Kidney International, Vol. 56 (1999), pp. 2292 2296 Effective correction of hyperhomocysteinemia in hemodialysis patients by intravenous folinic acid and pyridoxine therapy MALIK TOUAM, JOHANNA ZINGRAFF, PAUL JUNGERS, BERNADETTE CHADEFAUX-VEKEMANS, TILMAN DRÜEKE, and ZIAD A. MASSY Division of Nephrology, INSERM U507, and Biochemistry B Laboratory, Necker Hospital, Paris, France Effective correction of hyperhomocysteinemia in hemodialysis teinemia, attempts have been made to lower plasma thcy patients by intravenous folinic acid and pyridoxine therapy. concentrations in HD patients. Several studies in HD Background. Folic acid supplementation is only partially efpatients have shown that supplementation with pharmaficacious in correcting moderate elevation of plasma total homocysteine (thcy) concentrations observed in hemodialysis cological doses of folic acid (2.5 to 15 mg/day), either (HD) patients. Experimental and clinical data have suggested alone or combined with other B vitamins, reduced plasma that this partial efficacy may be due to impairment of folic thcy levels by approximately 30 to 40% [4 6]. However, acid metabolism to 5-methyltetrahydrofolate (MTHF) and of it became clear that folic acid supplementation was only MTHF transmembrane transport as well. To bypass these difpartially effective in reducing thcy levels, and only a ficulties, we assessed the efficacy of intravenous (i.v.) folinic acid, a ready precursor of MTHF, on reducing plasma thcy few HD patients entirely normalized their levels [4 6]. concentrations in HD patients. Recently, the use of 5-methyltetrahydrofolate (MTHF), Methods. In a cohort of 37 patients on intermittent HD the circulating reduced form of folate at an oral dose of treatment, plasma thcy concentrations were determined be- 15 mg/day (105 mg/week) for two months has been fore and during i.v. supplementation of folinic acid (50 mg once per week), together with i.v. pyridoxine (250 mg 3 times shown to lead to a significant decrease (70%) in plasma per week), to prevent vitamin deficiency, particularly in those thcy levels in 14 HD patients, with 5 patients achieving treated by recombinant erythropoietin. thcy levels below 15 m, which is the upper limit of Results. Folinic acid and pyridoxine i.v. supplementation was normal in this study [7]. The difference in efficacy obgiven for 11.2 2.45 months (range 7.5 to 17 months). The mean plasma thcy levels decreased significantly from 37.3 served between folic acid and MTHF supplementation 5.8 m at baseline to 12.3 5.4 m on folinic acid treatment may be due to better absorption of the latter. Folic acid (P 0.001). Moreover, 29 of the 37 patients (78%) had normal (pteroylglutamic acid) usually needs to undergo reduction plasma thcy levels at the end of follow-up (that is, 14.1 m, and methylation to MTHF in the small gut during intestimean 9.8 m, range 6.2 to 13 m). No adverse effects attributnal absorption before reaching the portal blood system, able to folinic acid treatment were observed during this time. Conclusions. Intravenous folinic acid therapy (50 mg) once whereas MTHF bypasses these latter steps [8, 9]. Higher per week associated with pyridoxine supplementation appears circulating MTHF levels with MTHF than with folic acid to be an effective and safe strategy to normalize plasma thcy supplementation may lead to the overcoming of the imlevels in the majority of chronic HD patients. pairment of MTHF transmembrane transport observed in HD patients [10] and to the deliverance of a sufficient amount of readily available substrate for methionine syn- Moderate elevation of plasma total homocysteine thase for the re-methylation of Hcy to methionine [11]. (thcy) concentrations is present in the majority of hemo- However, limited data in uremic rats [12] and uremic dialysis (HD) patients [1], and it is associated with an patients [13] have