Active learning for extracting surgomic features in robot-assisted minimally invasive esophagectomy: a prospective annotation study.

BACKGROUND: With Surgomics, we aim for personalized prediction of the patient's surgical outcome using machine-learning (ML) on multimodal intraoperative data to extract surgomic features as surgical process characteristics. As high-quality annotations by medical experts are crucial, but still a bottleneck, we prospectively investigate active learning (AL) to reduce annotation effort and present automatic recognition of surgomic features. METHODS: To establish a process for development of surgomic features, ten video-based features related to bleeding, as highly relevant intraoperative complication, were chosen. They comprise the amount of blood and smoke in the surgical field, six instruments, and two anatomic structures. Annotation of selected frames from robot-assisted minimally invasive esophagectomies was performed by at least three independent medical experts. To test whether AL reduces annotation effort, we performed a prospective annotation study comparing AL with equidistant sampling (EQS) for frame selection. Multiple Bayesian ResNet18 architectures were trained on a multicentric dataset, consisting of 22 videos from two centers. RESULTS: In total, 14,004 frames were tag annotated. A mean F1-score of 0.75 ± 0.16 was achieved for all features. The highest F1-score was achieved for the instruments (mean 0.80 ± 0.17). This result is also reflected in the inter-rater-agreement (1-rater-kappa > 0.82). Compared to EQS, AL showed better recognition results for the instruments with a significant difference in the McNemar test comparing correctness of predictions. Moreover, in contrast to EQS, AL selected more frames of the four less common instruments (1512 vs. 607 frames) and achieved higher F1-scores for common instruments while requiring less training frames. CONCLUSION: We presented ten surgomic features relevant for bleeding events in esophageal surgery automatically extracted from surgical video using ML. AL showed the potential to reduce annotation effort while keeping ML performance high for selected features. The source code and the trained models are published open source.

Surgomics: personalized prediction of morbidity, mortality and long-term outcome in surgery using machine learning on multimodal data.

BACKGROUND: Personalized medicine requires the integration and analysis of vast amounts of patient data to realize individualized care. With Surgomics, we aim to facilitate personalized therapy recommendations in surgery by integration of intraoperative surgical data and their analysis with machine learning methods to leverage the potential of this data in analogy to Radiomics and Genomics. METHODS: We defined Surgomics as the entirety of surgomic features that are process characteristics of a surgical procedure automatically derived from multimodal intraoperative data to quantify processes in the operating room. In a multidisciplinary team we discussed potential data sources like endoscopic videos, vital sign monitoring, medical devices and instruments and respective surgomic features. Subsequently, an online questionnaire was sent to experts from surgery and (computer) science at multiple centers for rating the features' clinical relevance and technical feasibility. RESULTS: In total, 52 surgomic features were identified and assigned to eight feature categories. Based on the expert survey (n = 66 participants) the feature category with the highest clinical relevance as rated by surgeons was "surgical skill and quality of performance" for morbidity and mortality (9.0 ± 1.3 on a numerical rating scale from 1 to 10) as well as for long-term (oncological) outcome (8.2 ± 1.8). The feature category with the highest feasibility to be automatically extracted as rated by (computer) scientists was "Instrument" (8.5 ± 1.7). Among the surgomic features ranked as most relevant in their respective category were "intraoperative adverse events", "action performed with instruments", "vital sign monitoring", and "difficulty of surgery". CONCLUSION: Surgomics is a promising concept for the analysis of intraoperative data. Surgomics may be used together with preoperative features from clinical data and Radiomics to predict postoperative morbidity, mortality and long-term outcome, as well as to provide tailored feedback for surgeons.

Transforming growth factor-ß1 promotes fibrosis but attenuates calcification of valvular tissue applied as a three-dimensional calcific aortic valve disease model.

