- Email: [email protected]
Potassium flux of erythrocytes in chronic hemodialysis patients
Clinica Chimica Acta 350 (2004) 189 – 193 www.elsevier.com/locate/clinchim
Potassium flux of erythrocytes in chronic hemodialysis patients Yasuaki Ozawaa,b, Yuji Imafukua, Sadataka Nishib, Hiroshi Yoshidaa,* a
Department of Clinical Laboratory Medicine, Fukushima, Medical University School of Medicine, 1 Hikarigaoka, Fukushima 960-1247, Japan b Hemodialysis Unit, Nishi Hospital, Fukushima 979-1521, Japan Received 2 June 2004; received in revised form 23 July 2004; accepted 29 July 2004
Abstract Background: In chronic hemodialysis patients, hyperkalemia is frequently observed. In these patients, erythrocytes were examined to know whether they participate in the regulation of K+ or not. Methods: Erythrocyte K+ release (DKr) was induced by the incubation of erythrocyte suspension at 4 8C for 24 h and the K+ influx followed at 37 8C for 3 h. K+ flux of erythrocytes or DKi/DKr ratio, which was reflected by Na+/K+-exchanging ATPase, was measured in chronic hemodialysis patients. K+ concentration was measured by ion-selective electrode method. Results: Non-diabetic hemodialysis patients classified into three groups according to their serum levels were compared for various factors. Among them, the DKi/DKr ratios in medium- and high-serum K+ groups were significantly lower than those in the low serum K+ group. The effect of hemodialysis on erythrocyte K+ flux was examined. After hemodialysis, the mean DKi/ DKr ratio increased significantly compared with that before the treatment. Erythrocyte K+ concentrations converted into a narrower range after hemodialysis. Conclusion: The reduced K+ flux in erythrocyte may play a part in the development of hyperkalemia in non-diabetic chronic hemodialysis patients. D 2004 Elsevier B.V. All rights reserved. Keywords: Erythrocyte K+ flux; Chronic renal failure; Hemodialysis patient; Serum K+
1. Introduction It has been known since the 1930s that cold storage of human whole blood increases the serum potassium ion (K+) concentration by the release of K+, predominantly from erythrocytes [1]. To evaluate K+ flux of * Corresponding author. Tel.: +81 24 547 1348; fax: +81 24 548 6016. E-mail address: [email protected] (H. Yoshida). 0009-8981/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2004.07.010
erythrocytes, we developed a simple method of measuring the release/influx ratio [2,3]. K+-release (DKr) was induced at 4 8C and K+-influx (DKi) at 37 8C. The K+ flux activity or DKi/DKr ratio was reflected predominantly by Na + /K + -exchanging ATPase activity [3]. Hyperkalemia is one of the causes of fatal cardiac complications in patients with chronic renal failure, and the measurement of serum K+ concentrations is very important [4]. In this investigation, we measured the
190
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
serum K+ and DKi/DKr ratios of erythrocytes in these patients and suggest that erythrocytes might participate in K+ metabolism in patients with chronic renal failure.
2. Materials and methods 2.1. Subjects and hemodialysis Forty-three patients with chronic renal failure who were being treated by hemodialysis at Nishi Hospital (male 19, female 24, age 27–84 years, mean 62.9 years) were included. Informed consent was obtained from all subjects, using a protocol that was approved by the ethics committee of the same hospital. None of the patients were administered steroid hormones, antihypertensive drugs including diuretics, Ca2+ antagonists or digitalis. Among the hemodialysis patients, 4 (male 1, female 3) were hospitalized, and the others were outpatients. All of them have been treated three times a week for 2–23 years (mean 7.2 years). The hemodialysis treatment usually takes 4–4.5 h (mean 4.2 h). The urine volume of these patients was b100 ml/day. The causes of chronic renal failure were chronic glomerulonephritis in 30 patients (male 12, female 18) and diabetes mellitus in 13 patients (male 7, female 6). Every hemodialysis patient with diabetes mellitus had complications of retinopathy and neuropathy. Four patients with diabetes mellitus received insulin therapy, and one was administered sulfonylurea. Thirty-seven had been injected with erythropoietin two to three times a week for N5 months. These patients were classified into three groups according to their K+ concentrations: low (K+b5.0 mmol/l), medium (K+z 5.0 or b5.5 mmol/l) and high (K+ z5.5 mmol/l). Blood samples for routine laboratory tests were collected just before hemodialysis. To examine the effect of hemodialysis on K+ flux and K+ concentration in erythrocytes, 11 non-diabetic patients were randomly selected. Hemodialyses were performed by using hemodialyser, BK-1.6U (polymethylmetacrylate membrane with 60-2 pore size, Toray Med., Tokyo).
