ASAIO /ourna/ 1993
The Role of Sequential Ultrafiltration and Varying Dialysate
Sodium on Vascular Stability During Hemodialysis
CHRISTOPHER L. PO, /VtARY
AFOLABI, AND RASIB M. RAJA
Hypotension is a very common problem during HD. We
studied vascular stability during sequential UF with constant
and variable dialysate Na. Ten chronic patients underwent
HD using two protocols. Protocol A was sequential UF
(50% weight loss in the first hour, 30% in the second, and
20% in the third hour) with a dialysate Na of 140 mEq/L.
Protocol B was sequential UF as in protocol A with varying
dialysate Na (150 mEq/L in the first hr, 142 in the second,
and 137 in the third). BP and pulse were monitored every
30 min. Hct, BUN, creatinine, osmolality, and serum
albumin were checked every hour. The plasma volume
decreased by 5.3% in protocol A as compared with 1.2%
in protocol B, as shown by the Hct values. The decrease in
serum osmolality during the first and second hours was
more marked in pro- tocol A than in protocol B (p <0.05),
but post-dialysis values were similar. The clinical
parameters (UF, BP, pulse rate) and serum Na showed no
difference. The intradialytic man- nitol infusion was 10
ml/HD and 0, whereas saline was 80 ml/HD and 10 for A
and B, respectively. These data sug- gest: (1) Sequential
UF with varying dialysate Na could ben- efit patients who
are hemodynamically unstable. (2) The need for mannitol
and saline may be more readily alleviated with protocol B
than with protocol A. (3) The beneficial ef- fect of varying
dialysate Na with sequential UF may be due to improved
plasma refilling and decreased early intradia- lytic osmolar
changes. ASAIO journal 1993: 39: M798— M800.
H ypotension is a very common problem during
hemodialy- sis. Modifying dialysate composition is
effective in decreas- ing the incidence, but it does not
prevent it. Lowering dialy- sate temperature, isolated and
sequential ultrafiltration, use
of varying dialysate sodium, changes in dialyzers, and
the use of volumetric dialysis machines may be
beneficial. 3-4 We studied the role of constant and
varying dialysate so-
dium on vascular stability with sequential ultrafiltration in
hemodialysis.
From Albert Einstein Medical Center, Kraftsow Division of
Ne- phrology, Philadelphia, Pennsylvania.
Reprint requests: Rasib M. Raja, MD, Head, Kraftsow Division
of Nephrology, Albert Einstein Medical Center, 5501 Old York
Road, Philadelphia, PA 19141.
M798
Materials and Methods
Ten chronic hemodialysis patients underwent treatments
with two dialysis protocols. Protocol A consisted of
sequen- tial ultrafiltration (50% weight loss in the first hour,
30% iFl the second, and 20% in the third hour) with a
dialysate so- dium of 140 mEq/L. Protocol B consisted of
sequential ultra- filtration as in protocol A with varying
dialysate sodium lev- els (150 mEq/L in the first hour,
142 mEq/L in the second hour, and 137 mEq/L in the third
hour). The patients had either an arteriovenous (AV) graft
or fistula for access, the dialysis machine was a COBE
Centry III (Lakewood, CO), and a Cuprophane dialyzer
(CF 15:11, KUF: 4.1 ml/mmHg/ hr, surface area: 1.1 m2;
Baxter Healthcare Corp., Deerfield, IL) was used in all
dialysis treatments. The dialysate tempera- ture was set at
37° C, with a dialysate flow of 500 ml/min; mean blood
flow was 270 ml/min. Blood pressure and pulse were
monitored every 30 min. Hematocrit, BUN, creatinine,
serum osmolality, sodium, and albumin were checked
be- fore dialysis and every hour thereafter. Hypotensive
epi- sodes (systolic BP <90 mmHg) were treated with
infusion of mannitol or saline.
Results
The mean age of the patients was 62 years (range, 48—
68 years). There were six women and four men. All
patients were receiving dialysis for >6 months. The
etiology of renal failure was diabetes mellitus in three
patients, nephrosclero- sis in six, and multiple myeloma in
one patient. All patients were dialyzed with an AV graft.
The mean weight loss was 1.8 kg (range, 1.3—2.2 kg).
There were no significant changes in blood pressure
or pulse rate (pre-dialysis, first, second, and third hour)
in pro-
tocols A and B (Table 1 and Table 2).
Table 3 shows the changes in serum osmolality. The
osmo- lality was significantly higher in protocol A than in
protocol B during the first and second hour, but not in
the third hour. The serum sodium was higher in protocol
A than in proto- col B during the first and second hour of
dialysis but did not
reach statistical significance (Table 4).
Changes in hematocrit showed no statistical significance
(Table 5). The decrease in plasma volume as calculated
from the changes in hematocrit was greater during each
hour of dialysis.
The patients in protocol A required more mannitol and
saline than did those in protocol B (Table 6).
 VARYING UF AND DIALYSATE SODIUM M799