Calcific aortic valve disease (CAVD) is characterized by valvular fibrosis and calcification and driven by differentiating valvular interstitial cells (VICs). Expression data from patient biopsies suggest that transforming growth factor (TGF)-ß1 is implicated in CAVD pathogenesis. However, CAVD models using isolated VICs failed to deliver clear evidence on the role of TGF-ß1. Thus, employing cultures of aortic valve leaflets, we investigated effects of TGF-ß1 in a tissue-based three-dimensional (3-D) CAVD model. We found that TGF-ß1 induced phosphorylation of Mothers against decapentaplegic homolog (SMAD) 3 and expression of SMAD7, indicating effective downstream signal transduction in valvular tissue. Thus, TGF-ß1 increased VIC contents of rough endoplasmic reticulum, Golgi, and secretory vesicles as well as tissue levels of RNA and protein. In addition, TGF-ß1 raised expression of proliferation marker cyclin D1, attenuated VIC apoptosis, and upregulated VIC density. Moreover, TGF-ß1 intensified myofibroblastic VIC differentiation as evidenced by increased α-smooth muscle actin and collagen type I along with diminished vimentin expression. In contrast, TGF-ß1 attenuated phosphorylation of SMAD1/5/8 and upregulation of ß-catenin while inhibiting osteoblastic VIC differentiation as revealed by downregulation of osteocalcin expression, alkaline phosphatase activity, and extracellular matrix incorporation of hydroxyapatite. Collectively, these effects resulted in blocking of valvular tissue calcification and associated disintegration of collagen fibers. Instead, TGF-ß1 induced development of fibrosis. Overall, in a tissue-based 3-D CAVD model, TGF-ß1 intensifies expressional and proliferative activation along with myofibroblastic differentiation of VICs, thus triggering dominant fibrosis. Simultaneously, by inhibiting SMAD1/5/8 activation and canonical Wnt/ß-catenin signaling, TGF-ß1 attenuates osteoblastic VIC differentiation, thus blocking valvular tissue calcification. These findings question a general phase-independent CAVD-promoting role of TGF-ß1.NEW & NOTEWORTHY Employing aortic valve leaflets as a tissue-based three-dimensional disease model, our study investigates the role of transforming growth factor (TGF)-ß1 in calcific aortic valve disease pathogenesis. We find that, by activating Mothers against decapentaplegic homolog 3, TGF-ß1 intensifies expressional and proliferative activation along with myofibroblastic differentiation of valvular interstitial cells, thus triggering dominant fibrosis. Simultaneously, by inhibiting activation of Mothers against decapentaplegic homolog 1/5/8 and canonical Wnt/ß-catenin signaling, TGF-ß1 attenuates apoptosis and osteoblastic differentiation of valvular interstitial cells, thus blocking valvular tissue calcification. These findings question a general phase-independent calcific aortic valve disease-promoting role of TGF-ß1.

The forkhead transcription factor Foxo3 negatively regulates natural killer cell function and viral clearance in myocarditis.

Aims: Foxo3 is a transcription factor involved in cell metabolism, survival, and inflammatory disease. However, mechanistic insight in Foxo3 effects is still limited. Here, we investigated the role of Foxo3 on natural killer (NK) cell responses and its effects in viral myocarditis. Methods and results: Effects of Foxo3 on viral load and immune responses were investigated in a model of coxsackie virus B3 myocarditis in wild-type (WT) and Foxo3 deficient mice. Reduced immune cell infiltration, viral titres, and pro-inflammatory cytokines in cardiac tissue were observed in Foxo3-/- mice 7 days post-infection (p.i.). Viral titres were also attenuated in hearts of Foxo3-/- mice at Day 3 while interferon-γ (IFNγ) and NKp46 expression were up-regulated suggesting early viral control by enhanced NK cell activity. CD69 expression of NK cells, frequencies of CD11b+CD27+ effector NK cells and cytotoxicity of Foxo3-/- mice was enhanced compared to WT littermates. Moreover, microRNA-155 expression, essential in NK cell activation, was elevated in Foxo3-/- NK cells while its inhibition led to diminished IFNγ production. Healthy humans carrying the longevity-associated FOXO3 single nucleotide polymorphism (SNP) rs12212067 exhibited reduced IFNγ and cytotoxic degranulation of NK cells. Viral inflammatory cardiomyopathy (viral CMI) patients with this SNP showed a poorer outcome due to less efficient virus control. Conclusion: Our results implicate Foxo3 in regulating NK cell function and suggest Foxo3 playing an important role in the antiviral innate immunity. Thus, enhanced FOXO3 activity such as in the polymorphism rs12212067 may be protective in chronic inflammation such as cancer and cardiovascular disease but disadvantageous to control acute viral infection.

Adiponectin modulates NK-cell function.

