Prediction of Postdialysis Serum Sodium Concentration and Transcellular Fluid Shift Without Measuring Body Fluid Volumes

Abstract: The postdialysis levels of serum sodium concentration, urea concentration, and osmolality, as well as the magnitude of both transcellular fluid shifts and sodium removal, were predicted based on computer modeling without measuring body fluid volumes. A 4-h hemodialysis was performed in five patients at a constant ultrafiltration rate of 0.5 L/h using dialysate with normal (141 mEq/L) or high (150 mEq/L) Na+ concentration. The serum sodium concentration, urea concentration, and osmolality, as well as intracellular and extracellular fluid volumes, were determined before and after hemodialysis. The model predictions without measurement of body fluid volumes were in excellent agreement with measured values, suggesting clinical validity. The model may be useful in clinical practice to control the postdialysis levels of sodium and water content by computerized hemodialysis.     

Theoretical Aspects of Various Ultrafiltration Methods in Artificial Kidney Therapy

A mathematical model including urea, creatinine and other osmotically important solutes (such as sodium, potassium and chloride) is applied to calculate volume shifts, caused by ultrafiltration, between the fluid compartments of the body. The volume shifts between the intracellular (ICV) and the extracellular (ECV) compartments are mainly caused by alteration of extracellular sodium concentration.
Various methods of achieving ultrafiltration, including conventional dialysis, initial ultrafiltration using Cuprophan (without dialysis) or hemofiltration, produce different responses. In choosing a method, one must consider that both a rapid decrease of ECV and a fast shift of water from ICV to ECV should be avoided.
In pure hemofiltration, ultrafiltrate is isotonic and water is removed from ECV only. Hemofiltration with dilution produces a very slow shift of water between ICV and ECV dependent on sodium concentration of plasma and diluting fluid. In initial ultrafiltration through Cuprophan, water is shifted from ICV to ECV. With ultrafiltration throughout the entire dialysis, there are pronounced shifts between ICV and ECV dependent on the difference of the sodium concentration between plasma and dialysate.     

Influence of high and low sodium dialysis on blood volume preservation.

Haemodialysis has a profound effect on fluid balance. Since fluid is initially withdrawn from the intravascular compartment, hypovolaemia is a frequent complication. A fluid shift from the overhydrated interstitium towards the intravascular compartment can counteract hypovolaemia. However, a fast decline in extracellular osmolality may cause an increase in the intracellular volume, reducing the available amount of fluid to compensate for the hypovolaemia. To overcome this problem, the use of alternating high and low sodium dialysate is advocated. In this study six patients were studied during standard haemodialysis (HD) and during dialysis with alternating high and low sodium dialysate (HLSD). Changes in intracellular fluid volume (IFV) and extracellular fluid volume (EFV) of tissue and blood were measured by means of a non-invasive electrical conductivity method. Changes in blood volume (BV) were studied by serial erythrocyte counts. Plasma sodium concentration was determined at regular intervals. The distribution volume of sodium during the high and low sodium episodes of HLSD was calculated according to a mathematical model. HLSD led to fluctations in plasma sodium concentration that induced changes in red cell volume, but not in IFV. Distribution of sodium was largely confined to blood. BV was better preserved during HLSD than during HD, probably due to a higher mean plasma sodium concentration. Postdialysis sodium concentration however, was not significantly different between HLSD and HD. These data suggest that the better BV preservation during HLSD results from an induced osmotic gradient across the capillary wall, rather than from an osmotic gradient across the cell membrane.     

