Hepatic biotransformation of gut microbiota-derived trimethylamine (TMA) to trimethylamine N-oxide (TMAO) may enhance cardiovascular risk [1, 2]. The source of TMA is fish-derived TMAO [3], meat-derived betaine/choline/phosphatidylcholine [1, 2], and the medicine l-carnitine [4]. There is no clear information on the metabolic fate of TMA/TMAO in patients receiving hemodialysis and treated with l-carnitine orally and intravenously. We report data showing that plasma TMA/TMAO concentrations in hemodialysis patients may be in the same range as those of healthy subjects consuming a typical diet in Japan.
Fourteen subjects, randomly selected from a cohort of 18 male and 6 female Japanese patients (47–89 years) with diabetes mellitus or chronic glomerulonephritis who had undergone hemodialysis for several years, were divided into five untreated controls, five treated with 1000 mg of l-carnitine (l-Cartin injection, Otsuka, Tokyo, Japan) intravenously on 3 days (Days 1, 3 and 5) over 1 week, and four treated with 300 mg of oral l-carnitine (l-Cartin tablets, Otsuka) three times per day for 7 days. Ten healthy volunteers (22–59 years) were also enrolled. This study was approved by the ethics committees of Minami Ikebukuro Clinic and Showa Pharmaceutical University. Total carnitine and TMA/TMAO concentrations were determined by Kainos laboratories (Tokyo, Japan) and by gas chromatography [3].
Carnitine concentrations for 2–8 weeks after treatment with l-carnitine were significantly higher than those in the untreated group (Figure 1). Plasma TMA concentrations after oral treatment with l-carnitine were higher than those of the untreated group. TMAO concentrations did not increase significantly after treatments and were in the range of daily interindividual variations. In separate observations in 22 of the 24-patient cohort, pre-dialysis plasma concentrations of carnitine (33 ± 4 µM, mean ± SD), TMA (34 ± 20 µM) and TMAO (189 ± 94 µM) were decreased by hemodialysis (12 ± 8, 25 ± 8 and 140 ± 77 µM, respectively). These concentration ranges were comparable with TMA (26 ± 17 µM) and TMAO (197 ± 100 µM) in healthy volunteers (n = 10). The present plasma levels of TMA and/or TMAO in Japanese patients were one or two orders of magnitude higher than the reported concentrations in Caucasians [1, 2], but were consistent with a review of patients undergoing hemodialysis [4].
Fig. 1.
Plasma concentrations of total l-carnitine (A), TMA (B) and TMAO (C) in untreated patients (circles) and in patients treated with l-carnitine intravenously (triangles) and orally (squares). The three bars are averages of the three groups (n = 4 or 5). ...
Some TMA formation in a Japanese cohort treated with l-carnitine orally was confirmed, but the levels of TMA/TMAO both pre-/post-dialysis in patients not receiving l-carnitine were in the same ranges as those of the control subjects.