Adiponectin (APN) has been shown to exert antiinflammatory effects in various disease models but little is known concerning its regulation of NK-cell function. Here, we show that the majority of human CD56(dim) NK cells express surface Adiponectin receptor (AdipoR) 1 and 2 while most CD56(high) NK cells are AdipoR-negative. Toll-like receptor (TLR) ligand-induced IFN-γ production was diminished by APN while it had no influence on NK-cell cytotoxicity. In contrast only a small subpopulation of murine NK cells expresses surface AdipoRs, but about 90% store them intracellularly. APN-deficient knockout (KO) mice had elevated frequencies of NK cells. However, cytotoxic degranulation of NK cells was decreased in APN knockout (APN-KO) animals. Accordingly, frequencies of CD11b(high) CD27(high) and CD94(high) effector NK cells and expression of NKG2D were lower in APN-KO mice. Upon CVB3 infection NK-cell function was restored in APN-KO mice. Our data suggest that in addition to its antiinflammatory effects APN also influences the numerical and differentiation status of NK cells, which may further impact the outcome of immune-mediated diseases in APN-KO mice.

One model to use them all: training a segmentation model with complementary datasets.

PURPOSE: Understanding surgical scenes is crucial for computer-assisted surgery systems to provide intelligent assistance functionality. One way of achieving this is via scene segmentation using machine learning (ML). However, such ML models require large amounts of annotated training data, containing examples of all relevant object classes, which are rarely available. In this work, we propose a method to combine multiple partially annotated datasets, providing complementary annotations, into one model, enabling better scene segmentation and the use of multiple readily available datasets. METHODS: Our method aims to combine available data with complementary labels by leveraging mutual exclusive properties to maximize information. Specifically, we propose to use positive annotations of other classes as negative samples and to exclude background pixels of these binary annotations, as we cannot tell if a positive prediction by the model is correct. RESULTS: We evaluate our method by training a DeepLabV3 model on the publicly available Dresden Surgical Anatomy Dataset, which provides multiple subsets of binary segmented anatomical structures. Our approach successfully combines 6 classes into one model, significantly increasing the overall Dice Score by 4.4% compared to an ensemble of models trained on the classes individually. By including information on multiple classes, we were able to reduce the confusion between classes, e.g. a 24% drop for stomach and colon. CONCLUSION: By leveraging multiple datasets and applying mutual exclusion constraints, we developed a method that improves surgical scene segmentation performance without the need for fully annotated datasets. Our results demonstrate the feasibility of training a model on multiple complementary datasets. This paves the way for future work further alleviating the need for one specialized large, fully segmented dataset but instead the use of already existing datasets.

SeeSaw: Learning Soft Tissue Deformation from Laparoscopy Videos with GNNs.

A major challenge in image-guided laparoscopic surgery is that structures of interest often deform and go, even if only momentarily, out of view. Methods which rely on having an up-to-date impression of those structures, such as registration or localisation, are undermined in these circumstances. This is particularly true for soft-tissue structures that continually change shape - in registration, they must often be re-mapped. Furthermore, methods which require 'revisiting' of previously seen areas cannot in principle function reliably in dynamic contexts, drastically weakening their uptake in the operating room. We present a novel approach for learning to estimate the deformed states of previously seen soft tissue surfaces from currently observable regions, using a combined approach that includes a Graph Neural Network (GNN). The training data is based on stereo laparoscopic surgery videos, generated semi-automatically with minimal labelling effort. Trackable segments are first identified using a feature detection algorithm, from which surface meshes are produced using depth estimation and delaunay triangulation. We show the method can predict the displacements of previously visible soft tissue structures connected to currently visible regions with observed displacements, both on our own data and porcine data. Our innovative approach learns to compensate non-rigidity in abdominal endoscopic scenes directly from stereo laparoscopic videos through targeting a new problem formulation, and stands to benefit a variety of target applications in dynamic environments. Project page for this work: https://gitlab.com/nct_tso_public/seesaw-soft-tissue-deformation.

Anatomy segmentation in laparoscopic surgery: comparison of machine learning and human expertise - an experimental study.

