Effect of Direction and Rate of Change of Calcium on Parathyroid Hormone Secretion in Uraemia

Transcription

1 Nephrol Dial Transplant (1989) 4: European Dialysis and Transplant Association-European Renal Association Nephrology Dialysis Transplantation Original Article Effect of Direction and Rate of Change of Calcium on Parathyroid Hormone Secretion in Uraemia J. Cunningham 1, P. Altmann 1, Janet H. Gleed 2, K. C. Butter 1, F. P. Marsh 1 and J. L. H. O'Riordan 2 'Department of Nephrology, The London Hospital, London, and 'Department of Medicine, The Middlesex Hospital, Mortimer Street, London, UK Abstract. We have studied the control of amino-terminal parathyroid hormone (PTH) secretion in haemodialysis patients in response to slow or fast calcium infusion and during acute hypocalcaemia. n nine patients, fast calcium infusion (0.4 mmol/kg bodyweight per hour) for 15 min increased ionised calcium and reduced PTH, with an initial t\ of 12.8 min. After the infusion had ceased, calcium decreased steadily, and PTH increased, mean PTH reaching baseline values when calcium was still significantly greater than pre-infusion values. During slow calcium infusion for 2.5 h (0.1 mmol/kg bodyweight per hour), parathyroid suppression was evident at 15 min, when the calcium increment was only 0.03 mm. After 60 min, PTH did not decrease further despite progressive hypercalcaemia. Hypocalcaemic haemodialysis led to rapid increases in PTH. After 15 min, the mean calcium decrement was 0.09mM (/ > <0.01) and the mean PTH increment was 283pg/ml (/ > <0.01). The parathyroid response was maximal at 30 min, and did not increase subsequently, despite progressive hypocalcaemia for a further 90 min. During recovery from hypocalcaemia, PTH reduced and, despite comparable hypocalcaemia, PTH during periods of increasing calcium was always lower at a given calcium concentration than while calcium ' was decreasing. This influence of the direction of change of calcium was not seen during hypocalcaemia. The results showed that even in advanced renal disease, the parathyroid glands are highly responsive to small initial increments (0.03 mm) and decrements (0.09 mm) in Correspondence and offprint requests to: Dr J. Cunningham, Department of Nephrology, The London Hospital, London El BB UK. blood calcium, though less so to further perturbation of blood calcium. During hypocalcaemia, the parathyroid glands respond to both the absolute value of blood calcium and also to the direction of change of calcium. Key words: Acute hypocalcaemia; Calcium infusion; Parathyroid glands; Uraemia ntroduction Despite important advances in treatment, secondary hyperparathyroidism continues to affect the majority of patients with end-stage renal disease. Real or threatened hypocalcaemia is thought to underlie the progressive hyperparathyroidism [1] which in some cases becomes autonomous, leading to serious biochemical and clinical problems. Because the dialysate contains calcium at a concentration of 1.5 to 1.75 mm (concentrations that are greater than normal blood ionised calcium), haemodialysis is associated with the movement of calcium into the blood during the period of dialysis [2]. n contrast, the interdialytic period is associated with an increasing tendency to hypocalcaemia as plasma inorganic phosphate increases. A previous study has focused on the relationship between parathyroid hormone (PTH) and calcium in dialysis patients during calcium infusion and the results indicated that the parathyroid glands are sensitive to both the absolute concentration of calcium in plasma and also to the rate and direction of movement of plasma calcium

