Clinical Science (1992) 82, 65 1-658 (Printed in Great Britain)

65 I

Platelet cytosolic free calcium concentration and parathyroid hormone: changing relationships with haemodialysis in end-stage renal disease R. J. FLUCK*, A. C. McMAHON*, F. M. ALAMEDDINEt, A. B. S. DAWNAYt, L. R. I. BAKER* and A. E. G. RAINE* Renal Research laboratory, Departments of”Nephro1ogy and tChemical Pathology, The Royal Hospital of S t Bartholomew, London, U.K. (Received 25 September/ I8 December I99 I; accepted I 3 January 1992)

1. Twelve patients receiving haemodialysis for end-stage

renal failure were studied at a single dialysis session. Platelet cytosolic calcium concentration, plasma ionized calcium concentration and serum parathyroid hormone concentration were measured before dialysis, mid-dialysis and 30 min after dialysis. 2. Plasma ionized calcium concentration increased towards dialysate calcium concentrations, falling insignificantly after cessation of dialysis. Serum parathyroid hormone concentration fell by 39%during dialysis, with incomplete recovery afterwards. There was no overall change in platelet cytosolic calcium concentration. 3. Patients were divided into two subgroups: low parathyroid hormone (serum parathyroid hormone concentration < 10 pmol/l) and high parathyroid hormone (serum parathyroid hormone concentration > 10 pmol/l). Before dialysis, values of platelet cytosolic calcium concentration or plasma ionized calcium concentration were not statistically different between the subgroups, but the platelet cytosolic calcium concentration was higher in the highparathyroid hormone subgroup during and after dialysis. 4. Before haemodialysis there was a linear correlation between plasma ionized calcium concentration and platelet cytosolic calcium concentration, which disappeared during dialysis. In contrast, there was no relationship between serum parathyroid hormone concentration and platelet cytosolic calcium concentration before dialysis, but after dialysis a hyperbolic relationship was evident. 5. These results suggest that uraemic toxins may interfere with cytosolic calcium homoeostasis, allowing passive diffusion of extracellular calcium to influence the resting concentration, and that this effect is reversible by haemodialysis.

INTRODUCTION Secondary hyperparathyroidism is common in endstage renal disease [ 11and may be implicated in the patho-

genesis of many of the complications of uraemia [2, 31. One possible mechanism of parathyroid hormone (PTH) toxicity in uraemia may be through a disturbance of intracellular calcium homoeostasis. In other settings, it has been postulated that cytosolic calcium overload in target organs may represent a final common pathway mediating organ dysfunction [4]. For example, disordered intracellular free calcium handling has been detected in the myocytes of cardiomyopathic hearts [ 5 ] .Furthermore, as cytosolic calcium plays a central role in cell signalling [6], any changes in the resting cytosolic free calcium concentration ([Ca’+ I,) and agonist-stimulated responses of [Ca2+],which accompany development of uraemia might be predicted to affect cellular function. Earlier studies have demonstrated a cori elation between serum PTH concentration and platelet [Ca’+], in mild to moderate chronic renal failure (CFW) [7], suggesting that PTH might directly influence [Ca*+],.In addition, suppression of PTH by vitamin D analogues resulted in a fall in platelet [Ca’+],. However, it remains uncertain whether similar relationships exist in end-stage renal disease. Haemodialysis itself causes an acute suppression of PTH [S], in response to increases in the plasma ionized calcium concentration as it equilibrates with the dialysate calcium concentration. The fall in serum PTH concentration is rapid, quickly reversed and is dependent on the rate and direction of the change in the plasma ionized calcium concentration [9]. In previous studies [lo] no change was observed in erythrocyte [Ca2+], during haemodialysis, but cell CaATPase activity increased to levels found in a nonuraemic control population [ 113. Other studies [ 121 demonstrated a mid-dialysis rise in platelet [Ca’+],, which fell after dialysis, although remaining at higher levels than those observed before dialysis. However, the relationship, if any, of these changes to factors which might modulate [Ca’+],, such as serum PTH concentration and extracellular calcium concentration, was not examined. The aims of this study were therefore to compare the effects of haemodialysis on platelet [Ca2+],,plasma ionized calcium