Fluid balance during haemodialysis and haemofiltration: the effect of dialysate sodium and a variable ultrafiltration rate

One of the main causes of hypotension during extracorporeal renal replacement therapy is an insufficient substitution of the ultrafiltrated plasma water by tissue water. To investigate the fluid balance and its effects on hypotension in dialysed patients, the following variables were studied: intracellular fluid volume (IFV) and extracellular fluid volume (EFV), blood volume (BV) and blood pressure. IFV and EFV were measured by means of non-invasive electrical conductivity measurements using four electrodes round the leg. Fifteen haemofiltration (HF) and 15 haemodialysis (HD) patients were studied. The latter group was dialysed in three ways: (1) conventionally, i.e. with dialysate sodium of 138 mmol/l (HD) (2) with a variable dialysate sodium (first half: 138 mmol/l; second half: 146 mmol/l) (HDS), and (3) with the same variable dialysate sodium and an ultrafiltration profile (two-thirds was withdrawn during the first half of treatment, the remainder during the second half) (HDSU). Hypotension frequency was less during HDS, HDSU, and HF compared to HD. This was caused by a more stable blood volume due to a better refill. During HD a fluid shift occurred from the EC to the IC compartment. The use of a high sodium dialysate concentration led to a transcellular fluid shift in the opposite direction. This fluid shift increased the refill, thereby stabilising blood volume. HF gave a better refill than HDS and HDSU, probably due to a reduced urea clearance.     

The amount of sodium removed by hemodialysis

The amount of sodium removed by hemodialysis was estimated, without using radioisotopes, as the change in total osmotically active cations, which is the product of the serum sodium concentration and urea-space. The extracellular and total body fluid volumes were measured using 35SO4 and 3H2O, respectively, in five stable hemodialysis patients under four different conditions. Urea-space determined, based on urea kinetics, was consistent with total body fluid volume measured by 3H2O. The amount of sodium removal, estimated as the change in the product of the serum (Na+) and urea-space, was equal to the change in the sodium content, which is the product of the serum (Na+) and extracellular fluid volume measured by 35SO4. Sodium removal may be divided into two components, diffusion and ultrafiltration.     

Effect of osmolar changes on plasma arginine vasopressin (PAVP) in dialysis patients.

The effect of changes in plasma osmolality and changes in plasma arginine vasopressin (PAVP) were analyzed in 10 stable chronic hemodialysis patients utilizing four protocols. During regular hemodialysis opposing influences on PAVP (decrease in blood pressure and intravascular volume and increase in serum calcium) resulted in no significant change in PAVP (by analysis of variance). In the second protocol low dialysate calcium (2.5 meq/l) isovolemic hemodialysis was used. PAVP and serum osmolality levels declined from 2.0 +/- 0.4 to 1.4 +/- 0.2 microU/ml (p less than 0.05), and 285 +/- 2.5 mOsm/l to 275 +/- 3.2 mOsm/l respectively. Removal of PAVP by hemodialysis did not occur as evidenced by no difference in arterial-venous PAVP levels and no "rebound" of PAVP for three hours after completion of dialysis (second protocol). Isovolemic low calcium high dialysate sodium (145 meq/l) hemodialysis was utilized in the third protocol. Serum osmolality and PAVP did not change. Addition of a very high dialysate sodium (155 meq/l) to isovolemic low calcium hemodialysis resulted in an increase in plasma sodium, osmolality and AVP (139.7 +/- 0.62 to 144 +/- 0.67 meq/l, 294 +/- 2.79 to 304.3 +/- 2.4 mOsm/l and 1.8 +/- 0.3 to 2.7 +/- 0.5 microU/ml (p less than 0.05 for each) respectively. In conclusion, PAVP responds to changes in plasma osmolality in chronic hemodialysis patients.     

Effect of dialysate composition on intercompartmental fluid shift

Effect of dialysate composition on intercompartmental fluid shift and hemodynamics was studied in 12 patients during 1.5 or 2 hours of hemodialysis without net ultrafiltration, using high (H;Na 154 mmol/liter), normal (N;Na 140 mmol/liter) or low (L:Na 126 mmol/liter) concentration dialysate. H dialysate was associated with a small (0.9%) increase in blood volume, a larger increase in plasma volume and a decrease in erythrocyte volume. L dialysate resulted in a 2.3% decrease in blood volume, a larger decrease in plasma volume and an increase in erythrocyte volume. N dialysate gave results which were intermediately between the other two dialysis conditions. There was no difference in the post-dialysis mean arterial pressure between the groups, although heart rate increased more during H dialysis than during the other two conditions. Change in blood and erythrocyte volume correlated significantly with change in plasma Na concentration and osmolality, but not with change in plasma urea concentration. We conclude that dialysate composition affects the movement of water into and out of the plasma and erythrocytes in a manner that can be accounted for by altered plasma concentrations of osmotically active substances.     

