Artijkid Orgcrns 16(6):547-552, Blackwell Scientific Publications, Inc., Boston 0 1992 International Society for Artificial Organs
Short Time Treatment with High-Efficiency Paired Filtration Dialysis for Chronic Renal Failure Roberto Marangoni, Renato Savino, Roberto Colombo, Roberto Cimino, and Fernando Civardi Nephrology and Dialysis Unir, Bollate Hospital, Milan, Italy
Abstract: With the aim of treating with short sessions even chronic uremic patients with cardiovascular instability, we have developed our previous findings about paired filtration dialysis (hemofiltration-hemodialysis in series) by a method we call high-efficiency paired filtration dialysis (HEPFD). We have treated with this method for 6 months 6 chronic anuric uremic patients previously treated for 6 months with hemofiltration (HF) because of their cardiovascular instability. By comparison with HF,
HEPFD shows much higher clearances of small molecules with consequent significant decrease of their predialytic values, improvement of KrIV value, lower P2-microglobulin clearance (its predialytic value remains unchanged), and the same tolerability. HEPFD allows short, efficient, and well-tolerated treatments even in patients with cardiovascular instability. Key Words: Hemofiltration-Paired filtration dialysis-Cardiovascular instability-Short treatments-Uremic toxins.
The current dialysis philosophy requires short, efficient, and well-tolerated treatments. Even traditional hemodialysis (HD) has shown a development toward this route employing highly permeable membrane dialyzers with a large surface, and bicarbonate as buffer. Nevertheless, we know that the majority of patients cannot undergo short treatments (that is, treatments with a duration of about 3 h) mainly because they do not tolerate high body weight loss per unit of time. In fact, it is generally accepted that short hemodialysis can be adopted only in the treatment of patients with low dry body weight, low interdialytic weight gain, and stable cardiovascular condition (1-3). The method that takes advantage of convection, hemofiltration (HF), gives a high tolerability, especially from the cardiovascular point of view, allowing high body weight loss per unit of time even in patients with cardiovascular instability (4-6). Nevertheless, mainly in order to remove low molecular weight uremic toxins, H F needs a large amount
of infusion fluid, which causes long session times especially in patients who for their critical cardiovascular condition do not tolerate treatments with a high blood flow (7-10). The method that takes advantage of convection and diffusion simultaneously, hemodiafiltration (HDF), gives a satisfactory tolerability and high clearance of small and medium-sized molecules (large-molecule clearances are, however, lower than those of HF), and therefore it enables us to increase the number of patients undergoing short treatments (1 1). However, in order to achieve a high efficiency per unit of time, HDF needs high blood flows that cannot be allowed in patients with cardiovascular instability or in patients with poor vascular access (12). Moreover, convection and diffusion taking place simultaneously can interfere with each other (13). With the aim of obtaining short, efficient, and welltolerated blood purification sessions, a method called paired filtration dialysis (PFD) that consists of HF-HD in series (Fig. 1) has been carried out (14-17). In this system, the convective section (HF) is used for removal of small and especially large molecules or for body weight loss while the diffusive section (HD) is used mostly for removal of small molecules. Transmembrane pressure (TMP) of the
Received September 1991; revised May 1992. Address correspondence and reprint requests to Dr. R. Marangoni at Bollate Hospital, via Piave 20, 20021 Bollate-MI, Italy. This work was presented in part at the VIIIth World Congress of the International Society for Artificial Organs, held August 19-23, 1991, Montreal, Canada.
54 7
R . MARANGON1 ET A L .
548
FIG. 1. Schematic representation of the paired filtration dailysis. See text.
HD section is kept about 0 mm Hg in order to avoid body weight loss in this section. Since the PFD original extracorporeal circuit did not allow, in our results, a satisfactory exploitation of the convective section (Table I ) , especially with respect to depurination because of the small surface of the hemofilter, we set out to improve the method, which we then called high-efficiency paired filtration dialysis (HEPFD). The changes consisted in increasing the hemofilter surface and the substitution fluid amount, which we variously modified together with the ultrafiltration rate (UFR), blood flow, and session time (Table 2), obtaining in this way different clearances of small and large molecules (18). Among the tested modifications, which we called HEPFD4, seemed to us to offer the best advantages: the highest exploitation of convection with satisfac-
TABLE 1. PFD original method Hernofilter": Hernodialyzer": Whole infusion fluid (NaCI 9%): UFR: Session time: I'
Polysulfone 0.4 rn? Hernophan 1.0-1.3 m? 4-5.000 rnl 41 -46 mlirnin 180 rnin
Kit SG3 by Sorin (Italy).
