Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jun 1.
Published in final edited form as: J Am Coll Nutr. 2013 Jun;32(3):187–193. doi: 10.1080/07315724.2013.791167

A Convenient Method for Measuring Blood Ascorbate Concentrations in Patients Receiving High-Dose Intravenous Ascorbate

Yan Ma 1,2, Garrett G Sullivan 2,3,4, Elizabeth Schrick 2, In-Young Choi 5,6,7, Zhuoya He 8, JoAnn Lierman 5, Phil Lee 5,7, Jeanne A Drisko 2,*, Qi Chen 1,2,*
PMCID: PMC3725640  NIHMSID: NIHMS469641  PMID: 23885992

Abstract

Objective

A simple method of using fingerstick blood glucose monitors (FSBG) to estimate blood ascorbate values after high-dose intravenous (IV) ascorbate infusion is evaluated as a substitution for HPLC measurement.

Methods

In 33 participants, readings from FSBG were taken before and after IV ascorbate infusions at various time points, with the post-infusion FSBG readings subtracted by the baseline glucose readings. The results of the subtractions (AAFSBG) were correlated with ascorbate concentrations detected by HPLC (AAHPLC).

Results

A linear regression was found between ascorbate concentrations detected by the fingersitck method (AAFSBG) and by HPLC (AAHPLC). The linear correlations were identical in healthy subjects, diabetic subjects and cancer patients. ANOVA analysis obtained an AAFSBG/AAHPLC ratio of 0.90, with 90% confidence interval of (0.69, 1.20). The corrections of AAFSBG improved similarity to AAHPLC, but did not significantly differ from the un-corrected values.

Conclusion

The FSBG method can be used as an approximate estimation of high blood ascorbate concentration after IV ascorbate (>50 mg/dL, or 2.8 mM) without correction. However this measurement is not accurate in detecting lower or baseline blood ascorbate. It is also important to highlight that in regard to glucose monitoring, FSBG readings will be erroneously elevated following intravenous ascorbate use and insulin should not be administered to patients based on these readings.

Keywords: Intravenous vitamin C, Ascorbate (Vitamin Cascorbate acid), blood glucose monitor, HPLC, complementary, alternative, integrative medicine

INTRODUCTION

For many years, high-dose intravenous (IV) ascorbate (Vitamin C or ascorbic acid) has been used by complementary, alternate and integrative medicine (CAIM) practitioners to treat various conditions including infections, cancer, autoimmune diseases and illnesses of uncertain origin [1–6]. Clinical use of IV ascorbate is growing as recent studies revealed the scientific basis for this use [4, 7–10]. Although the actual effective clinical dose of IV ascorbate has not yet been established by rigorous clinical studies, a target plasma concentration of 350 ~ 400 mg/dL (20~23 mM) is recommended based on clinical experience for effectiveness and safety considerations.

As the clinical use grows, monitoring of plasma ascorbate concentrations in patients receiving IV ascorbate remains of prime importance. High performance liquid-chromatography (HPLC) with coulometric electrochemical detection or UV detection has been the gold standard for detection of ascorbate in biological samples [11,12]. The HPLC method is highly sensitive and accurate, with great advantages in detecting low physiological ascorbate concentrations. However, because of the instability of ascorbate in biological fluids [13–15], samples must be immediately and properly processed to prevent ascorbate degradation [11,12]. Accurate measurement by HPLC requires strict sample processing/extraction methods, low-temperature (−80°C) storage and low-temperature sample handling (4 ~ −20°C) [11,12]. In addition, the sophisticated and expensive HPLC instruments are not often accessible on a daily basis on site where IV ascorbate infusions are performed. Blood samples need to be drawn and immediately prepared, frozen and shipped to laboratories equipped with HPLC instruments for further analysis. This process is complicated and costly and might ultimately lead to ascorbate degradation and faulty laboratory readings if samples are mishandled at any step along the chain.