2 340 J. Cunningham et al concentration [3]. n order to examine these events further, we have investigated the interactions between changes in blood ionised calcium concentration and parathyroid secretion in a group of haemodialysis patients and we have looked at this relationship under conditions of hypercalcaemia and hypocalcaemia. Methods Laboratory onised calcium was measured in fresh duplicate whole blood samples using a Kone Microlyte (Kone nstruments, Warrington, UK). The mean ± SD in 50 normal subjects was 1.22 ± 0.03 mm and in 28 randomly selected haemodialysis patients was 1.24 ± O.lOmM (measurements made immediately before dialysis). mmunoreactive PTH was measured using a homologous amino-terminal immuno-radiometric assay [4]. The limit of detection in plasma is 40 pg/ml and the upper limit in normal subjects is 120 pg/ml. Most dialysis patients have greatly elevated values of PTH using this assay [3,5]. Aluminium was measured in serum using a Varian 1275 atomic absorption spectrophotometer and a GTA95 electrothermal atomiser [6]. The detection limit is 1.8 ug/1 and the interassay coefficient of variation is 7.5%. Data Analysis Group data are expressed as the mean±sem, and compared using Student's t-test for paired data. Linear regression analysis was performed by the method of least squares. A value of/<0.05 was regarded as significant. Patients Fourteen patients maintained on long-term haemodialysis were studied. Their average age was 48 (24 60) years and they had been dialysed for a mean of 75 (27-156) months twice or thrice weekly using a dialysate containing calcium at a concentration of 1.75 mm. None of the patients had received vitamin D analogues for at least 4 months. Serum aluminium concentration was 61 ± 16 ug/ml. C. Acute hypocalcaemia. This was induced by haemodialysing nine patients for 2 h using a calcium-free dialysate. Four of these patients had also completed the calcium infusion protocols A and B. Nephross Andante hollowfibre dialysers (Organon Teknika) were used with a dialysate flow rate of 500 ml/min and blood flow of 3 ml/kg bodyweight/min. n addition, five patients began, but could not complete, this protocol because of the development of symptoms before the end of the 2 h hypocalcaemic challenge. D. Recovery from acute hypocalcaemia. After the end of the hypocalcaemic haemodialysis the same nine patients as in study C were followed for a further 1 h 30 min during which haemodialysis continued at the same blood flow rate using a normal (1.75 mm calcium) dialysate. These patients were therefore followed during a recovery period that was 'assisted' by transfer of calcium from the dialysate to the patient. Results Fast Calcium nfusion The initial calcium was mm and this increased to mm by the end of the 15-min infusion (Fig. 1). Baseline PTH was 554 ± 195 pg/ml and fell rapidly with an initial t of 12.8 min. Following the 15-min infusion, calcium immediately began to decrease, although had still not returned to baseline at the end of the 3 h 45 min study. PTH reached its nadir at 15 min but, despite the persistence of hypercalcaemia, the mean concentration started to increase again immediately after the end of the infusion. Although all the patients showed very marked early PTH suppression, the recovery of PTH following the infusion varied, most patients showing a prompt return of PTH towards baseline values, while some showed prolonged suppression for the duration of the protocol. n this study, neither the initial calcium concentration, nor the increment in calcium during the infusion, clearly predicted the magnitude of PTH suppression. Slow Calcium nfusion Protocols A. Fast calcium infusion. Nine patients underwent infusion of calcium gluconate at 0.4mmol/kg bodyweight/h for 15 min and were followed for a further 3 h 45 min. B. Slow calcium infusion. The same nine patients underwent calcium gluconate infusion at 0.1 mmol/kg bodyweight/h for 2 h 30 min. When the same nine patients were subjected to calcium infusion at 0.1 mmol/kg bodyweight per hour, blood ionised calcium increased from 1.27 ±0.03 to 1.68 ±0.08 mm over the course of the 2 h 30 min infusion (Fig. 2). n six of the patients, the rate of increase of calcium was relatively constant throughout the 2 h 30 min infusion, but in three there was acceleration of hypercalcaemia between 120 and 150 min. Baseline PTH was 515 ± 120 pg/ml and by 15 min this had decreased to