Medical, Osaka), and the plasma and buffy coat were removed after centrifugation at 1805g for 10 min. Hemodialysate (Kindaly solution AF2; Fuso Pharmaceutical, Osaka) containing Na+ 141 mmol/l, K+ 2.0 mmol/l, Cl 105 mmol/l, Ca2+ 1.5 mmol/l, Mg2+ 0.5 mmol/l, CH3COO 8 mmol/l, HCO3 30 mmol/l, and glucose 5.8 mmol/l was used for the preparation of erythrocyte suspensions. The pH of the solution was 7.37, and the osmotic pressure was 283 mosM/l. After washing three times by centrifugation, the erythrocyte count of the suspension was adjusted to 200104/Al. The erythrocyte suspension was divided into two plastic tubes (diameter 10 mm, length 75 mm); one was incubated at 4 8C for 24 h and the other at 4 8C for 24 h and then in a water bath at 37 8C for 3 h. After each incubation, they were centrifuged at 1805g for 10 min, and the supernatants were collected, stored at 4 8C and assayed simultaneously for K+. The centrifugation did not affect K+ release of erythrocytes. The difference in K+ concentration between pre- and postincubation at 4 8C for 24 h was designated as DKr and that between post-incubation at 4 8C and after rewarming at 37 8C for 3 h was designated as DKi. The K+ flux of erythrocytes was evaluated by calculating the ratio of influx after rewarming to release after cold storage (DKi/DKr). Erythrocyte DKi/ DKr or K+ flux of healthy subjects was 0.274F0.063 [3]. The coefficient of variation (CV) of the within-day imprecision (n=20) of the assay was 2.9%. 2.3. Measurement of serum electrolytes and other values Serum K+ and other electrolytes were measured by ion-selective electrodes (Jokoh electrolyte analyzer EX-180, Jokoh, Tokyo). Other biochemical values, including serum creatinine, were measured by means of a TBA-30FR (Toshiba Medical, Tokyo). Serum concentrations were expressed as the mean of results of four measurements in the preceding month. Serum K+ values of the patients were examined retrospectively for 6 months before the K+ flux experiment to determine whether they fluctuated or not.
2.2. Assay for K+ flux 2.4. Assay for erythrocyte K+ concentration +
K flux was assayed according to the previously described procedure [2]. Briefly, venous blood was collected in heparin-Li-containing tubes (Sekisui
To measure erythrocyte K+ concentrations, heparinized blood samples were drawn from 30 hemodial-
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
191
ysis patients (male 12, female 18). The blood cells, including erythrocytes, were counted using a Sysmex analyzer, and the samples were centrifuged at 1805g for 30 min. Plasma, buffy coat and the upper layer of erythrocytes were carefully removed, and 0.5 ml of packed erythrocytes was mixed with 9.5 ml of distilled water and allowed to hemolyze completely. After centrifugation, the K+ concentration in the supernatant was measured, and the erythrocyte K+ concentration was calculated. The amount of trapped plasma was regarded as constituting 2.5% of the packed erythrocyte volume. 2.5. Statistical analysis Data were analyzed using the paired and unpaired Student’s t-tests and Pearson’s correlation coefficient.
