The International Journal of Artificial Organs

Twenty-two patients were dialysed in a cross-over design using Hemophan® or cellulose acetate membranes. The dialysate buffer was acetate (n = 12) or bicarbonate (n = 10). Blood was sampled at 0, 15, 60 and 180 min and mean values were adjusted for changes in total protein in each sample. At 15 min during dialysis a decrease in leukocytes and platelets occurred with both membranes, irrespective of the buffer (Wilcoxon, p < 0.006). During dialysis, increases were found in granulocyte elastase inhibitor complex (E-α₁-PI), β-thromboglobulin and C3d. β₂-microgrobulin was not significantly changed in blood after dialysis with Hemophan® or cellulose acetate membranes with bicarbonate buffer. Side effects were more pronounced at 180 min during dialysis with bicarbonate in patients using cellulose acetate than with Hemophan® (p = 0.021, n = 8). Hemophan® seemed to be more favourable than cellulose acetate membranes in regard to leukopenia and E- α₁-PI. The dialysate buffer may also alter membrane biocompatibility.

We describe a cluster of four septicemias with pseudomonas, that occurred in a unit performing formaldehyde reuse of capillary dialyzers. Samples of blood, heparin solutions, dialysate and effluent of reused dialyzers, were evaluated bacteriologically and upon the adequacy of the reuse procedure. Pseudomonas aeruginosa, vesicularis and/or xanthomonas maltophilia were found on the blood cultures obtained during the septicemic reactions, and in the effluent of two reprocessed dialyzers not yet used (> 10⁴ CFU/ml). These two dialyzers had also extremely low formaldehyde concentrations (0.0014 and 0.005% versus the expected 4%). Membrane and antibiogram characteristics of a Pseudomonas aeruginosa strain, recovered from the blood cultures in one patient, and of a strain found in the effluent of one of the two contaminated reprocessed dialyzers, were the same. The problem was attributed to the inadequate mixing of the disinfectant with the tap water used in the automated reprocessing device, in the absence of an alarm disclosing this failure.

A new wet-state membrane characterization method, thermoporometry, was used to study the effect on membrane structure of commonly used sterilization methods for artificial kidney membranes. The porosity and pore size distribution of differently sterilized hollow fiber Hemophan hemodialysis membranes were determined. Also the effect of a glycerol treatment (before sterilization) on porosity and pore size distribution after sterilization was studied. Hemophan was found to have a pore size distribution of pores with radii between 1.5 and 12 nm. Most of the samples had a maximum pore volume at a pore radius of 2.5 nm, only the steam sterilized and non glycerol treated sample had a maximum pore volume at 1.5 nm. The porosity was found to vary between 14 and 31% and was dependent on the applied treatment

Extracorporeal lung assist (ELA) has been recommended for the treatment of ARDS if conventional therapy fails. However, the need for nearly complete anticoagulation is a major risk factor for hemorrhagic complications. We describe our experience with 13 ARDS patients treated with ELA using heparin-coated systems (Carmeda). Maintaining partial thromboplastin time and activated clotting time within or close to the normal range, even major surgery (20 thoracotomies and 2 laparotomies) could be performed without undue bleeding complications related to anticoagulation during extracorporeal support. Eight of the 13 patients survived. The use of heparin-coated systems allows prolonged ELA with nearly physiological coagulation function, permitting major surgical intervention. It enhances the safety margin of extracorporeal gas exchange and may ultimately extend its indications.

We investigated the efficacy of an anion-exchange adsorbent column (ASAHI BR-350, DIAMED) for removal of bilirubin and bile acids in five patients with intractable jaundice of various origin. Four litres of plasma were separated by membrane plasma separation (Plasmaflow OP-05) at a rate of 22.5 ml/min. The plasma was then perfused through an anion exchange adsorbent and returned to the venous blood line of the plasma separator. In some of the patients this procedure was combined with regular hemodialysis treatment. The concentration of total bilirubin was cut by 31 to 60%; total bile acids were reduced by 20 to 74%. Three patients recovered and had a favourable outcome. Two patients died despite the bilirubin adsorption treatment. The effects of the adsorbent column on specific blood parameters, including the coagulation system, were measured. Our data suggest that bilirubin adsorption should be examined further as a treatment for critically ill patients with intractable jaundice.

Mock loops are an important tool for in vitro investigations of artificial blood pumps. The simple windkessel, throttle, and atrium principle was used for the mock loop design presented. The components of the systemic and the pulmonary mock loop were designed according to calculated numerical simulation parameters. The loops offer a compact design and simple handling. For simulating biventricular assist or total artificial heart (TAH), both loops can be coupled correspondingly. The numerical simulation and the first results with the loops show very good similarity to physiological data of systemic and pulmonary circulation. The measurements of pump characteristics are significant for quantitative comparison of different pump sizes and types, or driving systems.

Circulatory maintenance with a left ventricular assist device (LVAD) alone during cardiac arrest until heart transplantation has been evaluated. To assess the effect on the autonomic nervous system, the sympathetic neurogram was analyzed by power spectrum and coherence function. LVAD were inserted between the left atrium and the descending aorta in seven adult mongrel dogs and ventricular fibrillation was induced electrically. Renal sympathetic nerve activity (RSNA) was detected by bipolar electrodes attached to the left renal sympathetic nerve. Values of squared coherence between the arterial pulse wave and RSNA were calculated. Under the condition of circulatory maintenance with only LVAD, coherence at the cardiac rhythm frequency was decreased, and coherence at the LVAD pumping rhythm frequency was increased. These results indicate that the arterial pulse wave observed during maintenance of the circulation with only LVAD contributed to the sympathetic neurogram.

Calibration of glucose sensors proved difficult for electrodes with immobilized glucose-oxidase. The correlation between the sensitivity of the electrodes in vitro and in vivo appeared to be poor. We developed a new type of glucose sensor, based on a microdialysis system, in which an oxygen electrode is used as detector outside the body and the enzyme glucose-oxidase dissolved in water is used as a dynamic selector. The enzyme solution is pumped through a hollow fiber placed subcutaneously, before the fluid passes the detector. The glucose sensor was tested in the subcutaneous abdominal tissue of 12 healthy volunteers and 12 type I diabetic patients. Blood glucose was clamped at two levels to permit a two-point calibration of the sensor in vivo. These values correlated well with the in vitro calibration factors (r=0.949). In subcutaneous tissue the sensor measures 43 ± 9% of the blood glucose value, using the in vitro calibration factor. No differences were detected between healthy volunteers and diabetic patients.

The aim of this study was to test different metals, widely employed in constructing prosthetic devices, by in vitro methods. Biological effects of such materials were analysed through four different assays on human lymphocytes and granulocytes. The lymphocyte proliferation assay gave quantitative results, while the viability test showed the morphological appearance of the cells correlated well with previous results. NK cytotoxicity and granulocyte chemokinesis tests provided interesting data on leucocyte performance when challenged with metals. Therefore the present study adds new basic information on cell behaviour when metal products are present in the body, e.g. around devices implanted in human tissues