In this study, we investigated a fast and easy method to assess blood ascorbate concentrations in patients receiving IV ascorbate. Commercially available fingerstick blood glucose (FSBG) monitoring is a fast way of testing blood glucose as commonly applied in diabetic patient care. In the clinical setting, we observed that glucose readings with glucometer FSBG testing resulted in very high readings following IV ascorbate infusions. By concurrent laboratory verification of serum glucose levels, these high FSBG readings have been found to be erroneously elevated. This suggests that ascorbate interferes with accurate glucose measurement, most likely because of the structural similarity of ascorbate and glucose, and interference of ascorbate with catalytic oxidation-based reactions [16, 17]. We became intrigued with the idea that this interference could be used as a quick estimation for the blood ascorbate level after IV ascorbate infusions as an alternative to HPLC. We hypothesize that the difference between FSBG readings before and after IV ascorbate infusion could be used as a rough estimate for blood ascorbate concentration in clinical IV ascorbate use.

MATERIALS AND METHODS

Subjects, blood sampling, and blood glucose readings taken from FSBG

A total of 33 subjects were included in this analysis. All study participants were consented prior to the study. Healthy and diabetic subjects were originally recruited for a study evaluating biomarkers of oxidative stress and IV ascorbate pharmacokinetics (KUMC IRB protocol HSC# 11119). For 19 of these subjects (9 healthy, 10 diabetic), blood samples for ascorbate and glucose were taken before IV ascorbate infusion, at the completion of IV ascorbate infusion, and then at 2 hrs, 2.5 hrs, 3 hrs, 3.5 hrs and 4 hrs post-infusion, in addition to the FSBG readings. The 14 oncology participants were receiving high-dose IV ascorbate treatment as part of outcome studies (KUMC IRB HSC#7823 and HSC#10006). For 13 of the oncology participants, blood samples for ascorbate were taken only at the completion of the IV infusion, in addition to the FSBG readings. For the remaining 1 oncology participant, blood samples for ascorbate were taken both before and at the completion of IV ascorbate infusions, in addition to the FSBG. The blood samples were put on ice and immediately transferred to the research laboratory and centrifuged at 4°C at 1500 × g for 10 min. Plasma was taken and stored at −80°C until analysis by HPLC for detection of ascorbate. Fingerstick glucose monitors (Accu-Chek by Roche, and Onetouch Ultra 2 by Lifescan) were used at each time point when blood was drawn, and readings were recorded.

HPLC analysis of blood ascorbate

The sample processing and HPLC analysis were performed according to a method established by Levine et al [11,12]. Plasma was diluted 5~1000 times in 90% methanol with 1 mM EDTA, vortex, and then centrifuged at 4°C at 20,000 × g for 15 min. Supernatant was analyzed by Waters e2695 HPLC (Waters, Milford, MA) with electrochemical detection (ESA, Chelmsford, MA). The column used was an ODS-DABS Ul Trasphere, 5μ, 4.6 mm × 25 mm (Beckman Coulter, Fullerton, CA). Mobile phase was run at a flow rate of 1 mL/min, containing 30% methanol, 0.05 M sodium phosphate monobasic, 0.05 M sodium acetate, 189 μM dodecyltrimethyl-ammonium chloride, and 36.6 μM tetraoctylammonium bromide. The Empower II software (Waters, MA) was used for instrument control and data analysis.

Data analysis and Statistics

To estimate the blood ascorbate concentrations from the FSBG readings, the FSBG readings of each sample point were subtracted by the baseline glucose readings of the relative subject: AAFSBG = FSBGX − FSBG0 (AAFSBG is the estimated blood ascorbate concentration, FSBGX is the FSBG reading at a given time point, FSBG0 is the FSBG reading before IV infusion of ascorbate).

A total of 114 paired data-points from 29 subjects (7 healthy, 8 diabetic and 14 cancer) were used for correlation analysis and ANOVA analysis. To assess if the new method FSBG is comparable to the standard HPLC method, a linear mixed effects models was performed on log-transformed concentrations with sample, method, and the interaction of sample and method as fixed effects, and measurements within each subject as repeated measures. Point estimates and 90% confidence interval (CI) for differences on the log scale was exponentiated to obtain estimates for ratios of geometric means on the original scale. HPLC method was used as the reference in the comparisons.