3. Results Non-diabetic hemodialysis patients classified into three groups according to their serum K+ concentrations were compared for various factors, including the DKi/DKr ratio. As shown in Table 1, no significant differences were observed in age, duration of hemodialysis treatment, serum pH and Na+ and
Table 1 Parameters of three groups of non-diabetic hemodialysis patients with different serum K+ concentrations Mean serum K+ (mmol/l)
N (male, female) Age (years) Duration of hemodialysis treatment (years) Serum pH Mean serum Na+ (mmol/l) Erythrocyte K+ (mmol/l) DKi/DKr
Low (K+b0) Middle (K+=5.0 or b5.5)
High (K+z5.5)
10 (3, 7) 10 (5, 5) 66.2F14.32# 57.1F15.38 7.0F6.78 10.5F6.36
10 (4, 6) 64.1F16.16 6.0F2.81
7.36F0.038 140.2F1.64
7.33F0.033 140.9F2.77
7.35F0.043 138.8F2.62
110.0F8.54
109.4F3.88
110.0F5.21
0.269F0.041 0.228F0.041* 0.213F0.032**
* Pb0.05 vs. low. ** Pb0.005 vs. low. # MeanFS.D.
Fig. 1. Relationship between the DKi/DKr ratio and the mean serum K+ concentration in non-diabetic (A) and diabetic (B) hemodialysis patients. The serum K+ concentration is expressed as the mean of four measurements in the preceding month before estimation of the ratio.
erythrocyte K+; however, the DKi/DKr ratios in the medium and high K+ groups were significantly lower than those in the low K+ group ( Pb0.05, Pb0.005, respectively). K+ concentrations in sera of the patients in each group did not change to other groups in the 6 months of observation. The relationships between the DKi/DKr ratio and mean serum K+ concentration in diabetic and nondiabetic hemodialysis patients were compared. As shown in Fig. 1A, a negative correlation (r= 0.482, Pb0.01) was observed between the DKi/DKr ratio and the mean serum K+ concentration in 30 non-diabetic
192
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
hemodialysis patients (male 12, female 18). However, a significant correlation was not observed in the 13 diabetic hemodialysis patients (male 7, female 6) (Fig. 1B). No significant differences were observed between male and female patients and between patients receiving erythropoietin or not. To investigate the effect of hemodialysis, the DKi/ DKr ratio and erythrocyte K+ concentration before and after hemodialysis were measured in 11 nondiabetic patients. The mean DKi/DKr ratio of erythrocytes collected after hemodialysis (0.274F0.040) increased significantly ( Pb0.05) compared with that of the erythrocytes collected before the treatment (0.263F0.036) (Fig. 2). The erythrocyte K+ concentrations before hemodialysis treatment were scattered over a wide range, and they converged into a narrower range after hemodialysis (Fig. 3). Among 11 cases examined, decreases after hemodialysis were observed in seven cases and, in contrast, increases in four cases. Totally, the mean values did not differ significantly before (113.5F8.82 mmol/l) or after hemodialysis (110.0F3.71 mmol/l). No significant differences in the changes between DKi/DKr ratio and
Fig. 3. Comparison of the erythrocyte K+ concentration in hemodialysis patients before and after hemodialysis.
erythrocyte K+ concentration were observed after the treatment.
4. Discussion
Fig. 2. Comparison of DKi/DKr ratios in hemodialysis patients before and after hemodialysis.