BACKGROUND: Lack of anatomy recognition represents a clinically relevant risk in abdominal surgery. Machine learning (ML) methods can help identify visible patterns and risk structures; however, their practical value remains largely unclear. MATERIALS AND METHODS: Based on a novel dataset of 13 195 laparoscopic images with pixel-wise segmentations of 11 anatomical structures, we developed specialized segmentation models for each structure and combined models for all anatomical structures using two state-of-the-art model architectures (DeepLabv3 and SegFormer) and compared segmentation performance of algorithms to a cohort of 28 physicians, medical students, and medical laypersons using the example of pancreas segmentation. RESULTS: Mean Intersection-over-Union for semantic segmentation of intra-abdominal structures ranged from 0.28 to 0.83 and from 0.23 to 0.77 for the DeepLabv3-based structure-specific and combined models, and from 0.31 to 0.85 and from 0.26 to 0.67 for the SegFormer-based structure-specific and combined models, respectively. Both the structure-specific and the combined DeepLabv3-based models are capable of near-real-time operation, while the SegFormer-based models are not. All four models outperformed at least 26 out of 28 human participants in pancreas segmentation. CONCLUSIONS: These results demonstrate that ML methods have the potential to provide relevant assistance in anatomy recognition in minimally invasive surgery in near-real-time. Future research should investigate the educational value and subsequent clinical impact of the respective assistance systems.

The Dresden Surgical Anatomy Dataset for Abdominal Organ Segmentation in Surgical Data Science.

Laparoscopy is an imaging technique that enables minimally-invasive procedures in various medical disciplines including abdominal surgery, gynaecology and urology. To date, publicly available laparoscopic image datasets are mostly limited to general classifications of data, semantic segmentations of surgical instruments and low-volume weak annotations of specific abdominal organs. The Dresden Surgical Anatomy Dataset provides semantic segmentations of eight abdominal organs (colon, liver, pancreas, small intestine, spleen, stomach, ureter, vesicular glands), the abdominal wall and two vessel structures (inferior mesenteric artery, intestinal veins) in laparoscopic view. In total, this dataset comprises 13195 laparoscopic images. For each anatomical structure, we provide over a thousand images with pixel-wise segmentations. Annotations comprise semantic segmentations of single organs and one multi-organ-segmentation dataset including segments for all eleven anatomical structures. Moreover, we provide weak annotations of organ presence for every single image. This dataset markedly expands the horizon for surgical data science applications of computer vision in laparoscopic surgery and could thereby contribute to a reduction of risks and faster translation of Artificial Intelligence into surgical practice.

Adiponectin is a negative regulator of antigen-activated T cells.

Adiponectin (APN), a cytokine constitutively produced in fat tissue, has been shown to exert anti-inflammatory effects in various disease models. While the influence of APN on monocytic cells has been extensively studied in vitro, little is known about its role in T cells. In this study, we show that while <10% of human peripheral blood T cells express adiponectin receptors (AdipoRs) on their surface, most T cells store AdipoRs in intracellular compartments. AdipoRs colocalized with immune regulatory molecules CTLA-4 and TIRC7 within clathrin-coated vesicles. After stimulation, the expression of adiponectin receptor 1 (AdipoR1) and AdipoR2 was upregulated on the surface of antigen-specific T cells, as determined by tetramer or CD137 staining, and AdipoR1 and AdipoR2 coexpressed with CTLA-4. Addition of APN resulted in a significant diminution of antigen-specific T-cell expansion. Mechanistically, APN enhanced apoptosis and inhibited proliferation of antigen-specific T-cell lines. Further, APN directly inhibited cytokine production in response to antigen stimulation. In line with the in vitro data, APN-deficient (knockout, KO) mice had higher frequencies of CD137(+) T cells upon Coxsackie B virus infection. Altogether, our data suggest that APN is a novel negative T-cell regulator. In contrast to the CTLA-4 ligand B7 only expressed on APCs, APN is abundant in human plasma.

Adiponectin expression in patients with inflammatory cardiomyopathy indicates favourable outcome and inflammation control.

