Evaluating extremely low plasma ascorbate levels and reduction of plasma 1 ascorbate levels by dialysis in Japanese hemodialysis patients

Background: Low plasma ascorbate levels in hemodialysis patients have been reported 29 worldwide; hence, many end-stage kidney disease patients are forced to restrict their diets, 30 especially potassium-rich fruits and vegetables, to prevent hyperkalemia. In this study, we 31 aimed to clarify whether plasma ascorbate levels are low in Japanese dialysis patients and 32 whether plasma ascorbate levels fluctuate before and after dialysis. In addition, we aimed to 33 clarify whether there are clinical test items that have a causal relationship with plasma ascorbate 34 levels. 35 Methods: Plasma ascorbate levels in 27 chronic kidney disease (CKD) stage G3–G5 patients 36 (mean age 84 years) and pre- and post-dialysis plasma ascorbate levels in 19 CKD stage G5D 37 hemodialysis patients (mean age 79 years) were determined using high-performance liquid 38 chromatography and electrochemical detection. 39 Results: Pre-dialysis plasma ascorbate levels in hemodialysis patients (12.0 ± 1.4 µ M) were 40 significantly lower (by 56%) than those in CKD stage G3–G5 patients (27.1 ± 2.7 µ M). After 41 dialysis, there was a 40% reduction in plasma ascorbate levels. Moreover, pre-dialysis ascorbate 42 levels correlated significantly with plasma potassium levels. Conclusions: The study results indicate that Japanese hemodialysis patients have lower plasma 44 ascorbate levels than CKD stage G3–G5 patients and that these low plasma ascorbate levels in 45 hemodialysis patients were further reduced by hemodialysis. To avoid the development of 46 scurvy in hemodialysis patients, it is necessary to take sufficient ascorbate from supplements 47 or medicines. 48 ascorbate and other clinical

Vitamin C (L-ascorbic acid) is a water-soluble micronutrient and antioxidant that scavenges 55 reactive oxygen species [1][2][3]. Under physiological pH conditions, ascorbic acid most 56 commonly exists in its mono-anion form, ascorbate [4]. In addition to its antioxidant property, 57 ascorbate contributes to numerous well-defined enzymatic reactions involving collagen 58 hydroxylation, carnitine and norepinephrine biosynthesis, tyrosine metabolism, and peptide 59 After reduction, the reaction mixture was diluted with 5% metaphosphoric acid containing 0.5 135 mmol/L EDTA and analyzed for total ascorbate by high-performance liquid chromatography 136 coupled with electrochemical detection. Separation was achieved on an Atlantis dC18 5-µm 137 column (4.6 × 150 mm) combined with an Atlantis dC18 5-µm guard column (4.6 × 20 mm) 138 from Nihon Waters (Tokyo, Japan). The mobile phase consisted of 50 mM phosphate buffer 139 (pH 2.8), 540 µM EDTA, and 2% methanol at a flow rate of 1.3 mL/min, and electrical signals 140 were recorded using an electrochemical detector with a glassy carbon electrode at +0.6 V. All 141 electrical signal data from the electrochemical detector were collected using Waters Empower 142 2 software (Nihon Waters). The value of DHA was determined by subtracting ascorbate from 143 total ascorbate. The ascorbate level of urine is immediately affected after a meal. Therefore, we 144 evaluated the plasma ascorbate level in the samples whose urinary ascorbate was <0.5 mM. 145 146

Statistical analysis 155
The results and clinical characteristic data are expressed as means ± standard error of the mean 156 (SEM). The probability of statistical differences between experimental groups was determined 157 by Welch's t-test, paired t-test, and one-way analysis of variance (ANOVA) followed by 158 Tukey-Kramer test. We verified that Pearson correlation coefficient between ascorbate 159 concentration and clinical characteristics data is different from zero. Statistical differences were 160 considered significant at p < 0.05. 161 162

