CIN2001: Differences in solute clearance between High Flux High Efficiency dialysis and on line Hemodiafiltration

Differences in solute clearance between High Flux High Efficiency dialysis and on line Hemodiafiltration

Garcia-Valdecasas Bernal J. C., Quiroga Subirana P.A., Abaigar P., Garcia-Valdecasas Valverde J. , Kimmel P.L.

Hospital Universitario San Cecilio. Facultad de Medicina. Granada.

E-mail: jgvaldecasas@hsc.sas.junta-andalucia.es

INTRODUCTION

According to technological progress, hemodialysis has been transforming towards more efficient and better tolerated modalities. In this way, several blood purification techniques have appeared in order to obtain both better quantity of treatment and dialysis quality, which actually are the aim treatment for patients with chronic renal failure.

These new therapeutic modalities must be done with high permeability membranes, volumetric control machines and using bicarbonate as a buffer. In this way, more comfortable and biocompatibility dialysis, less morbidity intradialysis and better nutrition state are obtained and, as a consequence, better rehabilitation and higher survival are reached.

Albertini assayed a new way of treatment that allowed dialysis efficiency to increase. It consisted in using high blood flows, between 600 and 800 ml/min, high hydraulic permeability membranes and the association of both difussive and convective movement of solutes (hemodiafiltration, HDF). Canaud started to do the hemodifilatration on line (HDF on line) in 1988, althought with unexpected results because of the problem of transfering endotoxines to the patients, problem that could be solved using a ultrapure liquid for substitution (4,10,13). This kind of HDF uses generally a ultrafiltration between 6 to 10 liters of liquid which has been made with previous filtrated dialysis fluid.The development of this technique, which allows an steril infussion fluid to be produced from dialysate liquid, has led to reduce the cost of treatment and thus to simplify and to generalize the use of HDF.

In the seventies and eighties, several studies showed that HDF was more efficient than hemodialysis because the adition of convection to difussion in molecule removal in the dialytic procedure. This favoured not only a good clearance of small molecules but too the large ones. However its efficiency is more difficult to note when we compare with high flux dialysis, because both use high permeable membranes and high blood and dilaysate flows (22, 23).

The aim of the present study is to compare solute clearances (urea, creatinine, phosphorus and beta 2 microglobuline), the removal of these molecules towards dialysate fluid and quantifying the amount of dialysis delivered (Kt/V), when high flux and high permeabilty membranes are used in two hemodialysis modalities: highflux hemodialysis (HD-HF)and hemodialysis on line (HDF on line).

PATIENTS AND METHODS.

A group of six patients were studied, four men and two woman (TABLE I), average age 56,0±14.7 years (range 29-76), average height 160,8±6,1 cms, the real time on dialysis was 240 minutes and the average dry weight 64,8±8,6 kg (range 51,5-75,5) and they had an ultrafiltration lost of 2433 ml± 324 ml. The average hematocrit was 36,6± 0,8 (range 35,8-38,1) Patients had neither residual function nor clinical cardiovascular problems and good autologues fistules. Corporal water volume was calculated according to Watson-Watson´s equation: average 34,3 liters (range 27,4-39,5).

TABLA I . DATOS GENERALES DE LOS PACIENTES

Paciente	Sexo	Edad (años)	Talla (cm)	Peso Seco (Kg)	Pérdida de Peso (ml)	Volumen corporal (Litros)	Tiempo HD (min)	Hematocrito (%)
Nº 1	H	76	162	70.0	2.500± 300	36.3	240	36.3
Nº 2	M	29	151	51.5	2.100± 300	27.4	240	35.8
Nº 3	H	54	171	75.5	2.600± 350	39.5	240	38.1
Nº 4	M	68	157	58.0	2.300± 300	28.5	240	35.8
Nº 5	M	51	161	71.0	2.700± 400	34.9	240	36.5
Nº 6	H	58	163	58.0	2.400± 320	39.3	240	37.3

All patients had to undergo hemodialysis treatment with highflux dialysers of 1,8 square meters (Fresenius HF 80) and the dialysis machine Fresenius 4000 Blood flow (QB) at 300, 400, 450, 500, 600 ml/min were used so on highflux hemodilaysis technique (HD-HF) as on hemodiafiltration on line (HDF on line). Dilaysate flow (QD) rate was used at 500 ml/min to evaluate the effect of dialysate flow on the solute clearances in all cases but when blood flow rates were 500 and 600 ml/min respectively, dialysate flow was 800 ml/min both on HD-HF and on HDF on line. A third of real QB was applied as substitution flow on HDF on line. All patients were on dialysis sessions of 240 minutes.