Regulation of fluid and electrolyte balance

Adequate tissue perfusion and cellular function is dependent on the maintenance of effective circulatory volume and serum osmolality, respectively. As sodium is the principal extracellular cation with the inability to pass freely across the cellular membrane, it therefore has the greatest effect on extracellular fluid osmolality. The extracellular fluid (ECF) volume can increase or decrease independent of the surrounding osmolality, indicating that control of plasma osmolality and volume occur through distinct physiological processes. Disorders in sodium balance with consequent effect on osmolality come about mainly due to disturbances in water homeostasis rather than an abnormality of sodium intake or excretion.     

Regulation of fluid and electrolyte balance

Adequate tissue perfusion and cellular function is dependent on the maintenance of effective circulatory volume and serum osmolality, respectively. As sodium is the principal extracellular cation with the inability to pass freely across the cellular membrane, it therefore has the greatest effect on extracellular fluid osmolality. The extracellular fluid (ECF) volume can increase or decrease independent of the surrounding osmolality indicating that control of plasma osmolality and volume occur through distinct physiological processes. Disorders in sodium balance with consequent effect on osmolality come about mainly due to disturbances in water homeostasis rather than an abnormality of sodium intake or excretion.     

Simple model of intra-extracellular potassium kinetics and removal applied to constant and potassium-profiled dialysis

AIM: The incidence rate for sudden death in hemodialysis patients ranges between 2% and 7%. This phenomenon is frequently due to cardiac arrhythmias. In particular, the process of potassium (K(+)) depuration performed during hemodialysis has been found to be related to arrhythmia onset. The main aim of this study was to introduce a simple double-pool mathematical model of K(+) kinetics to investigate the effects of dialysate K(+) concentration on intracellular and extracellular K(+) removal. The secondary aim was to evaluate the K(+) removed from the different body pools in 2 different types of K(+) dialysate: constant and profiled. METHODS: Our model evaluated K(+) removal and body water in the intracellular and extracellular spaces using plasma, erythrocytes and spent dialysate K(+) concentration, and intracellular and extracellular volume (t=0) in 6 patients (4 females and 2 males). All patients were treated with acetate-free biofiltration with a constant K(+) dialysate concentration (AFB) and with a profiled one (AFB-K). Moreover, the electrolyte concentration (sodium, calcium and bicarbonate) and pH were analyzed in all sessions. RESULTS: A similar total potassium removal was evaluated by the model, starting from a similar final K(+) plasma reduction. At 10 minutes, the model assessed a higher K(+) removal in the extracellular space during AFB (26.6% vs. 7.7%, p<0.001) involving a lower K(+) concentration (5.0 +/- 0.5 in AFB and 5.2 +/- 0.6 in AFB-K, p<0.05) and consequently a higher cell hyperpolarization (-73.4 +/- 3.9 mV vs. -72.1 +/- 2.4 mV, p=0.05). No differences in pH, intracellular and extracellular Na+ or plasma Ca(2+) were highlighted between AFB and AFB-K. CONCLUSIONS: The model we developed allows us to evaluate K(+) removal and body water in the intracellular and extracellular spaces during treatment. The assessment of this information may have a relevant role toward an understanding of the causes of the Nernst potential changes during hemodialysis that are often related to the onset of arrhythmias.     

Hemodynamic monitoring during hemodialysis.