In measuring extracellular K+ concentration of erythrocyte suspension incubated at 4 8C for K+ release followed by incubation at 37 8C for the influx of erythrocytes, the amount of K+ influx was consistently reduced compared with that of the release [2]. These results led us to speculate that some component in erythrocytes was damaged by the cold exposure and that the data obtained by assays performed in such cold conditions might sometimes be inaccurate. We therefore developed a simple method to estimate K+ flux of erythrocytes by measuring the DKi/DKr ratio, which reflects Na+/ K+-exchanging ATPase activity [3]. Serum K+ concentrations are recognized to be higher in chronic hemodialysis patients; however, among the patients in our hospital, they varied and were classified into three groups according to their concentrations. Because diabetic patients receive anti-
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
diabetic therapies including insulin that might affect K+ metabolism [5], they were not included in the comparative analyses shown in Table 1. In nondiabetic patients, no significant differences were observed in various factors. However, the K+ flux was the lowest in high serum K+ group, which suggests that erythrocytes participate in the regulation of K+ metabolism in chronic hemodialysis patients. As erythrocyte K+ concentrations were not different among the three groups, it could be speculated that the K+ flux of erythrocytes could indicate that of other cells including muscle cells. In patients with chronic renal diseases, Na+/K+exchanging ATPase activity in erythrocytes was decreased [6]. We observed in this study that the K+ flux of erythrocytes was reduced in hemodialysis patients with high serum K+ concentrations. In the hemodialysis patients, a negative correlation between the mean serum K+ concentration and DKi/DKr ratio was observed in the non-diabetic patients, but not in the diabetic patients. These findings suggest that some factors in sera with high K+ concentrations affect the K+ flux of erythrocytes in non-diabetic patients. However, it appears more complicated in diabetic patients. Active transport systems in various cells including erythrocytes may play a role in the regulation of Na+ and K+ metabolism. It was reported that the elevation of serum K+ and intracellular Na+ concentrations activates Na+/K+-exchanging ATPase of erythrocytes [7], but the effects of the changes of internal K+ and external Na+ concentrations were considered to be very small [8]. However, our results showed that the DKi/DKr ratio in non-diabetic hemodialysis patients with higher serum K+ concentrations was the lowest, and the mean concentrations of erythrocyte K+ were almost the same in the three groups despite different serum K+ concentrations. We suspect that reduced erythrocyte K+ flux activity could be one of the causes of hyperkalemia in nondiabetic hemodialysis patients; however, other mechanisms might participate in the regulation of intracellular K+ concentration. In cases with reduced K+ flux of erythrocytes, Na + flux might be also decreased, which might affect physiological conditions including blood pressure. We have observed a
193
correlation between K+ flux and hypertension in our preliminary experiment (data not shown). After hemodialysis treatment, various pathological conditions are corrected to varying degrees, which include abnormal serum electrolyte concentrations, metabolic acidosis, water excess, and accumulation of uremic toxins. Our results showed that the DKi/DKr ratios were increased significantly and the erythrocyte K+ concentrations were reduced to a narrower range after hemodialysis treatments. The convergence of the internal K+ concentrations seems to be induced as a response to the surrounding conditions. It is reasonable to speculate that some pathologic factors in the sera could affect the K+-flux activity of erythrocytes as indicated by the low DKi/DKr ratios. We conclude from our results that erythrocytes participate partly in the regulation of serum K+ concentration in non-diabetic chronic hemodialysis patients.
References [1] Harris JE. The influence of the metabolism of human erythrocytes on the potassium content. J Biol Chem 1941; 141:579 – 95. [2] Ozawa Y, Sato S, Nishi S, Shishido S, Imafuku Y, Nagai T, et al. Effect of temperature and drugs on potassium flux in human erythrocytes. In: Ohba Y, Kanno T, Okabe H, Totani M, Nakahara K, Ichihara K, editors. Quality control in the clinical laboratory ’95. Tokyo7 Excepta Medica, 1995. p. 191 – 6. [3] Ozawa Y, Sato S, Matsumoto K, Nishi S, Shishido S, Imafuku Y, et al. Determination of potassium flux activity of viable human erythrocytes by measuring the release-influx ratio. Clin Chim Acta 1999;279:125 – 32. [4] Kunie CL, Charney AN. Potassium and renal failure. Compr Ther 1981;7:29 – 33. [5] Rahmani-Jourdheuil D, Mourayre Y, Vague P, Boyer J, JuhanVague I. In vivo insulin effect on ATPase activities in erythrocyte membrane from insulin-dependent diabetics. Diabetes 1987;36:991 – 5. [6] Cole CH. Decreased ouabain-sensitive adenosine triphosphatase activity in the erythrocyte membrane of patients with chronic renal disease. Clin Sci Mol Med 1973;45:775 – 84. [7] Sachs JR, Welt LG. The concentration dependence of active potassium transport in the human red blood cell. J Clin Invest 1967;46:65 – 76. [8] Knight A, Taylor RC, Welt LG. Autoregulation of sodium and potassium concentration in erythrocytes. Clin Res 1971;19:86.