After a function was established from the correlation analysis, additional 22 pairs of data-points from 4 additional subjects (2 healthy, 2 diabetic) were used to test accuracy and reliability of the function.

RESULTS

Correlation and difference between the results obtained from FSBG measurement and HPLC detection

Ascorbate concentrations from the FSBG measurement were calculated using: AAFSBG = FSBGX- FSBG0 ( AAFSBG is the estimated blood ascorbate concentration from FSBG, FSBGX is the FSBG reading at a given time point, FSBG0 is the FSBG reading before IV infusion of ascorbate). Total of 114 pairs of data from 29 subjects (7 healthy, 8 diabetic and 14 cancer) were obtained by both FSBG method (AAFSBG) and HPLC method (AAHPLC). With 14 subjects, we were able to obtain both AAFSBG and AAHPLC values at all the indicated time points from 0 hr (beginning of ascorbate IV infusion) to 6 hrs after ascorbate IV infusion (Fig 1A). A similarity of the time-course curves is found between values from HPLC method and those from the FSBG method (Fig 1A).

Fig. 1. Correlation and difference between FSBG and HPLC measurement for blood ascorbate concentrations.

Fig. 1

Open in a new tab

A. Time-dependent plasma ascorbate values after ascorbate IV infusion detected by fingerstick blood glucose monitor (FSBG) or by high-performance liquid chromatography (HPLC). The beginning of the IV infusion was defined as 0 hr. Each point represents data from 14 participants ± standard deviation (SD). The IV does was 1g/kg body weight of the participant, and was complete approximately at 2 hrs. B. Linear correlation between blood ascorbate concentrations detected by FSBG (AAFSBG) and by HPLC (AAHPLC) in 29 subjects. C. Linear correlation between AAFSBG and AAHPLC in different groups of subjects. Dark circles, healthy subjects; open squares, diabetic subjects; and snow flaks, cancer patients. D. Log-transformed linear correlation between Ln (AAFSBG) and Ln ( AAHPLC) in 29 subjects. E. Time-dependent plasma ascorbate values of 4 additional participants after ascorbate IV infusion for the validation of the FSBG method. FSBGcorrected represents the correction of AAFSBG from previous regression analysis (function gained from Fig 1B: AAFSBGcorrected = 0.9535 AAFSBG + 29.778). FSBG/0.9 represents the correction of AAFSBG using the AAFSBG/AAHPLC ratio of 0.9.

A linear relationship was found between AAFSBG and AAHPLC for all the subjects at all time points (Fig 1B, r2=0.79). The liner correlation held true in healthy subjects (n=7, r2=0.76), diabetic subjects (n=8, r2=0.73), and cancer patients (n=14, r2=0.71) (Fig 1C). The linearity was identical between the three groups as tested by ANOVA analysis (P>0.05). All the data points were then log-transformed, and linear relationship still held true (Fig 1D, r2=0.88). However, when Ln(AAHPLC) was 2, equivalent to blood ascorbate concentration ~7.4 mg/dL or ~420 μM, the FSBG method could not detect the ascorbate concentration (Fig 1D).

ANOVA analysis using a linear mixed effects model was performed on log-transformed data. The model assumptions were evaluated graphically, that the residuals were normally-distributed and had constant variance. Both assumptions were met. There was no interaction between sample and methods (P = 0.54). Method effect was not significant (P=0.31), whereas sample effect was highly significant (P <0.0001). The difference between FSBG method and HPLC method on the log scale is −0.09 with 90% confidence interval (−0.38, 0.18). Back-transformed to the original scale, the AAFSBG/AAHPLC estimated ratio was 0.90 with 90% confidence interval (0.69, 1.20). Taken as a whole, the two different methods had no significant difference in detecting ascorbate concentration.

