Artificial intelligence-based analysis of body composition predicts outcome in patients receiving long-term mechanical circulatory support.

BACKGROUND: Obesity is a known cardiovascular risk factor and associated with higher postoperative complication rates in patients undergoing cardiac surgery. In heart failure (HF), conflicting evidence in terms of survival has been reported, whereas sarcopenia is associated with poor prognosis. An increasing number of HF patients require left ventricular assist device (LVAD) implantations. The postoperative mortality has improved in recent years but is still relatively high. The impact of body composition on outcome in this population remains unclear. The aim of this investigation was to examine the preoperative computed tomography (CT) body composition as a predictor of the postoperative outcome in advanced HF patients, who receive LVAD implantations. METHODS: Preoperative CT scans of 137 patients who received LVADs between 2015 and 2020 were retrospectively analysed using an artificial intelligence (AI)-powered automated software tool based on a convolutional neural network, U-net, developed for image segmentation (Visage Version 7.1, Visage Imaging GmbH, Berlin, Germany). Assessment of body composition included visceral and subcutaneous adipose tissue areas (VAT and SAT), psoas and total abdominal muscle areas and sarcopenia (defined by lumbar skeletal muscle indexes). The body composition parameters were correlated with postoperative major complication rates, survival and postoperative 6-min walk distance (6MWD) and quality of life (QoL). RESULTS: The mean age of patients was 58.21 ± 11.9 years; 122 (89.1%) were male. Most patients had severe HF requiring inotropes (Interagency Registry for Mechanically Assisted Circulatory Support [INTERMACS] profile I-III, 71.9%) secondary to coronary artery diseases or dilated cardiomyopathy (96.4%). Forty-four (32.1%) patients were obese (body mass index ≥ 30 kg/m2 ), 96 (70.1%) were sarcopene and 19 (13.9%) were sarcopene obese. Adipose tissue was associated with a significantly higher risk of postoperative infections (VAT 172.23 cm2 [54.96, 288.32 cm2 ] vs. 124.04 cm2 [56.57, 186.25 cm2 ], P = 0.022) and in-hospital mortality (VAT 168.11 cm2 [134.19, 285.27 cm2 ] vs. 135.42 cm2 [49.44, 227.91 cm2 ], P = 0.033; SAT 227.28 cm2 [139.38, 304.35 cm2 ] vs. 173.81 cm2 [97.65, 254.16 cm2 ], P = 0.009). Obese patients showed no improvement of 6MWD and QoL within 6 months postoperatively (obese: +0.94 ± 161.44 months, P = 0.982; non-obese: +166.90 ± 139.00 months, P < 0.000; obese: +0.088 ± 0.421, P = 0.376; non-obese: +0.199 ± 0.324, P = 0.002, respectively). Sarcopenia did not influence the postoperative outcome and survival within 1 year after LVAD implantation. CONCLUSIONS: Preoperative AI-based CT body composition identifies patients with poor outcome after LVAD implantation. Greater adipose tissue areas are associated with an increased risk for postoperative infections, in-hospital mortality and impaired 6MWD and QoL within 6 months postoperatively.

Comparison of feasibility and results of frailty assessment methods prior to left ventricular assist device implantation.

AIMS: Assessing frailty and sarcopenia is considered a valuable cornerstone of perioperative risk stratification in advanced heart failure patients. The lack of an international consensus on a diagnostic standard impedes its implementation in the clinical routine. This study aimed to compare the feasibility and prognostic impact of different assessment tools in patients undergoing continuous-flow left ventricular assist device (cf-LVAD) implantation. METHODS AND RESULTS: We prospectively compared feasibility and prognostic values of six frailty/sarcopenia assessment methods in 94 patients prior to cf-LVAD implantation: bioelectrical impedance analysis (BIA), computed tomography (CT)-based measurement of two muscle areas/body surface area [erector spinae muscle (TMESA/BSA) and iliopsoas muscle (TPA/BSA)], physical performance tests [grip strength, 6 min walk test (6MWT)] and Rockwood Clinical Frailty Scale (RCFS). Six-month mortality and/or prolonged ventilation time >95 h was defined as the primary endpoint. BIA and CT showed full feasibility (100%); physical performance and RCFS was limited due to patients' clinical status (feasibility: 87% grip strength, 62% 6MWT, 88% RCFS). Phase angle derived by BIA showed the best results regarding the prognostic value for 6 month mortality and/or prolonged ventilation time >95 h (odds ratio (OR) 0.66 [95% confidence interval (CI): 0.46-0.92], P = 0.019; area under the curve (AUC) 0.65). It provided incremental value to the clinical risk assessment of EuroSCORE II: C-index of the combined model was 0.75 [95% CI; 0.651-0.848] compared with C-index of EuroSCORE II alone, which was 0.73 (95% CI: 0.633-0.835). Six-month survival was decreased in patients with reduced body cell mass derived by BIA or reduced muscle area in the CT scan compared with patients with normal values: body cell mass 65% (95% CI: 51.8-81.6%) vs. 83% (95% CI: 74.0-93.9%); P = 0.03, TMESA/BSA 65% (95% CI: 51.2-82.2%) vs. 82% (95% CI: 73.2-93.0%); P = 0.032 and TPA/BSA 66% (95% CI: 53.7-81.0%) vs. 85% (95% CI: 75.0-95.8%); P = 0.035. CONCLUSIONS: Bioelectrical impedance analysis parameters and CT measurements were shown to be suitable to predict 6-month mortality and/or prolonged ventilation time >95 h in patients with advanced heart failure prior to cf-LVAD implantation. Phase angle had the best predictive capacity and sarcopenia diagnosed by reduced body cell mass in BIA or muscle area in CT was associated with a decreased 6 month survival.