3 Control of PTH Secretion in Uraemia " calcium infusion T \ psi,. T " s a ' f/ / \ T» E, E 8 1.5" 1.4" 1.2" 800" 700" 600" 500" " ; > '. noimal-;cajnge ' ' ' ' '' ' ' > ' _ ' ' ' ' ; r calcium.: is infusion: n i \^ 1 T y 1, Fig. 1. Changes in ionised calcium and PTH during and after infusion of calcium gluconate at 0.4mmol/kg bodyweight per min for 15 min. Data are mean + SEM from nine patients. 'Significantly different from baseline; significantly different from peak of Ca 2+ or nadir of PTH (recovery only) pg/ml (P< 0.022) in response to an increment in ionised calcium of only 0.03mM. Thereafter PTH continued to decline, reaching 321 ±75 pg/ml by 60 min, by which time calcium had increased from to 1.41 ±0.04mM (/ > <0.001). From then on PTH did not change even though calcium continued to increase, the mean calcium eventually reaching 1.68 mm at the end of the 2 h 30-min infusion. Again, neither the initial calcium concentration, nor its increment, was related to the changes in PTH. Response to Acute Hypocalcaemia Blood calcium (initially 1.21 ± 0.03 mm) decreased at a near constant rate throughout the 2h hypocalcaemic i [ N N calcium infusion T k Fig. 2. Changes in ionised calcium and PTH during infusion of calcium gluconate at 0.1 mmol/kg bodyweight per min for 2 h 30 min. Data are mean±sem from the same nine patients illustrated in Fig. 1. Significantly different from baseline. challenge, to mm after 15 min (P< 0.01) and eventually to mm (P < 0.001) at the end of the procedure (Fig. 3). PTH was initially pg/ml, and increased very rapidly during the early part of the hypocalcaemic stimulus, reaching 725 ± 168 pg/ml by 15 min and 876 ± 191 pg/ml by 30 min. At these times, the decrements in ionised calcium, though significant, were only 0.09 and 0.14 mm respectively. The parathyroid response was maximal after 30 min of progressive hypocalcaemia and did not change thereafter, even though calcium continued to decrease during the remainder of the 2 h dialysis. Baseline PTH was closely related to the maximum stimulated PTH (r=0.88, P=0.002), and both were inversely related to the change in calcium during the 2 h challenge (/-=0.83, P<0.006 and r = 0.81, /><0.008 respectively). ft 1

4 342 J. Cunningham et al. :...-. hypocalcaemic dialysis ' ' " : : :- : : :-:- : h6rmai:ran e-;.- :. >;.-:- > :. :.-:.-:.- : " E 1 i z " " 500" hypocalcaemic dialysis time (min) Fig. 3. Changes in ionised calcium and PTH during hypocalcaemia produced by haemodialysis using a calcium-free dialysate for 2 h. Data are mean±sem from nine patients. The slight decrease in mean PTH after 30 min was not significant. 'Significantly different from baseline. Recovery from Acute Hypocalcaemia n the 90 min after induction of hypocalcaemia, calcium increased steadily, assisted by the influence of continuing haemodialysis using dialysate containing calcium at a concentration of 1.75 mm (Fig. 4). Mean ionised calcium at the beginning of the recovery period was mm and it increased over the ensuing 90min to 1.07±0.02mM (P< 0.002). Although normocalcaemia was not achieved during this period, PTH declined progressively, eventually to pg/ml, a value substantially less than the mean of 789 ± 191 pg/ml that prevailed immediately before the initiation of the recovery period (P< 0.001), and close to the baseline value at the beginning of the hypocalcaemic challenge ( and 469 ± 145 pg/ml respectively for baseline and postrecovery). 1 calciurr O 1 K z " 800" 700" T ^^^^ time (min) Fig. 4. Changes in ionised calcium and PTH during recovery from hypocalcaemia. Patients continued haemodialysis for 90 min after switching from dialysate containing 0 mm calcium to 1.75 mm calcium. Data are mean +SEM from nine patients. 'Significantly different from pre-hypocalcaemic dialysis baseline; "significantly different from prerecovery level. nfluence of Direction of Change of Calcium The observations made within the region of hypercalcaemia are shown in Fig. 5a. During and after the rapid infusion of calcium, the concentration of PTH at a given value of calcium was similar whether calcium was increasing or decreasing. The relationship was therefore little affected by the direction of change of calcium under these circumstances. During hypocalcaemia, however, the increase in PTH above baseline was much greater during the period of decreasing calcium than during the subsequent recovery period at points of comparable hypocalcaemia (Fig. 5b). For example at a blood calcium of 1.07 mm (decrement of 0.14 mm), PTH was pg/ml when calcium was falling, and only 469 ± 145 pg/ml during recovery from hypocalcaemia (P<0.018).

