Early tracheostomy after initiation of venovenous extracorporeal membrane oxygenation is associated with decreased duration of extracorporeal membrane oxygenation support.

Timing of tracheostomy placement for patients with respiratory failure requiring venovenous extracorporeal membrane oxygenation support is variable and continues to depend on surgeon preference. We retrospectively reviewed all consecutive adult patients supported with peripheral venovenous extracorporeal membrane oxygenation for acute respiratory distress syndrome at a single institution with the hypothesis that early tracheostomy (within 7 days of extracorporeal membrane oxygenation initiation) decreases the duration of extracorporeal membrane oxygenation support. The primary endpoint was duration of extracorporeal membrane oxygenation support. Secondary endpoints included mortality, overall and intensive care unit length of stay, duration of mechanical ventilation, and time from extracorporeal membrane oxygenation initiation to liberation from ventilator, intensive care unit discharge, and hospital discharge. Overall and extracorporeal membrane oxygenation-associated hospital costs were compared. A total of 50 patients were identified for inclusion (early n = 21; late n = 29). Baseline characteristics including indices of disease severity were similar between groups. Duration of extracorporeal membrane oxygenation support was significantly shorter in the early tracheostomy group (12 vs. 21 days; p = 0.005). Median extracorporeal membrane oxygenation-related costs were significantly decreased in the early tracheostomy group ($3,624 vs. $5,603, p = 0.03). Early tracheostomy placement is associated with decreased time on extracorporeal membrane oxygenation support and reduced extracorporeal membrane oxygenation-related costs in this cohort. Validation in a prospective cohort or a clinical trial is indicated.

Early Decompressive Laparotomy for Intra-Abdominal Hypertension Following Initiation of Venovenous Extracorporeal Membrane Oxygenation.

Patients supported with venovenous extracorporeal membrane oxygenation are at risk for intra-abdominal hypertension and abdominal compartment syndrome. Flow through the return cannula may be compromised in these patients, resulting in inadequate support and end-organ malperfusion. Early decompressive laparotomy can mitigate these complications and potentially improve outcomes. Here we review a series of nine patients undergoing early decompressive laparotomy for abdominal compartment syndrome at a single institution and propose an algorithmic approach to the management of these patients.

Occult Chronic Thromboembolic Disease in Patients Presenting for Surgical Pulmonary Embolectomy.

BACKGROUND: Pulmonary embolism is a significant cause of morbidity and mortality. An important subpopulation of patients presenting with acute pulmonary embolism have concurrent undiagnosed chronic thromboembolic disease. We sought to further characterize this subpopulation, specifically in a cohort of patients referred for surgical pulmonary embolectomy. METHODS: A retrospective review of all patients requiring surgical pulmonary embolectomy at a single center (2013-2017) was performed. Chart review and blinded examination of presenting computed tomographic pulmonary angiograms were done to identify clinical and radiographic features associated with acute on chronic disease. A predictive tool utilizing readily accessible variables was then constructed for risk stratification. RESULTS: A total of 90 patients were identified for analysis; 34 were treated by specifically trained pulmonary thromboendarterectomy surgeons. Twelve patients in the subgroup treated by pulmonary thromboendarterectomy surgeons were found to have acute on chronic disease confirmed by intraoperative endarterectomized specimen. On univariate analysis, degree of pulmonary artery and inferior vena cava dilation and duration of respiratory symptoms were statistically significant variables. Severe right ventricular hypertrophy, elevation in N-terminal pro-B-type natriuretic peptide, and unprovoked etiology were also associated with acute on chronic disease. A predictive tool weighing the presence of these features was created allowing for stratification into low-, moderate-, and high-risk patients. The positive predictive value of a high-risk score was 100%. CONCLUSIONS: In this cohort, the prevalence of acute on chronic thromboembolic disease was greater than 1 in 3. Readily available clinical and radiographic features were used to create a predictive tool for risk stratification.

Ambulation With Femoral Arterial Cannulation Can Be Safely Performed on Venoarterial Extracorporeal Membrane Oxygenation.

BACKGROUND: Venoarterial extracorporeal membrane oxygenation (VA-ECMO) support can be associated with significant deconditioning due to the requirement for strict bedrest as a result of femoral arterial cannulation. To address this issue, we evaluated our experience with ambulation in patients with peripheral femoral cannulation for VA-ECMO. METHODS: All patients that were peripherally cannulated for VA-ECMO over a 2-year period were retrospectively reviewed. Patients that ambulated at least once while supported with VA-ECMO were included in the analysis. The primary outcomes were safety and feasibility of ambulation, defined as the absence of major bleeding, vascular, or decannulation events. RESULTS: Of 104 patients placed on VA-ECMO, 15 ambulated with a femoral arterial cannula. Forty-six percent of patients were placed on VA-ECMO for decompensated heart failure, and 54% for massive pulmonary embolism. Twenty-seven percent of patients were cannulated during active cardiopulmonary resuscitation. The median length of time from cannulation to out of bed was 3 (range, 0 to 26) days. The median length of time from cannulation to initial ambulation was 4 (range, 1 to 42) days. The median distance of the first postcannulation walk was 300 feet. Neither flow nor speed decreased during or after ambulation. There were no major bleeding events, vascular complications, or decannulation events associated with ambulation. The median intensive care unit length of stay and hospital length of stay were 12 and 21 days, respectively. One-year survival was 100% for ambulating patients. CONCLUSIONS: Ambulating patients supported with VA-ECMO, despite femoral arterial cannulation, appears feasible and safe in carefully selected patients.

Minimally Invasive Left Ventricular Assist Device Implantation: Implementation Early in a Surgical Career.

OBJECTIVE: Several centers have presented minimally invasive surgical approaches to centrifugal left ventricular assist device implantation. Although minimally invasive implantation has been successfully performed by experienced surgeons, at large implanting centers, it is unknown whether these techniques are widely adoptable. We evaluated the experience of a surgeon early in his career with conventional and minimally invasive approaches to device implantation. METHODS: All consecutive left ventricular assist device implantations by a single surgeon in the first year of practice (2015-2016) were retrospectively reviewed. Patients were stratified by standard approach, conventional full sternotomy versus a minimally invasive approach, left anterior thoracotomy and upper hemisternotomy. Demographics, perioperative variables, and short-term outcomes were compared using Wilcoxon rank-sum test. RESULTS: Thirteen patients were identified: six performed via the standard approach and seven performed via the minimally invasive approach. Preoperative demographics were comparable in both groups. However, there was significantly more preoperative right ventricle dysfunction in the minimally invasive group (P = 0.01). Although operative time was significantly longer in the minimally invasive cohort, there was a trend toward decreased cardiopulmonary bypass time. Six-month survival in both groups was 100%. CONCLUSIONS: Compared with conventional sternotomy, minimally invasive ventricular assist device implantation, performed by a surgeon in his first year of practice, had similar perioperative outcomes and excellent survival. Based on these data, minimally invasive implantation may be a feasible strategy for device implantation even early in a surgeon's career.