Correction of the AAFSBG values

We hypothesized the difference between AAFSBG and AAHPLC could be corrected with the function obtained from the linear regression in Fig 1B: y=0.9535x + 29.778. The corrected AAFSBG-corrected values were compared to AAHPLC values. Because ANOVA analysis showed an AAFSBG/AAHPLC ratio of 0.90, we also compared the AAFSBG/0.90, values with the AAHPLC values. The AAFSBG-corrected values better agreed with the AAHPLC values ( P=0.999) than either AAFSBG or AAFSBG/0.9 (P=0.11). When compared to AAHPLC, paired t-test showed that all the corrected values and the un-corrected AAFSBG has no significant difference. The P value between AAFSBG and AAHPLC was 0.068, between AAFSBG-corrected and AAHPLC was 0.999, and between AAFSBG/0.90 and AAHPLC it was 0.112. This indicates the corrected data resembles AAHPLC more closely. When the 2 correction methods were compared to each other, they were identical as the P value was 0.548. When the corrected values were compared to the un-corrected values, not significant difference was found either (P= 0.548 and 0.182). This no-difference before and after correction was also evident as shown in Fig 1A. This indicates that corrections were not necessary for the FSBG measurement for blood ascorbate concentration.

Validation of the estimation method of FSBG

To further validate this estimation method and the correction equation gained from the liner correlation, 18 pairs of data-points from 4 more subjects (2 normal and 2 diabetic) were used. The same data were plotted as time-dependent plasma ascorbate values with FSBG method, HPLC method, and the corrections of FSBG values (Fig 1E). No significant difference between AAFSBG and AAHPLC (P=0.14) indicated again that the FSBG method can be a useful estimation. The corrections (both AAFSBG-corrected and AAFSBG/0.9) improved the similarity to AAHPLC by increased P values (P=0.87 and 0.45 respectively), but still showed no significant difference when compare to AAFSBG (P=0.60 and 0.37 respectively).

Consistent with results described above, the corrections did not improve the compliance with HPLC data, thus were not necessary in assessing ascorbate levels. When plasma ascorbate level was above 50 mg/dL, the error of FSBG methods was 13% (±8%) using HPLC results as standards. When plasma ascorbate levels were below 1 mg/dL (baseline), the error was large (>100%) that the FSBG methods cannot detect the ascorbate concentration correctly. Moreover, there were 3 data pairs that had errors >100% in the range of 40 – 80 mg/dL, resulting in blood ascorbate determination discrepancy between the two methods of approximately 100 mg/dL. These data points were from one subject and may have been confounded because the subject ate just before these 3 samples were taken. Therefore, the blood glucose levels might be different from the baseline glucose level that was used to calculate the ascorbate levels.

Taken together, the AAFSBG values can provide a rough estimation for blood ascorbate concentrations after IV ascorbate infusion. Corrections examined here did not improve the compliance with HPLC results, and thus were indicated not necessary. The FSBG method slightly underestimates blood ascorbate concentration compared to HPLC method (AAFSBG/AAHPLC ratio is 0.90 from this study). The FSBG method works better at a high blood ascorbate concentration (> 50 mg/dL, or 2,800 μM) after IV ascorbate infusion. When blood ascorbate concentration is below ~7.4 mg/dL (or 420 μM), the FSBG method cannot detect the concentration correctly.

DISCUSSION

Recent studies have revealed the scientific basis and the scope for the use of vitamin C beyond its use as a nutritional supplement. IV ascorbate but not oral vitamin C produced plasma concentrations sufficiently high (300 – 20,000 μM) to encounter pro-oxidant effect in the extracellular milieu, thus is able to deliver treatment for diseases in which peroxide may have a therapeutic role, such as cancer or infections [7–9]. High-dose IV ascorbate is widely adopted in the CAIM community [1–6]. About 10,000 patients per year in US are using IV ascorbate, with an average dose of 28 grams/injection and 22 total doses per patients [4]. However the dosage used varies largely from 1 to 200 grams per treatment. A standard blood concentration level needs to be established for therapeutic purposes under different clinical conditions. Unfortunately, the recommended therapeutically effective blood levels of ascorbate are currently only dependent on limited human and animal pharmacokinetics data [8–10,18], and largely depend on CAIM practitioners’ experience [4]. Clinical studies are needed to establish standards for administration of high-dose IV ascorbate.