Potassium flux of erythrocytes in chronic hemodialysis patients Yasuaki Ozawaa,b, Yuji Imafukua, Sadataka Nishib, Hiroshi Yoshidaa,* a
Department of Clinical Laboratory Medicine, Fukushima, Medical University School of Medicine, 1 Hikarigaoka, Fukushima 960-1247, Japan b Hemodialysis Unit, Nishi Hospital, Fukushima 979-1521, Japan Received 2 June 2004; received in revised form 23 July 2004; accepted 29 July 2004
Abstract Background: In chronic hemodialysis patients, hyperkalemia is frequently observed. In these patients, erythrocytes were examined to know whether they participate in the regulation of K+ or not. Methods: Erythrocyte K+ release (DKr) was induced by the incubation of erythrocyte suspension at 4 8C for 24 h and the K+ influx followed at 37 8C for 3 h. K+ flux of erythrocytes or DKi/DKr ratio, which was reflected by Na+/K+-exchanging ATPase, was measured in chronic hemodialysis patients. K+ concentration was measured by ion-selective electrode method. Results: Non-diabetic hemodialysis patients classified into three groups according to their serum levels were compared for various factors. Among them, the DKi/DKr ratios in medium- and high-serum K+ groups were significantly lower than those in the low serum K+ group. The effect of hemodialysis on erythrocyte K+ flux was examined. After hemodialysis, the mean DKi/ DKr ratio increased significantly compared with that before the treatment. Erythrocyte K+ concentrations converted into a narrower range after hemodialysis. Conclusion: The reduced K+ flux in erythrocyte may play a part in the development of hyperkalemia in non-diabetic chronic hemodialysis patients. D 2004 Elsevier B.V. All rights reserved. Keywords: Erythrocyte K+ flux; Chronic renal failure; Hemodialysis patient; Serum K+
1. Introduction It has been known since the 1930s that cold storage of human whole blood increases the serum potassium ion (K+) concentration by the release of K+, predominantly from erythrocytes [1]. To evaluate K+ flux of * Corresponding author. Tel.: +81 24 547 1348; fax: +81 24 548 6016. E-mail address: [email protected] (H. Yoshida). 0009-8981/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2004.07.010
erythrocytes, we developed a simple method of measuring the release/influx ratio [2,3]. K+-release (DKr) was induced at 4 8C and K+-influx (DKi) at 37 8C. The K+ flux activity or DKi/DKr ratio was reflected predominantly by Na + /K + -exchanging ATPase activity [3]. Hyperkalemia is one of the causes of fatal cardiac complications in patients with chronic renal failure, and the measurement of serum K+ concentrations is very important [4]. In this investigation, we measured the
190
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
serum K+ and DKi/DKr ratios of erythrocytes in these patients and suggest that erythrocytes might participate in K+ metabolism in patients with chronic renal failure.
2. Materials and methods 2.1. Subjects and hemodialysis Forty-three patients with chronic renal failure who were being treated by hemodialysis at Nishi Hospital (male 19, female 24, age 27–84 years, mean 62.9 years) were included. Informed consent was obtained from all subjects, using a protocol that was approved by the ethics committee of the same hospital. None of the patients were administered steroid hormones, antihypertensive drugs including diuretics, Ca2+ antagonists or digitalis. Among the hemodialysis patients, 4 (male 1, female 3) were hospitalized, and the others were outpatients. All of them have been treated three times a week for 2–23 years (mean 7.2 years). The hemodialysis treatment usually takes 4–4.5 h (mean 4.2 h). The urine volume of these patients was b100 ml/day. The causes of chronic renal failure were chronic glomerulonephritis in 30 patients (male 12, female 18) and diabetes mellitus in 13 patients (male 7, female 6). Every hemodialysis patient with diabetes mellitus had complications of retinopathy and neuropathy. Four patients with diabetes mellitus received insulin therapy, and one was administered sulfonylurea. Thirty-seven had been injected with erythropoietin two to three times a week for N5 months. These patients were classified into three groups according to their K+ concentrations: low (K+b5.0 mmol/l), medium (K+z 5.0 or b5.5 mmol/l) and high (K+ z5.5 mmol/l). Blood samples for routine laboratory tests were collected just before hemodialysis. To examine the effect of hemodialysis on K+ flux and K+ concentration in erythrocytes, 11 non-diabetic patients were randomly selected. Hemodialyses were performed by using hemodialyser, BK-1.6U (polymethylmetacrylate membrane with 60-2 pore size, Toray Med., Tokyo).