QB given by dialysis machine was modified to real QB according to negative arterial pressure (Table II)(24):

TABLA II. VALORES DE FLUJOS SANGUINEOS Y DE INFUSION ATENDIENDO A LA PRESION EN CIRCUITO EXTRACORPOREO. RECIRCULACION.

FLUJO MAQUINA (ml/min)	300	400	450	500	600
PRESION ARTERIAL NEGATIVA (mmHg)	120 ± 12	163 ± 17	203 ± 22	228 ± 24	280 ± 30
FLUJO REAL (ml/min)	274 ± 5	349 ± 9	387 ± 12	429 ± 16	513 ± 22
FLUJO DE INFUSION (ml/min)	91 ± 2	116 ± 3	129 ± 4	143 ± 5	171 ± 7
RECIRCULACION (%)	4.2 ± 0.5	6.6 ± 0.6	8.9 ± 0.7	10.7 ± 1.0	13.4 ± 1.3

when QB device < 400 ml/min : real QB = QB machine - 0,22*A.P.

when QB device > 400 ml/min : real QB = QB machine - 0,31*A.P.

Plasma samples were obtained to analyze urea, creatinine, phosphorus and beta 2 microglobuline KT/Vs at the begining and two minutes after dialysis procedure was finished. Urea Kt/V was calculated according to second generation Daugirdas` equation:

Kt/V = -Ln (ureapost/ureapre) - 0,008*D.t. + (4-3,75*ureapos/ureapre)*lost weight/dry weight.

Kt/V = -Ln(ureapost/ureapre)– 0,008*td+(4-3,5*ureapost/ureapre) * pérdida de peso/peso seco)

While Kt/V of the rest molecules were calculated according to a single pool model:

Kt/V = Ln (solute pre/solute post).

Blood samples were obtained 30 minutes after the begining of the dilaysis at arterial and venous lines and in peripheral veins to determine creatinine, phosphorus, urea and beta 2 microglobuline clearances and the recirculation of the vascular access. Thus, urea and creatinine clearances were measured in total blood and phosphorus and beta 2 microglobuline in plasma water, appling next equations (ultrafiltration rate was eliminated for simplifing):

Kurea = QBr*BUN_a(1-Hct+ßurea*Hct)-QBr*BUN_v(1-Hct+ßurea*Hct)
BUN_a(1-Hct+ßurea*Hct)

Kcreatinine = QBr*Creat_a(1-Hct+ßcreat*Hct)-QBr*Creat_v(1-Hct+ßcreat*Hct)
Creat_a(1-Hct+ßcreat*Hct)

Kphosph = QBr*Hct(Fosf_a – Fosf_v)
Fosf_a*100

Kbeta2-microglobuline = QBr*Hct(Beta2m_a – Beta2m_v)
Beta2m_a

Meanings: QBr, real blood flow. Solute_a, solute at arterial line. solute_v. solute at venous line. Hct, hematocrit. Burea, intracelular urea concentration, 0,859.Bcreat, intracelular creatinine concentration. These clearances were corrected attending to recirculation (TABLE II), according to next equation:

Krecirc = K*[1-1.05*Recirc(1- ) soluto_v
soluto_v

Meanings: Krecir, clearance solute after correcting to recirculation. Recir, % of obtained recirculation to each blood flow.

Values are showing as average ± SD (standard deviation). Samples were analized with paired t Student test and we consider significant a p value < 0,05

RESULTS:

Urea

Figure 1 shows urea clearances both on highflux dialysis and on hemodiafiltration on line with different blood and dialysate flows that were performed.

In table III it can be observed with a parallel clearance increment with QB increment that becomes a 60,2 % higher on HD-HF and a 58, 9 % on HDF on line with respect to basal values at 300 ml/min (p< 0.001).

