Mid-dilution: the perfect balance between convection and diffusion.

Although hemodiafiltration (HDF) offers the advantage of increased convective clearance for middle molecules, there is still controversy as to whether reinfusion should occur pre- or post-filter. Mid-dilution hemodiafiltration (MD HDF) is a new HDF technique that uses a special dialyzer, MD190, which allows both pre- and post-reinfusion. While externally the dialyzer looks similar to conventional hemodialyzers, the internal fibers are divided into two bundles by a special annular header that first lets the blood pass through the peripheral bundle in post-dilution, mix with the reinfusion fluid at the opposite end of the dialyzer and then proceed (after pre-dilution) to the dialyzer blood exit. The dialyzer is able to support substantially higher reinfusion rates (10-12 l/h). We have compared the removal characteristics of several small solutes and larger middle-molecular-weight toxins by examining instantaneous clearance at 45 min, the dialysis reduction ratio and total mass removal (by spilling) in a three-center prospective cross-over study. Twenty patients were randomized to a treatment sequence of one-week high-flux bicarbonate hemodialysis (HD) followed by MD HDF, or vice versa. The parameters evaluated included urea, creatinine, beta2-microglobulin, angiogenin, leptin, retinol-binding protein, and the effects on sodium, potassium, bicarbonate and calcium. Blood flow rates ranged between 300-450 ml/min (mean 359 +/- 44 HD, 367 +/- 35 MD HDF). The mean reinfusion for MD HDF was 166 +/-17 ml/min. MD HDF had a significantly better instantaneous clearance for urea (328 +/- 28 vs 277 +/- 40); creatinine (292 +/- 32 vs. 212 +/- 66); phosphate (324 +/- 38 vs. 242 +/- 63); beta2-microglobulin (249 +/- 27 vs. 100 +/- 24); angiogenin (173 +/- 27 vs. 28 +/- 32); and leptin (202 +/- 29 vs. 63 +/- 43). Treatments were well tolerated with no adverse reactions occurring during any of the treatments. The MD HDF filter's unique configuration is designed to deliver high-efficiency HDF with a significant improvement in small and middle molecule removal. MD HDF supports substantially higher ultrafiltration rates, and as such, results in a higher removal of middle-molecular-weight toxins.

Effects of the infusion mode on bicarbonate balance in on-line hemodiafiltration.

BACKGROUND: Electrolyte and acid-base balance may be differently affected by the infusion mode in on-line hemodiafiltration (HDF). We studied the effects of the different infusion modes on bicarbonate transport across the dialyzer membrane, and thus on the final bicarbonate balance of the HDF sessions. METHODS: Instantaneous HCO3- transfer across the dialyzer membrane, blood bicarbonate profile and the total balance of the sessions were studied in six dialysis patients under the same operating conditions over 36 HDF sessions, in order to compare the effects of predilution HDF (pre-HDF), postdilution HDF (post-HDF), and mixed HDF on the final bicarbonate balance. RESULTS: The final HCO3- balance was more positive in post-HDF vs pre-HDF (142 +/- 36 vs 99 +/- 41 mmol/session, p<0.05), with a final blood HCO3- concentration of 26.6 +/- 1.0 vs 25.4 +/- 1.1 mmol/L, (p<0.05). Mixed HDF yielded intermediate results (balance: 119 +/- 42 mmol/session, final HCO3- 26.2 (1.2 mmol/L). These differences were seen to result from the increased HCO3- concentration of blood entering the filter in predilution, due to the infused HCO3-, enhancing convective loss and reducing the driving force for diffusive HCO3- gain. CONCLUSIONS: Bicarbonate concentration in dialysate-reinfusate is critical in order to obtain an adequate end of session HCO3- balance in on-line HDF. The predilution method produced the lowest cumulative net HCO3- gain between the three studied infusion modes. Our data suggest that, under the same operating conditions and excluding the effect of ultrafiltration, dialysate HCO3- should be increased by about 2 mmol/L in pre-HDF, and 1 mmol/L in mixed HDF, to yield the same final balance as in post-HDF.