Medical, Osaka), and the plasma and buffy coat were removed after centrifugation at 1805g for 10 min. Hemodialysate (Kindaly solution AF2; Fuso Pharmaceutical, Osaka) containing Na+ 141 mmol/l, K+ 2.0 mmol/l, Cl 105 mmol/l, Ca2+ 1.5 mmol/l, Mg2+ 0.5 mmol/l, CH3COO 8 mmol/l, HCO3 30 mmol/l, and glucose 5.8 mmol/l was used for the preparation of erythrocyte suspensions. The pH of the solution was 7.37, and the osmotic pressure was 283 mosM/l. After washing three times by centrifugation, the erythrocyte count of the suspension was adjusted to 200104/Al. The erythrocyte suspension was divided into two plastic tubes (diameter 10 mm, length 75 mm); one was incubated at 4 8C for 24 h and the other at 4 8C for 24 h and then in a water bath at 37 8C for 3 h. After each incubation, they were centrifuged at 1805g for 10 min, and the supernatants were collected, stored at 4 8C and assayed simultaneously for K+. The centrifugation did not affect K+ release of erythrocytes. The difference in K+ concentration between pre- and postincubation at 4 8C for 24 h was designated as DKr and that between post-incubation at 4 8C and after rewarming at 37 8C for 3 h was designated as DKi. The K+ flux of erythrocytes was evaluated by calculating the ratio of influx after rewarming to release after cold storage (DKi/DKr). Erythrocyte DKi/ DKr or K+ flux of healthy subjects was 0.274F0.063 [3]. The coefficient of variation (CV) of the within-day imprecision (n=20) of the assay was 2.9%. 2.3. Measurement of serum electrolytes and other values Serum K+ and other electrolytes were measured by ion-selective electrodes (Jokoh electrolyte analyzer EX-180, Jokoh, Tokyo). Other biochemical values, including serum creatinine, were measured by means of a TBA-30FR (Toshiba Medical, Tokyo). Serum concentrations were expressed as the mean of results of four measurements in the preceding month. Serum K+ values of the patients were examined retrospectively for 6 months before the K+ flux experiment to determine whether they fluctuated or not.
2.2. Assay for K+ flux 2.4. Assay for erythrocyte K+ concentration +
K flux was assayed according to the previously described procedure [2]. Briefly, venous blood was collected in heparin-Li-containing tubes (Sekisui
To measure erythrocyte K+ concentrations, heparinized blood samples were drawn from 30 hemodial-
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
191
ysis patients (male 12, female 18). The blood cells, including erythrocytes, were counted using a Sysmex analyzer, and the samples were centrifuged at 1805g for 30 min. Plasma, buffy coat and the upper layer of erythrocytes were carefully removed, and 0.5 ml of packed erythrocytes was mixed with 9.5 ml of distilled water and allowed to hemolyze completely. After centrifugation, the K+ concentration in the supernatant was measured, and the erythrocyte K+ concentration was calculated. The amount of trapped plasma was regarded as constituting 2.5% of the packed erythrocyte volume. 2.5. Statistical analysis Data were analyzed using the paired and unpaired Student’s t-tests and Pearson’s correlation coefficient.
