Diffusive clearance of small and middle-sized molecules in combined dialyzer flow configurations.

Clearance of low (LMW) and middle molecular weight (MMW) solutes was investigated in vitro for different dialyzer configurations and mutual flow directions. Single pass tests were performed with two low flux Fresenius F6HPS hemodialyzers placed in series (12 tests) and in parallel (6 tests), and results were compared with those for one single dialyzer (2 tests). Either high concentrated (45mS/cm) bicarbonate dialysis fluid (surrogate LMW) or trisodiumphosphate (surrogate MMW) concentration (31mS/cm) was used as blood substitution fluid. Standard blood and dialysate flows of 250 and 500ml/min, respectively, were prescribed. Clearance was derived from conductivity measurements in blood and dialysate compartment, correcting for the overall ultrafiltration rate of 0.1-0.5l/h. In a single dialyzer, changing the counter current flow to co-current deteriorates diffusive clearance by 14% (LMW) and 18% (MMW). Compared to one single dialyzer using counter current flow, clearance increases by 3 to 8% (LMW) and by 15 to 18% (MMW) using two dialyzers in parallel and in series, respectively. As a consequence, the benefit by using a second dialyzer is more prominent for larger molecules. Moreover, pressure profiles drawn for the different configurations show the impact of limited convection on diffusive clearance.

Citrate anticoagulation and adverse events.

Several patients with heparin intolerance were dialysed with tri-sodium citrate as anticoagulant without acute clinical problems (good tolerance). After some weeks however problems arose. In all patients an alkalosis developed: the pre dialysis bicarbonate level rose progressively from 27 mmol/l to 40 mmol/l. This could be tempered by lowering the dialysis fluid bicarbonate concentration from 37 mmol/l to 25 mmol/l. A second problem was a progressive rise in pre dialysis sodium level from a mean of 136 mmol/l to 150 mmol/l. Adapting the dialysis fluid sodium concentration from 140 mmol/l towards 132 mmol/l could solve this. The third problem was a progressive rise in serum aluminium level in patients from 3 microg/l to 38 microg/l. After excluding water, concentrate, dialysis fluid, drug intake, etc... as possible sources, we controlled the aluminium level in the glass bottle containing tri-sodium citrate. We noted the very high value of 35,300 microg/l. After replacing the glass bottles with polyvinylchloride bags with a negligible aluminium content, the serum aluminium levels returned back to normal. It is known that citrate chelates the aluminium present in the glass of bottles or vials.

Quality monitoring of care: ask the patient.

In 1998 and 1999, we organised a questionnaire to be filled out by chronic haemodialysis patients. The first part dealt with general information data on the specific patient and the second part dealt with quality assessment data given as experienced by the individual patient. This second part included a ranking of preference by the patients and aspects patients wanted to see improved in the future. As we implemented the EDTNA/ERCA Clinical Standards and Core Curriculum, the monitoring tool, as described in this paper, was developed in order to monitor the effect of implementation. As a result of this two-year monitoring process of quality of care, we can follow up the care provided by our multidisciplinary team. It has proven a very useful tool to find out what is to stay as it is and what is to be improved. Simply closing the loop!

Heparinisation in chronic haemodialysis treatment: bolus injection or continuous homogeneous infusion?

After several incidents where patients either bled after dialysis or suffered severe anaphylactic reactions caused by intravenous heparin injections, we decided to administer heparin as a continuous homogeneous intravenous infusion. This had been usual practice in our unit in the past. Following the shocking experience of the death of a patient due to an anaphylactic reaction during his first dialysis, we wanted to warn colleagues for the potential dangers of administering intravenous heparin as a bolus injection.

Predialysis urea concentration is sufficient to characterize hemodialysis adequacy.