Table 1. Changes in Blood Pressure (mmHg)

ol fluid loss, autonomic neuropathy, left ventricular
A B p Value dysfunc- tion, changes in plasma osmolality, and decrease
in vascular refilling 5t6
Pre dialysis 157/82 160/81 >0.05
1st hr 146/80 151/81 >0.05 Our patients did not present significant changes in
2nd hr 159/84 153/81 >0.05 heart rate. This may be due to the autonomic impairment
3rd hr 156/82 144/81 >0.05 that occurs in a substantial proportion of hemodialysis
patients. However, Assali and colleagues have shown that
changes in cardiac rate are of minor importance during
Table 2. Changes in Pulse (beats/min) hypovolemia.’ There were more episodes of hypotension
A B p Value requiring man- nitol and saline in protocol A than there
were in protocol B, although there were no significant
Pre dialysis 77 85 >0.05 changes in blood pres- sure between the two groups.
1st hr 88 85 >0.05
2nd hr 84 85 >0.05 This could be due to the fact that the final blood pressure
3rd hr 85 82 >0.05 recording does not take into
account the readings during hypotensive episodes.
During a 3 hr hemodialysis, ultrafiltration without atten-
Table 3. Changes in Serum Osmolality (mosm/L) dant hypotension is a tribute to the effectiveness of
A B p Value
normal compensatory hemodynamic mechanisms,
especially to the effectiveness of plasma refilling.’A
Pre dialysis 310 310 >0.05 decrease in plasma osmo- lality is observed and is a
1st hr 304 311 <0.05 major factor in fluid removal be- tween interstitial and
2nd hr 296 305 <0.05
3rd hr 297 296 >0.05
vascular compartments. The rate of ultrafiltration and
dialysis solution sodium concentration are also important
co-determinants of the plasma refilling rate. 9 ' Our study
Table 4. Changes in Serum Sodium (mEq/L) showed that plasma refilling could be bet- ter achieved
with protocol B than with protocol A. This was shown by
A B p Value the changes in plasma osmolality and serum so- dium
Pre dialysis 137 136 >0.05 concentrations, and the decrease in plasma volume as
1st hr 139 143 >005 computed by the hematocrit values. The patients in
2nd hr 138 143 >005 protocol A required more mannitol and saline, especially
3rd hr 139 140 >0.05 during the second and third hours, which confirmed that
dialysis hypo- tension most often occurs during the later
Table S. Changes in Hematocrit (Hct) and Plasma Volume stages of hemodi- alysis, and that sequential
(PV) with Both Protocols (%) ultrafiltration with varying dialy- sate sodium levels would
be important in this regard.
A B

Hct PV Hct PV p Value Conclusion

Predialysis 28.1 — 28.0 — >0.05
1sthr 29.5 —5.0 27.2 +2.8 >0.05 Sequential ultrafiltration with varying dialysate sodium
2ndhr 29.7 —5.6 27.8 — >0.05 concentrations could benefit patients who are hemodynami-
3rdhr 29.3 —5.3 28.5 —1.2 >0.05 cally unstable and may be preferable to sequential
ultrafiltra- tion with a constant dialysate sodium level.
Table 6. Mannitol and Saline Infusion (ml) During Dialysis
The need for mannitol and saline may be alleviated with
protocol B rather than with protocol A. The beneficial
effects of varying dialy- sate sodium from high to low
with sequential ultrafiltration may be due to improved
plasma refilling and decreased early intradialytic
osmolar changes.
Mannitol Saline Acknowledgment
A B A B
Supported by the Women’s League for Medical Research at Al- bert
Einstein Medical Center.
1st hr 50 0 0 0
2nd hr 50 0 700 100
3rd hr 0 0 100 0 References
Total 100 0 800 100
1. Daugirdas jT: Dialysis hypotension: a hemodynamic analysis.
kidney Int 39: 233-246, 1991.
2. Wehle B, Asaba H, Castenfors j, et al: Hemodynamic
Discussion changes during sequential ultrafiltration and dialysis. Kidney
Hypotension is a frequent complication of Int 15: 411—418, 1979.
3. Kirschbaum B: Hemodialysis with variable rate ultrafiltration
hemodialysis and represents a major problem in patients (abstract). The National Kidney Foundation Annual Meeting,
who are hemody- namically unstable. This has been Baltimore, Maryland, November 13, 199 , A7.
related to the magnitude
M800 PO, AFOLABI, AND RADA
4. Raja R, Kramer M, Barber K, Chen S: Sequential changes
in dialysate sodium during hemodialysis. ASAIO 29: 649-
651, 988.
S. Fleming Sj, Wilkinson jS, Aldridge C, et al: Blood volume
change during isolated ultrafiltration and combined
ultrafil- tration-dialysis. Nephrol Dial Transplant 3: 272-
276, 1988.
6. Daul AE, Wang XL, Michel MC, Brodde O-E: Arterial
hypoten- sion in chronic hemodialyzed patients. Kidney
lnl32: 728— 735, 1987
7. Assali NS: The contribution of the beta-adrenergic system
to
the cardiovascular response to hypovolemia. Am j
Obstet Cynecol 147: 423—429, 1983.
8. Rodriguez M, Pederson JA, Llach F: Effect of dialysis and
ultrafil- tration on osmolality, colloid osmotic pressure and
vascular refilling rate. Kidney Int 28: 808-813, 1985.
9. Stiller S, Thommes A, Konigs F, Schallenberg U, Mann H:
Char- acteristic profiles of circulating blood volume during
dialysis therapy. Irans Am Soc Artif Intern Organs 35: 530-
532, 1989.
10.Kooman QP, GladZiwa U, Bocker G, et al: Role of the
venous system in hemodynamics during ultrafiltration and
bicarbon- ate dialysis. Kidney Int 42: 718—726, 1992.