tory clearances of large (Pz-microglobulin: B2-M) and small molecules without cutting down the clearances of the last ones in the diffusive section, satisfactory whole efficiency even with blood flows lower than the standard ones, and well-equilibrated shortening of the session times to allow a tolerable body weight loss in the majority of the patients. On the basis of these findings, we wanted to compare HEPFD4 with HF for efficiency and tolerability, aiming at verifying the possibility of using short sessions in patients with such serious cardiovascular instability as to be obliged to leave HD. MATERIALS AND METHODS
Six chronic anuric uremic patients (Table 3) had to leave bicarbonate HD treatment because of cardiovascular instability. At least 30% of their sessions involved symptomatic hypotension, and fall of maximum arterial pressure under 90 mm Hg was accompanied by one or more of the following symptoms: asthenia, nausea, vomiting, vertigo, or cramps even with body weight loss of 700 t50 glh. These patients have been treated for 6 months with HF (Table 4) (during this treatment no patient showed cardiovascular instability) and successively for 6 months with HEPFD4 (Fig. 2; Table 5 ) .
TABLE 2. Modifiixtions to PFD Session time (rnin) I80 180 210 210
Blood flow 200 3 00 250 250
Infusion rate (mlhin)
so 2 5 61 2 7 43 3 64 2 6
*
Ultrafiltration rate 61 73 55 74
Whole infusion (ml)
2 6
9.000
7 3 7
1 1.000 9.000 13.000
2 2 2
Method denomination HEPFD HEPFD HEPFD HEPFD
I 2 3 4
HIGH-EFFICIENC Y PAIRED FILTRATION DIALYSIS TABLE 3. Population baseline data Patient
1
2
3
4
5
6
Age (years) Sex Diagnosis Dry body weight (kg) Fluid overl. (kg) (mean)
28 m GN
57 f GN
66 CPN
74 f DN
61 f DN
58 m PKD
69 2.3
58 2.5
61 2.4
72 2.7
75 2.9
78 3.0
m
G N , glomerulonephritis; PKD, polycystic kidney disease; DN, diabetic nephropathy; C P N , cystopyelonephritis.
TABLE 4. Technical data of hernofiltration ( H F ) Monitors Hemofil ter U FR Session duration Mean body weight loss Substitution fluid (mmoliL) NaK' Ca? Mg! *
c1
Lactate Reinfubion fluid (postdilutional method) Blood flow
Sifra S I F 907; Gambro H F Polysulfone 2.0 m? 110 2 10 mlimin 250 2 13 min 700 +- 50 mlih
TABLE 5. Technical dutu oj'high-efjkiency paired jiltration dialysis (HEPFD4) ( H F + H D in series) H F monitor: PFD Sorin Hernofilter": PS 0.7 m'
26 L 250 mlimin
In these patients we have determined urea, creatinine, uric acid, and P2-M clearances during HF and HEPFD4 treatment; removal amount of urea, creatinine, uric acid, and P2-M in one HF and in one HEPFD4 session; mean arterial pressure (MAP) and heart rate (HR) modifications during HF and
H D monitor: Fresenius 2008 Dialyzer": Hemophan 1.3 m?
Formulae (mmoliL) Diulysute Na+ 1 K' 2.25 Ca' 0.75 Mg" 112 CI 35 HCOICH3COO-
Suhstitlrtion Pirid Na+ 140
K' Ca?+ Mg? CI Lac late +
Reinfusion fluid: 13 L Dialysate Row: 500 rnlimin Blood flow: 250 mlirnin Session time: 210 min UFR: 74 2 7 mlimin Mean body weight loss: 910 "
140 (138) I (21 2.25 (2) 0.75 112 (110) 35
549
+
~
138
2 I .75 0.50 109.50 31 4
* 60 mlih
Kit SG6 by Sovin (Italy).
HEPFD4 treatment; urea, creatinine, uric acid, and P2-M predialytic serum levels after 6 months of HF and after 6 months of HEPFD4 treatment; and KtlV after 6 months of HF and after 6 months of HEPFD4 treatment. The statistics were calculated using the Student's r test; p values < 0.05 were accepted as significant. RESULTS The results obtained show that small-molecule clearances (urea, creatinine, uric acid) are significantly higher in HEPFD than in HF (Fig. 3); the diffusive section of HEPFD contributes more than the convective section to the whole system clear-
FIG. 2. Working HEPFD (HF-HD in series).