Intradialytic monitoring of hemodynamic parameters is an active area of research; future developments in this field will decrease intradialytic morbidity and the mortality of end-stage renal disease patients treated by hemodialysis. Recent investigations have been assisted by the development of devices that can continuously and noninvasively measure hematocrit and plasma protein concentration during the treatment. Intradialytic morbidity, fluid overload, and hypertension in chronic hemodialysis patients have been shown to be associated with either large or small intradialytic decreases in blood or plasma volume that can be routinely measured by these devices. The use of intradialytic changes in blood volume as a feedback control parameter to vary the ultrafiltration rate and dialysate sodium concentration, so called profiling, is now possible, but further research in this area is necessary to show how to optimize the control algorithms. Other, more preliminary studies suggest that monitoring of central blood volume, extracellular volume, and cardiac output during hemodialysis may permit improved hemodynamic stability during treatment and better control of blood pressure. Although optimal application of these techniques and devices remains to be shown, their routine use during maintenance hemodialysis therapy will likely be the standard of care in the near future.     

KT/V and protein catabolic rate determination from serial urea measurement in the dialysate effluent stream

A bloodless technique of evaluating protein catabolic rate (PCR) and KT/V (K, clearance; T, dialysis time; V, urea distribution volume) in hemodialysis patients is presented based on serial measurement of urea in the dialysate effluent stream. PCR follows from equating urea generation and urea removal over a 7 day cycle, changes in body stores being comparatively negligible: PCR = 0.026 [U1 + U2 + U3]/BWdry + 0.17, where U1 is the amount of urea in mmol appearing in the dialysate for each session in the 7 day period. KT/V is obtained from the slope of the natural logarithm of spent dialysate urea concentration-time plot: KT/V = [- slope.T + 3.delta BW/BWdry]/[1 - 0.01786.T(hr], where delta BW = amount ultrafiltered in liters. The dialysate-based approach was validated and compared with conventional urea kinetic modeling (UKM) for 17 patients studied for three consecutive dialyses. The dialysate-based and UKM values of PCR agreed well when in vivo clearance values based on total dialysate collection were used for UKM. KT/V values agreed poorly on a session-by-session basis but were nearly equivalent when averaged for the three dialyses of the week. These findings lay the foundation for UKM automation with a urea sensor in the effluent dialysate stream.     

Electrolyte and pH dependence of heart rate during hemodialysis: a computer model analysis

The influence of hemodialysis-induced modifications in extracellular fluid characteristics on heart rate was investigated by using a detailed computer model of sinus-node electrical activity. Changes similar to those occurring in the course of hemodialysis in extracellular concentrations of sodium (from 138 to 140 mM), potassium (from 6 to 3.3 mM), and calcium (from 1.2 to 1.5 mM) ions as well as in pH (from 7.31 to 7.4) and intracellular volume were simulated. The model predicted that such changes may largely influence the rhythm of the sinoatrial node pacemaker, causing the heart rate to range from 69 to 86 bpm. Heart rate increases after removing potassium (up to 7 bpm) and also after calcium perfusion (up to 11 bpm) whereas restoring pH slows heart beat (up to 6 bpm). Extracellular sodium has no significant influence, but the heart rate strictly depends on intracellular sodium concentration (5 bpm/mM). A complex dependence of heart rate on electrolytes and pH was also recognized. Providing extracellular potassium concentration is maintained above 5 mM, heart rate exhibits low sensitivity to changes in calcium and potassium. When potassium concentration is reduced below 4.5 mM, heart rate sensitivity to calcium and potassium increases significantly to 10 and 30 bpm/mM, respectively. A sustained increase in heart rate always corresponds to an increase in intracellular sodium concentration.     

Hypotension during acetate and bicarbonate dialysis in patients with acute renal failure

