ASAIO transactions最新文献

The presence of biofilm is thought to accompany infection or colonization of chronic peritoneal dialysis (PD) catheters, and to be an important pathogenetic factor in the recurrence or persistence of peritonitis. In the current study, the characteristics of identifiable biofilm associated with the PD catheter were studied using scanning and transmission electron microscopy in six consecutive cases requiring catheter removal for a variety of indications. Biofilm characteristics in each case were rated in blinded fashion by three independent observers, and findings were then correlated with the clinical histories and microbiologic findings. Surprisingly, two of the three cases with the most severe biofilm formation occurred in patients with no history or microbiologic findings of recent infection, and the positive findings of leukocytes, macrophages, fibrillar matrix, and other structures on these catheters did not correlate with detectable infection. In addition, extracellular spherical lipoid structures and intracellular lipoid vacuoles in mesothelial-like cells were prominent in four of six cases, did not correspond to the presence of infection, and suggested possible mesothelialization of the catheter. The findings of this study do not necessarily controvert the microbial origin of some components of the biofilm, or the possible role of biofilm in some cases of persisting peritoneal infection. However, it is clear that many important components of the biofilm arise, not from microorganisms, but rather from host origin in the absence of detectable infection. Moreover, such "endogenous" biofilm production can result in extensive accumulation of catheter-associated matrix.

A blind, randomized study of the effects of a heparin-coated bypass circuit on thromboembolus formation and hematologic and hemostatic parameters was carried out on 12 dogs. The dogs were anesthetized and bypassed for 24 hours using a centrifugal pump circuit with a mean flow of 475 ml/minute. Six of the dogs were bypassed with a heparin coated circuit and six with an uncoated circuit. No systemic heparin was administered to any of the dogs at any time during the procedure. Both the coated and uncoated circuits remained patent at low flows. Thrombus formation, however, primarily around the pump axis, was observed in both groups. Four of the dogs (two from either group) had no evidence of lung emboli. The remaining eight dogs all showed pulmonary emboli. Thromboemboli did not lead to clinical complications or hemodynamic disturbances. Hematologic and hemostatic parameters showed a reduction in hemoglobin, erythrocytes, thrombocytes, leukocytes, and antithrombin III, which was most pronounced in the uncoated group. This study showed that a low flow bypass circuit remains patent for 24 hours whether or not the circuit is coated with heparin. Although the heparin coating limits the reduction in antithrombin III and blood elements, it does not eliminate the risk of thromboembolus formation.

The authors began ventricular assist pumping as treatment for postcardiotomy cardiogenic shock in November 1988 with a new automated pulsatile support system, the ABIOMED BVS 5000 Bi-Ventricular Support System. Five patients (0.6% of total cardiac surgery patients) have been placed on support, four after coronary artery bypass grafting, and one after bypass grafting and mitral valve repair. All patients were refractory to pharmacologic and intraaortic balloon pump therapy. Three patients had left ventricular support and two had biventricular support. Four patients were successfully weaned, and three are long-term survivors. Duration of support ranged from 39 to 118 hours (mean, 89.4 hours). Resternotomy was performed in four patients: twice for hemostasis, once for tamponade, and once for inadequate left ventricular drainage. Three patients, two nonsurvivors and one survivor, had perioperative myocardial infarctions. No device related thromboembolic complications, hemolysis, or infection were experienced. Follow-up at more than 1 year demonstrated that all patients are in NYHA Class 1. Ventricular assist pumping with the ABIOMED BVS 5000 Bi-Ventricular Support System is an effective treatment for postcardiotomy cardiogenic shock.

The pumping capabilities of nine unconditioned canine rectus abdominus muscles (93-163 gm) and six latissimus dorsi muscles (99-146 gm) were measured. The muscles were wrapped around a 100 ml ellipsoidal pouch in a mock circulatory system in which the afterload was 100 mmHg. Pouch diastolic pressure was kept low by an electrically controlled inlet valve to maximize muscle capillary blood flow. Immediately before tetanic contraction of the pouch-encircling muscle, the inlet valve opened for 450 msec to increase pouch pressure to 100 mmHg, thereby providing a high preload and ensuring a forceful muscle contraction. The motor nerves to the muscles were stimulated with 450 msec trains of 0.1 msec stimuli, using a frequency of 40/sec. The train rates (muscle contractions/min) were 10-50/min. In this circulatory model it was found that the maximum output for both muscle types occurred between 20 and 40 contractions/min. It was also found that for both muscle types, the maximum output (L/min) was dependent upon muscle weight. The data revealed that an output of 4 ml/min was obtained per gram of muscle. The power (mW/gm) developed was related to the output (L) in L/min. For the rectus muscle W = 0.47L, and for the latissimus muscle W = 0.41L mW/gm. Pumping periods lasted approximately 4 hours, with no evidence of fatigue. When viewed as a potential cardiac assist device, the muscles were able to provide a flow equivalent to approximately 25% of the cardiac output. However, it is important to note that the pumping capability is directly related to muscle weight, indicating that a higher output can be achieved with a larger muscle.