Artif'Orgnns, V o l . 16. N o . 6 , 1992
550
R . MARANGONI ET AL.
250
ml/nin
Ht = 28% total proteins = 7’ g/dl 200
150
FIG. 3. Urea, creatinine, uric acid, and 02M clearances determined after 60 min in HF and in HEPFD4 (mean SD).
*
i 00
50
0
creatinine
urea
HF HEFFD4 * HEPFD4 vs HF p< 0.001 ** HF YS HEPFD4 p< 0.001
ances. P2-M clearance, which is due to convection only, is obviously significantly higher in H F than in HEPFD (Fig. 3). Consequently, small-molecule removal is significantly higher in HEPFD (Fig. 4) while large-molecule removal is higher in H F (Fig. 5 ) . HEPFD tolerability, especially from the cardiovascular point of view, is at least the same as of H F (Fig. 6). Predialytic serum levels (determined before the beginning of the first weekly treatment) of urea,
E’7- K
uric acid
nHEPFD4 (diffusive)
(whole)
creatinine, and uric acid show a significant decrease after 6 months of HEPFD treatment (Table 6). Even KtlV values show a significant improvement after 6 months of HEPFD treatment (Table 6 ) . DISCUSSION AND CONCLUSION According to our experiences, HEPFD has severa1 advantages. It can give high small-molecule re-
grams 50
-r *
mean of 3 determination in each of 8 patients
Ht
=
28% tc&d proteins
=
7
g/dl
20
*
FIG. 4. Amount of removal of BUN, creatinine, and uric acid in only one session of HF and HEPFD4 (mean t SD).
HEPFDI vs HF p c 0.001
10
n
BUN
Artf Organs, V d . 16, N o . 6 , 1992
=
creatinine
HF
HEFFD4
uric acid
HIGH-EFFICIENC Y PAIRED FILTRATION DIALYSIS grhms
0.5
I
I
mean of 3 determinations in each of 6 patients
I *
0.4 -
Ht =
*
28% t.otal nroteinr
.
HF
0.3 FIG.,5. Amount of removal of p2-M in only one session of HF and HEPFD4 (mean SD).
55 1
=
7 g/dl
vs HEPFDB P
Short Time Treatment with High-Efficiency Paired Filtration Dialysis for Chronic Renal Failure Roberto Marangoni, Renato Savino, Roberto Colombo, Roberto Cimino, and Fernando Civardi Nephrology and Dialysis Unir, Bollate Hospital, Milan, Italy
Abstract: With the aim of treating with short sessions even chronic uremic patients with cardiovascular instability, we have developed our previous findings about paired filtration dialysis (hemofiltration-hemodialysis in series) by a method we call high-efficiency paired filtration dialysis (HEPFD). We have treated with this method for 6 months 6 chronic anuric uremic patients previously treated for 6 months with hemofiltration (HF) because of their cardiovascular instability. By comparison with HF,
HEPFD shows much higher clearances of small molecules with consequent significant decrease of their predialytic values, improvement of KrIV value, lower P2-microglobulin clearance (its predialytic value remains unchanged), and the same tolerability. HEPFD allows short, efficient, and well-tolerated treatments even in patients with cardiovascular instability. Key Words: Hemofiltration-Paired filtration dialysis-Cardiovascular instability-Short treatments-Uremic toxins.
