The machine and the therapy concept.

Maintenance haemodialysis became established in mainstream clinical practice in the 1960s. For pragmatic reasons, diffusive dialysis was the technique which underpinned its success. Over the next 15 years it was shown that short- and medium-term survival depended only on a critical level of urea clearance being achieved. Uncomplicated technology with negligible capacity for middle molecule removal could deliver this and the case for developing more sophisticated machines able to broaden the spectrum of solute removal was unconvincing. Dialysis-related amyloidosis which was recognised in the mid-1980s as a devastating complication in long survivors disturbed this complacency. The journey to develop machines which could deliver broad-spectrum solute removal while exposing patients only to ultrapure fluids and biocompatible materials is described elsewhere in this text. The Lister Renal Unit was established in 1988. A fruitful collaboration between the multidisciplinary clinical team and engineering colleagues in the R&D Department of Fresenius contributed to a steady and in-depth understanding of the effect of superimposing convection on diffusive dialysis. From the outset only high-flux dialysis using ultrapure fluids was employed. Haemodiafiltration (HDF) was introduced in 1993. This paper summarises our observations regarding the relative contributions of natural renal function and convective blood purification to long-term outcomes. We have recently reported a 19-year experience which has allowed us to more clearly define the rationale for HDF in modern clinical practice. HDF is an engineering triumph which is likely to universally supersede diffusive dialysis. The challenge for clinicians moving forward is to learn in which treatment schedules this technology can best be deployed to improve the health prospects of patients with kidney failure.

The effect of exercise during haemodialysis on solute removal.

BACKGROUND: Urea rebound results as urea re-equilibrates between intracellular and intravascular compartments post haemodialysis. The mechanism of the rebound is thought to be due to either a reduced diffusion rate or blood flow. It is hypothesized that low blood flow in the skeletal muscles might be responsible. We tested this by studying the effect of exercise during dialysis on the removal of urea, creatinine and potassium. METHODS: Eleven patients (aged 32-78 years) on haemodialysis (4-58 months) were studied on paired dialysis sessions; one with exercise and the other as a control. Patients pedalled on a cycle for 5-20 min at submaximal workload followed by 10 min rest to achieve a total of 60 min exercise. Plasma concentrations of urea, creatinine and potassium were measured pre-, post- and 30-min post dialysis. The post-dialysis rebound (% rebound) and reduction ratios (RR) of the solutes and equilibrated (two-pool) urea Kt/V were calculated for comparison. RESULTS: The rebound of all three solutes was reduced significantly following exercise. The rebound of urea decreased from 12.4 to 10.9% (median, P<0.01 Wilcoxon signed rank test), creatinine from 21.2 to 17.2% (P<0.001) and potassium from 62 to 44% (P<0.05). Kt/V and RR increased significantly as a result: Kt/V urea from 1.00 to 1.15 (P=0.001), RR urea from 0.63 to 0.68 (P<0.001); Kt/V creatinine from 0.71 to 0.84 (P<0.01); and RR creatinine from 0.51 to 0.57 (P<0.05). CONCLUSION: Exercise increased the efficiency of dialysis by reducing the rebound of solutes due to increased perfusion of the skeletal muscles.

Fluid balance modelling in patients with kidney failure.

In patients with kidney failure, adequate control of fluid status remains one of the most difficult routine issues to be addressed in the modern style of dialysis. This is primarily due to the lack of quantitative methods for the assessment of fluid status and the reliance on subjective criteria. Fluid is removed from the blood during dialysis treatments using a process called ultrafiltration. The last decade has seen considerable developments in blood volume monitoring (BVM) technology which has enabled responses to ultrafiltration to be continually monitored on an individual basis. This has enabled feedback control of patients' blood volume to be applied with partial success, reducing the number of symptoms. The feedback control algorithms employed have been relatively unsophisticated, using simple proportional control with no attempt to include models of the patient fluid dynamics. This paper describes the development of some prototype fluid kinetic models which may be used in a more advanced control system. Initial results demonstrate the importance of active control processes in the patients' physiological compensatory mechanisms.

Is there a rationale for rationing chronic dialysis? A hospital based cohort study of factors affecting survival and morbidity.