Rate dependence of acute PTH release and association between basal plasma calcium and set point of calcium-PTH curve in dialysis patients.

BACKGROUND: In vivo, the control of calcium-mediated acute PTH release during induced hypo- or hypercalcaemia is linked not only to plasma calcium concentration per se but also to the rate and direction of calcium change. In fact, during induced hypocalcaemia, the predominant mechanism that causes PTH to be released is the reduction of plasma Ca(2+) irrespective of the absolute starting concentration of ionized calcium. This mechanism, which is rate-dependent and even activated in conditions of hypercalcaemia, may be involved in the association, reported in several papers, between the basal Ca(2+) and the set point of the calcium-PTH curve. METHODS: The calcium-PTH relationship was studied in 12 dialysis patients under conditions of induced low and high predialysis plasma Ca(2+). At each level of basal Ca(2+), dynamic tests were conducted using two methodological approaches. In method A patients underwent low (0.5 mmol/l) calcium dialysis in the stimulation test and high (2 mmol/l) calcium dialysis in the inhibition test, while the dialysate calcium (CaD) was kept constant during each test. In this way a higher but variable rate of change in plasma Ca(2+) was achieved. In method B, CaD was progressively decreased (stimulation test) and increased (inhibition test) during the tests in order to obtain a lower but more constant rate of change in plasma Ca(2+). Consequently, for each patient, four calcium-PTH curves were produced: low basal Ca(2+) with methods A and B, and high basal Ca(2+) with methods A and B. RESULTS: Basal plasma Ca(2+) was similar in A and B at low (1.16+/-0.02 vs 1.15+/-0.02 mmol/l) and high (1.25+/-0.02 vs 1.26+/-0.02 mmol/l) basal plasma Ca(2+). The set point was higher in A than in B both at low (1.12+/-0.02 vs 1.10+/-0.02 mmol/l, P=0.01) and high (1.20+/-0.02 vs 1.16+/-0.02 mmol/l, P=0.03) basal Ca(2+) as was the slope (542+/-41 vs 426+/-44%/mmol, P=0.02; 615+/-73 vs 389+/-25%/mmol, P=0.01). No significant difference was found between A and B as regards minimal PTH and plasma Ca(2+) at minimal PTH (Camin) in both calcaemic states. Maximal PTH was slightly higher in B at low (510+/-97 vs 548+/-107 pg/ml, P=NS) and high basal plasma Ca(2+) (410+/-97 vs 464+/-108 pg/ml, P=0.02). Plasma calcium at maximal PTH (Camax) was significantly higher in A (1.1+/-0.03 vs 0.99+/-0.02 mmol/l, P=0.001) at high basal plasma Ca(2+). The set point was strictly related to basal plasma Ca(2+) in both methods, but the slope of the linear regression was significantly steeper with method A. The set point was predicted to increase by 0.881 (CI 0.772-0.990) mmol/l for each mmol/l of increase in basal plasma Ca(2+) with method A and by 0.641 (CI 0.546-0.737) mmol/l for each mmol/l of increase in basal plasma Ca(2+) with method B. CONCLUSIONS: (i) Higher and variable rates of change in plasma Ca(2+) produce a higher set point value and a steeper slope of the calcium-PTH curve when compared to lower and more constant rates of calcium change. (ii) The different slope of the linear correlations between basal plasma Ca(2+) and set point in the two methods suggests that the rate-dependent mechanism of acute PTH release plays a significant role in the association between set point and basal plasma Ca(2+). (iii) The significance of the set point is questionable when the calcium-PTH curve is carried out in vivo.

Mixed predilution and postdilution online hemodiafiltration compared with the traditional infusion modes.