Bicarbonate dialysate is claimed to be superior to acetate for both chronic and acute hemodialysis. We compared acetate and bicarbonate dialysates in 30 acute renal failure patients during 120 dialyses. 4 patients were diabetic and 2 had liver failure. Patients were dialyzed alternating acetate and bicarbonate dialysate in a double-blind cross-over manner; each patient was his own control. BUN, creatinine, Na+, K+, osmolality, delta osmolality, % ultrafiltration, arterial blood gases, pre, post and lowest dialysis mean arterial blood pressure, dialysis with hypotensive episodes and symptoms of hypotension were recorded. The measurements obtained for each patient during dialyses with acetate and bicarbonate were compared. There was no difference in predialysis chemistries, osmolality or osmolality fall, no change in mean arterial blood pressure or hypotensive episodes and symptoms and ultrafiltration. PCO2 and pH were slightly lower for the acetate group at the 2nd h but not at the end of dialysis. 4 patients had serum acetate determinations, all metabolized acetate normally. These findings contradict recent suggestions that severely ill patients should not be dialyzed against acetate. Since acetate is technically much easier to use and has no clinical drawbacks, it does not need to be replaced with bicarbonate in acute patients. Other factors must be more important than acetate in generating hypotension during acute dialysis.     

The effect of plasma calcium on plasma ADH levels in anephric patients.

Ten anephric patients were studied before and during hemodialysis. The extracorporeal circuit was primed with 5% albumin in 0.9% sodium chloride. Ultrafiltration volume removed by the hemodialyzer was replaced continuously. Modifications of a standard chronic renal failure dialysate were used to minimize changes in plasma urea while varying plasma sodium and calcium in opposite directions. Plasma ionized calcium concentrations in two patients confirmed other studies demonstrating a correlation between plasma total calcium and ionized calcium under these conditions. Plasma ADH determined by bioassay did not correlate with plasma osmolality, plasma sodium concentration, plasma potassium concentration, blood pressure, or pulse rate. The change in plasma ADH during dialysis was significantly correlated only with the change in plasma calcium (r = 0.47, p less than 0.05). The data support the hypothesis that plasma calcium plays a role in the regulation of ADH release in man, independent of the renin-aldosterone system.     

Unraveling the relationship between macula densa cell volume and luminal solute concentration/osmolality

At the macula densa, flow-dependent changes in luminal composition lead to tubuloglomerular feedback and renin release. Apical entry of sodium chloride in both macula densa and cortical thick ascending limb (cTAL) cells occurs via furosemide-sensitive sodium-chloride-potassium cotransport. In macula densa, apical entry of sodium chloride leads to changes in cell volume, although there are conflicting data regarding the directional change in macula densa cell volume with increases in luminal sodium chloride concentration. To further assess volume changes in macula densa cells, cTAL-glomerular preparations were isolated and perfused from rabbits, and macula densa cells were loaded with fluorescent dyes calcein and 1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene p-toluenesulfonate. Cell volume was determined with wide-field and multiphoton fluorescence microscopy. Increases in luminal sodium chloride concentration from 0 to 80 mmol/l at constant osmolality led to cell swelling in macula densa and cTAL cells, an effect that was blocked by luminal application of furosemide. However, increases in luminal sodium chloride concentration from 0 to 80 mmol/l with concomitant increases in osmolality caused sustained decreases in macula densa cell volume but transient increases in cTAL cell volume. Increases in luminal osmolality with urea also resulted in macula densa cell shrinkage. These studies suggest that, under physiologically relevant conditions of concurrent increases in luminal sodium chloride concentration and osmolality, there is macula densa cell shrinkage, which may play a role in the macula densa cell signaling process.     

Amelioration of hemodialysis-associated hypotension by the use of cool dialysate

The effect of a reduction in dialysate temperature on BP during hemodialysis was studied in seven patients with end-stage renal disease suffering frequently from intradialytic hypotension. Each patient received six dialyses using 34.4 degrees C dialysate. These treatments were preceded (six dialyses) and also followed (six dialyses) by control periods using a 36.7 degrees C bath. Symptomatic hypotension was defined as systolic BP below 100 mm Hg associated with typical symptoms of hypotension requiring treatment with intravenous (IV) fluid. Cool dialysate reduced the frequency of symptomatic hypotension from 0.58 to 0.05 episodes per dialysis (P = less than 0.016). In addition, the rate of fall of mean BP during treatment was substantially slowed with the reduction in dialysate temperature (P = 0.002). Cool dialysate (34.4 degrees C) substantially ameliorates hemodialysis-associated hypotension.