TABLA III. INCREMENTOS DE LOS ACLARAMIENTOS ESTUDIADOS

QB (ml/min)	300	400	450		500	500	600	600
QD(ml/min)	500	500	500		500	800	500	800
Urea
Estandar vs. HD-HF	0 %	21.9%		32.1%	40.3%	43.4%	54.3%	60.2%
Estandar vs. HDF en línea	19.9%	36.2%		44.1%	48.0%	51.8%	54.3%	58.9%
HD-HF vs. HDF en línea	19.9%	11.8%		9 %	5.4%	5.8%	0.0%	-0.8%
Creatinina
Estandar vs. HD-HF	0%	13.8%		19.1%	23.9%	28.8%	31.5%	36.3%
Estandar vs. HDF en línea	16.4%	26.6%		29.5%	30.8%	31.9%	33.0%	37.0%
HD-HF vs. HDF en línea	16.4%	11.3%		8.7%	5.6%	2.4%	1.2%	0.5%
Fósforo
Estandar vs. HD-HF	0%	11.2%		16.6%	20.5%	22.8%	28.0%	31.3%
Estandar vs. HDF en línea	19.1%	28.9%		32.5%	35.5%	37.2%	38.1%	41.0%
HD-HF vs. HDF en línea	19.1%	15.9%		13.6%	12.4%	11.7%	7.8%	7.4%
Beta - 2 - microglobulina
Estandar vs. HD-HF	0%	28.2%		41.5%	51.7%	59.6%	70.7%	78.1%
Estandar vs. HDF en línea	48.5%	75.2%		88.5%	98.9%	108.4%	120.5%	130.6%
HD-HF vs. HDF en línea	48.5%	36.7%		33.2%	31.1%	30.6%	29.2%	29.4%

QB= Flujo de sangre; QD= Flujo de líquido de diálisis

HD-HF= Hemodiálisis de alto flujo; HDF= Hemodiafiltración

In the same table, it may be observed a different relationship between both techniques according to QB, whereas urea clearence is a 19, 9 % higher (p<0,01) at 300 ml/min blood flow on HDF on line than the one on HD-HF, this better efficiency of the HDF on line is slowly shortening as the blood and substitution flux increase, so at 600 ml/min blood flow the statistic associations between both modalities not only disappear but also urea clearance is even slightly superior on highflux hemodialysis.

. En la TABLA IV se indican los valores obtenidos de urea pre y post diálisis.

TABLA IV . VALORES DE LOS SOLUTOS ESTUDIADOS

QB (ml/min)	300	400	450	500	500	600	600
QD(ml/min)	500	500	500	500	800	500	800
Urea
Pre HD-HF	147.6 ± 4.2	144.8 ± 3.2	142.3 ± 3.9	148.5 ± 3.9	151.7 ± 2.9	143.1 ± 3.8	141.3± 3.7
Post 2´ HD-HF	50.2 ± 4.1	38.7 ± 3.1	33.0 ± 3	30.6 ± 3.3	30.0 ± 4.1	24.6 ± 7	22.7± 6
Pre HDF en línea	145.2 ± 5.3	150.5 ± 3.5	146.6 ± 6	144.4 ± 4.3	142.8 ± 4.7	146.3 ± 6	138.9± 5
Post 2´ HDF en línea	41.1 ± 4.0	32.9 ± 4.2	30.2 ± 4	26.4 ± 5.2	25.1 ± 3.8	24.9 ± 6	23.2± 5
Creatinina
Pre HD-HF	10.7± 1.3	11.2± 1.7	10.4± 1.5	10.6± 1.1	11.1± 1.7	11.3± 1.4	10.8± 1.9
Post 2´ HD-HF	3.8± 0.9	3.5± 0.7	3.1± 1.1	3.1± 0.8	3.1± 0.7	3.0± 0.7	2.8± 0.6
Pre HD-HF	11.3± 1.6	10.8± 1.8	10.6± 1.3	10.8± 1.7	11.21.9	10.4± 1.6	10.6± 1.5
Post 2´ HDF en línea	3.5± 0.6	3.1± 0.8	2.9± 0.3	2.9± 0.9	3.0± 0.8	2.7± 0.5	2.7± 0.8
Fósforo
Pre HD-HF	4.8± 0.9	4.9± 0.8	5.1± 1.1	5.0± 0.9	5.3± 0.8	5.2± 0.9	4.8± 0.7
Post 2´ HD-HF	2.4± 0.3	2.3± 0.5	2.3± 0.4	2.2± 0.3	2.3± 0.3	2.2± 0.4	2.0± 0.2
Pre HDF en línea	5.1± 0.9	4.9± 1.2	5.3± 0.9	4.8± 0.9	5.2± 1.1	5.0± 1.1	4.9± 1.1
Post 2´ HDF en línea	2.3± 0.3	2.1± 0.5	2.3± 0.4	2.0± 0.2	2.1± 0.3	2.0± 0.3	1.9± 0.4
Beta – 2 – microglobulina
Pre HD-HF	37.1± 3.5	33.6± 4.3	35.4± 4.1	36.4± 3.9	35.8± 4.1	33.3± 3.6	38.9± 4.1
Post 2´ HD-HF	15.3± 2.3	11.1± 2.1	10.1± 2.7	10.0± 1.9	9.6± 1.6	8.6± 2.3	9.8± 2.8
Pre HDF en línea	34.9± 4.3	39.8± 4.5	35.6± 3.8	32.9± 4.1	34.7± 4.3	38.9± 3.7	36.2± 4.5
Post 2´ HDF en línea	9.6± 0.9	8.7± 1.1	7.1± 1.8	6.0± 1.2	5.6± 1.3	5.8± 1.5	4.9± 0.8