3. Results Non-diabetic hemodialysis patients classified into three groups according to their serum K+ concentrations were compared for various factors, including the DKi/DKr ratio. As shown in Table 1, no significant differences were observed in age, duration of hemodialysis treatment, serum pH and Na+ and
Table 1 Parameters of three groups of non-diabetic hemodialysis patients with different serum K+ concentrations Mean serum K+ (mmol/l)
N (male, female) Age (years) Duration of hemodialysis treatment (years) Serum pH Mean serum Na+ (mmol/l) Erythrocyte K+ (mmol/l) DKi/DKr
Low (K+b0) Middle (K+=5.0 or b5.5)
High (K+z5.5)
10 (3, 7) 10 (5, 5) 66.2F14.32# 57.1F15.38 7.0F6.78 10.5F6.36
10 (4, 6) 64.1F16.16 6.0F2.81
7.36F0.038 140.2F1.64
7.33F0.033 140.9F2.77
7.35F0.043 138.8F2.62
110.0F8.54
109.4F3.88
110.0F5.21
0.269F0.041 0.228F0.041* 0.213F0.032**
* Pb0.05 vs. low. ** Pb0.005 vs. low. # MeanFS.D.
Fig. 1. Relationship between the DKi/DKr ratio and the mean serum K+ concentration in non-diabetic (A) and diabetic (B) hemodialysis patients. The serum K+ concentration is expressed as the mean of four measurements in the preceding month before estimation of the ratio.
erythrocyte K+; however, the DKi/DKr ratios in the medium and high K+ groups were significantly lower than those in the low K+ group ( Pb0.05, Pb0.005, respectively). K+ concentrations in sera of the patients in each group did not change to other groups in the 6 months of observation. The relationships between the DKi/DKr ratio and mean serum K+ concentration in diabetic and nondiabetic hemodialysis patients were compared. As shown in Fig. 1A, a negative correlation (r= 0.482, Pb0.01) was observed between the DKi/DKr ratio and the mean serum K+ concentration in 30 non-diabetic
192
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
hemodialysis patients (male 12, female 18). However, a significant correlation was not observed in the 13 diabetic hemodialysis patients (male 7, female 6) (Fig. 1B). No significant differences were observed between male and female patients and between patients receiving erythropoietin or not. To investigate the effect of hemodialysis, the DKi/ DKr ratio and erythrocyte K+ concentration before and after hemodialysis were measured in 11 nondiabetic patients. The mean DKi/DKr ratio of erythrocytes collected after hemodialysis (0.274F0.040) increased significantly ( Pb0.05) compared with that of the erythrocytes collected before the treatment (0.263F0.036) (Fig. 2). The erythrocyte K+ concentrations before hemodialysis treatment were scattered over a wide range, and they converged into a narrower range after hemodialysis (Fig. 3). Among 11 cases examined, decreases after hemodialysis were observed in seven cases and, in contrast, increases in four cases. Totally, the mean values did not differ significantly before (113.5F8.82 mmol/l) or after hemodialysis (110.0F3.71 mmol/l). No significant differences in the changes between DKi/DKr ratio and
Fig. 3. Comparison of the erythrocyte K+ concentration in hemodialysis patients before and after hemodialysis.
erythrocyte K+ concentration were observed after the treatment.
4. Discussion
Fig. 2. Comparison of DKi/DKr ratios in hemodialysis patients before and after hemodialysis.