Mathematical description of urea kinetics for a week showed that, under steady state conditions (i.e., total removal equals total synthesis), any predialysis urea concentration is expressed as a linear function of specific urea generation (G/V) and of dialysis schedule timing and sessional Kt/V (product of clearance, K, and session time, t, divided by the urea distribution volume, V). It also predicts that TACurea is proportional to the predialysis concentrations. The ratio between the two depends linearly on delivered weekly dialysis dose ([wDD] = T(G/V)/TACurea, with T the number of hours in 1 week). These hypotheses have been tested by retrospectively analyzing urea kinetc modelling data that include all predialysis and post dialysis concentrations of 163 patient-weeks. All patients were anuric, and dialysis frequency was thrice weekly. Accuracy is assessed with regression analysis between database numbers and computed values. The theoretical ratio between midweek concentration and TACurea (1.43) is close to the computed ratio (1.46, r2 = 0.909). TACurea (slope = 1.002, r2 = 0.997), specific generation rate G/V as a precursor to PCRn (slope = 1.007, r2 = 0.985), and wDD (slope = 1.002, r2 = 0.909) are all accurately computed from predialysis concentrations. To aid in the determination of the ratio for the different predialysis, concentrations using wDD a nomogram is included.

Blood trauma in plastic haemodialysis cannulae.

To investigate the haemolysis in haemodialysis cannulae, an in-vitro set up is built, using a unipuncture dialysis system. This system is connected to a bag with fresh calf's blood, by the cannula under test, mounted in a large bloodline (5 mm diameter). The blood characteristics are kept constant by means of a bicarbonate dialysate in the dialyser. During a 6 h period, haematological parameters are regularly samples. Flow through the cannulae is recorded, which is about 500 mL/min. Four different cannulae are tested and compared to the results obtained without any cannula in the circuit. In all cases a linear increase in plasma free haemoglobin levels is found after 6 h. The cannulae can be ranked from 8F catheter over 13G, 14G to 16G cannula, the latter producing the highest degree of haemolysis. When using plastic cannulae at high blood flows, their haemolytic effect may not be neglected.

Automated on-line bicarbonate concentrate production: microbiological, chemical and economic benefits.

We all know that "bicarbonate dialysate" became the world wide used fluid for use in haemodialysis and related techniques. To prepare bicarbonate dialysate we actually have to use two separate concentrates: one containing sodium bicarbonate, solely or in combination with other electrolytes other than calcium, and a second concentrate containing mostly all the other electrolytes (sodium, magnesium, potassium, chloride and/or some acetate).

Red cell injury assessed in a numeric model of a peripheral dialysis needle.

The highest shear stresses in a dialysis system are expected to be found in the needle, where the largest velocity-diameter ratio appears. Shear is a known source of hemolysis and related patients' discomfort. To assess the magnitude of blood cell injury and the location of its sources, a finite element model is used to calculate three-dimensional velocities and shear stresses in peripheral dialysis needles, concentrically placed in a rigid wall fistula. The boundary conditions consist of time dependent in vivo measured pressures. Cell damage is computed for different cell tracks into the needle by means of Wurzinger's empirical formula, which expresses the hemoglobin (Hb) release as a function of shear stress and shearing time. Near the needle wall, velocities are low and shear stresses high, resulting in a significantly higher level of cell damage: 0.1% vs 0.001% in bulk flow for a mean flow of 91 ml/min into a 14G needle with a peak velocity of 220 cm/sec. The deviation from the classic Poiseuille velocity profile is shown. Less than 5% of the flow passes through this high damage path. A vortex at the inner side of the needle has a cumulative damage of 0.007% per 0.23 sec trip around the vortex.

Continuous monitoring of the spent dialysate urea level using a disposable biosensor. Clinical evaluation.

The first real-time, online measurements obtained using a screen-printed planar sensor to continually monitor urea in the spent dialysate are described. Evaluation of the sensor performance in the real clinical situation involves an extensive comparison of pre and post dialysis characteristics. Steady-state response and calibration curves show excellent reproducibility (typically within 5%; maximum deviation, 15%). The real-time results show that the characteristics predicted from the predialysis characterization are preserved during and after the prolonged period of continuous contact with the dialysate without needing recalibration. An operational stability of more than 10 hours of continuous use has been observed. Stability including intermittent wet-storage was found to be more than 5 days. Zero-baseline stability greater than 0.2 mmol/l (1.2 Og/dl) during the 4 hr treatment period and a resolution of 0.2 mmol/l were achieved, with a full-scale range of 15 mmol/l (90 mg/dl). A response time of less than 3 minutes allows dialysate urea levels to be monitored as a function of dialysis parameters such as varying blood flow. Parallel analysis of dialysate samples taken at 15 min intervals shows good correlation with the recorded data, although correspondence of the absolute values requires that the real-time data be rescaled. Because no degradation of the sensor performance was observed in vitro, this deviation can be explained only in terms of a dialysate based interference, probably caused by changes in pH and buffer.