QB= Flujo de sangre; QD= Flujo de líquido de diálisis

HD-HF= Hemodiálisis de alto flujo; HDF= Hemodiafiltración

Pre= Pre hemodiálisis; Post 2´ = 2 minutos posterior a la hemodiálisis

Urea pre and post dialysis are shown in table IV. Figure 2 and Table V display urea Kt/V values and their increments at several blood and dialysate flows. There are an urea Kt/V increment as the blood flow increases as it was observed with urea clearances and they rise to values of 69,4 % and 65,7 % higher on HD-HF and on HDF on line with respect to basal values at 300 ml/min. Therefore a different behaviour was observed between both dialysis modalities according to blood flow: at low blood flows (QB equal to 300 ml/min), the Kt/V obtained with HDF on line is a 19, 9 % higher (p>0,01) than the one on HD-HF. However, as blood flow increases the statistical difference disappeared and at 600 ml/min even so HD-HF is better in 2,2 % to HDF on line.

TABLA V. INCREMENTOS DE LOS Kt/V ESTUDIADOS

QB (ml/min)	300	400	450		500	500	600	600
QD(ml/min)	500	500	500		500	800	500	800
Urea
Estandar vs. HD-HF	0 %	22.2 %		35.2 %	46.3 %	50.0 %	62.9 %	69.4 %
Estandar vs. HDF en línea	16.7%	40.7 %		45.4 %	57.4 %	61.1 %	63.9 %	65.7 %
HD-HF vs. HDF en línea	16.7%	15.1%		7.5%	7.6%	7.4%	0.6%	-2.2%
Creatinina
Estandar vs. HD-HF	0%	11.7 %		16.5 %	19.4 %	24.2 %	28.1 %	30.1 %
Estandar vs. HDF en línea	14.6 %	22.3 %		26.2 %	28.1 %	29.1 %	30.1 %	31.1 %
HD-HF vs. HDF en línea	14.6%	7.8%		9.0%	6.5%	3.0%	1.5%	1.5%
Fósforo
Estandar vs. HD-HF	0%	10.1 %		15.9 %	18.8 %	20.2 %	24.6 %	27.5 %
Estandar vs. HDF en línea	15.9 %	23.2 %		26.1 %	29.0 %	31.9 %	33.3 %	37.7 %
HD-HF vs. HDF en línea	15.9 %	11.8%		8.7%	8.5%	9.6%	7.0%	7.9%
Beta - 2 – microglobulina
Estandar vs. HD-HF	0%	24.7 %		41.6 %	44.9 %	48.3 %	51.6 %	55.0 %
Estandar vs. HDF en línea	44.9 %	70.7 %		80.1 %	91.0 %	100.1 %	114.6 %	123.6 %
HD-HF vs. HDF en línea	44.9%	36.9%		28.8%	31.8%	37.9%	40.7%	44.2%

QB= Flujo de sangre; QD= Flujo de líquido de diálisis
HD-HF= Hemodiálisis de alto flujo; HDF= Hemodiafiltración

Creatinine

Figure 3 delays creatinine clearance obtained at different blood and dialysate flows in both modalities of dialysis. We can observe a parallel clearance increment at the same time with blood flow increment. These clearances reached values a 36,3 % and 37 % higher both in HD-HF and in HDF on line respectively versus values at 300 ml/min blood flow (p<0,001).