Impact of left ventricular inspection employing cardiopulmonary bypass on outcome after implantation of left ventricular assist device.

BACKGROUND: Cardiopulmonary bypass (CPB) during left ventricular assist device (LVAD) implantation provides circulatory support and allows for safe inspection of the left ventricle (LV), whereas circulatory support by veno-arterial extracorporeal life support (va-ECLS) or off-pump implantation may reduce postoperative bleeding and inflammatory response. METHODS: Retrospective analysis of 616 consecutive adult patients who received an LVAD via median sternotomy between January 1, 2015 and December 31, 2019. All patients undergoing concomitant intracardiac procedures other than closure of persistent foramen ovale or atrial septal defect and redo surgeries were excluded from the analysis. The remaining patients (n = 222) were divided into two groups and 1:1 propensity score-matched regarding preoperative parameters: patients who underwent LVAD implantation with LV inspection employing CPB (CPB group, n = 62) and without LV inspection on va-ECLS or off-pump (non-CPB group, n = 62). RESULTS: The groups were well balanced with regard to preoperative baseline characteristics (standard difference <0.1). Patients in the CPB group required more blood transfusions (median 2 vs. 0 units, p = 0.031) during surgery and in the first 24 h afterwards. The median intensive care unit stay was longer in the CPB group (18 vs. 11 days, p = 0.021). The CPB group showed an absence of perioperative stroke and a smaller number of events per patient-year for postoperative ischemic stroke (0.02 vs. 0.12, p = 0.003). 30-day survival (87% vs. 87.1%) and 1-year survival (80.3% vs. 74%) were similar in both groups (p = 0.78). CONCLUSION: Visual LV inspection on CPB may reduce the risk of postoperative ischemic stroke. Despite the negative effects of employing CPB in lieu of other intraoperative strategies, survival was similar in both groups.

Impact of Muscle Mass as a Prognostic Factor for Failed Waiting Time Prior to Heart Transplantation.

Objectives: Clinical deterioration during the waiting time impairs the prognosis of patients listed for heart transplantation. Reduced muscle mass increases the risk for mortality after cardiac surgery, but its impact on resilience against deterioration during the waiting time remains unclear. Methods: We retrospectively analyzed data from 93 patients without a VAD who were listed in Eurotransplant status "high urgent (HU)" for heart transplantation between January 2015 and October 2020. The axial muscle area of the erector spinae muscles at the level of thoracic vertebra 12 indexed to body surface area (TMESA/BSA) measured in the preoperative thoracic computed tomography scan was used to measure muscle mass. Results: Forty patients (43%) underwent emergency VAD implantation during the waiting time and four patients (4%) died during the waiting time. The risk of emergency VAD implantation/death during the waiting time decreased by 10% for every cm2/m2 increase in muscle area [OR 0.901 (95% CI: 0.808-0.996); p = 0.049]. After adjusting for gender [OR 0.318 (95% CI: 0.087-1.073); p = 0.072], mean pulmonary artery pressure [OR 1.061 (95% CI: 0.999-1.131); p = 0.060], C-reactive protein [OR 1.352 (95% CI: 0.986-2.027); p = 0.096], and hemoglobin [OR 0.862 (95% CI: 0.618-1.177); p = 0.360], TMESA/BSA [OR 0.815 (95% CI: 0.698-0.936); p = 0.006] remained an independent risk factor for emergency VAD implantation/death during the HU waiting time. Conclusion: Muscle area of the erector spinae muscle appears to be a potential, easily identifiable risk factor for emergency VAD implantation or death in patients on the HU waiting list for heart transplantation. Identifying patients at risk could help optimize the outcome and the timing of VAD support.

Safety of Contemporary Heart Failure Therapy in Patients with Continuous-Flow Left Ventricular Assist Devices.