Fibrin sheet covering subclavian or femoral dialysis catheters.

The existence of a sheet around a single lumen dialysis catheter tip, which provokes a valve mechanism, is proved by the observation that several times during the replacement procedure of a dialysis catheter, a sheet surrounding the surface of the catheter is removed with the dialysis catheter. This sheet is grey, approximately 1 mm thick and 30 mm long and consists of fibrin and thrombocytes. Bacteriological examinations were always negative. The existence of the sheet in vivo is demonstrated by digitalized angiography during the removal procedure for single lumen dialysis catheters. Rarely, only the sheet is removed with the catheter. It all other instances, the sheet is stripped off and remains in the subcutaneous tunnel or in the vascular bed without causing much clinical discomfort in most patients. Occasionally an episode of cough, dyspnea, hypotension, retrosternal oppression or hemoptae after removing the single lumen dialysis catheter, suggest pulmonary embolism or lung infarction.

Severe electrolyte disturbances in lethal subclavian catheter dialysis.

Syncope, irreversible shock, and respiratory and circulatory arrest developed in a 54-year-old diabetic man chronically dialyzed with a subclavian catheter (SC) minutes after the end of a dialysis session. Resuscitation attempts remained unsuccessful. During the resuscitation attempt, a blood analysis showed severe hyponatremia, acidosis, and hypochloremia. Respiratory and cardiac arrest developed during dialysis in a 64-year-old woman on chronic SC dialysis. Resuscitation was unsuccessful; chloremia levels were 79 mEq/L, and calcemia levels were 20 mg%. Both patients were dialyzed with a standard dialysate solution. The reasons for the electrolyte disturbances could not be explained technically. The autopsy showed myocardial perforation by the SC and accumulation in the pericardium of the fluids administered during the resuscitation attempt (e.g., glucose 5%, bicarbonate, Ca gluconate, human albumines), thus explaining the erroneous electrolyte results. The reason for the perforation was a too-rigid central femoral vein catheter, erroneously labeled a subclavian catheter by the supplying firm. Because of a syndrome of progressive vena subclavia and vena cava sclerosis with insufficient arterial phase flow and venous-phase bleeding around the puncture site during single-needle dialysis, the original SC had to be replaced by a longer one with the tip located in the atrium (this SC was actually a femoral catheter). Analysis of the fluid aspirated through the SC can determine the diagnosis in sudden death of SC dialysis patients.

A disposable urea sensor for continuous monitoring of hemodialysis efficiency.

A new sensor for the on-line measurement of urea in the dialysate output is described. The sensor is based on a differential measurement of conductivity changes induced by the urease-catalyzed hydrolysis of urea. The use of screen printing for the batch-fabrication of the basic transducers results in cheap disposable devices. In addition, the sensor has been designed to fit into a standard male luer-adapter, and can be plugged directly into the dialysate line. The in vitro response is linear to urea concentrations exceeding 6620mM. A resolution of 20020 microM has been achieved with a baseline stability of 50 microM/hr. Interferences caused by fluctuations in the ionic strength and the consequent conductivity changes are effectively suppressed by the differential sensor pair. The efficiency of this suppression is expressed in a common mode rejection ratio of typically 40 to 50. Preliminary ex vivo results show the feasibility of the concept. The sensor principle is not restricted to urea but can be extended to other molecules of interest for hemodialysis monitoring, such as creatinine and L- and D-amino acids.

Comparison of a new and a standard single-needle dialysis system using a mathematical model.