OBJECTIVES: To determine factors influencing survival and need for hospitalisation in patients needing dialysis, and to define the potential basis for rationing access to renal replacement therapy. DESIGN: Hospital based cohort study of all patients starting dialysis over a 4 year recruitment period (follow up 15-63 months). Groups were defined on the basis of age, comorbidity, functional status, and whether dialysis initiation was planned or unplanned. SETTING: Renal unit in a district general hospital, which acts as the main renal referral centre for four other such hospitals and serves a population of about 1.15 million people. SUBJECTS: 292 patients, mean age 61.3 years (18-92 years, SD 15.8), of whom 193 (66%) were male, and 59 (20%) were patients with diabetes. Dialysis initiation was planned in 163 (56%) patients and unplanned in 129 (44%). MAIN OUTCOME MEASURES: Overall survival, 1 year survival, and hospitalisation rate. RESULTS: Factors affecting survival in the Cox's proportional hazard model were Karnofsky performance score at presentation (hazard ratio 0.979, 95% confidence interval 0.972 to 0. 986), comorbidity severity score (1.240, 1.131 to 1.340), age (1.036, 1.018 to 1.054), and myeloma (2.15, 1.140 to 4.042). The Karnofsky performance score used 3 months before presentation was significant (0.970, 0.956 to 0.981), as was unplanned presentation in this model (1.796, 1.233 to 2.617). Using these factors, a high risk group of 26 patients was defined, with 19.2% 1 year survival. Denying dialysis to this group would save 3.2% of the total cost of the chronic programme but would sacrifice five long term survivors. Less rigorous definition of the high risk group would save more money but lose more long term survivors. CONCLUSIONS: Severity of comorbid conditions and functional capacity are more important than age in predicting survival and morbidity of patients on dialysis. Late referral for dialysis affects survival adversely. Denial of dialysis to patients in an extremely high risk group, defined by a new stratification based on logistic regression, would be of debatable benefit.

Low serum vitamin B12 levels in chronic high-flux haemodialysis patients.

The occurrence of vitamin B12 (B12) deficiency in chronic haemodialysis patients and the need for its supplementation in these patients are still matters of debate. We measured serial predialysis serum B12 levels, at 3- to 6-month intervals, in 67 unselected patients on our high-flux haemodialysis programme. Over a 12-month period, there was a significant fall in serum B12 from 497 +/- 200 (SD) to 391 +/- 131 ng/l (p < 0.001). 22 patients developed subnormal serum B12 levels and were commenced on hydroxocobalamin supplements. We were unable to demonstrate B12 clearance during dialysis using blood side studies. Measurement of B12 in the dialysate showed that 0-4.5 microg B12 was cleared per dialysis. Using these B12 measurements, in vivo B12 clearance was estimated at 9.1 ml/min. Dietary studies on 24 unselected patients showed borderline or low B12 intake in 4 patients. Absorption studies by whole-body counting on 6 patients using 57Co and 58Co showed normal B12 absorption. The same radioisotope studies demonstrated no B12 adsorption to the dialyser membrane. This study demonstrates that low serum B12 levels occur in high-flux haemodialysis patients and that losses during dialysis and dietary deficiency may be contributing factors.

The post-hemodialysis rebound: predicting and quantifying its effect on Kt/V.

Immediately after hemodialysis, the urea concentration rebounds upwards as urea continues to be transferred into the arterial circulation from peripheral body compartments. This rebound takes at least 30 minutes to complete. Hemodialysis is quantified as the Kt/V, calculated prom pre- and post-dialysis urea samples. Unless the post-dialysis sample is taken at least 30 minutes after dialysis, the Kt/V will be overestimated. This overestimation will be relatively greater in short high-efficiency dialyses, which have greater post-dialysis rebounds. We propose a method of correction that uses only the conventional pre- and immediate post-dialysis samples and is based on the physiologically-appropriate patient clearance time (tp). This is the time needed to clear all body compartments when the dialyzer clearance is infinite. The tp can be calculated from the pre-, immediate post- and 30-minute post-dialysis urea concentrations and was 35 minutes (SD 16) in 29 patients undergoing short (149 min) hemodiafiltration and standard (243 min) hemodialysis the following week. There was no significant difference between tp values calculated during the two treatments. Standard Kt/V can be corrected by multiplying by t/(t + tp) and dialysis time should be increased by tp x Kt/V minutes to compensate for the rebound. Despite individual variations in tp, a value of tp = 35 was sufficient to correct Kt/V in all patients. Kt/V corrected in this way agreed with Kt/V calculated using a 60-minute post-dialysis sample (r = 0.856, P < 0.001). The method predicted the 60-minute post-rebound concentration (SE 0.5 mM, r = 0.983, P < 0.001) and the addition of 35 minutes to the treatment time corrected for the rebound in both conventional and short treatments. Similar simple equations corrected the error in V caused by rebound effects.

Alkalosis and hypomagnesaemia: unwanted effects of a low-calcium CAPD solution.