Despite the lack of a standardized therapeutic level, the blood ascorbate monitoring in high-dose IV ascorbate therapy is vital in the planning and management of ascorbate dosing by healthcare providers. HPLC separation with electrochemical detection is the gold standard for determining blood ascorbate concentration, which is reliably accurate and reproducible. However, the HPLC detection for ascorbate demands instruments and specifically-trained personnel which are less likely to be available for private practices without access to a general clinical laboratory. Given that ascorbate is unstable and is rapidly oxidized in aqueous systems [13–15, 19], it is recommended that blood samples containing ascorbate should be processed under 4°C immediately after blood draw, and plasma thus obtained should be immediately processed to extract ascorbate at 4°C [11,12]. If HPLC analysis of the plasma is not immediately available, plasma storage should be at −80°C [11,12]. Hemolysis, which could lead to ascorbate degradation during blood draw and/or sample processing, should also be carefully avoided [12]. Currently, many clinical samples are shipped to specific labs equipped with HPLC instruments for the ascorbate analysis. The turnaround time and handling conditions are not always ideal, which can lead to inaccurate readings.

Our study provides an easy, on-site measurement for blood ascorbate levels after high-dose IV ascorbate administration. Figure 2 represents a diagram for the different and simplified procedures of the new FSBG method, in comparison with the traditional HPLC method. As compared with the HPLC detection, FSBG test is simple and fast and blood draw and handling are avoided. Sample preparation and storage are not required, and the results are available in minutes. The results from the FSBG test hold a liner relationship to the results from HPLC detection. In our study, participants were divided into 3 groups – normal subjects, diabetic, and cancer patients. The FSBG worked in all three groups without notable difference (Fig 1D). There was only an in-significant difference in slopes of the liner regression. Studies have reported that in diabetic patients, the baseline blood ascorbate levels can be lower than healthy subjects [20]. However this difference cannot be detected in this study as the normal baseline levels of ascorbate is ~50μM and FSBG method is not sensitive and not meant for detection of ascorbate concentrations lower than 7.4 mg/dL (or 420μM) (Fig 1C). In fact, the in-sensitivity of FSBG to physiological ascorbate concentrations is desirable for the reliability of FSBG in measuring blood glucose. In contrast, when high-dose IV ascorbate is used, blood concentrations could easily elevate to > 50 mg/dL [4, 10]. Our study showed in these settings, the FSBG method could provide a quick and useful reference for practitioners to decide IV ascorbate dosing for a particular patient.

Fig. 2. Comparison of the procedures between HPLC method and FSBG method for detection of blood ascorbate concentration after intravenous ascorbate infusion.

Fig. 2

Open in a new tab

The HPLC method requires blood-draw, multi-step low temperature sample handling and HPLC equipment and software for data analysis. To avoid ascorbate degradation, hemolysis during blood-draw and sample handling should also be carefully avoided. The FSBG method takes pre- and post-infusion readings. Ascorbate concentration can be resulted simply by subtraction of pre-infusion reading from post-infusion reading.

However in spite of its convenience, there are drawbacks for the use of FSBG test for ascorbate levels. It is not as accurate as HPLC method and it cannot detect low concentrations of ascorbate. Our data showed that the FSBG method can estimate ascorbate concentrations > 50 mg/dl (or 2,800 μM), but cannot correctly measure concentrations at baseline (< 1 mg/dL or 57 μM), or when blood ascorbate levels are only slightly elevated (e.g. < 7.4 mg/dL or 420μM). The minimum level that can be detected by the FSBG method cannot be concluded from these data, as concentrations between 7.4 mg/dl and 50 mg/dl were not tested. This was due to the original design of the human studies was limited to pre-decided high dosages of IV ascorbate and not using a moderate dose. Also, the FSBG method is an approximate estimation rather than a precise measurement. The benefit is for its accessibility, simplicity and speed.