suggested that the intestinal absorpincreased risk of cardiovascular disease in such patients tion of MTHF is impaired in chronic renal failure. In light [2, 3]. In view of the atherogenic role of hyperhomocysof this information, it is conceivable that the intravenous (i.v.) administration of MTHF could be preferable to Key words: dialysis, folic acid, methyltetrahydrofolate, total homocysteine. in such patients. oral administration to reach higher plasma MTHF levels In our center, because we have observed that a high Received for publication December 15, 1998 and in revised form June 14, 1999 percentage of HD patients did not appear to be taking Accepted for publication August 3, 1999 oral vitamin supplementation, we have recently started 1999 by the International Society of Nephrology to prescribe routinely i.v. folinic acid supplementation 2292
Touam et al: Treatment of hyperhomocysteinemia in dialysis 2293 in HD patients to prevent folate deficiency, particularly Table 1. Characteristics of 37 hemodialysis patients before (Ti) and during (Tf) intravenous folinic acid treatment in those treated by recombinant erythropoietin. Folinic acid, which is the formyl form of tetrahydrofolate, is the Ti Tf only form available for i.v. administration in France. (N 37) (N 37) P Folinic acid is rapidly converted to MTHF in plasma [14, Plasma thcy lm 37.3 5.8 12.3 5.4 0.001 Plasma thcy 14.1 m 15]. Because we routinely measured the plasma thcy % of patients 0 78 0.001 concentration once a year in HD patients, we took the Plasma folate nm 24.3 3.7 40.6 6.8 0.001 opportunity to assess the efficacy of i.v. folinic acid on Erythrocyte folate nm 321 36 1825 356 0.001 Plasma vitamin B 12 pm 192 30 358 61 0.001 reducing plasma thcy concentrations in such patients. Plasma albumin g/liter 35.1 3.4 36.2 3.1 NS Abbreviation is: thcy, total homocysteine. METHODS Thirty-seven HD patients were selected, provided they were being treated by regular bicarbonate HD treatment state before and during folinic acid therapy, immediately (the mean time of follow-up since the first HD session before the a HD session. Plasma thcy concentrations was 32.4 months, range 9 to 204 months), were treated were measured using the radioenzymatic method [16]. by i.v. supplementation of folinic acid, and had plasma Normal thcy concentrations were 8.2 2.2 m (upper thcy concentration assessment before and during folinic limit of normal range, 14.1 m). Plasma and erythrocyte acid treatment. Five of the 37 patients had an oral supple- folate concentrations, as well as plasma vitamin B 12 conmentation of folic acid (5 mg/day for 7 2.3 months) centrations, were determined using a commercial radioprior to i.v. folinic acid administration, with only minor isotope assay (Chiron Diagnostic, St Ouen-l Aumône, effects on the plasma thcy concentration [from 27 France; normal range, 7 to 34 nm for plasma folate, 300 11.4 to 22.5 7.2 m (P NS) before and during folic to 1400 nm for erythrocyte folate, and 150 to 750 pm for acid oral supplementation, respectively]. This inefficacy plasma vitamin B 12 ). was due to either poor compliance or reduced intestinal Results are expressed as mean sd. Student s paired absorption, as attested by a negligible increase of plasma and unpaired t-test and chi-square test were used to folate (26 15 vs. 31 10 nm, P NS) and erythrocyte evaluate differences between groups. The Spearman folate (310 56 vs. 450 26 nm, P NS) levels observed Pearson test was used to evaluate the relationship beafter oral folic acid supplementation. The remaining 32 tween numeric variables. P values 0.05 were considpatients either were not taking oral folic acid supplemen- ered to be significant. tation for socioeconomic reasons or were referred to our nephrology division to dialysis after we started to prescribe i.v. folinic acid supplementation routinely without RESULTS