AIMS: Circulating adiponectin (APN) is an immunomodulatory, pro-angiogenic, and anti-apoptotic adipocytokine protecting against acute viral heart disease and preventing pathological remodelling after cardiac injury. The purpose of this study was to describe the regulation and effects of APN in patients with inflammatory cardiomyopathy (DCMi). METHODS AND RESULTS: Adiponectin expression and outcome were assessed in 173 patients with DCMi, 30 patients with non-inflammatory DCM, and 30 controls. Mechanistic background of these findings was addressed in murine experimental autoimmune myocarditis (EAM), a model of human DCMi, and further elucidated in vitro. Adiponectin plasma concentrations were significantly higher in DCMi compared with DCM or controls, i.e. 6.8 ± 3.9 µg/mL vs. 5.4 ± 3.6 vs. 4.76 ± 2.5 µg/mL (P< 0.05, respectively) and correlated significantly with cardiac mononuclear infiltrates (CD3+: r(2)= 0.025, P= 0.038; CD45R0+: r(2)= 0.058, P= 0.018). At follow-up, DCMi patients with high APN levels showed significantly increased left ventricular ejection fraction improvement, decreased left ventricular end-diastolic diameter, and reduced cardiac inflammatory infiltrates compared with patients with low APN levels. A multivariate linear regression analysis implicated APN as an independent prognostic factor for inhibition of cardiac inflammation. In accordance with these findings in human DCMi, EAM mice exhibited elevated plasma APN. Adiponectin gene transfer led to significant downregulation of key inflammatory mediators promoting disease. Mechanistically, APN acted as a negative regulator of T cells by reducing antigen specific expansion (P< 0.01) and suppressed TNFα-mediated NFκB activation (P< 0.01) as well as release of reactive oxygen species in cardiomyocytes. CONCLUSION: Our results implicate that APN acts as endogenously upregulated anti-inflammatory cytokine confining cardiac inflammation and progression in DCMi.

The combination approach with Rhokinase inhibition and mechanical circulatory support in myocardial ischemia-reperfusion injury: Rho-kinase inhibition and ventricular unloading.

BACKGROUND: It remains unclear whether the Rho-kinase (ROCK) inhibition in combination with mechanical circulatory support (MCS) had a synergic protective effect on myocardial ischemia (MI)/reperfusion injury in therapeutic strategies for acute myocardial infarction (AMI). We report the results of an approach using a rat model consisting of a miniaturized cardiopulmonary bypass (CPB) and AMI. METHODS: A total of 25 male Wistar rats were randomized into 5 groups: (1) Sham: a suture was passed under the left anterior descending artery (LAD) creating no MI. A vehicle solution (0.9% saline) was injected intraperitoneally. (2) Myocardial ischemia (MI) + vehicle (MI + V): LAD was ligated for 30 min and reperfused for 120 min, followed by administration of vehicle solution. (3) MI + fasudil (MI + F): the work sequence of group 2, but the selective ROCK inhibitor fasudil (10 mg/kg) was administered instead. (4) MI + V + CPB: CPB was initiated 15 min after the ligation of the LAD to the end of the reperfusion, in addition to the work sequence in group 2. (5) In the MI + F + CPB group, the work sequence of group 4, but with fasudil administration (10 mg/kg). RESULTS: Measurements of cardiac function through conductance catheter indicated that the drop of + dP/dt after reperfusion was moderately limited in MI + F + CPB (vs. MI + V, dP/dt p = 0.22). The preload recruitable stroke work was moderately improved in the MI + F + CPB (p = 0.23) compared with the corresponding control animals (MI + V). Phosphorylated protein kinase B expression in the MI + V + CPB and MI + F + CPB was higher than that in MI + V (p = 0.33). CONCLUSION: Therefore, fasudil administration with MCS resulted in a moderately better left ventricular performance.

AdipoRon Attenuates Inflammation and Impairment of Cardiac Function Associated With Cardiopulmonary Bypass-Induced Systemic Inflammatory Response Syndrome.