Key words: cytosolic calcium, Fura-2, haemodialysis, ionized calcium, parathyroid hormone. Abbreviations: [Ca”],, cytosolic free calcium concentration; CRF, chronic renal failure; PTH. parathyroid hormone. Correspondence: D r R. J. Fluck, Department of Nephrology, The Royal Hospital of St Bartholornew, West Smithfield, London EClA 7BE, U.K.

R. J. Fluck et al.

652

concentration and serum PTH concentration, examining the relationships between these variables immediately before, during and after a haemodialysis treatment session in patients with end-stage renal failure.

MATERIALS AND METHODS Patients Twelve patients with end-stage renal failure on regular haemodialysis were studied at a single haemodialysis session. Patient details, including time on dialysis, length of haemodialysis session and medication, are given in Table 1. No patient was taking calcium-channel blockers, and the use of vitamin D analogues and phosphate binders is detailed in Table 1. One patient had tertiary hyperparathyroidism secondary to a forearm implant of parathyroid tissue after parathyroidectomy. All patients gave their informed consent. No modifications were made to individual patients' dialysis schedules. All patients used a bicarbonate-based dialysate (140 mmol/l Na+, 1.0 mmol/l K + , 1.5 mmol/l CaZ+,30 mmol/l HCO;, 105 mmol/l Cl-). All patients used cuprophane membranes. Arterialized blood was withdrawn via the fistula access at three time points: (a) immediately before dialysis, (b) mid-dialysis and (c) 30 min after dialysis. Blood pressure and heart rate were also measured at each time point, using an automated sphygmomanometer.

Analytical methods Serum PTH concentration. This was measured by using a specific two-site immunochemiluminometric assay for intact PTH (Magic-Lite; Ciba Corning, Halstead, Essex, U.K.) [13]. Blood samples were taken into plain tubes, allowed to clot for 30-60 min and centrifuged at 2000 g. Duplicate serum samples were immediately frozen and stored at -20°C until assayed. Each set of three samples from a patient were assayed in duplicate within the same batch. The normal range of serum PTH concentration is 0.9-5.4 pmol/l. Plasma ionized calcium concentration. Whole-blood samples (2 ml) were taken anaerobically into lithium heparin tubes and were stored at +4"C until assayed in duplicate on a Kone Microlyte calcium analyser. The normal range of plasma ionized calcium concentration is 1.22 k 0.03 mmol/l [9]. Values were adjusted for a pH of 7.4 by an internal correction. Other biochemical parameters. Serum urea, electrolyte, creatinine, calcium and phosphate concentrations and serum alkaline phosphatase activity were analysed by standard autoanalyser techniques. Platelet isolation. Whole blood (10 ml) was taken into tubes containing 2 ml of acid citrate dextrose anticoagulant (85 mmol/l sodium citrate, 78 mmol/l citric acid, 111 mmol/l glucose). After centrifugation at 2000 gfor 4 min at 2 4 T , platelet-rich plasma was aspirated into plain conical tubes and was divided into two equal portions. Fura 2 penta-acetoxymethyl ester (1 pl/ml of platelet-rich

Table I. Patient details. Abbreviations: CGN, chronic glomerulonephritis; CPN, chronic pyelonephritis: PCKD, polycystic kidney disease; SLE, systemic lupus erythematosus; DHT, dihydrotachysterol; EPO, erythropoietin: PTX. total parathyroidectomy; implant, parathyroid tissue implant; BPC, British PharmaceuticalCodex 1973. Age (years)

Sex

Diagnosis

PTX

Implant Time on haernodialysis (months)

41 60 56

M M F

Presumed CGN Wegener's Presumed CGN

Yes Yes No

Yes No

Presumed CGN Myeloma

No No

PCKD PresumedCGN

Yes

SLE

No Yes No

No

35

F

67

M

59 71*

F

25

M M

47

F

Eclampsia

Yes

45 48 49

F M M

Presumed CPN Diabetic CGN

No No No

*Tertiary hyperparathyroid patient.