The current dialysis philosophy requires short, efficient, and well-tolerated treatments. Even traditional hemodialysis (HD) has shown a development toward this route employing highly permeable membrane dialyzers with a large surface, and bicarbonate as buffer. Nevertheless, we know that the majority of patients cannot undergo short treatments (that is, treatments with a duration of about 3 h) mainly because they do not tolerate high body weight loss per unit of time. In fact, it is generally accepted that short hemodialysis can be adopted only in the treatment of patients with low dry body weight, low interdialytic weight gain, and stable cardiovascular condition (1-3). The method that takes advantage of convection, hemofiltration (HF), gives a high tolerability, especially from the cardiovascular point of view, allowing high body weight loss per unit of time even in patients with cardiovascular instability (4-6). Nevertheless, mainly in order to remove low molecular weight uremic toxins, H F needs a large amount
of infusion fluid, which causes long session times especially in patients who for their critical cardiovascular condition do not tolerate treatments with a high blood flow (7-10). The method that takes advantage of convection and diffusion simultaneously, hemodiafiltration (HDF), gives a satisfactory tolerability and high clearance of small and medium-sized molecules (large-molecule clearances are, however, lower than those of HF), and therefore it enables us to increase the number of patients undergoing short treatments (1 1). However, in order to achieve a high efficiency per unit of time, HDF needs high blood flows that cannot be allowed in patients with cardiovascular instability or in patients with poor vascular access (12). Moreover, convection and diffusion taking place simultaneously can interfere with each other (13). With the aim of obtaining short, efficient, and welltolerated blood purification sessions, a method called paired filtration dialysis (PFD) that consists of HF-HD in series (Fig. 1) has been carried out (14-17). In this system, the convective section (HF) is used for removal of small and especially large molecules or for body weight loss while the diffusive section (HD) is used mostly for removal of small molecules. Transmembrane pressure (TMP) of the
Received September 1991; revised May 1992. Address correspondence and reprint requests to Dr. R. Marangoni at Bollate Hospital, via Piave 20, 20021 Bollate-MI, Italy. This work was presented in part at the VIIIth World Congress of the International Society for Artificial Organs, held August 19-23, 1991, Montreal, Canada.
54 7
R . MARANGON1 ET A L .
548
FIG. 1. Schematic representation of the paired filtration dailysis. See text.
HD section is kept about 0 mm Hg in order to avoid body weight loss in this section. Since the PFD original extracorporeal circuit did not allow, in our results, a satisfactory exploitation of the convective section (Table I ) , especially with respect to depurination because of the small surface of the hemofilter, we set out to improve the method, which we then called high-efficiency paired filtration dialysis (HEPFD). The changes consisted in increasing the hemofilter surface and the substitution fluid amount, which we variously modified together with the ultrafiltration rate (UFR), blood flow, and session time (Table 2), obtaining in this way different clearances of small and large molecules (18). Among the tested modifications, which we called HEPFD4, seemed to us to offer the best advantages: the highest exploitation of convection with satisfac-
TABLE 1. PFD original method Hernofilter": Hernodialyzer": Whole infusion fluid (NaCI 9%): UFR: Session time: I'
Polysulfone 0.4 rn? Hernophan 1.0-1.3 m? 4-5.000 rnl 41 -46 mlirnin 180 rnin
Kit SG3 by Sorin (Italy).
tory clearances of large (Pz-microglobulin: B2-M) and small molecules without cutting down the clearances of the last ones in the diffusive section, satisfactory whole efficiency even with blood flows lower than the standard ones, and well-equilibrated shortening of the session times to allow a tolerable body weight loss in the majority of the patients. On the basis of these findings, we wanted to compare HEPFD4 with HF for efficiency and tolerability, aiming at verifying the possibility of using short sessions in patients with such serious cardiovascular instability as to be obliged to leave HD. MATERIALS AND METHODS
Six chronic anuric uremic patients (Table 3) had to leave bicarbonate HD treatment because of cardiovascular instability. At least 30% of their sessions involved symptomatic hypotension, and fall of maximum arterial pressure under 90 mm Hg was accompanied by one or more of the following symptoms: asthenia, nausea, vomiting, vertigo, or cramps even with body weight loss of 700 t50 glh. These patients have been treated for 6 months with HF (Table 4) (during this treatment no patient showed cardiovascular instability) and successively for 6 months with HEPFD4 (Fig. 2; Table 5 ) .
TABLE 2. Modifiixtions to PFD Session time (rnin) I80 180 210 210
Blood flow 200 3 00 250 250
Infusion rate (mlhin)
so 2 5 61 2 7 43 3 64 2 6
*
Ultrafiltration rate 61 73 55 74
Whole infusion (ml)
2 6
9.000
7 3 7
1 1.000 9.000 13.000
2 2 2
Method denomination HEPFD HEPFD HEPFD HEPFD
I 2 3 4
HIGH-EFFICIENC Y PAIRED FILTRATION DIALYSIS TABLE 3. Population baseline data Patient
1
2
3
4
5
6
Age (years) Sex Diagnosis Dry body weight (kg) Fluid overl. (kg) (mean)
28 m GN
57 f GN
66 CPN
74 f DN
61 f DN
58 m PKD
69 2.3
58 2.5
61 2.4
72 2.7
75 2.9
78 3.0
m
G N , glomerulonephritis; PKD, polycystic kidney disease; DN, diabetic nephropathy; C P N , cystopyelonephritis.