Background Cardiac surgery using cardiopulmonary bypass (CPB) frequently provokes a systemic inflammatory response syndrome, which is triggered by TLR4 (Toll-like receptor 4) and TNF-α (tumor necrosis factor α) signaling. Here, we investigated whether the adiponectin receptor 1 and 2 agonist AdipoRon modulates CPB-induced inflammation and cardiac dysfunction. Methods and Results Rats underwent CPB with deep hypothermic circulatory arrest and were finally weaned from the heart-lung machine. Compared with vehicle, AdipoRon application attenuated the CPB-induced impairment of mean arterial pressure following deep hypothermic circulatory arrest. During the weaning and postweaning phases, heart rate and mean arterial pressure in all AdipoRon animals (7 of 7) remained stable, while cardiac rhythm was irretrievably lost in 2 of 7 of the vehicle-treated animals. The AdipoRon-mediated improvements of cardiocirculatory parameters were accompanied by increased plasma levels of IL (interleukin) 10 and diminished concentrations of lactate and K+. In myocardial tissue, AdipoRon activated AMP-activated protein kinase (AMPK) while attenuating CPB-induced degradation of nuclear factor κB inhibitor α (IκBα), upregulation of TNF-α, IL-1ß, CCL2 (C-C chemokine ligand 2), nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, and inducible nitric oxide synthase. Correspondingly, in cultured cardiac myocytes, cardiac fibroblasts, and vascular endothelial cells, AdipoRon activated AMPK, upregulated IL-10, and attenuated activation of nuclear factor κB, as well as upregulation of TNF-α, IL-1ß, CCL2, NADPH oxidase, and inducible nitric oxide synthase induced by lipopolysaccharide or TNF-α. In addition, the treatment of cardiac myocytes with the AMPK activator 5-aminoimidazole-4-carboxamide 1-ß-D-ribofuranoside resulted in a similar inhibition of lipopolysaccharide- and TNF-α-induced inflammatory cell phenotypes as for AdipoRon. Conclusions Our observations indicate that AdipoRon attenuates CPB-induced inflammation and impairment of cardiac function through AMPK-mediated inhibition of proinflammatory TLR4 and TNF-α signaling in cardiac cells and upregulation of immunosuppressive IL-10.

Active learning using deep Bayesian networks for surgical workflow analysis.

PURPOSE: For many applications in the field of computer-assisted surgery, such as providing the position of a tumor, specifying the most probable tool required next by the surgeon or determining the remaining duration of surgery, methods for surgical workflow analysis are a prerequisite. Often machine learning-based approaches serve as basis for analyzing the surgical workflow. In general, machine learning algorithms, such as convolutional neural networks (CNN), require large amounts of labeled data. While data is often available in abundance, many tasks in surgical workflow analysis need annotations by domain experts, making it difficult to obtain a sufficient amount of annotations. METHODS: The aim of using active learning to train a machine learning model is to reduce the annotation effort. Active learning methods determine which unlabeled data points would provide the most information according to some metric, such as prediction uncertainty. Experts will then be asked to only annotate these data points. The model is then retrained with the new data and used to select further data for annotation. Recently, active learning has been applied to CNN by means of deep Bayesian networks (DBN). These networks make it possible to assign uncertainties to predictions. In this paper, we present a DBN-based active learning approach adapted for image-based surgical workflow analysis task. Furthermore, by using a recurrent architecture, we extend this network to video-based surgical workflow analysis. To decide which data points should be labeled next, we explore and compare different metrics for expressing uncertainty. RESULTS: We evaluate these approaches and compare different metrics on the Cholec80 dataset by performing instrument presence detection and surgical phase segmentation. Here we are able to show that using a DBN-based active learning approach for selecting what data points to annotate next can significantly outperform a baseline based on randomly selecting data points. In particular, metrics such as entropy and variation ratio perform consistently on the different tasks. CONCLUSION: We show that using DBN-based active learning strategies make it possible to selectively annotate data, thereby reducing the required amount of labeled training in surgical workflow-related tasks.

Targeting of cell-free DNA by DNase I diminishes endothelial dysfunction and inflammation in a rat model of cardiopulmonary bypass.

The use of cardiopulmonary bypass (CPB) results in the activation of leukocytes, release of neutrophil extracellular traps (NETs) and severe inflammation. We hypothesize that targeting of circulating cell-free DNA (cfDNA) by DNases might represent a feasible therapeutic strategy to limit CPB-associated side effects. Male Wistar rats (n = 24) underwent CPB with deep hypothermic circulatory arrest (DHCA) and were divided into 3 groups: control (group 1), one i.v. bolus DNase I before CPB start (group 2) and a second DNase I dose before reperfusion (group 3). We found a positive correlation between plasma cfDNA/NETs levels and compromised endothelial vasorelaxation after CPB. DNase I administration significantly diminished plasma cfDNA/NETs levels. Further, a dose-dependent improvement in endothelial function accompanied by significant reduction of circulating intercellular adhesion molecule (ICAM)-1 was observed. Rats of group 3 had significantly reduced plasma IL-6 levels and downregulated expression of adhesion molecules resulting in impaired leukocyte extravasation and reduced MPO activity in lungs. Mechanistically, digestion of NETs by DNase I significantly diminished NETs-dependent upregulation of adhesion molecules in human endothelial cells. Altogether, systemic DNase I administration during CPB efficiently reduced cfDNA/NETs-mediated endothelial dysfunction and inflammation and might represents a promising therapeutic strategy for clinical practice.

