknown causes at 20% (N.=46). SVD occurred in 24 patients (10%) (median delay between implantation and diagnosis: 62.6 (36.5) months). Reintervention was required in 20 cases (13 redo surgery, 7 percutaneous transcatheter valve intervention). The 8 years actuarial global survival was 54.7±4.9%, freedom from valve related mortality 67.5±4.9% and freedom from SVD 72±8%. The estimated freedom from SVD was significantly (P=0.007) longer in larger prosthesis (diameter >21 mm, 77±11%) compared to the smaller devices (≤21 mm, 59±13%.). Multivariate analysis identified smaller prostheses and age at implantation as independent predictors of SVD. CONCLUSIONS: The Mitroflow LXA showed evidence of early SVD in this cohort. A close follow-up of these patients is strongly advised.

Bioprosthetic replacement of the ascending thoracic aorta: what are the options?

Increasing patient age and improved durability of latest generation bioprostheses have stimulated the use of bioprosthetic devices in the setting of ascending aortic replacement as an alternative to mechanical valved conduits or aortic valve-sparing procedures. We performed an English literature review to assess different surgical options that have been described for bioprosthetic replacement of the ascending aorta. Reported options include: (1) composite valved conduits using a stented bioprosthesis; (2) composite valved conduits using a stentless bioprosthesis; (3) total xenopericardial valved conduits. Composite valved grafts using stented bioprostheses offer a safe and durable option for bioprosthetic replacement of the ascending aorta. Other options are of more recent use and await medium-term results.

Composite freestyle stentless xenograft with Dacron graft extension for ascending aortic replacement.

BACKGROUND: The present study was undertaken to evaluate clinical, hemodynamic, and morphologic results of composite stentless xenograft with polyethylene terephthalate fiber (Dacron; DuPont, Wilmington, DE) graft extension for combined replacement of the aortic valve, root, and ascending aorta. METHODS: Between 1997 and 2008, 55 consecutive patients (33 men, 71 +/- 11 years) underwent ascending aortic replacement using Medtronic Freestyle with Dacron graft extension (DuPont). Indications included aneurysm (n = 31, 56%), dissection (n = 16, 29%), and endocarditis (n = 8, 15%). Associated procedures were performed in 25 patients (46%). Preoperative logistic EuroSCORE averaged 34% +/- 28%. Mean cardiopulmonary bypass and aortic cross-clamp times were 244 +/- 134 minutes and 162 +/- 69 minutes, respectively. RESULTS: Clinical follow-up was 100% complete and averaged 2 +/- 3 years. Early mortality was 0% (n = 0) in patients with a preoperative EuroSCORE of less than 20 (n = 26, mean expected mortality, 13% +/- 5%) and 31% (n = 9) in those with preoperative logistic EuroSCORE of at least 20 (n = 29, mean expected mortality, 52% +/- 28%). One- and 3-year survival rates were 83% +/- 5% and 78% +/- 7%, respectively. No major thromboembolic or spontaneous bleeding events were recorded. One patient (2%) required late reoperation for prosthetic valve endocarditis. Echocardiographic follow-up showed no valve dysfunction and low mean transvalvular gradients (7 +/- 5 mm Hg). A 64-channel computed tomographic scan was performed in 33 patients at 32.4 +/- 34 months and revealed two small pseudoaneurysms in a single patient. CONCLUSIONS: Composite Freestyle with Dacron graft extension appears to be a safe option for bioprosthetic replacement of the aortic root and tubular ascending aorta. However, long-term results using this composite graft will have to be determined.