TABLE 4. Technical data of hernofiltration ( H F ) Monitors Hemofil ter U FR Session duration Mean body weight loss Substitution fluid (mmoliL) NaK' Ca? Mg! *
c1
Lactate Reinfubion fluid (postdilutional method) Blood flow
Sifra S I F 907; Gambro H F Polysulfone 2.0 m? 110 2 10 mlimin 250 2 13 min 700 +- 50 mlih
TABLE 5. Technical dutu oj'high-efjkiency paired jiltration dialysis (HEPFD4) ( H F + H D in series) H F monitor: PFD Sorin Hernofilter": PS 0.7 m'
26 L 250 mlimin
In these patients we have determined urea, creatinine, uric acid, and P2-M clearances during HF and HEPFD4 treatment; removal amount of urea, creatinine, uric acid, and P2-M in one HF and in one HEPFD4 session; mean arterial pressure (MAP) and heart rate (HR) modifications during HF and
H D monitor: Fresenius 2008 Dialyzer": Hemophan 1.3 m?
Formulae (mmoliL) Diulysute Na+ 1 K' 2.25 Ca' 0.75 Mg" 112 CI 35 HCOICH3COO-
Suhstitlrtion Pirid Na+ 140
K' Ca?+ Mg? CI Lac late +
Reinfusion fluid: 13 L Dialysate Row: 500 rnlimin Blood flow: 250 mlirnin Session time: 210 min UFR: 74 2 7 mlimin Mean body weight loss: 910 "
140 (138) I (21 2.25 (2) 0.75 112 (110) 35
549
+
~
138
2 I .75 0.50 109.50 31 4
* 60 mlih
Kit SG6 by Sovin (Italy).
HEPFD4 treatment; urea, creatinine, uric acid, and P2-M predialytic serum levels after 6 months of HF and after 6 months of HEPFD4 treatment; and KtlV after 6 months of HF and after 6 months of HEPFD4 treatment. The statistics were calculated using the Student's r test; p values < 0.05 were accepted as significant. RESULTS The results obtained show that small-molecule clearances (urea, creatinine, uric acid) are significantly higher in HEPFD than in HF (Fig. 3); the diffusive section of HEPFD contributes more than the convective section to the whole system clear-
FIG. 2. Working HEPFD (HF-HD in series).
Artif'Orgnns, V o l . 16. N o . 6 , 1992
550
R . MARANGONI ET AL.
250
ml/nin
Ht = 28% total proteins = 7’ g/dl 200
150
FIG. 3. Urea, creatinine, uric acid, and 02M clearances determined after 60 min in HF and in HEPFD4 (mean SD).
*
i 00
50
0
creatinine
urea
HF HEFFD4 * HEPFD4 vs HF p< 0.001 ** HF YS HEPFD4 p< 0.001
ances. P2-M clearance, which is due to convection only, is obviously significantly higher in H F than in HEPFD (Fig. 3). Consequently, small-molecule removal is significantly higher in HEPFD (Fig. 4) while large-molecule removal is higher in H F (Fig. 5 ) . HEPFD tolerability, especially from the cardiovascular point of view, is at least the same as of H F (Fig. 6). Predialytic serum levels (determined before the beginning of the first weekly treatment) of urea,
E’7- K
uric acid
nHEPFD4 (diffusive)
(whole)
creatinine, and uric acid show a significant decrease after 6 months of HEPFD treatment (Table 6). Even KtlV values show a significant improvement after 6 months of HEPFD treatment (Table 6 ) . DISCUSSION AND CONCLUSION According to our experiences, HEPFD has severa1 advantages. It can give high small-molecule re-
grams 50
-r *
mean of 3 determination in each of 8 patients
Ht
=
28% tc&d proteins
=
7
g/dl
20
*
FIG. 4. Amount of removal of BUN, creatinine, and uric acid in only one session of HF and HEPFD4 (mean t SD).
HEPFDI vs HF p c 0.001
10
n
BUN
Artf Organs, V d . 16, N o . 6 , 1992
=
creatinine
HF
HEFFD4
uric acid
HIGH-EFFICIENC Y PAIRED FILTRATION DIALYSIS grhms
0.5
I
I
mean of 3 determinations in each of 6 patients
I *
0.4 -
Ht =
*
28% t.otal nroteinr
.
HF
0.3 FIG.,5. Amount of removal of p2-M in only one session of HF and HEPFD4 (mean SD).
55 1
=
7 g/dl
vs HEPFDB P