In the use of the FSBG method for detection of high ascorbate concentrations, a correct glucose value for subtraction is very important. If blood glucose level might significantly change after the baseline glucose reading (e.g. due to food ingestion by the patient during the IV ascorbate infusion), then ascorbate measurement by FSBG could be confounded. A correct glucose level at the time of measurement should be detected by a conventional laboratory and be used rather than the baseline glucose level taken before.

Equally of great importance, practitioners should be aware that patients receiving IV ascorbate will have erroneous elevations of FSBG readings for several hours after infusions and should not be given insulin therapy based on these readings. This could result in a precipitous drop in blood glucose from insulin therapy based on an erroneous FSBG reading. After IV ascorbate infusions, only blood levels sent to a conventional laboratory should be used to monitor blood glucose and practitioners should not rely on FSBG.

CONCLUSION

The FSBG method can be used as an approximate estimation of high blood ascorbate concentration after IV ascorbate (>50 mg/dL, or 2.8 mM). Corrections are not necessary. However this measurement is not accurate in detecting lower or baseline physiologic blood ascorbate. In regard to glucose monitoring, FSBG readings will be erroneously elevated following IV ascorbate use and insulin should not be administered to patients based on these readings.

Acknowledgments

The research was partially supported by an NIH grant (R21 DK081079), and by a grant from The Hilton Family Foundation. The Hilton Family Foundation had no input in trial design, evaluation of data or writing of the report.

ABBREVIATIONS

CAIM

Complementary, Alternative and Integrative Medicine

FSBG

fingerstick blood glucose monitors

IV

intravenous

G6PD

glucose-6-phosphate dehydrogenises

HPLC

high performance liquid-chromatography

CI

confidence interval

Footnotes

Author Disclosure Statement

The authors have no conflicts of interest to report.