We can observe similar findings as in urea clearances, so creatinine clearance presents different behaviour between both techniques according to QB applyes, while creatinine clearance is higher on HDF on line than on HD-HF at 300 ml/min blood flow (16,4 %, p< 0,019
This HDF on line high efficiency disappears as soon as blood flow and substitution fluid are increased. At 500 ml/min blood flow and at 800 ml/min dialysate flow there are no statistical differences between both dilaysis modalities, although in this case HD-HF never reached HDF on line.

Table V shows pre and post dialysis values obtained. Creatinine Kt/V values and their variations at several blood and dilysate flows are displayed in Figure 4 and Table V. In the same manner that ocurred with urea clearances, those Kt/V increase according to blood flow increment, in such a way they rise at 30,1-31,1 % higher on HD-HF and HDF on line if they were compared with basal values at 300 ml/min blood flow. Besides, blood flow induces a different replay between both modalities: at low blood flow (300 ml/min), Kt/V on HDF on line is a 14, 6 % (p< 0,01) higher than on HD-HF; however this efficiency is reduced so that there are no statistical differences between both techniques at 500 ml/min blood flow, althought HDF on line is slightly higher than HD-HF (1,5 %, ns).

Phosphorus

Figure 5 shows phosphorus clearences obtained with both modalities at different blood and dialysate flows. We can observe a phosphorus clearance increment in parallel with blood flow. They are both at 31,3% and 41% (p<0,001) higher on HD-HF and HDF on line compared with basal values (TableIII). This solute presents different behaviour than urea and creatinine clearances according to blood flow applied, so at 300 ml/min blod flow phosphorus clearance on HDF on line is a 19,1% (p<0,01) superior to the obtained on HD-HF. On the contrary that was observed with urea and creatinine, the higher efficiency of HDF on line was maintained as blood flow and substitution fluid rates were increased, although these differences are slower, so the phosphorus clearance was a 7,4 % (p<0,05) higher with HDF on line versus HD-HF at 600 ml/min and 800 ml/min blood flow.

Pre and postdialysis values are showed in Table IV. figure 6 and Table V display the phosphorus Kt/V and its variations at several blood and dialysate flows. There is an increment of phosphorus Kt/v in parallel with blood flow increments, which reached a 25,75% and 37,7 % higher on HD-HF and HDF on line when they are compared with basal values (p<0,001). In the same way the phosphorus clearance has a different beahviour of its Kt/V in comparing with urea and creatinine. It was observed that the phosphorus Kt/V is always superior with HDF on line versus HD-HF, at 300 ml/min is a 15,9 % higher and a 7,9 % (p<0,05) at 600 ml/min and 800 ml/min blood flow.

Beta 2 microglobuline

Beta 2 microglobuline clearances with both modalities of dialysis at different blood and dialysate flows are showed in figure 7. Those clearances increase in parallel with blood flow and they are at 7,8 % and a 130,6 % (p<0,001) higher on HD-HF and HDF on line respectively in comparing with basal values at 300 ml/min blood flow (Table III). < > Clearance of this solute has a different behaviour to the urea and and creatinine clearances, so that beta 2 microglobuline clearance on HDF on line at 300 ml/min blood flow is a 48,5 % superior (p<0,001) to the one on HDF. High efficiency of the HDF on line over HD-HF is mantained at all blood and dilysate flows, and so ata 600 ml/min blood flow and at 800 ml/min dilysate flow, HDF on line reaches a clearance of beta 2 microglobuline a 29,4 % (p<0,001) hgiher than HD-HF.

Table V shows pre and post dialysis values. Figure 8 and Table V display the Kt/V beta 2 microglobuline indice and its increments to different blood and dilysate flows.