Extracorporeal membrane oxygenation (ECMO) is being employed with increasing frequency for the treatment of neonates with severe cardiac or respiratory failure. The risks related both to carotid artery and jugular vein ligation continue to cause concern. Use of umbilical vessels for vascular access in ECMO could eliminate many of these risks. The experience to date with this approach is summarized, along with case reports of three patients treated at our center in whom the umbilical vein was cannulated to augment venous drainage. One patient died of causes unrelated to umbilical vein cannulation. One had an uneventful ECMO course and is a normal survivor, and one developed a tension hemopericardium as a complication of the umbilical vein cannulation, but is a normal survivor. Potential risks and benefits of this approach are reviewed.

Simultaneous measurement of transcapillary ultrafiltration (TCUF), lymphatic absorption rate (LAR), and intraperitoneal volume (IPV) was performed by means of intraperitoneally-administered polydisperse dextran 70 in nine CAPD patients during a 4 hr dialysis dwell (glucose 1.36%). The recovery of dextran was 88 +/- 1%. LAR, calculated as the amount of dextran lost, divided by the dialysate dextran concentration, was 1.30 +/- 0.12 ml/min. The time course of TCUF could be described as a hyperbola. Therefore, the application of the Lineweaver-Burke plot made it possible to calculate TCUFmax (median 641 mL) and its half-time (t50: median 211 min). delta IPV4h, calculated as the difference between the Lineweaver-Burke adjusted TCUF4h and LA4h, was correlated with measured delta IPV4h after drainage (r = 0.89, p less than 0.001). The latter was dependent upon LAR (r = -0.71) and effective peritoneal surface area, as represented by mass transfer area coefficients (MTC) of low molecular weight solutes (creatinine r = -0.76, glucose r = -0.81). High MTC values of these solutes were exponentially related to a short t50 (creatinine r = -0.76). LAR was correlated with the MTC of intraperitoneally administered inulin (r = 0.83). Dextran 70 had no measurable effect on solute transport. It is concluded that dextran 70 is a useful marker for the measurement of peritoneal fluid kinetics, even during CAPD with a low glucose concentration in the dialysate.

The influence of centrifugal blood pumps on the elasticity of erythrocytes was tested in an in vitro set-up. Fresh bovine blood was pumped by vaneless and impeller blood pumps with low and high hemolytic potential (1L priming volume, flow 5 L/min versus 150 mmHg, pumping time 6 hours). The elasticity of the red blood cells was measured by laser diffraction. Starting with an elasticity of 238 +/- 39, the overall change during 6 hours was 12 +/- 38 compared to a change of the control value of -13 +/- -58. Even a pump with very high hemolytic potential (Hemolysis index 21.9) did not cause relevant changes of elasticity. It is concluded that 1) elasticity is not a useful parameter to use in the evaluation of centrifugal pumps, and 2) the mechanical trauma caused by centrifugal pumps produces no relevant permanent alteration of the elastic properties of red blood cells, at least under in-vitro conditions.

Theoretical advantages and risks exist for the use of both the centrifugal and roller pump systems in neonatal extracorporeal membrane oxygenation (ECMO). The authors studied the pressure-volume-flow relationships in clinically configured ECMO systems using these two pumps and a simulated patient to characterize differences in the circuit mechanics of the two systems, and thereby improve the design of subsequent clinical comparative trials of the pumps themselves. The relationship between flow and pressure generated across the pump was identical for the two systems. Within the range of clinically used flows, there was a direct relationship between pump revolution and flow with the roller pump, and between pump revolution and pressure generated for the centrifugal pump. Flow was limited in both systems by restrictions on negative pressure generating capacity. In the roller pump circuit, the venous reservoir (bladder box) assembly interrupted flow when negative pressure exceeded -20 mmHg; in the centrifugal pump system, forward flow stopped when negative pressure exceeded -100 mmHg. Volume had no detectable effect on the patient-pump inlet pressure gradient until critically low volumes were reached. At that point, removal of a few milliliters of volume led to large increases in the pressure gradient. The authors conclude that differences in pressure-volume-flow relationships between roller and centrifugal pump ECMO systems are due to the presence of the bladder box in the roller pump circuit. The advantages and disadvantages of the greater negative pressure in the centrifugal pump system require further study.

A new type of pump, the differential displacement pump, has recently been developed. The new pump and a conventional pneumatic displacement pump were evaluated with respect to pump flow, filling pressures, stroke frequency, and inflow pattern under equivalent test conditions. The pumps were connected via inlet and outlet cannulas to a Donovan hydraulic circulation analogue. The differential displacement pump showed a high sensitivity to filling pressure in the range of 2.3-9.6 L/min/mmHg, and a high efficiency as long as the stroke volume was submaximal. The inflow was continuous and pulsatile with a maximal forward flow velocity during systole as long as the maximal stroke volume was not employed. In contrast, the conventional displacement pump showed low sensitivity to filling pressures, a discontinuous inflow, with a maximal inlet cannula forward flow during diastole and a substantial flow reversal during systole. These results indicate that the differential displacement pump is a better concept with respect to sensitivity to venous return, and that it provides an inflow pattern and regulation of pump flow similar to that of the biologic heart.