In measuring extracellular K+ concentration of erythrocyte suspension incubated at 4 8C for K+ release followed by incubation at 37 8C for the influx of erythrocytes, the amount of K+ influx was consistently reduced compared with that of the release [2]. These results led us to speculate that some component in erythrocytes was damaged by the cold exposure and that the data obtained by assays performed in such cold conditions might sometimes be inaccurate. We therefore developed a simple method to estimate K+ flux of erythrocytes by measuring the DKi/DKr ratio, which reflects Na+/ K+-exchanging ATPase activity [3]. Serum K+ concentrations are recognized to be higher in chronic hemodialysis patients; however, among the patients in our hospital, they varied and were classified into three groups according to their concentrations. Because diabetic patients receive anti-
Y. Ozawa et al. / Clinica Chimica Acta 350 (2004) 189–193
diabetic therapies including insulin that might affect K+ metabolism [5], they were not included in the comparative analyses shown in Table 1. In nondiabetic patients, no significant differences were observed in various factors. However, the K+ flux was the lowest in high serum K+ group, which suggests that erythrocytes participate in the regulation of K+ metabolism in chronic hemodialysis patients. As erythrocyte K+ concentrations were not different among the three groups, it could be speculated that the K+ flux of erythrocytes could indicate that of other cells including muscle cells. In patients with chronic renal diseases, Na+/K+exchanging ATPase activity in erythrocytes was decreased [6]. We observed in this study that the K+ flux of erythrocytes was reduced in hemodialysis patients with high serum K+ concentrations. In the hemodialysis patients, a negative correlation between the mean serum K+ concentration and DKi/DKr ratio was observed in the non-diabetic patients, but not in the diabetic patients. These findings suggest that some factors in sera with high K+ concentrations affect the K+ flux of erythrocytes in non-diabetic patients. However, it appears more complicated in diabetic patients. Active transport systems in various cells including erythrocytes may play a role in the regulation of Na+ and K+ metabolism. It was reported that the elevation of serum K+ and intracellular Na+ concentrations activates Na+/K+-exchanging ATPase of erythrocytes [7], but the effects of the changes of internal K+ and external Na+ concentrations were considered to be very small [8]. However, our results showed that the DKi/DKr ratio in non-diabetic hemodialysis patients with higher serum K+ concentrations was the lowest, and the mean concentrations of erythrocyte K+ were almost the same in the three groups despite different serum K+ concentrations. We suspect that reduced erythrocyte K+ flux activity could be one of the causes of hyperkalemia in nondiabetic hemodialysis patients; however, other mechanisms might participate in the regulation of intracellular K+ concentration. In cases with reduced K+ flux of erythrocytes, Na + flux might be also decreased, which might affect physiological conditions including blood pressure. We have observed a
193
correlation between K+ flux and hypertension in our preliminary experiment (data not shown). After hemodialysis treatment, various pathological conditions are corrected to varying degrees, which include abnormal serum electrolyte concentrations, metabolic acidosis, water excess, and accumulation of uremic toxins. Our results showed that the DKi/DKr ratios were increased significantly and the erythrocyte K+ concentrations were reduced to a narrower range after hemodialysis treatments. The convergence of the internal K+ concentrations seems to be induced as a response to the surrounding conditions. It is reasonable to speculate that some pathologic factors in the sera could affect the K+-flux activity of erythrocytes as indicated by the low DKi/DKr ratios. We conclude from our results that erythrocytes participate partly in the regulation of serum K+ concentration in non-diabetic chronic hemodialysis patients.
References [1] Harris JE. The influence of the metabolism of human erythrocytes on the potassium content. J Biol Chem 1941; 141:579 – 95. [2] Ozawa Y, Sato S, Nishi S, Shishido S, Imafuku Y, Nagai T, et al. Effect of temperature and drugs on potassium flux in human erythrocytes. In: Ohba Y, Kanno T, Okabe H, Totani M, Nakahara K, Ichihara K, editors. Quality control in the clinical laboratory ’95. Tokyo7 Excepta Medica, 1995. p. 191 – 6. [3] Ozawa Y, Sato S, Matsumoto K, Nishi S, Shishido S, Imafuku Y, et al. Determination of potassium flux activity of viable human erythrocytes by measuring the release-influx ratio. Clin Chim Acta 1999;279:125 – 32. [4] Kunie CL, Charney AN. Potassium and renal failure. Compr Ther 1981;7:29 – 33. [5] Rahmani-Jourdheuil D, Mourayre Y, Vague P, Boyer J, JuhanVague I. In vivo insulin effect on ATPase activities in erythrocyte membrane from insulin-dependent diabetics. Diabetes 1987;36:991 – 5. [6] Cole CH. Decreased ouabain-sensitive adenosine triphosphatase activity in the erythrocyte membrane of patients with chronic renal disease. Clin Sci Mol Med 1973;45:775 – 84. [7] Sachs JR, Welt LG. The concentration dependence of active potassium transport in the human red blood cell. J Clin Invest 1967;46:65 – 76. [8] Knight A, Taylor RC, Welt LG. Autoregulation of sodium and potassium concentration in erythrocytes. Clin Res 1971;19:86.