In the same way as it was observed with clearances of this solute, there is a Kt/V increment when blood flow increases, so it is a 55% and 123,6 % (p<0,001) higher on HDF and HDF on line respectively when comparing with basal values.
Besides, the beta 2 microglobuline Kt/V is different to the urea and creatinine ones. It was observed that this Kt/V was always higher on HDF on line versus HD-HF: at 300 ml/min blood flow is a 44,9% higher and at 600 ml/min blood flow and 800 ml/min dialysate flow (44,2 %, p<0,001)

DISCUSION

We have studied clearances an Kt/v of urea,creatinine,phosphorus and beta 2 microglobuline comparatively at at several blood and dilysate flows, between 300 and 600 ml/min and between 500 and 800 ml/min respectively. This study suggests that the dialysis efficiency, which is evaluated towards clearances and Kt/V of some molecules, is increased both with blood and dialysate flow increments which allow the quantity of dilaytic treatment to improve without increasing the dialyser membrane surface or the treatment time. It hab been choosen dialysers with high flux membranes which are known as those with a ultrafiltration coefficient rate higher than ml/mmHg*h*m² (Norma UNE 111-325-89). Those dialysers allow to do high efficiency dialysis (Kt/V higher than 0,35 ml/min2, mass coefficiente -KoA- superior to 800 ml/min), which not only led higher survival but better quality of live than conventional dialysis (25,26).

When blood solute clearances are studied, the main mistake that has been made is related to blood flow. Dilaysis machine blood flow never reflects the reality and this is a consequence of the different needle gauge and calipre lines in the extracorporeal circuit and the diameter of segmente pump , which produces negative pressure at arterial extracorporeal circuit. This arterial pressure is inversely proportional to the section of calipre segment pump and idrectly proportional to the flow obtained from the pump. In fact, it is necessary to modify the flow showed by the monitor in order to avoid clearance overvaluations. Venous postive pressure does not affect the real flow (24). On the other hand, clearance is increased by recirculation which depends on blood flow and so it must be taken into account to avoid clearance overevaluation.

Clearance of small solutes such as urea and creatinine have a different behaviour to middle molecules like beta 2 microblobuline both on HD-HF and HDF on line. Blood and dilaysate flows increment Allows the removal of solutes towards dialyser to increase, this removal behaviours in a different way according to the treatment modality applies. HDF on line is better than HD-HF in clearance of small solutes at middle flows, that means blood flow values between 300 - 450 ml/min. However, when blood flow is higher than 500 ml/min, clearance rises with both techniques, although this difference is minimized at higher blood flows and it disappeard at 600 ml/mi blood flow. This fact can be due to the interference of high convection on the diffusion at this flows. Data existing in medical literature shows a high efficiency of hemodiafiltration on line with respect to conventional dialysis and there is no contradition with our results because in all essays blood flow applied were less than 400 ml/min. Maduell shows a slight difference between hemodiafiltration on line and high flux hemodialysis in a previous essay and it was a consequence of using blood flow at 450 - 500 ml/min, that is according to our results and those from other authors (23,31,32).

On the contrary, when a high molecular weight solute is evaluated, as for example beta 2 microglobuline, which has its main removal towards convective movement and hardly by diffuse one, we not only observed that clearance increases as blood and dilaysate flow increments, but also is superior HDF on line versus HD-HF. Phosphorus has a intermediate behaviour between both kind of molecules, that means it has a mixed removal mechanism.

When we observe the quantity of dialysis (Kt/V), it occurs the same. Our results show that there is a different behaviour of urea and creatinine Kt/V both with respect to phosphorus and beta 2 microglobuline. Whereas blood and dilysate flows led to improve dilaysis quantity of all solutes evaluated in this study, and the HDF on line always offers higher and best phosphorus and beta 2 microglobuline Kt/V than HD-HF, we find a different behaviour with small solutes as urea and creatinine. At 450 ml/min or higher blood flow, quntity of dialyisis of this solutes is quite similar in both modalities of treatment. This is a consequence of clearance behaviour. According to our results, such in a treatment (HD-HF) as in the other (HDF on line), urea Kt/v higher than 1,3 are obtained in all patients when blood flow is at least 400 ml/min, value that should be reached if we want to follow the present recommendations of the American multicenter study or the guidelines for hemodialysis of the National Kidney Foundation (DOQQI) (34). Otherwise, considering a Kt/V of 1,2 or higher as acceptable according to the study HEMO (35,36), HDf on line enables all patients to have this value at 300 ml/min or at higher blood flow and a third of this one of reinfusion substitution fluid, but HD-HF reaches only this values at 400 ml/min or higher blood flow.