BACKGROUND: On postdilution hemodiafiltration (post-HDF), convective removal of medium-high molecular weight solutes is, at the highest ultrafiltration rates, limited by high blood viscosity and protein concentration. Prefilter reinfusion (pre-HDF) may overcome this problem, but plasma dilution may affect the overall efficiency of the technique. In this study, an experimental system of online HDF with combined prefilter and postfilter infusion (mixed HDF) was evaluated and compared with the traditional predilution and postdilution modes. METHODS: Removal of urea (U), creatinine (Cr), phosphate (Phos), and beta(2)-microglobulin (beta(2)m), ultrafiltration coefficients of the dialyzer (K(UF)), and rheologic conditions of the blood circuit were evaluated during the three infusion modes (a total of 36 runs lasting 180 min), performed with a polysulfone hemofilter 1.8 m(2), blood flow (Q(b)) 400 mL/min, dialysate flow (Q(d)) 700 mL/min, and infusion rate 120 mL/min (pre-HDF and post-HDF), or 60 + 60 mL/min (mixed HDF). RESULTS: The mean effective U and Cr clearances and urea index of dialysis dose (eKt/V) were significantly higher on post-HDF than on pre-HDF (K(WB) (U) 210 vs. 193 mL/min, K(DQ) (Cr) 152 vs. 142 mL/min, eKt/V 1.41 vs. 1.30), while mixed HDF did not show significant differences versus post-HDF (K(WB) (U) 201 mL/min, K(DQ) (Cr) 149 mL/min). K(DQ) for Phos and beta(2)m were higher on post-HDF in only absolute values. Similar differences were found for instantaneous dialyzer clearances (K(I)) at 60, 120, and 180 minutes of the sessions, with a common trend to decrease with time. K(UF) and the apparent beta(2)m sieving coefficient showed their lowest values toward the end of post-HDF sessions. Increasing filtration fractions (FFs) were associated with increasing transmembrane pressure (TMP) and solute clearances up to FF values of 0.45. These were values achieved in only post-HDF, at which point the curve of the relationship between TMP and FF assumed its steepest exponential trend. CONCLUSIONS: Mixed HDF, by better preserving the characteristics of water and solute transport of the membrane, ensured safer operating conditions than post-HDF, while achieving similar removal of small- and large-size solutes. Optimizing the ratio of prefilter/postfilter infusion and the total infusion according to the relationships found in our study between solute clearances, FF, and TMP, convective flux and transport may avoid excessive hemoconcentration and dangerous pressure gradients.

Calcium-PTH relationship in dialysis patients: the pitfalls of using a constant dialysate calcium concentration during the dynamic tests.

The calcium-PTH relationship in uremic patients has been often studied during dialysis sessions with high or low dialysate calcium concentration (CaD). This method has been used because it is less complex and invasive than i.v. infusion of calcium salts and calcium chelating agents. However, the constancy of CaD during the tests does not allow for the control of the serum calcium profile and, given that the blood calcium concentration is only one factor of a more complex calcium-related mechanism of the PTH release, the calcium-PTH curve may become dependent on the unpredictable rate at which the ionized calcium changes. Dynamic testing of the parathyroid gland was performed in 9 dialysis patients comparing constant CaD of 1.0 and 2.0 mmol/l (A) with a linear change in CaD (B). The rate of serum calcium change remained constant over time only in experiment B. The total decrease in calcemia (0-0.38 +/- 0.03 vs -0.14 +/- 0.1 mmol/l) and PTHmax (748.25 +/- 124.76 vs 374.89 +/- 53.03 pg/ml) were significantly higher in B, whereas the total increase in calcemia (+0.26 +/- 0.03 vs +0.28 +/- 0.02 mmol/l) and the minimum value of PTH (59.15 +/- 9.53 vs 55.64 +/- 9.08 pg/ml) were similar in both experiments. The calcium-PTH curves were clearly different in A and B. The setpoint and the slope were significantly higher in A (1.196 +/- 0.01 vs 1.142 +/- 0.02 mmol/l; 840.54 +/- 96.85 vs 542.43 +/- 112.26%/mmol). For similar serum calcium values (1.084 +/- 0.01 vs 1.059 +/- 0.02 in the stimulation test and 1.325 +/- 0.02 vs 1.336 +/- 0.02 mmol/l in the inhibition test) the PTH secretion was significantly different (335.86 +/- 44.36 vs 647.65 +/- 104.09 in the stimulation test and 76.35 +/- 12.57 vs 105.03 +/- 20.59 pg/ml in the inhibition test). In conclusion, the way of inducing serum calcium change affected the calcium-PTH curve and the value of the set point and the slope was a function of the way in which the blood calcium changes were achieved. The modulated CaD dialysis was shown to be a more correct method of studying the calcium-PTH relationship in dialysis patients, as well as an alternative to the more complex and invasive infusional methodology.