previous oral folic acid supplementation. The effects of folinic acid and pyridoxine are shown The patients were 21 men and 16 women with a mean in Table 1. Data from individual patients are shown in age of 56.7 18.1 years (mean sd). Primary renal Figure 1. The mean plasma thcy levels were significantly disease was chronic glomerulonephristis (5 HD patients), lower (by 67%) during than at start of folinic acid treat- chronic interstitial nephritis (7), hypertensive nephro- ment. Moreover, 29 of 37 HD patients (78%) had achieved sclerosis (4), polycystic kidney disease (3), diabetic nephropathy normal plasma thcy levels (that is, 14.1 m, mean 9.8 m, (4), systemic diseases (6), other and undeter- range 6.2 to 13.0 m). It is noteworthy that both patients mined (8). They were hemodialyzed using high-flux with systemic disease (N 6) and patients who were under synthetic membrane (10 HD patients) or a low-flux syn- oral folic acid supplementation with only minor effects thetic or cellulosic membrane (27). Thirty-three of them on plasma thcy concentration (N 5) normalized their were receiving recombinant erythropoietin therapy. The plasma thcy levels. Eight patients, in whom plasma thcy percentage of urea reduction was routinely checked at concentrations were not normalized, experienced a sub- monthly intervals and was always above 70%. stantial decrease in thcy levels (47%; Table 2). Except All patients received an i.v. supplementation of 50 mg for a longer period of folinic acid treatment in patients folinic acid [Calcium folinate (racemic mixture of l- and who did normalize plasma thcy than in patients who did d-isomers); Dakota Pharm, Creteil, France] at the end not and a tendency in males (18 out of 21 patients) to of the first HD session of each week. They also routinely normalize thcy levels more frequently than females (11 received i.v. pyrodixine (250 mg 3 times per week). Two out of 16), other characteristics of two groups were simi- of them also received oral vitamin B 12 (1 mg/day). Folinic lar (Table 2). thcy levels during folinic acid treatment acid i.v. supplementation was given for 11.2 2.5 months were negatively correlated with the duration of folinic (range 7.5 to 17 months). acid treatment (r 0.46, P 0.004) and positively Blood was drawn while the patient was in the fasting with thcy levels before folinic acid treatment (r 0.42,
2294 Touam et al: Treatment of hyperhomocysteinemia in dialysis P 0.001). No adverse effects could be attributed to folinic acid treatment during the period of follow-up. Fig. 1. Plasma total homocysteine, plasma folate, and erythrocyte folate concentrations of individual patients before (Ti) and during (Tf) intravenous folinic acid treatment. DISCUSSION Our data show that a once weekly i.v. administration of 50 mg folinic acid together with pyridoxine is highly effective in correcting the hyperhomocysteinemia of HD patients, as attested by normalization of plasma thcy levels in 29 (78%) and the substantial decrease (47%) in 9 patients. We found no evidence of toxicity associated with this treatment regimen. Our results emphasize the importance of folate metabolism in the regulation of plasma thcy levels in HD patients. The difficulty of reaching acceptable circulating MTHF levels in HD patients, which in turn are capable of decreasing plasma thcy levels in HD patients, appears to play a central role. The reduction of MTHF intestinal absorption and/or production could be one of the factors explaining this latter difficulty [8, 9, 12, 13]. An alternative mechanism could be that HD patients require pharmacological MTHF concentrations to overcome the impairment of MTHF transmembrane transport [10]. It is possible that only oral, or even better i.v., MTHF supplementation could deliver such quantity. The fact that a once weekly i.v. MTHF supplementation (50 mg) leads to normalization of