BACKGROUND: Limited data are available concerning the safety, optimal administration and benefits of contemporary heart failure therapy in patients after left ventricular assist device (LVAD) implantation. METHODS: Between 2015 and 2019, 257 patients underwent LVAD implantation and were included in this observational study. Oral heart failure therapy was initiated and uptitrated during the further course. After propensity matching and excluding patients with immediate postoperative treatment in an affiliated center with different medical standards, hospitalization rates and mortality within 12 months after LVAD implantation were compared between 83 patients who received medical therapy including an angiotensin receptor neprilysin inhibitor (ARNI) and 83 patients who did not receive an ARNI. RESULTS: The overall use of heart-failure medications after 12 months was high: prescriptions: beta-blockers, 85%; angiotensin inhibiting drugs, 90% (angiotensin-converting-enzyme inhibitors 30%, angiotensin receptor blockers 23%, ARNI 37%); mineralocorticoid receptor antagonists, 80%. No serious drug-related adverse events occurred. The conditional 1-year survival in the group with ARNIs was 97% (95% CI: 94%-100%) compared to 88% in the group without an ARNI (95% CI: 80%-96%); P = 0.06. CONCLUSIONS: Contemporary heart failure therapy is safe in patients with LVADs. No increase in serious adverse events was seen in patients receiving ARNIs. No significant difference in the conditional 1-year survival was seen between the ARNI group and the nonARNI group.

Prediction of survival of patients in cardiogenic shock treated by surgically implanted Impella 5+ short-term left ventricular assist device.

OBJECTIVES: Short-term mechanical circulatory support is a life-saving treatment for acute cardiogenic shock (CS). This multicentre study investigates the preoperative predictors of 30-day mortality in CS patients treated with Impella 5.0 and 5.5 short-term left ventricular assist devices. METHODS: Data of patients in CS (n = 70) treated with the Impella 5 (n = 63) and 5.5 (n = 7) in 2 centres in Berlin between October 2016 and October 2019 were collected retrospectively. RESULTS: CS was caused by acute myocardial infarction (n = 16), decompensated chronic heart failure (n = 41), postcardiotomy syndrome (n = 5) and acute myocarditis (n = 8). Before implantation 12 (17%) patients underwent cardiopulmonary resuscitation and 32 (46%) patients were ventilated. INTERMACS level 1, 2 and 3 was established in 35 (50%), 29 (41%) and 6 (9%) of patients, respectively. The mean preoperative lactate level was 4.05 mmol/l. The median support time was 7 days (IR= 4-15). In 18 cases, the pump was removed for myocardial recovery, in 22 cases, durable left ventricular assist devices were implanted, and 30 patients died on support. The overall 30-day survival was 51%. Statistical analysis showed that an increase in lactate per mmol/l [odds ratio (OR) 1.217; P = 0.015] and cardiopulmonary resuscitation before implantation (OR 16.74; P = 0.009) are predictors of 30-day survival. Based on these data, an algorithm for optimal short-term mechanical circulatory support selection is proposed. CONCLUSIONS: Impella treatment is feasible in severe CS. Severe organ dysfunction, as well as the level and duration of shock predict early mortality. An algorithm based on these parameters may help identify patients who would benefit from Impella 5+ support.

Safety of bioelectrical impedance analysis in advanced heart failure patients.

BACKGROUND: Cardiac cachexia and frailty are major complications of advanced heart failure (AHF). Bioelectrical impedance analysis (BIA) may provide valuable information regarding fluid balance, muscle mass and prognosis. The main concerns regarding the use of BIA in AHF patients remain arrhythmias and electromagnetic interferences with cardiac implantable electronic devices (CIEDs). Reliable data regarding patients on continuous-flow ventricular assist device (cf-VAD) remain scarce. The aim of this study is to evaluate the safety of BIA in AHF patients on pro-arrhythmogenic therapy with an implanted CIED and/or with a cf-VAD. METHODS: We prospectively performed 217 BIA measurements in 143 AHF patients at risk of severe arrhythmias due to inotropic support/a history of ventricular arrhythmias and/or treated with CIED, including 104 patients with an ICD, CRT or pacemaker and 95 patients with a cf-VAD. All patients were under continuous Electrocardiogram (ECG) monitoring and clinical surveillance for 24 hours. RESULTS: No adverse events were observed during the 217 BIA measurements: No rhythm disturbances were documented in the telemetric monitoring during or within 30 minutes after the measurement. CIEDs showed no malfunction, regardless of the location measured or the device manufacturer. In particular, no inappropriate shocks were observed. No alarms, flow disturbances, or malfunctions of the cf-VAD occurred during or after the measurements. CONCLUSION: We consider BIA a safe measurement with major clinical relevance in our cohort of AHF patients, despite an increased arrhythmic potential on inotropic support or the presence of implanted electronic devices (ICD, CRT, pacemaker and cf-VAD).