Two single-needle dialysis systems are compared for their hydrodynamic and kinetic behavior, the classic double-headed pump system and a newer bidirectional blood pump. First, discharges and pressures are calculated in a time-space grid to learn about the flow characteristics of the single-needle systems. These parameters are important for they control ultrafiltration and clearance rates in the artificial kidney. Also, the influence of different artificial kidneys and expansion chambers on the flow parameters is studied. Second, starting from these results, the removal of urea and creatinine is simulated using a double-pool model for both a normal dialysis session and an ultrashort daily autodialysis. As a result, the use of a single marker molecule to study or characterize solute removal is questioned. The new blood pump is able to perform dialysis filtering equivalently to the classic system but has its own specific advantages.

Polyunsaturated fatty acids of the n-3 class in chronic dialysis.

N-3 PUFAs are believed to 1) reduce serum lipids, 2) modify the complement activating system, 3) increase red blood cell elasticity, and 4) decrease thrombotic events. These four parameters were investigated in 19 chronic stabilized dialysis patients assessed by 1) the control of the serum lipids, 2) the degree of leukopenia 10 minutes after the start of dialysis, 3) the degree of red cell fragmentation (RCF) during dialysis, and 4) the reusability of the dialyzer. The patients were their own controls during four periods: 1) no PUFAs for longer than 6 months, 2) 4 x 425 mg of PUFAs/day for 6 weeks, 3) 6 x 425 mg PUFAs/day for 6 weeks, and 4) no PUFAs during at least 6 weeks. Previous diets and medications remained unchanged. It was concluded that there was no significant effect on serum lipids in the given dosages, and no effect on the degree of leukopenia, red blood cell elasticity or bleeding, and/or antithrombotic activity. No change was seen in insulinemia, fibrinogenemia, or arterial pressure. No serious side effects were noted other than a "fishy aftertaste."

(A)typical symptoms during single needle dialysis.

In 5 elderly patients, an abnormally high occurrence of some symptoms was noted during dialysis. All patients were dialyzed with biocompatible membranes, bicarbonate dialysate, and a blood flow of 250 to 300 ml/min by a single needle system, on a fistula 14 Gauge catheter-needle. These symptoms were: 1) "angina," resistant to O2 and nitrates, with biochemical stigmata of infarction, but without electrocardiogram (ECG) localization; 2) intractable persistent hypotension, not hypovolemic, lasting 1 or 2 days; 3) esophagal spasms, with inability to swallow solid food. Because we knew that these symptoms were compatible with hemolysis (biochemically proven by the increase in serum LDH during dialysis and by a fall in haptoglobin) due to red cell fragmentation (RCF), we switched these patients from fistula dialysis (A) to central catheter dialysis (B), with the same apperture, blood flow, etc. The total number of sessions of A versus B were 512 and 891; the mean LDH ratios (serum LDH postdialysis divided by predialysis) were 1.8 and 1.0 (= no RCF); angina events were 132 (26%) for A, and 25 (3%) for B; persistent hypotension was seen 37 (7%) times in A and 5 (0.6%) times in B; esophagal spasms were noted 65 (13%) times for A, and 0 times for B. This clinical improvement was so overwhelming that 3 patients refused to be dialyzed again using their well functioning fistulae. This study also proved the need for a better designed and manufactured peripheral dialysis catheter-needle.

Limitations of short dialysis are the indications for ultrashort daily auto dialysis.

Dialysis efficiency can be increased while shortening the duration of the treatment by the use of highly permeable dialysis membranes, larger surfaces, and higher blood and dialysate flows. However, vascular repletion of water and solutes, compared with the amount removed by rapid dialysis, is a limitation of short dialysis. This can be overcome by ultrashort daily dialyses, preferably performed by the patient alone. A safe, efficient, and user-friendly apparatus has been developed to permit Ultrashort Daily Auto Dialysis (UDAD).

A mysterious extrafraction in the gamma area of protein electrophoresis in dialysis patients is found to be fibrinogen.

The appearance of a heretofore unnoticed extrafraction in the gamma protein electrophoresis of 17 dialyzed patients was linked to the modified Shaldon dialysis catheter. Medication interference, infection, and immunological or hematological disease as causative factors were ruled out. After initial contradictory results, the extrafraction was determined to be fibrinogen, present because of the slow release of heparin contained in the catheter, so that electrophoresis was performed on plasma instead of on serum.