plasma thcy levels in a much higher percentage of HD patients than the daily oral administration of folic acid (17.5 to 105 mg/week) [4 6] or MTHF (105 mg/week) [7] is in line with this explanation. In vitro, the accumulation of factors related to the uremic state has been shown to be associated with the inhibition of MTHF absorption and transmembrane transport, which can be improved by dialysis treatment, suggesting that these defects are due to the uremic environment rather than an acquired intrinsic defect of cells in uremia [10, 12]. However, the fact that 22% of HD patients did not entirely normalize plasma thcy levels suggests that factors other than uremia are implicated in MTHF metabolism abnormalities, although we cannot exclude the possibility that some HD patients (for genetic or environmental reasons) require higher circulating MTHF levels or longer periods of folinic acid treatment than those achieved in this study to normalize plasma thcy. The close relationship between thcy concentrations and the duration of folinic acid treatment is an argument in favor of this explanation. A comparison between plasma and/or erythrocyte folate levels in patients treated by i.v. folinic acid and in patients treated with oral folic acid could not be done in this study, as measurement of plasma and/or erythrocyte folate was performed only in patients on i.v. folinic acid therapy and in noncompliant patients. However, by using i.v. folinic acid, we could show that a twofold increase of plasma folate levels was sufficient to normalize plasma
Touam et al: Treatment of hyperhomocysteinemia in dialysis 2295 Table 2. Characteristics of hemodialysis (HD) patients who did (responders) or did not (partial responders) normalize plasma total homocysteine (thcy) concentrations under intravenous folinic acid treatment Responders (N 29) Partial responders (N 8) Ti Tf Ti Tf Age years 57 18 54 19 Males/females 18/11 3/5 Time since first HD months 33 35 29 11 High flux membrane % 27 25 Treatment duration months 11.7 2.4 9.4 1.6 d Plasma thcy lm 36.6 5.1 9.8 1.8 a 40.1 7.6 21.4 3.9 b,c Plasma folate nm 26.7 6.2 42 3.1 a 23.4 8.1 41 6 b Erythrocyte folate nm 328 33 1700 408 a 330 46 1570 520 b Plasma vitamin B 12 pm 196 27 361 46 a 180 32 350 75 b Plasma albumin g/liter 34.5 4.2 36 3.7 35.7 2.5 36.5 3.4 Abbreviations are: Ti, before folinic acid treatment; Tf, under folinic acid treatment; thcy, total homocysteine. a P 0.001 vs. Ti of responders b P 0.001 vs. Ti of partial responders c P 0.05 vs. Tf of responders d P 0.02 vs. Tf of responders thcy levels in 78% of HD patients. Our data are in although we do not see any obvious rationale for such contrast with previous reports, which demonstrated that an explanation. oral folic acid supplementation (up to 15 mg/day) led to A spontaneous increase of plasma vitamin B 12 levels a 20- to 50-fold increase of plasma folate levels, but could was observed in all but two of our patients. The latter not normalize plasma thcy levels in more than 10 to were on oral vitamin B 12 (1 mg/day) supplements for low 30% of thus treated dialysis patients [4, 6]. Because high plasma vitamin B 12 levels ( 150 pm). This spontaneous oral doses of folic acid ( 2 to 3 mg) increase both folic increase was almost certainly due to a better nutritional acid and MTHF in the serum [17, 18], it is possible that status under erythropoietin therapy and due to the start this discrepancy is due to an alteration of the reduction of dialysis therapy in some of them. Whereas vitamin of folic acid to MTHF in the plasma of HD patients. B 12 deficiency could increase plasma thcy levels, vitamin Recently, Livant et al reported the presence of an un- B 12 supplementation could do the opposite in HD patients identified inhibitor(s) of plasma folate conjugase (the with a vitamin B 12 deficiency [24]. However, it enzyme responsible for reducing