We studied 43 CAPD patients for 4 months during the change from a high-calcium dialysis fluid (Baxter PD1) to a low-calcium fluid (Baxter PD4), which also contained low magnesium (0.25 mmol/l) and high lactate concentrations (40 mmol/l). Serum calcium fell significantly as did the incidence of hypercalcaemia, whilst the proportion of patients taking calcium-containing phosphate binders increased. There was a non-significant increase in serum i-PTH levels but the proportion with i-PTH > 150 pg/ml (normal range 10-65 pg/ml) increased significantly. There was a significant fall in serum magnesium level and seven patients developed hypomagnesaemia. Serum bicarbonate increased significantly and progressively and 17 patients were alkalotic at 4 months, five severely (bicarbonate 35-40 mmol/l). One patient developed recurrent episodes of painful subcutaneous and periarticular calcification, which may have been related to the alkalosis. Initial serum bicarbonate levels correlated significantly with dialysis adequacy assessed by daily Kt/V (r = 0.458, P = 0.002). The relationship to adequacy was abolished during the period of use of the high-lactate dialysis fluid. Use of low-magnesium CAPD fluids must be supported by regular monitoring of serum magnesium levels. The high lactate concentration in such fluids may not be appropriate and is potentially hazardous when individualization of dialysis dose demands the use of relatively high exchange volumes. Low serum bicarbonate levels in CAPD patients reflect inadequate dialysis, which use of these fluids serves to mask.

Maintaining adequacy in CAPD by individualizing the dialysis prescription.

Urea kinetic modelling (UKM) has been proposed as a tool for auditing the adequacy of CAPD and a total fractional daily urea cleared volume (Kt/V) of 0.25 suggested as the minimum adequate level. At the start of CAPD the kidneys contribute significantly to the total clearance and Kt/V often falls below 0.25 as renal function declines. We performed 3-monthly UKM measurements in 56 CAPD patients. These results were used to individualize exchange volume and frequency in an attempt to achieve a Kt/V > 0.25 and compensate for declining renal function in all patients over a study period of 1 year. The mean Kt/V was maintained over 0.29 over the study period. During this time the residual renal component of Kt/V fell significantly from 0.09 (SD +/- 0.07) to 0.06 +/- 0.08 (P < 0.001) while the dialysis component increased significantly from 0.20 +/- 0.05 to 0.24 +/- 0.05 (P < 0.005). This was achieved by increasing the mean daily exchange volume from 8.12 +/- 1.22 to 10.39 +/- 2.68 litres (P < 0.001). After a year, 15 patients had Kt/V < or = 0.25 despite maximum practical exchange volumes. Twelve patients dropped out of the study due to death (4), transplantation (2), and transfer to haemodialysis (6 patients, of whom 4 had frank uraemic toxicity). In most CAPD patients it is possible to compensate for declining renal function by increasing exchange volume, at least over 1 year. However, CAPD was unable to provide Kt/V > 0.25 in 40% of patients, despite individualization of the dialysis prescription.

Kinetic modelling and underdialysis in CAPD patients.

Kinetic analysis was performed in all 58 patients undergoing standard CAPD. The urea distribution volume was estimated from anthropomorphic measurements (Watson formulae). Normalized protein catabolic rate (NPCR), daily protein leak (PL), urea and creatinine Kt/Vs, clearances and peritoneal mass transfer coefficients (Kp) were calculated from measurements on serum, 24-h urine and PD fluid effluent. The mean total (renal+PD) daily creatinine and urea Kt/Vs (KT/V) were 0.31 (range 0.15-0.79) and 0.31 (0.18-0.65). There was no relationship between KT/V and serum urea or Kp. The strongest determinant of the urea KT/V was the residual renal urea clearance (KrU)(R = 0.79, P < 0.001) which decreased with time on dialysis (R = -0.38, P < 0.005). There was a significant correlation between the hospital admissions per year and both the urea and creatinine KT/V and KrU (R = -0.30, -0.32, P < 0.05). Patients with urea KT/V < 0.25 (n = 22) had more hospital admissions/year than those with KT/V > 0.25 (mean of 2.6 versus 1.5, P < 0.05). NPCR correlated with urea KT/V (R = 0.62, P < 0.001) but not with serum albumin or the PL. Patients identified by UKM to be less well dialysed have a lower residual renal function and are more likely to be hospitalized. Undernutrition in CAPD patients appears to be related to underdialysis rather than protein loss.

Haemodialysis recirculation detected by the three-sample method is an artefact.