5 Control of PTH Secretion in Uraemia 343 pg/ml) E Z 500" 400" recovery la) calcium infusion A ionised calcium (mm) 400" 200" 100" recovery» \ hypocalcaemic dialysis A A ionised calcium (mm) Fig. 5. Effect of direction of change of calcium on PTH. On both abscissa and ordinate, zero refers to the baseline value at the onset of the study. Each point represents the mean value from nine patients. The recovery phase is shown by the broken line, (a) Hypercalcaemia fast infusion and recovery. The relationship between ionised calcium and PTH was similar with rising and falling calcium, (b) Hypocalcaemia hypocalcaemic dialysis and recovery. PTH concentrations were increased more during the period of falling ionised calcium than during recovery. nfluence of Aluminium There was no discernible influence of the patients' serum aluminium concentrations on either baseline or perturbed calcium and PTH values in any of the studies presented. here. Discussion These studies have demonstrated very marked and rapid responses of the parathyroid glands to upward and downward perturbation of blood calcium, and to clearcut changes in circulating PTH within 10 and 15min bl respectively. ncreases in ionised calcium as small as 0.03 mm and decreases of 0.09 mm were associated with, respectively, mean changes in PTH of 22% and + 64% and it is likely that even smaller increments and decrements in calcium would have been associated with perceptible PTH responses had the studies been designed specifically to show this. Although the initial disappearance half-time for PTH during fast calcium infusion was quite short (12.8 min), this was still somewhat slower than was measured using the same assay during calcium infusion in subjects with normal renal function [3,7]. The results are consistent with studies using a cytochemical bioassay for PTH which have also shown some prolongation of the serum \.\ of PTH bioactivity in patients with uraemia [8]. Our results also point to important influences of the direction of change in calcium, and its rate, on PTH, in addition to control by the absolute concentration of ionised calcium per se. This was shown here when comparing the response during and after the rapid induction of hypocalcaemia the stimulus to the parathyroid glands provided by hypocalcaemia was much greater when calcium was declining than when it was increasing and under these circumstances, the direction of change of blood calcium concentration was an important additional modulator of PTH secretion. n contrast, the effect of direction of change was not seen in relation to hypercalcaemia, although this has been reported previously [3]. Although we did not examine the influence of the rate of decrease of blood calcium, EDTA infusions in normal subjects have shown that a rapid decrease in serum calcium provided a more potent stimulus to PTH secretion than did a slow decrease [9]. Collectively these results suggest that when responding to deviations of blood calcium from normal, hyperplastic parathyroid glands may be influenced both by the magnitute of the deviation and its rate, in a manner that allows the glands to sense and respond vigorously to a worsening threat to calcium homeostasis, and to attenuate that response when calcium is moving towards normal, possibly helping to avoid overshoot. The data from the slow calcium infusions and acute hypocalcaemia point to very pronounced early responses of PTH to initial deviation of calcium concentrations from baseline values, with later stabilisation. For example, during the long calcium infusion, the suppression of PTH was maximal at 60 min when the calcium increment was 0.15mM and PTH did not respond to further increases in calcium to a final increment of 0.40 mm. Likewise, during progressive hypocalcaemia the PTH response was maximal at 30 min and did not increase further despite significant additional decreases in ionised calcium. These observations in patients with hyperplastic parathyroid glands are in keeping with earlier work in vivo [10,11] and in vitro [12], in which the