polyglutamate folate to remains unclear whether an administration of vitamin monoglutamate folate) in HD patients [19]. However, B 12 in addition to folate therapy might further improve we could not use the determination of plasma and/or the reduction of plasma thcy observed in HD patients erythrocyte folate levels to verify our hypothesis that it without a vitamin B 12 deficiency, as was the case in the is important to reach an acceptable circulating MTHF majority of our patients. Moreover, it has been shown level, as the folate assays currently available for clinical in a recent study that the addition of 1 mg/day vitamin practice measure all folate derivatives, including MTHF B 12 to an oral dose of 15 mg/day folic acid led to an [20]. Thus, it may be difficult to evaluate the efficacy increase of plasma vitamin B 12 levels, which was of the of various folate treatment strategies, based solely on same order of magnitude as the elevation of plasma currently available commercial assays, and a direct determination vitamin B 12 levels in this work. However, this combina- of circulating MTHF is warranted to assess tion therapy was only partially effective in reducing thcy such efficacy [14, 15, 17, 18, 21]. levels in that study, as only a few HD patients entirely The addition of i.v. pyridoxine therapy to i.v. folinic normalized their levels [4]. acid supplementation could have contributed to the effectiveness Besides the absence of plasma MTHF measurements of our strategy to normalize plasma thcy in all of our patients, this work has another important levels in HD patients. However, we previously observed limitation, namely the lack of a controlled and prospec- that oral pyridoxine supplementation (70 mg/day) has tive study design. Nevertheless, it reports for the first no significant effect on fasting total Hcy levels in nondialyzed time an effective and probably safe strategy that leads chronic renal failure patients [22]. Moreover, it to normalization of plasma thcy levels in the majority was demonstrated that an oral pyridoxine supplementation of HD patients and offers new insights into the potential of 300 mg/day had no effect on fasting total Hcy role of MTHF as a determinant of plasma thcy levels levels in a mixed group of HD and PD patients [23]. It in such patients. Whether a higher dose and/or a longer remains possible, however, that i.v. pyridoxine supple- time period of i.v. MTHF administration once weekly mentation has a greater efficacy than oral pyridoxine, may lead to the normalization of plasma thcy levels in
2296 Touam et al: Treatment of hyperhomocysteinemia in dialysis the entire HD patient population, whether such treattions. Altern Med Rev 3:208 220, 1998 9. Kelly GS: Folates: Supplemental forms and therapeutic applica- ment may be safe after long-term utilization, and 10. Jennette JC, Goldman D: Inhibition of the membrane transport whether the normalization of hyperhomocysteinemia of folates by anion retained in uremia. J Lab Clin Med 86:834 843, may be helpful in decreasing the cardiovascular morbid- 1975 11. Malinow MR: Homocysteine and arterial occlusive diseases. J ity and mortality of such patients deserve further study. Intern Med 236:603 617, 1994 Until such additional data become available, it is reasonabsorption 12. Said HM, Vaziri ND, Kariger RK, Hollander D: Intestinal of 5-methyltetrahydrofolate in experimental uremia. able to wait before proposing i.v. folinic acid as routine Acta Vitaminol Enzymol 6:339 346, 1984 therapy in HD patients. 13. Reteif FP, Heyens AP, Oosthuizen M, van Reenen OR: Aspects of folate metabolism in renal failure. Br J Hematol 36:405 415, Reprint requests to Ziad A. Massy, M.D., INSERM U507, Necker 1977 Hospital, 161 rue de Sèvres, 75730 Paris, Cedex 15, France. 