Haemodialysis session time (h)

Drugs

30

5 4 4

50 2

5 4

Calcichew, vitamin BPC A l OH gel, warfarin, digoxin, testosterone Alucaps, dipyridamole, warfarin, ranitidine, EPO, DHT A l O H gel, DHT, ferrous sulphate Alucaps, quinine, folic acid, ferrous sulphate Prednisolone, ranitidine, nystatin Alucaps, DHT, ferrous sulphate Alucaps, aspirin, ferrous sulphate, prednisolone Calcichew, alucaps, calcitriol, captopril, isosorbide dinitrate Calcium carbonate Alucaps, DHT, isophane insulin (Mixtard) Alucaps, ferrous sulphate, calcium carbonate

87 132

3 156 9

3.5 5

4

140

5

2

4 4 4

3 5

Cytosolic calcium and haemodialysis

plasma, 1 mmol/l stock solution) was added to one sample to give a final concentration of 1 pnol/l, the second sample was left unloaded. Both samples were incubated at 37°C for 60 min. At the end of the incubation, the samples were centrifuged at 1 6 0 0 g for 10 min and the supernatant of platelet-poor plasma was discarded. The platelet pellets were resuspended to a final volume of 6-8 ml in a physiological saline buffer [145 mmol/l NaCl, 5 mmol/l KCl, 1 mmol/l MgSO,, 10 mmol/l 4-(2-hydroxyethyl)-lpiperazine-ethane sulphonic acid, 11.1 mmol/l glucose, pH 7.41. The cells were allowed to equilibrate at room temperature for 5-10 min. Portions (2 ml) were incubated at 37°C for 5 min and were transfered to a quartz cuvette in a Shimadzu RF5000 fluorimeter (V. A. Howe, London,

U.K.). Measurement of resting platelet [Ca2+li.[Ca"], was measured as previously described [7] with modifications as follows. In brief, cells were alternately excited at 340 nm and 380 nm, and light emission was measured at 500 nm to give a ratio of fluorescence. The extracellular calcium concentration was set to 1 mmol/l and extracellular leakage of Fura-2 was quenched with 0.2 mmol/l MnCI,. Calibration was achieved by adding 0.4 mmol/l diethylenetriaminopenta-acetate(calcium salt) to chelate MnC& followed by 0.075 mmol/l digitonin to lyse the cells and give the maximum ratio in the presence of 1 mmol/l calcium. EGTA (50 mmol/l) with Tris buffer (pH 7.4) was added to set the cuvette ionized calcium concentration at 0 mmol/l. Duplicate loaded samples were measured as described. Unloaded cells were also subjected to the above protocol to allow the subtraction of background fluorescence. Readings were made at 37°C with continuous stirring of the samples using a Peltier heater/stirrer block (V. A. Howe). Ratios were recalculated after the subtraction of background and the [Ca2'], was calculated as:

653

buted data were given as medians with interquartile ranges, and were analysed by Wilcoxon signed-rank testing with non-linear regression where appropriate. No corrections were made for multiple comparisons.

RESULTS All data were normally distributed except serum PTH concentrations, which followed a log-linear distribution.

Changes in plasma ionized calcium concentration, serum PTH concentration and platelet [Ca"], during haemodialysis There was a mid-dialysis rise in plasma ionized calcium concentration towards dialysate concentrations, followed by an insignificant fall at cessation of dialysis (Fig. 1). There were significant differences in the plasma ionized calcium concentration before and mid-dialysis (1.20 2 0.12 versus 1.33zk 0.07 mmol/l, P< 0.005, n = 10) and before and after dialysis (1.2020.12 versus 1.312 0.07 mmol/l, P < 0.05, iz = 10). There was a fall of 39% in the median serum PTH concentration during haemodialysis ( P