6 344 J. Cunningham et al parathyroid responses to changing calcium appeared to be most sensitive at and around the physiological level of calcaemia. n vitro studies using dispersed parathyroid cells from patients with uraemia have shown a similar pattern, albeit with an altered set point in some cases [12]. We found that neither initial PTH nor the PTH response to changes in blood calcium was related to the initial calcium ion concentration. This observation differs from earlier work by Adami et al [3] in which patients with the greatest baseline plasma calcium showed the greatest degree of PTH suppression during calcium infusion. The explanation for the discrepancy is not clear, but may be connected with the much greater heterogeneity of Adami's patients with respect to their baseline total calcium concentrations, which ranged from 1.67 to 2.87 mm. n contrast the range of ionised calcium in our patients was only mm. During acute hypocalcaemia, patients with increased initial PTH tended to demonstrate the greatest PTH increment and also the smallest calcium decrement during hypocalcaemic stress. We found that both high unstimulated PTH and maximum PTH during hypocalcaemia were quantitively related to resistance to hypocalcaemic stress, both relationships being independent of the serum aluminium concentrations. Thus, patients with more severe hyperparathyroidism, and by implication those with the greatest gland mass [13], were better able to withstand hypocalcaemic stress, perhaps because of their enhanced PTH response. Two previous studies of hypocalcaemic challenge in uraemia have also shown enhanced PTH responses in patients with high baseline PTH [14,15] but, in contrast to our own observations, high PTH in their patients did not correlate with greater resistance to hypocalcaemic stress. One of those studies in addition assessed the aluminium status of the patients and, as found here, reported no relationship between aluminium status and PTH response [14]. Finally, our results may have a bearing on the pathogenesis of secondary hyperparathyroidism in maintenance haemodialysis patients. These patients undergo periodic increases in blood ionised calcium during haemodialysis, with accompanying decreases in PTH [3], alternating with slow reversal during the interdialytic period. Our studies have demonstrated the ability of relatively small movements of blood ionised calcium, such as may be encountered during intermittent haemodialysis, to trigger major changes in parathyroid activity, and it is possible that these periodic stimuli provide a potent stimulus for the development of parathyroid hyperplasia. Acknowledgements. This work ws supported by a grant from the National Kidney Research Fund. References 1. Slatopolsky E, Caglar S, Pennel JP et al. On the pathogenetis of hyperparathyroidism in chronic experimental renal insufficiency in the dog. J Clin nvest 1971; 50: Regan RJ, Peacock M, Rosen SM, Robinson PJ, Horsman A. Effect of dialysate calcium concentration on bone disease in patients on hemodialysis. Kidney nt 1976; 10: Adami S, Muirhead N, Manning RM et al. Control of secretion of parathyroid hormone in secondary hyperparathyroidism. Clin Endocrinol 1982; 16: Manning RM, Hendy GN, Papapoulos SE, O'Riordan JLH. Development of homologous immunological assays for human parathyroid hormone. J Endocrinol 1980; 85: Muirhead N, Catto GRD, Edward N, Adami S, Manning RM, O'Riodan JLH. Suppression of secondary hyperparathyroidism in uraemia: acute and chronic studies. Br MedJ 1984; 288: D'Haese PC, van de Vyver FL, de Wolff FA, de Brae ME. Measurement of aluminum in serum, blood, urine and tissues of chronic hemodialysed patients by use of electrothermal atomic absorption spectrometry. Clin Chem 1985; 31: Papoulos SE, Manning RM, Hendy GN, Lewin G, O'Riordon JLH. Studies of circlating parathyroid hormone in man using a homologous amino-terminal specific immunoradiometric assay. Clin Endocrinol 1980; 13: Goltzman D, Gomolin H, DeLean A, Wexler M, Meakins L. Discordant disappearance of bioactive and immunoreactive parathyroid hormone after parathyroidectomy. J Clin Endocrinol Metab 1984; 58: Brent GA, LeBoff MS, Seely EW, Brown EM. The influence of the rate of change of serum calcium on the magnitude of the intact parathyroid hormone response in humans. J Bone Min Res ; [Suppll]:Abstrl Fox J, Heath H. The 'calcium clamp': Effect of constant hypocalcemia on parathyroid hormone secretion. Am J Physiol 1981; 240: E649-E Mayer GP, Hurst JG. Sigmoidal relationship between parathyroid hormone secretion and plasma calcium concentration in calves. Endocrinology 1978; 103: Brown EM, Wilson RE, Eastman RC, Pallotta J, Marynick P. Abnormal regulation of parathyroid hormone release by calcium in secondary hyperparathyroidism due to chronic renal failure. J Clin Endocrinol Metab 182; 54: McCarron DA, Muther RS, Lenfesty B, Bennett WM. Parathyroid function in persistent hyperparathyroidism: relationship to gland size. Kidney nt 1982; 22: Kraut JA, Shinaberger JH, Singer FR et al. Parathyroid gland responsiveness to acute hypocalcemia in dialysis osteomalacia. Kidney nt 1983; 23: Voights A, Felsenfeld AJ, Andress D, Llach F. Parathyroid hormone and bone histology: response to hypocalcemia in osteitis fibrosa. Kidney nt 1984; 25: Received for publication Accepted