14. McGuire BW, Sia LL, Leese PT, Gutierrez ML, Stokstad EL: E-mail: massy@citi2.fr Pharmacokinetics of leucovorin calcium after intravenous, intramuscular, and oral administration. Clin Pharm 7:52 58, 1988 REFERENCES 15. Zittoun J, Tonelli AP, Marquet J, De Gialluly E, Hancock C, Yacobi A, Johnson JB: Pharmacokinetic comparison of leucovorin 1. Robinson K, Gupta A, Dennis V, Arheart K, Chaudhary D, and levoleucovorin. Eur J Clin Pharmacol 44:569 573, 1993 Green R, Vigo P, Mayer A, Selhub J, Kunter M, Jacobsen DW: 16. Chadefaux B, Coudé M, Hamet P, Aupetit J, Kamoun P: Rapid Hyperhomocysteinemia confers an independent increased risk of determination of total homocysteine in plasma. Clin Chem 35:2002, atherosclerosis in end-stage renal disease and is closely linked to 1989 plasma folate and pyridoxine concentrations. Circulation 94:2743 17. Lucock M, Wild J, Smithells R, Hartley R: Biotransformation 2748, 1996 of pteroylmonoglutamic acid during absorption: Implications of 2. Bostom AG, Shemin D, Verhoef P, Nadeau MR, Jacques PF, Michaelis-Menten kinetics. Eur J Clin Nutr 43:631 635, 1989 Selhub J, Dworkin L, Rosenberg IH: Elevated fasting total 18. Kelly P, McPartlin J, Scott J: A combined high-performance plasma homocysteine levels and cardiovascular disease outcomes liquid chromatographic-microbiological assay for folic acid. Anal in maintenance dialysis patients: A prospective study. Arterioscler Biochem 238:179 183, 1996 Thromb Vasc Biol 17:2554 2558, 1997 19. Livant EJ, Tamura T, Johnston KE, Vaughan WH, Bergman 3. Moustapha A, Naso A, Nahlawi M, Gupta A, Arheart K, Jacob- SM, Forehand J, Walthaw J: Plasma folate conjugase activities sen DW, Robinson K, Dennis V: Prospective study of hyperhomocysteinemia and folate concentrations in patients receiving hemodialysis. J Nutr as an adverse cardiovascular risk factor in end-stage Biochem 5:504 508, 1994 renal disease. Circulation 97:138 141, 1998 20. Longo DL, Herbert V: Radioassay for serum and red cell folate. 4. Bostom AG, Shemin D, Lapane KL, Hume AL, Yoburn D, Na- J Clin Lab Invest 87:138 151, 1976 deau MR, Bendich A, Selhub J, Rosenberg I: High dose B-vitamin 21. Bagley P, Selhub J: A common mutation in the methylenetetrahy- treatment of hyperhomocysteinemia in dialysis patients. Kid- drofolate reductase gene is associated with an accumulation of ney Int 49:147 152, 1996 formylated tetrahydrofolates in red blood cells. Proc Natl Acad 5. Janssen MJ, Van Guldener C, De Jong GM, Van Den Berg M, Sci USA 95:13217 13220, 1998 Stehouver CDA, Donker AJM: Folic acid treatment of hyperhomocysteinemia 22. Chauveau PH, Chadefaux B, Coude M, Aupetit J, Kamoun P, in dialysis patients. Miner Electrolyte Metab 22:110 Jungers P: Long-term folic acid (but not pyridoxine) supplementa- 114, 1996 tion lowers elevated plasma homocysteine level in chronic renal 6. Van Guldener C, Janssen MJ, Lambert J, ter Wee PM, Jacobs failure. Miner Electrolyte Metab 22:106 109, 1996 C, Donker AJM, Stehouwer CDA: No change in impaired endothelial 23. Arnadottir M, Brattström L, Simonsen O, Thysell H, Hultberg function after long-term folic acid therapy of hyperhomocys- B, Andersson A, Nilsson-Ehle P: The effect of high-dose pyri- teinemia in hemodialysis patients. Nephrol Dial Transplant 13:106 doxine and folic acid supplementation on serum lipid and plasma 112, 1998 homocysteine concentrations in dialysis patients. Clin Nephrol 7. Perna A, Ingrosso D, De Santo NG, Galletti P, Brunone M, 40:236 240, 1993 Zappia V: Metabolic consequences of folate-induced reduction of 24. Dierkes J, Domrose U, Ambrosch A, Schneede J, Guttormsen hyperhomocysteinemia in uremia. J Am Soc Nephrol 8:1899 1905, AB, Neumann KH, Luley C: Supplementation with vitamin B12 1997 decreases homocysteine and methylmamonic acid but also serum 8. Perry J, Chanarin I: Intestinal absorption of reduced folate compounds in man. Br J Haematol 18:329 339, 1970 folate in patients with end-stage renal disease. Metabolism 48:631 635, 1999