164
Clinical characteristics of CKD stage G3-G5 and hemodialysis patients 165 A total of 34 CKD stage G3-G5 and 19 CKD stage G5D hemodialysis patients were enrolled. 166 Since seven CKD stage G3-G5 patients showed that urine ascorbate level was higher than 0.5 167 mM, suggesting supplemental intake of ascorbate just before collecting blood and urine, they 168 were excluded from analysis. Therefore, the results of 27 CKD stage G3-G5 patients were used 169 for analysis. Clinical characteristics of CKD stage G3-G5 and hemodialysis patients are shown 170 in Table 1. The levels of the following were beyond the normal range and differed between 171 CKD stage G3-G5 and hemodialysis patients: albumin, blood urea nitrogen, creatinine, 172 phosphorus, and β2-microglobulin. Albumin values of both CKD stage G3-G5 and 173 hemodialysis patients were below the normal range and blood urea nitrogen, creatinine, and β2-174 microglobulin values of both CKD stage G3-G5 and hemodialysis patients were higher than 175 the normal range. Phosphorus values of hemodialysis-only patients were higher than the normal 176 range. 177 Plasma ascorbate levels in CKD stage G3-G5 patients and pre-and post-dialysis plasma 179 ascorbate levels in hemodialysis patients 180 The plasma ascorbate levels in 27 CKD stage G3-G5 and pre-and post-dialysis plasma 181 ascorbate levels in 19 hemodialysis patients were measured (see Fig. 1). Pre-dialysis plasma 182 ascorbate levels in hemodialysis patients (12.0 ± 1.4 µM) were significantly lower (by 56%) 183 than that in CKD stage G3-G5 patients (27.1 ± 2.7 µM) (Fig. 1a). Moreover, after dialysis, 184 there was a 40% reduction in the plasma ascorbate levels (7.2 ± 0.9 µM) (

Relationships between clinical characteristics and plasma ascorbate levels 190
We then analyzed the relationships between clinical characteristics and pre-dialysis plasma 191 ascorbate levels in hemodialysis patients (Fig. 2). In Pearson correlation coefficient, pre-192 dialysis ascorbate levels correlated significantly with those of plasma potassium levels (positive 193 correlation; Pearson correlation coefficients (r) = 0.6; p = 0.006) (  198 199 In this study, we revealed that Japanese hemodialysis patients have low plasma ascorbate levels 200 compared to non-hemodialysis CKD patients and these low plasma ascorbate levels in 201 hemodialysis patients were further reduced by a single hemodialysis treatment session. 202