In conclusion, our results show that blood and dialysate increments allow the removal of solutes of low and high molecular weight to improve as much on high flux dilaysis as on hemodiafiltration on line. At equal or low 450 ml/min blood flow hemodiafiltration on line leads higher clearance and Kt/V of solutes studied - (urea, creatinine, phosphorus, and beta 2 microglobuline)- than high flux dialysis. However, at higher flows than this one the difference dissappears both in urea and creatinine clearances. It can be due to the differences between diffusion and convection of these solutes at this flow. But, this fact does not occur when we analize phosphorus and beta 2 microglobuline clearances, being HDF on line better technique than high flux dialysis.

REFERENCES

Gotch F.A. Dialysis of the future. Kidney Int. 32 (Suppl. 24) S100-S106, 1988.

Farrell P.C. Adequacy of dialysis: Marker molecules and kinetic modeling.Artif. Organs 10(3), 195-199, 1986.

Greksa J.E., Laos J.B. Operational considerations of reduced-time hemodialysis therapy. Contemp. Dial. Nephrol. 8(3), 8-12, 1987.

Von Albertini B., Miller J.H., Gardner P.W., Shinaberger J.H. High-flux hemodiafiltration: under six hours/week treatment. Trans. Am. Soc. Artif. Intern. Organs 30, 227-231, 1984.

Kjellstrand C, Mauer S, Buselmeier T: Haemodialysis of premature and newborn babies. Nephrol Dial Transplant 10:349-352, 1973.

Kaiser B, Potter D, Bryant R, Vreman H, Weiner M: Acid-base changes and acetate metabolism during routine and high-efficiencicy hemodialysis in children. Kidney Int 19:70-79, 1981.

Colling AJ: High-efficiency treatments using conventional equipment. En: Hemodialysis high-efficiency treatments. Churchill livingtone Inc, pp. 91-104, 1993.

Funck-Brentano J.L., Man N.K., Sausse A. Effect of more porous dialysis membrane on neuropathic toxins. Kidney Int. 7 (Suppl. 2) S52-S57, 1975.

Alvarez Ude F., Ward M., Elliot R.W., Uldall P.R., Wilkinson R., Appelton D.R., Kerr D.N.S., Petrella E., Gentile M., Romagnoni M., Orlandini G., Luciani L.,Ferrandes C.,D'Amico G. A comparison of short and long haemodialysis. Proc. Eur. Dial. Transpl. Assoc. 12, 606-614, 1975.

Canaud B., Flavier J.L., Argilés A., Stec F., NGuyen Q.V., Bouloux Ch., Garred L.J., Mion C. Hemodiafiltration with on-line production of substitution fluid: long-term safety and quantitative assessment of efficacy.Contrib. Nephrol., Basel, Karger, vol 108, 12-22, 1994.

Gould M.J. Bacterial endotoxins in Hemodialysis LAL UPDATE Vol 7, (4): 1-5, 1989.

Frinak S, Polaschegg HD, Levin NW, Pohlod DJ, Dumler F, Saravolatz LD. Filtration of dialysate using an on-line dialysate filter. Int. J. Lf Artif. Organs. Vol 14 (11), 691-697, 1991.

Bommer J. Becker KP, Urlaschek R. Potential transfer of endotoxin acrrs High-Flux Polysulfone membranes. J. Am. Soc. Nephrol. 5, 883-888, 1996.

Lewer WW, Wizemann v, Goubeau G: Simultaneous hemodiafiltration-hemodialysis in the treatament of the uremic patients. Clin Nephrol 9: 115, 1978.

Held PJ, Blagg CR, Liska DV, Hakim R, Levin N: The dose of hemodialysis according to dialysis prescription in Europe and the United States. Kidney Int 42 (Suppl. 38): 19-21, 1992.

Schaefer S, Von Herrth D: Hemofiltration: The past and present, in Karger, Polyamide: The evolution of a snthetic membrane for therapy, contributions to Nephrology 96, 64-76, Basel, 1992.