References

  • 1.Dusing RW, Drisko JA, Grado GG, Levine M, Holzbeierlein JM, Van Veldhuizen P. Imaging studies that can be used for complementary & alternative medicine clinical studies. In: Moyad M, editor. Urologic Cliniccs of North America Devoted to Complementary and Alternative Medicine in Urology. 2011. [DOI] [PubMed] [Google Scholar]
  • 2.Drisko JA, Chapman J, Hunter V. The use of antioxidants with first-line chemotherapy in two cases of ovarian cancer. J Am College Nutrition. 2003;22(2):118–123. doi: 10.1080/07315724.2003.10719284. [DOI] [PubMed] [Google Scholar]
  • 3.Gonzalez MJ, Miranda-Massari JR, Mora EM, Guzman A, Riordan NH, Riordan HD, Casciari JJ, Jackson JA, Roman-Franco A. Orthomolecular oncology review: ascorbic acid and cancer 25 years later. Integr Cancer Ther. 2005;4:32–44. doi: 10.1177/1534735404273861. [DOI] [PubMed] [Google Scholar]
  • 4.Padayatty SJ, Sun AY, Chen Q, Espey MG, Drisko J, Levine M. Vitamin C: intravenous use by complementary and alternative medicine practitioners and adverse effects. PLos One. 2010;5:e11414. doi: 10.1371/journal.pone.0011414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Levy TE. Vitamin C, Infectious Diseases, and Toxins: Curing the Incurable. Philadelphia: Xlibris; 2002. [Google Scholar]
  • 6.Rozanova TN, Zhang JZ, Heck DE. Catalytic therapy of cancer with ascorbate and extracts of medicinal herbs. Evid Based Complement Alternat Med. 2010;7:203–212. doi: 10.1093/ecam/nem159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chen Q, Espey MG, Krishna MC, Mitchell JB, Corpe CP, Buettner GR, Shacter E, Levine M. Pharmacologic ascorbic acid concentrations selectively kill cancer cells: action as a pro-drug to deliver hydrogen peroxide to tissues. Proc Natl Acad Sci U S A. 2005;102:13604–13609. doi: 10.1073/pnas.0506390102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chen Q, Espey MG, Sun AY, Lee JH, Krishna MC, Shacter E, Choyke PL, Pooput C, Kirk KL, Buettner GR, Levine M. Ascorbate in pharmacologic concentrations selectively generates ascorbate radical and hydrogen peroxide in extracellular fluid in vivo. Proc Natl Acad Sci U S A. 2007;104:8749–8754. doi: 10.1073/pnas.0702854104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Chen Q, Espey MG, Sun AY, Pooput C, Kirk KL, Krishna MC, Khosh DB, Drisko JA, Levine M. Pharmacologic doses of ascorbate act as a prooxidant and decrease growth of aggressive tumor xenografts in mice. Proc Natl Acad Sci U S A. 2008;105:11105–11109. doi: 10.1073/pnas.0804226105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hoffer LJ, Levine M, Assouline S, Melnychuk D, Padayatty SJ, Rosadiuk K, Rousseau C, Robitaille L, Miller WH., Jr Phase I clinical trial of i.v. ascorbic acid in advanced malignancy. Ann Oncol. 2008;19:1969–1974. doi: 10.1093/annonc/mdn377. [DOI] [PubMed] [Google Scholar]
  • 11.Washko PW, hartzell WO, Levine M. Ascorbic acid analysis using high-performance liquid chromatography with coulometric electrochemical detection. Anal Biochem. 1989;181:276–282. doi: 10.1016/0003-2697(89)90243-1. [DOI] [PubMed] [Google Scholar]
  • 12.Espey MG, Chen Q, Sun AY, Kim HS, Padayatty S, Wang Y, Tu H, Margolis S, Levine M. Ascorbate Methodologies and Considerations. In: Das DK, editor. Handbook of the methods for studying redox signaling. New York: Mary Ann Liebert; 2009. pp. 24–31. [Google Scholar]
  • 13.Kuge H, Rasch R, Brux B, Frunder H. Formation and structure of the ascorbate radical. Biochim Biophys Acta. 1967;141:260–265. [PubMed] [Google Scholar]
  • 14.Njus D, Wigle M, Kelley PM, Kipp BH, Schlegel HB. Mechanism of ascorbic acid oxidation by cytochrome b(561) Biochemistry. 2001;40:11905–11911. doi: 10.1021/bi010403r. [DOI] [PubMed] [Google Scholar]
  • 15.Goldenberg H. Vitamin C: from popular food supplement to specific drug. Forum Nutr. 2003;56:42–45. [PubMed] [Google Scholar]
  • 16.Chang G, Tatsu Y, Goto T, Imaishi H, Morigaki K. Glucose concentration determination based on silica sol-gel encapsulated glucose oxidase optical biosensor arrays. Talanta. 2010;83:61–65. doi: 10.1016/j.talanta.2010.08.039. [DOI] [PubMed] [Google Scholar]
  • 17.Wu L, Zhang X, Ju H. Amperometric glucose sensor based on catalytic reduction of dissolved oxygen at soluble carbon nanofiber. Biosens Bioelectron. 2007;23:479–484. doi: 10.1016/j.bios.2007.06.009. [DOI] [PubMed] [Google Scholar]
  • 18.Padayatty SJ, Sun H, Wang Y, Riordan HD, Hewitt SM, Katz A, Wesley RA, Levine M. Vitamin C pharmacokinetics: implications for oral and intravenous use. Ann Intern Med. 2004;140:533–537. doi: 10.7326/0003-4819-140-7-200404060-00010. [DOI] [PubMed] [Google Scholar]
  • 19.Levine M, Padayatty S, Wang YH, Corpe CP, Lee J, Wang J, Chen Q, Zhang LQ. Vitamin C. In: Saunders SM, editor. Biochemical, Physiological, Molecular Aspects of Human Nutrition. 2. St. Louis MO: Elsevier Inc; 2006. pp. 760–796. [Google Scholar]
  • 20.Ginter E, Zdichynec B, Holzerová O, Tichá E, Kobza R, Koziaková M, Cerná O, Ozdín L, Hrubá F, Nováková V, Sasko E, Gaher M. Hypocholesterolemic effect of ascorbic acid in maturity-onset diabetes mellitus. Int J Vitam Nutr Res. 1978;48 (4):368–373. [PubMed] [Google Scholar]

RESOURCES