Discussion
Moreover, we found that ascorbate levels in hemodialysis patients correlated with those of 203 plasma potassium levels. In general, hemodialysis patients are forced to restrict their diets, 204 especially potassium-rich fruits and vegetables, to prevent hyperkalemia, which is a risk factor 205 for dialysis morbidity and mortality [22]. Most of these fruits and vegetables also contain high 206 amounts of ascorbate. Thereby, plasma ascorbate levels in Japanese hemodialysis patients 207 might be correlated with plasma potassium levels. 208 The recommended dietary allowance (RDA) of vitamin C for a healthy adult is 100 mg per day 209 in Japan and 90 mg and 75 mg per day for men and women, respectively, in the United States 210 to prevent scurvy [23]. The average concentration of ascorbate in the plasma of healthy humans 211 is 40-60 µM [24,25]. When the plasma ascorbate concentration drops to below 11 µM, there 212 is a risk of developing scurvy, which is thus conventionally considered deficient [24,25]. In 213 our previous report regarding chronic obstructive pulmonary disease (COPD) and plasma 214 ascorbate levels, we reported that plasma ascorbate levels were significantly lower in COPD 215 patients (mean age 72.7 ± 6.9 years) than those in healthy elderly people (mean age 68.8 ± 3.8 216 years) using the same procedure and method as here [20]. Plasma ascorbate levels in COPD 217 patients and healthy elderly people were 31.2 ± 2.2 µM and 42.3 ± 2.9 µM, respectively. 218 Furthermore, the observed plasma levels in non-hemodialysis CKD patients and hemodialysis 219 patients in the present study were lower than those in COPD patients. Plasma ascorbate levels 220 in non-hemodialysis CKD patients and hemodialysis patients were 27.1 ± 2.7 µM and 12.0 ± 221 1.4 µM, respectively. Since there is a risk of developing scurvy when the plasma ascorbate 222 concentration drops to below 11 µM [24,25], many Japanese hemodialysis patients are likely 223 develop scurvy (see Fig. 3). Worldwide, many hemodialysis patients have developed scurvy 224 [10,26,27]. 225 Moreover, we tried to compare the percentages of DHA per total ascorbate in plasma from 226 healthy elderly people, COPD patients, non-hemodialysis CKD patients, and hemodialysis 227 patients, and found that the percentages of DHA in non-hemodialysis CKD patients (33.5%) 228 and hemodialysis patients (37.4%) were notably higher percentages than those in COPD 229 patients (12.4%) and healthy elderly people (10.0%) (see Fig. 3) [20]. High percentage of DHA 230 in non-hemodialysis CKD patients and hemodialysis patients may reflect a higher oxidative 231 stress levels in their body. 232 Wang et al. [16] reported that plasma ascorbate concentrations were reduced by a median of 233 33% following dialysis. Deicher et al. [28] have also reported that hemodialysis causes a 50-234 75% decrease in plasma ascorbate levels. In the present study, plasma ascorbate levels reduced 235 to 40% by hemodialysis. Thus, hemodialysis certainly reduces plasma ascorbate concentration 236 in hemodialysis patients. 237 For a long time, there has been concern about the accumulation and deposition of oxalate with 238 increased intake of vitamin C because oxalate is a breakdown product of vitamin C and is 239 heavily excreted by the kidneys [29]. Oxalate crystallization occurs at levels above 30 mM [30] 240 and high plasma oxalate levels were seen in hemodialysis patients [31][32][33]. However, in a recent 241 prospective case series exploring high-dose intravenous vitamin C (15-100 g) administration, 242 increased vitamin C intake was not associated with any cases of symptomatic renal stones and 243 kidney injury [34]. Moreover, significant side effects of vitamin C are not reported in any of 244 the mentioned controlled trials, including the most recent VITAMIN randomized trial [35]. 245 CKD patients with higher levels of plasma calcium, phosphate, and parathyroid hormone have 246 a high risk of death because CKD often causes abnormal calcium and phosphate metabolism 247 and hyperparathyroidism [36][37][38][39][40]. Therefore, it is important to control the plasma calcium, 248 phosphate, and parathyroid hormone levels in the non-hemodialysis CKD and hemodialysis 249 patients [41]. Through a systematic review and meta-analysis, Ke et al. [42] have reported that 250 vitamin C supplementation in CKD patients has no positive effect that influence the plasma 251 phosphate or parathyroid hormone levels, but it increase plasma calcium levels in the short term. 252 In the present study, we could not detect any correlation between plasma ascorbate and plasma 253 calcium, phosphate, and parathyroid hormone levels in Japanese hemodialysis patients. 254 Meanwhile, we only found the positive correlation between plasma ascorbate and plasma 255 potassium levels in Japanese hemodialysis patients. Perhaps Japanese hemodialysis patients 256 that have dietary potassium restrictions to prevent hyperkalemia may limit their consumption 257 of fresh vegetables and fruits that are rich in ascorbate. Thereby, there is a possibility that 258 Japanese hemodialysis patients have low plasma ascorbate levels. 259 Recently, the increase of frailty in the elderly has become a social problem globally. The 260 Dialysis Morbidity and Mortality Wave 2 cohort study revealed that >60% of end-stage kidney 261 disease patients over the age of 40 met a definition of frailty, which impairs the prognosis [43].    Table 2. ascorbate in the plasma of healthy humans is 40-60 µM (blue zone). There is a risk of 442 developing scurvy when the plasma ascorbate concentration drops to below 11 µM (red zone). 443 Values are expressed as a mean ± SEM. P < 0.05 by one-way ANOVA followed Tukey-Kramer 444 test. 445