Quantitative Analysis of Clot Deposition on Extracorporeal Life Support Membrane Oxygenators Using Digital and Scanning Electron Microscopy Imaging Techniques

Device-induced thrombosis remains a major complication of extracorporeal life support (ECLS). To more thoroughly understand how blood components interact with the artificial surfaces of ECLS circuit components, assessment of clot deposition on these surfaces following clinical use is urgently needed. Scanning electron microscopy (SEM), which produces high-resolution images at nanoscale level, allows visualization and characterization of thrombotic deposits on ECLS circuitry. However, methodologies to increase the quantifiability of SEM analysis of ECLS circuit components have yet to be applied clinically. To address these issues, we developed a protocol to quantify clot deposition on ECLS membrane oxygenator gas transfer fiber sheets through digital and SEM imaging techniques. In this study, ECLS membrane oxygenator fiber sheets were obtained, fixed, and imaged after use. Following a standardized process, the percentage of clot deposition on both digital images and SEM images was quantified using ImageJ through blind reviews. The interrater reliability of quantitative analysis among reviewers was evaluated. Although this protocol focused on the analysis of ECLS membrane oxygenators, it is also adaptable to other components of the ECLS circuits such as catheters and tubing. Key features • Quantitative analysis of clot deposition using digital and scanning electron microscopy (SEM) techniques • High-resolution images at nanoscale level • Extracorporeal life support (ECLS) devices • Membrane oxygenators • Blood-contacting surfaces Graphical overview.

Comparison of hemodynamic features and thrombosis risk of membrane oxygenators with different structures: A numerical study

PURPOSE

The geometric structure of the membrane oxygenator can exert an impact on its hemodynamic features, which contribute to the development of thrombosis, thereby affecting the clinical efficacy of ECMO treatment. The purpose of this study is to investigate the impact of varying geometric structures on hemodynamic features and thrombosis risk of membrane oxygenators with different designs.

METHODS

Five oxygenator models with different structures, including different number and location of blood inlet and outlet, as well as variations in blood flow path, were established for investigation. These models are referred to as Model 1 (Quadrox-i Adult Oxygenator), Model 2 (HLS Module Advanced 7.0 Oxygenator), Model 3 (Nautilus ECMO Oxygenator), Model 4 (OxiaACF Oxygenator) and Model 5 (New design oxygenator). The hemodynamic features of these models were numerically analyzed using the Euler method combined with computational fluid dynamics (CFD). The accumulated residence time (ART) and coagulation factor concentrations (C[i], where i represents different coagulation factors) were calculated by solving the convection diffusion equation. The resulting relationships between these factors and the development of thrombosis in the oxygenator were then investigated.

RESULTS

Our results show that the geometric structure of the membrane oxygenator, including the location of the blood inlet and outlet as well as the design of the flow path, has a significant impact on the hemodynamic surroundings within the oxygenator. In comparison to Model 4, which had the inlet and outlet located in the center position, Model 1 and Model 3, which had the inlet and outlet at the edge of the blood flow field, exhibited a more uneven distribution of blood flow within the oxygenator, particularly in areas distant from the inlet and outlet, which was accompanied with lower flow velocity and higher values of ART and C[i], leading to the formation of flow dead zones and an elevated risk of thrombosis. The oxygenator of Model 5 is designed with a structure that features multiple inlets and outlets, which greatly improves the hemodynamic environment inside the oxygenator. This results in a more even distribution of blood flow within the oxygenator, reducing areas with high values of ART and C[i], and ultimately lowering the risk of thrombosis. The oxygenator of Model 3 with circular flow path section shows better hemodynamic performance compared to the oxygenator of Model 1 with square circular flow path. The overall ranking of hemodynamic performance for all five oxygenators is as follows: Model 5 > Model 4 > Model 2 > Model 3 > Model 1, indicating that Model 1 has the highest thrombosis risk while Model 5 has the lowest.

CONCLUSION

The study reveals that the different structures can affect the hemodynamic characteristics inside membrane oxygenators. The design of multiple inlets and outlets can improve the hemodynamic performance and reduce the thrombosis risk in membrane oxygenators. These findings of this study can be used to guide the optimization design of membrane oxygenators for improving hemodynamic surroundings and reducing thrombosis risk.

Comparison of the effect of membrane sizes and fibre arrangements of two membrane oxygenators on the inflammatory response, oxygenation and decarboxylation in a rat model of extracorporeal membrane oxygenation

Background

Extracorporeal membrane oxygenation (ECMO) has gained widespread acceptance for the treatment of critically ill patients suffering from cardiac and/or respiratory failure. Various animal models have been developed to investigate the adverse effects induced by ECMO. Different membrane oxygenators have been used with varying priming volumes and membrane surfaces (Micro-1, small animal membrane oxygenator (SAMO)).

Methods

Sixteen male Lewis rats (350–400 g) were randomly assigned to receive ECMO with Micro-1 or SAMO (n = 8, respectively). Venoarterial ECMO was established after cannulation of the femoral artery and the jugular vein. The cardiac output was measured using a left-ventricular conductance catheter. The oxygen fraction of the ECMO was set to 1.0, 0.75, 0.5 and 0.21 after a stabilisation period of 15 min. Further, arterial blood gas analyses were performed at baseline, and during the first hour every 15 min after commencing the ECMO, and subsequently every 30 min. Dilutional anaemia was calculated using haemoglobin concentration at baseline, and 15 min after the start of ECMO therapy. Moreover, inflammation was determined by measuring tumour necrosis factor alpha, interleukin-6 and -10 at baseline and every 30 min.

Results

Animals of the Micro-1 group showed a significantly lower dilutional anaemia (ΔHaemoglobin t₀ – t_0.25: SAMO 6.3 [5.6–7.5] g/dl vs. Micro-1 5.6 [4.6–5.8] g/dl; p = 0.028). Further, significantly higher oxygen partial pressure was measured in the SAMO group, at an oxygen fraction of 0.75, 0.5 and 0.21 (380 [356–388] vs. 314 [263–352] mmHg, p = 0.002; 267 [249–273] mmHg vs. 197 [140–222] mmHg, p = 0.002; 87 [82–106] mmHg vs. 76 [60–79] mmHg, p = 0.021, respectively). However, no differences were found regarding the oxygen fraction of 1.0, in terms of carbon-dioxide partial pressure and cardiac output. Moreover, in the Micro-1 group tumour necrosis factor alpha was increased after 60 min and interleukin-6 after 120 min.

Conclusion

While the dilutional anaemia was increased after commencing the ECMO, the oxygenation was augmented in the SAMO group. The inflammatory response was elevated in the Micro-1 group.