The efficiency of haemodialysis may be limited by recirculation of blood between venous and arterial needles. Recirculation can be detected directly using a saline dilution method but is most commonly calculated from the urea concentrations of simultaneous samples from venous and arterial lines and a peripheral vein (three-sample method). The methods detect markedly different rates of recirculation in similar study populations. To investigate the possibility that the methods detect different phenomena, we performed both tests on 16 haemodialysis patients at various extracorporeal blood flow rates (Qb). The saline dilution method showed no recirculation in any of the patients, whereas the three-sample method indicated recirculation in all patients. The three-sample method indicated a mean recirculation fraction of 12.5% (SD 6.1) and was not influenced by changing Qb, suggesting that it was not detecting fistula recirculation. The three-sample method detects a solute concentration difference between arterial blood and peripheral blood during dialysis. There appears to be a disequilibrium between a central pool, represented by the arterial sample, and a poorly perfused peripheral pool, relatively isolated from the dialysis process, represented by the peripheral venous sample. The three-sample method for detecting recirculation should be abandoned.

Posture and central venous pressure measurement in circulatory volume depletion.

The effect of posture on central venous pressure (CVP) was studied in 16 patients with circulatory volume depletion before and after fluid replacement. At presentation, measurement of CVP when supine did not reflect circulatory volume depletion, with a mean (SEM) of 0.1 cm H2O (0.6), but when sat at 45 degrees the CVP showed a striking fall in all patients to -9.7 cm H2O (1.1). After fluid replacement, the CVP was 2.3 cm H2O (0.4) when supine, and -0.4 cm H2O (0.4) at 45 degrees. In the assessment of circulatory volume depletion, CVP should be measured with the patient sat at 45 degrees, if possible: measurement of CVP in a supine patient may not detect or severely underestimate circulatory volume depletion.

Renal biopsy in patients with unexplained renal impairment and normal kidney size.

We report renal biopsy findings in 109 patients with unexplained renal impairment (serum creatinine greater than 0.15 mmol/l) and normal-sized non-obstructed kidneys. The most common histological lesions were interstitial nephritis, rapidly progressive glomerulonephritis and a variety of other types of glomerulonephritis. The groups could not be distinguished by the presence or absence of hypertension, haematuria, proteinuria, or features of systemic disease. However interstitial nephritis was found more frequently in patients presenting with one or none of these features and rapidly progressive glomerulonephritis in patients presenting with three or more. All four patients with none of these features had interstitial lesions. Fifty-two per cent of patients with interstitial nephritis improved and 60 per cent of the patients with rapidly progressive glomerulonephritis who received immunosuppressive treatment improved or remained stable with treatment. The benefits of a biopsy diagnosis were almost wholly confined to these two groups. Complications were recorded in nine patients - prolonged macroscopic haematuria in six and symptomatic perirenal haematomata in three. Six required blood transfusion. One required nephrectomy to control haemorrhage and subsequently died. Percutaneous renal biopsy is not without risk in patients with renal impairment but the benefits of diagnosing interstitial nephritis and rapidly progressive glomerulonephritis outweigh the disadvantages.

Dialysis-induced change in erythrocyte volume: effect on change in blood volume calculated from packed cell volume.

Blood volume (BV) change during hemodialysis is often monitored by packed cell volume (PCV). This assumes erythrocyte volume is constant. We tested this by dialyzing 5 patients for 2 hours against high (154 mmol/l), normal (140 mmol/l) and low (126 mmol/l) dialysate sodium concentrations. Erythrocyte water content, calculated from measured blood and plasma water contents, decreased with high and increased with low dialysate sodium concentrations. Erythrocyte volume, calculated from mean corpuscular hemoglobin concentration (MCHC) decreased 3.8% with high concentration dialysate and increased 2.5% when dialysate concentration was low. These changes correlated significantly (r = 0.80, p less than 0.01) with alterations in plasma sodium. Mean corpuscular volume (MCV), measured with a Coulter-S Plus Counter did not alter because of a methodological artefact. BV change can be calculated from PCV when plasma concentrations of osmotically active substances are changed only if allowance is made for altered erythrocyte volume.

Blood volume change during isolated ultrafiltration and combined ultrafiltration-dialysis.

Despite extensive study of the difference in the cardiovascular response to ultrafiltration in isolation and during haemodialysis, there is still disagreement in the literature as to whether or not there is a difference in the efficiency of plasma refilling between the two procedures. We believe that this results both from the use of inadequate techniques to measure change in blood volume and shortcomings in experimental design. We have employed novel methods in ten patients to study plasma refilling during isolated ultrafiltration and dialytic ultrafiltration. Measurement of blood pressure and heart rate confirmed the inferior cardiovascular tolerance to haemodialysis observed by others. However, no difference in the rate of plasma refilling could be demonstrated. Our observations suggest that the superior tolerance to isolated ultrafiltration is not explained by differences in vascular refilling.

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.