First level prevention instead of third level intervention-review of research to improve biocompatibility and performance of capillary membrane apheresis in critically ill patients.

In intensive care medicine, convection-based apheresis is of growing interest. Applying extracorporeal systems in the critically ill patient can cause severe complications like nosocomial infections and bleeding, which can be worsened or even initialized by the anticoagulation protocol used. Furthermore, the filter modules (hemo- and plasmafilters) often tend to a fast blockage. A decrease in sieving performance due to membrane fouling may be tolerable for some time, but the complete blockage of high percentages of hollow fibers, which is named "clotting," often requires the immediate exchange of the filter. Extracorporeal detoxification and high clearance renal replacement regimes both require high blood flow and filtration rates. As a consequence, filter clotting and anticoagulation-associated bleeding are the most sensitive aspects in these applications. We were interested in the paradox phenomenon of the parallel occurrence of intra vitam bleeding and filter clotting in critically ill patients. Through stepwise investigations based on in vitro and animal experiments, we identified a stasis of blood flow followed by blood cell sedimentation and aggregation ("clogging") as the main factor of hollow fiber blockage in hemo- and plasma filters. As a result, various aspects which increase the risk of stasis inside the hollow fibers were investigated, for example, patient's hemorheology, configuration of an extracorporeal treatment system including interaction of catheter features with the filtration procedure, and basic therapeutic approaches such as colloidal volume substitutes and tolerated acidosis. Finally, an etiological triad for the blockage of hollow fibers due to filter clogging and consecutive filter failure was formed.

Adverse influence of mixed acidemia on the biocompatibility of continuous veno-venous hemofiltration with respect to the lungs.

Experimental data indicate that hypercapnic adidosis has anti-inflammatory effects. These anti-inflammatory effects may even be a beneficial property in case of low tidal volume ventilation with consecutive hypercapnic acidosis. It is unclear whether these anti-inflammatory effects predominate in critically ill patients who suffer from multiple pro- and anti-inflammatory insults like extracorporeal organ support (pro-inflammatory), metabolic acidosis (pro- and anti-inflammatory), as well as hypoxia (pro-inflammatory). Eighteen pigs were randomized into three groups, mechanically ventilated and connected to a continuous veno-venous hemofiltration (CVVH) as pro-inflammatory insult. A reference group with normal acid-base state obtained normoventilation; a normoxemic acidemia group obtained normoxemic, mixed acidemia due to infusion of lactic and hyperchloremic acid and low tidal volume ventilation, and in a hypoxemic acidemia group the mixed acidemia was paralleled by hypoxemia. Lung histology including pulmonary leukocyte invasion, blood gases, blood cell counts, and hemodynamics were examined. The histological examination of the lungs of acidemic pigs showed a suppressed invasion of leukocytes and thinner alveolar walls compared with normoventilated and with hypoxemic pigs. Enhanced congestion and alveolar red blood cells (RBCs) combined with an increase of the pulmonary artery pressure were observed in acidemic pigs in comparison with the reference group. Normoxemic acidemia reduced the pro-inflammatory reaction to the CVVH and mechanical ventilation in the ventilated lung areas in the form of pulmonary leukocyte invasion. However, this did not result in reduced scores for lung injury. Instead, an increased score for criteria which represent lung injury (congestion and alveolar RBCs) was observed in acidemic pigs.

Impact of different types of resuscitation fluids on coagulation and continuous venovenous hemofiltration hemocompatibility in a porcine model.

Intensive therapy demanding diseases (organ failure or sepsis) are assumed to be the etiology behind a decreased biocompatibility of extracorporeal systems for renal replacement therapy (RRT). There are also potential interactions between different components of the overall therapy. Volume substitutes are known to influence hemorheology and coagulation. To define a potential net effect of volume substitutes on the hemocompatibility of an RRT, we chose an animal model without interfering pathophysiologies. According to the problem of early filter failure and coagulation disorders in critically ill patients, we focused on the hypothesized interaction between RRT and different volume substitutes with respect to blood cell counts, coagulation parameters and required heparin dose. Forty-eight pigs were assigned to four groups of fluid therapy with either normal saline (NaCl), 6%HES130kD/0.4 (HES130), 6%HES200kD/0.5 (HES200) or 4%gelatin (GEL). Six pigs of each fluid group underwent continuous venovenous hemofiltration (CVVH), the remaining six served as the control group. Anticoagulation was performed with continuous heparin infusion. CVVH was run in a recirculation-mode for 4.5 h to force hemocompatibility reactions, thereafter in a standard-mode for 2 h. During the CVVH-treatment GEL reduced platelet counts and fibrinogen concentration and additionally lowered ATIII levels. Heparin requirements did not differ between different volume substitutes or CVVH and control groups. Severe pathophysiologies are not the only reason for a reduced hemocompatibility of CVVH treatment. Interaction of a particular volume substitute with CVVH should be considered when interpreting study results and evolving new strategies.