Short-term effects of low-calcium dialysis solutions on calcium mass transfer, ionized calcium, and parathyroid hormone in CAPD patients.

Patients on CAPD using calcium carbonate (CaCO3) as phosphate binder might benefit from low-calcium (Ca) concentration dialysis solutions; however, no data are available for the effects of this regimen on Ca metabolism. We studied 10 patients on stable CAPD regimens with standard dialysis solutions (Ca 7 mg/dL) who were taking CaCO3 to control hyperphosphatemia (mean daily doses 4.5 +/- 2.4 g). Hypercalcemic episodes had been recorded in 6 patients. Standard dialysis solutions were replaced with solutions containing 5 mg/dL of Ca. Calcium and phosphate peritoneal mass transfer (MT), serum concentrations of total Ca, ionized Ca (Ca++), phosphate, intact PTH, and mid-molecular PTH, were evaluated before and 48 hours after change of dialysate. The switch to low-Ca solutions was accompanied by significant changes in calcium mass transfer (Ca MT) (+9.84 +/- 48.22 versus -96.74 +/- 48.32 mg/day, p less than .001). Ca MT was significantly (p less than .05) correlated with the serum/dialysate Ca gradient. There was no difference in phosphate MT. Serum Ca++ significantly (p less than .05) decreased from 5.20 +/- 0.32 to 4.88 +/- 0.36 mg/dL, and intact PTH significantly increased (81.5 +/- 139 versus 112.4 +/- 168 pg/mL, p less than .05). It is concluded that dialysis solutions with Ca 5 mg/dL result in a negative peritoneal Ca MT and can be useful to prevent and treat hypercalcemia in CAPD patients taking CaCO3 as phosphate binder. A careful monitoring of ionized calcium, PTH, and phosphate is suggested when an extensive and long-term use of this solution is considered.

Beta-2 microglobulin in patients on peritoneal dialysis and hemodialysis.

Serum beta 2 microglobulin (beta 2 mu) levels were determined in 62 patients on chronic dialysis, divided according to the type of dialysis--cuprophane hemodialysis, chronic ambulatory peritoneal dialysis (CAPD), or CAPD started after 76 +/- 47 months on cuprophane hemodialysis--and to residual urine output greater than 400 mL/day or less than 10 mL/day. In addition, for patients on CAPD, peritoneal excretion, peritoneal clearance, and urinary excretion of the protein were determined. In anuric patients serum beta 2 mu levels were significantly higher in HD than in CAPD. In patients with residual urine output, serum concentrations of the microprotein were similar in HD and in CAPD. Significant differences were observed in beta 2 mu serum levels and peritoneal clearances in patients switched to CAPD from hemodialysis as compared to those starting with CAPD. Peritoneal clearances of the microprotein was slightly and non-significantly greater in patients with urine output than in anuric patients.

Acquired immunodeficiency syndrome transmitted by transplanted kidney: clinical course during maintenance haemodialysis.

We describe the clinical course of a 39-year-old man who developed AIDS during maintenance haemodialysis. The infection had been transmitted by cadaveric kidney transplantation from a donor with a history of i.v. drug abuse. Fever, splenomegaly, hypergammaglobulinaemia and thrombocytopenia were the first signs, and appeared when haemodialysis was resumed 26 months after transplantation. Twenty-eight months later the patient developed a cerebral abscess due to toxoplasma infection. A striking improvement was obtained with cotrimoxazole, but the treatment had to be stopped because of severe leukopenia. The patient died due to relapse of the cerebral abscess. End-stage renal failure does not seem to have modified the clinical course of AIDS in our patient.