Characterization of the Epicardial Adipose Tissue in Decellularized Human-Scaled Whole Hearts: Implications for the Whole-Heart Tissue Engineering.

Whole-organ engineering is an innovative field of regenerative medicine with growing translational perspectives. Recent reports suggest the feasibility of decellularization and repopulation of entire human size hearts. However, little is known about the susceptibility of epicardial adipose tissue (EAT) to decellularization. In this study, human size hearts of ovine donors were subjected to perfusion-based decellularization using detergent solutions. Upon basic histological evaluation and total DNA measurement myocardial regions prove largely decellularized while EAT demonstrated cellular remnants, further confirmed by transmission electron microscopy. Western blot analysis showed a significant reduction in lipid-associated and cardiac proteins. However, gas chromatography revealed unchanged proportional composition of fatty acids in EAT of decellularized whole hearts. Finally, cell culture medium conditioned with EAT from decellularized whole hearts had a significant deleterious effect on cardiac fibroblasts. These data suggest that perfusion decellularization of human size whole hearts provides inconsistent efficacy regarding donor material removal from myocardial regions as opposed to EAT.

Adiponectin attenuates profibrotic extracellular matrix remodeling following cardiac injury by up-regulating matrix metalloproteinase 9 expression in mice.

Adiponectin (APN) is a multifunctional adipocytokine that inhibits myocardial fibrosis, dilatation, and left ventricular (LV) dysfunction after myocardial infarction (MI). Coxsackievirus B3 (CVB3) myocarditis is associated with intense extracellular matrix (ECM) remodeling which might progress to dilated cardiomyopathy. Here, we investigated in experimental CVB3 myocarditis whether APN inhibits adverse ECM remodeling following cardiac injury by affecting matrix metalloproteinase (MMP) expression. Cardiac injury was induced by CVB3 infection in APN knockout (APN-KO) and wild-type (WT) mice. Expression and activity of MMPs was quantified by qRT-PCR and zymography, respectively. Activation of protein kinases was assessed by immunoblot. In cardiac myocytes and fibroblasts APN up-regulates MMP-9 expression via activation of 5' adenosine monophosphate-activated protein kinase (AMPK) and extracellular signal-regulated kinase (ERK)1/2 which function as master regulators of inflammation-induced MMP-9 expression. Correspondingly, APN further increased up-regulation of MMP-9 expression triggered by tumor necrosis factor (TNF)α, lipopolysaccharide (LPS) and R-848 in cardiac fibroblasts. In vivo, compared to WT mice cardiac MMP-9 activity and serum levels of carboxy-terminal telopeptide of type I collagen (ICTP) were attenuated in APN-KO mice in subacute (day 7 p.i.) CVB3 myocarditis. Moreover, on day 3 and day 7 post CVB3 infection splenic MMP-9 expression was diminished in APN-KO mice correlating with attenuated myocardial immune cell infiltration in subacute CVB3 myocarditis. These results indicate that APN attenuates adverse cardiac remodeling following cardiac injury by up-regulating MMP-9 expression in cardiac and immune cells. Thus, APN mediates intensified collagen cleavage that might explain inhibition of LV fibrosis and dysfunction.

Whole-Heart Construct Cultivation Under 3D Mechanical Stimulation of the Left Ventricle.

Today the concept of Whole-Heart Tissue Engineering represents one of the most promising approaches to the challenge of synthesizing functional myocardial tissue. At the current state of scientific and technological knowledge it is a principal task to transfer findings of several existing and widely investigated models to the process of whole-organ tissue engineering. Hereby, we present the first bioreactor system that allows the integrated 3D biomechanical stimulation of a whole-heart construct while allowing for simultaneous controlled perfusion of the coronary system.