Henderson LW: Biophysics of ultrafiltration and hemofiltration. En replacement of renal function by dialysis. Drukker W, Parsons FM, Maher JF, Boston y M Nijoff, (EDS.) pp 242-264, 1983.

Henderson LW: Middle molecules re-examined. Nephron 22: 146-149,1978.

Bergstrom J, Fust P, Zimmerman L: Uremic middle molecules exist and are biologicallyactive. Clin Nephrol 11:229-232, 1979.

Mrowka Ch, Schiffl H: Comparative evaluation of B2-microglobulin removal by different hemodialysis membranes: a six year follow-up. Nephron 63:368-369, 1993.

Hoenich NA, Woffindin C, Matterws JNS, Goldfinch ME, Tumbull J: Clinical comparison of hig-flux cellulose acetate and syntetic membranes. Nephro Dial Transplant 9: 60-66, 1994.

Kerr PG, Argilés A, Canaud B, Flavier JL, Mion CM: Accuracy of Kt/V estimations high-flux haemodiafiltration using percent reduccion of urea: incorporation of urea reboud. Nephrol 12: 11, 1992.

Ahrenholz P, Winkler RE, Ramlow W, Tiess M, Müller W. On-line hemodiafiltration With pre-and postdilution: A comparison of the efficacy. Int J Artif Organs. 20 (2): 81-90, 1997.

Boch JP, Barlee V, Garcia-Valdecasas J. Blood flow measurement during hemodialysis. Adv. Renal Replac. Thec. 1, 83-88, 1994.

Levin N, Dumler F, Zasuwa G, Stalla K: Mortality comparison between conventional and high flux dialysis. J Am Soc Nephrol. :365-367, 1990.

Collins A, Liao M, Umen A: High-efficiency bicarbonate hemodialysis, has a Iower fisk of death than standard acetate dialysis, abstract. J Am Soc Nephrol. 2: 318, 19991.

Henderson HL: Biophysics of ultrafiltration and hemofiltration, in Kluwer Academic Publishers, Replacement of renal fuction by dialysis, 114-145 Dordrecht, 1996.

Waniewski J, Werynski A, Ahrenholz P, Lucjanek P, Judycki W, Esther G: Theoretical basis and experimental verification of the impact of ultrafiltration on dialyzer clearence. Artificial Organs 15(2): 70-77, 1991.

Aljama P, Amate JM, Conde JL. Criterios de clasificación de las membranas. Nefrología (XVI) Suplemt 4: 50-63, 1996.

García H, Hernández-Jaras, Maduell F, Yago M, Calvo C, Navarro V, Villatoro J. Eficacia de la hemodiafiltración en línea (HDF) comparada con la hemodiálisis de alto flujo (HD). XVIII (6): 476-482, 1998.

Jindal KK, Mcdoudall J, Woods B, Nowaroski L, Goldstein MB: Study of the basic principles determining the performance of several high-flux dialyzers. Amer J Kidney Dis. 14(6): 507-511, 1989.

Pedrini L, De Cristofaro V, Samà F, Pagliari B, Masa A, Filippini M: Low-and hig-molecular-weight (MW) solute renoval in pre- and post-ditlution on-line hemodiafiltration (HDF). Abstracts, 14^th International Congress of Nephrology. 108, 38-45, Basel, 1994.

Held PJ, Port FK, Wolfe RA, Stannard DC, Carrol CE, Daugirdas JT, Bloembergen WE, Greer JW, Hankim RM: The dose of hemodialysis and patient mortality. Kidney Int 50: 550-556, 1996.

NKF-DOQI Hemodialysis Adequacy Word Group Memberchip. Guidelines for hemodialysis adequacy. Am J Kidney Dis 30 (Supl 2): S22-S63, 1997.

Eknoyan G. Levery AS, Beck GJ, Agodoa LY, Daugirdas JT, Kusec JW, Levin NW, Schulman G. Hemodialysis (HEMO) study: Rationale for selection on interventions. Semin Dial 9: 24-33, 1996.

Daugirdas JT, Depner TA, Gotch F, Greene T, Keshaviah P, Levin NW, Shulman G: Comparison of methods to predict equilibrated Kt/V in the HEMO pilot study. Kidney Int 52: 1395-1405, 1997.