The extracellular isoform of superoxide dismutase has a significant impact on cardiovascular ischaemia and reperfusion injury during cardiopulmonary bypass.

OBJECTIVES: Cardiac surgery with cardiopulmonary bypass (CPB) provokes ischaemia and reperfusion injury (IRI). Superoxide is a main mediator of IRI and is detoxified by superoxide dismutases (SODs). Extracellular SOD (SOD3) is the prevailing isoform in the cardiovascular system. Its mutation is associated with elevated risk for ischaemic heart disease as epidemiological and experimental studies suggest. We investigated the influence of SOD3 on IRI in the context of CPB and hypothesized a protective role for this enzyme. METHODS: Mutant rats with loss of SOD3 function induced by amino acid shift, SOD3-E124D, (SOD3 mutant; n = 9) were examined in a model of CPB with deep hypothermic circulatory arrest provoking global IRI and compared with SOD3 competent controls (n = 8) as well as sham animals (n = 7). SOD3 plasma activity was photometrically measured with a diazo dye-forming reagent. Activation of cardioprotective rescue pathways (p44-42 MAPK and STAT3), cleavage of PARP-1, expression of SOD isoforms (SOD1, 2 and 3) and nitric oxide metabolism were analysed on the protein level by western blot. To evaluate whether SOD3 inactivity directly affects the myocardium, we isolated adult cardiac myocytes, which underwent hypoxia prior to protein analyses. RESULTS: Relative SOD3 plasma activity in SOD3 mutant rats was significantly decreased by at least 50% compared with that in SOD3 competent controls (prior to euthanasia P = 0.008). Effectively, physiological parameters [heart rate and mean arterial pressure (MAP)] indicated a trend toward impaired handling of ischaemia and reperfusion in SOD3 mutants: after reperfusion, mean heart rate was 46 bpm lower (P = 0.083) and MAP 8 mmHg lower (P = 0.288) than that in SOD competent controls. Decreased SOD3 activity led to reduced activation of cardioprotective rescue pathways in vivo and in vitro: relative activation of p44-42 MAPK (P = 0.074) and STAT3 (P = 0.027) was more than 30% decreased in heart and aortic tissue of SOD3 mutants (activity normalized to sham control as 1). After CPB, cleavage of PARP-1 was doubled in the control group (P = 0.017), but increased 3-fold in SOD3 mutants (P = 0.002). Furthermore, 3-nitrotyrosine as a measure of decreased nitric oxide bioavailability and other SOD isoforms (SOD1 and 2) were increased. CONCLUSIONS: Collectively, SOD3 has a significant cardioprotective role in cases of IRI and directly affects the myocardium as hypothesized. Exploration of intervention strategies targeting SOD3 may provide therapeutic options against IRI and associated systemic inflammation.

Modulation of Immunologic Response by Preventive Everolimus Application in a Rat CPB Model.

Everolimus (EVL) is widely used in solid organ transplantation. It is known to have antiproliferative and immunosuppressive abilities via inhibition of the mTOR pathway. Preventive EVL administration may lower inflammation induced by cardiopulmonary bypass (CPB) and reduce systemic inflammatory response syndrome (SIRS). After oral loading with EVL 2.5 mg/kg/day (n = 11) or placebo (n = 11) for seven consecutive days, male Wistar rats (400-500 g) were connected to a miniaturised heart-lung-machine performing a deep hypothermic circulatory arrest protocol. White blood cells (WBC) were significantly reduced in EVL-pretreated animals before start of CPB with a preserved reduction by trend at all other time points. Ischemia/reperfusion led to decreased glucose levels. Application of EVL significantly increased glucose levels after reperfusion. In addition, potassium levels were significantly lower in EVL-treated animals at the end of reperfusion. Immunoblotting revealed increased S6 levels after CPB. EVL decreased phosphorylation of S6 in the heart and kidney, which indicates an inhibition of mTOR pathway. Moreover, EVL significantly modified phosphorylation of AKT, while decreasing IL2, IL6, RANTES, and TNFα (n = 6). Preventive application of EVL may modulate inflammation by inhibition of mammalian target of rapamycin (mTOR) pathway and reduction of proinflammatory cytokines. This may be beneficial to evade SIRS-related morbidities after CPB.