of natto and chlorella intake on blood vitamin K levels and coagulation activity
Hideo Shimodaira, PhD *, Tatsuhiko Kudo MD **
*Hachioji Pharmaceutical Center.
**: Cardiovascular Surgery, Tokyo Medical University Hachioji Medical Center.
A critical aspect of anticoagulation therapy with warfarin involves maintaining the patient's blood coagulation system within the therapeutic range for an extended period. Food containing significant amounts of vitamin K is known to act antagonistically upon warfarin, and natto has been shown to affect blood coagulation. However the lack of standards for restricting natto intake currently makes it difficult to provide appropriate instructions to patients undergoing warfarin therapy.
To improve our understanding and to work toward an appropriate standard, we undertook a study of the effects of one-time intake of small amounts (10 g and 30 g) of natto and chlorella 9 g on vitamin K blood concentrations and the coagulation system. Our subject group was composed of thirteen healthy individuals.
We found that even small quantities (10 g) of natto greatly increased by 225% (from 0.57}0.18 ng/ml to 1.28}0.46 ng/ml) vitamin K2 blood concentrations (MK-7: menaquinone-7) for a period of over 48 hours. Also, vitamin K1 concentrations increased by 152% (from 0.46}0.28 ng/ml to 0.70}0.21 ng/ml) four hours after the intake of chlorella. We propose from these observations that intake of even small amounts of natto as well as chlorella should be restricted during warfarin coagulation therapy.
Warfarin anticoagulation therapy is in wide use for the treatment of patients at risk for thrombosis and embolism 1). The goal of warfarin therapy is to reduce blood coagulation within the predetermined therapeutic range and to maintain this reduced state of blood coagulation for extended periods by adjusting the warfarin dosage 2). Fluctuations in coagulation need to be avoided. Additionally, to ensure effective anticoagulation therapy, it should be required to carefully evaluate the presence of various factors that affect clinical efficacy of warfarin.
Warfarin expresses its anticoagulation action by working antagonistically upon vitamin K, inhibiting the synthesis of vitamin K-dependent blood coagulation factors 3). It has been reported that the consumption of food containing significant quantities of vitamin K, natto in particular should be restricted in patients on warfarin anticoagulation therapy 4). Natto is a common Japanese food item made from fermented soybeans 5).
We first reported in a previous paper 6) that natto intake may seriously affect anticoagulation efficacy of warfarin. We also reported in a more recent issue 7) that a healthy subject consumed 100g of natto showed a thrombotest value suppressed below 40% by warfarin increased to be@86%. Kaneki et al. found that 80 g of natto led to rapid increases in vitamin K2 blood concentrations 8). Compared to pre-consumption values, vitamin K2 blood concentrations were 2.5 times higher when measured again after seven days of natto intake. Most commercially-available natto comes in 50-g packs, but the effect of consuming small amounts of natto is unclear. Consequently, there is currently no general consensus regarding the restriction of natto intake for patients on warfarin treatment.
In the present study, we examined changes in vitamin K blood concentrations in healthy individuals after consumption of small amounts of natto.
We also investigated the effects of chlorella intake on vitamin K levels. Chlorella is increasingly consumed as the current trend toward health-conscious diets expanded in Japan. Chlorella is a type of Chlorophyta, commercially grown for the manufacture of food products rich in proteins and vitamins. It contains significant amounts of vitamin K, and therefore, Chlorella intake may result in interference of warfarin therapy.
Natto (in the form of small natto beans) was purchased from Kume Quality
Products (Kuji-Gun Ibaragi Pref.). Table
1 shows the subject profiles for the test group of thirteen healthy
Table 1. Subject Profile
Average age: 28.5 years.
Average body weight: 58.2kg.
There were no significant differences between Groups A and B.
Group Sex Age (years) Body weight (kg)
A male 34 56
A male 22 83
A male 37 61
A female 30 48
A female 30 54
A female 24 44
A female 25 51
A male 28 90
A male 22 63
A male 37 68
A female 25 43
A female 30 52
A female 26 43
mean}SD 28.5}5.1 58.2}14.8
For all subjects, periodic medical examinations showed no abnormalities in test results, including hepatic and renal function. Figure 1(PDF) shows the protocol for the intake of natto and collection of blood samples. This trial was a cross-over study. Group A was composed of three males and four females (see Table 1). In this group, the subjects were asked to avoid consuming food containing significant quantities of vitamin K for a two-week wash-out period before the start of the trial. During this period, these seven patients ate a regular breakfast, including 10 g of natto, prepared by the hospital at eight in the morning. Blood samples were collected just before breakfast and at 4, 24, and 48 hours following breakfast. Following two weeks of another wash-out period, they were asked to consume 30 g of natto. Blood samples were collected in a same manner. Excessive exercise and alcohol consumption were prohibited two days before and after consuming natto. Group B was composed of three men and three women. In this group, subjects were asked to consume 30 g of natto for the first time, and then 10g for the second time. The same procedures used for Group A were applied.
The subjects included six males (three in Group A, and another three in Group B: see Table 1). Grossmin manufactured by Chlorella Industry Inc. (Minato-ku, Tokyo) was used. Every 100 g of Grossmin contains 22-30 g of chlorella extract. The standard amount is 15 to 45 granules per day (each granule weighs 200 mg). This time, subjects were asked to consume 45 granules (9.0 g) at eight o'clock in the morning before a regular breakfast prepared by the hospital. Four blood samples were collected at set intervals, just as in the above section on the natto trial.
Determination of the levels of various types of vitamin K (phylloquinone: VK1, menaquinone-4: MK-4, and menaquinone-7: MK-7) and thrombotest (an indicator of the coagulation system) were carried out. In the six men consuming 30 g of natto (n=6), changes in their coagulation systems were assessed by measuring the activity of protein C and blood coagulation factors VII, IX, and X. Vitamin K blood concentrations were measured by high-performance liquid chromatography (HPLC) 9) , and the thrombotest values were determined by the Owren method (Nycomed Pharma AC, Oslo, Norway) 10). Protein C concentrations were measured by the chromogenic-activated partial-thromboplastic time method (aPTT) (Dade International Inc., USA) 11), and the activity of blood coagulation factors VII, IX, and X were measured by the fluorescent PT method 12).
Test results before and after consumption of test food were compared and analyzed using a Wilcoxon test, with risk values (p) of less than 0.05 considered to be statistically significant.
in vitamin K levels after natto or chlorella intake
Figure 2 shows changes in VK1 blood concentrations following intake of natto or chlorella. VK1 blood concentrations were 0.53}0.26 ng/ml (mean}SD) just before intake of 10 g of natto, and the concentrations slightly increased to 0.72}0.25 ng/ml four hours after consumption, then decreased up to 24 hours after consumption. Similarly, just before consuming 30 g of natto, VK1 blood concentrations were 0.46}0.28 ng/ml, which increased to 0.70}0.21 ng/ml four hours after consumption, then decreased up to 24 hours after consumption. There were no significant differences between VK1 blood concentrations before and 24 hours after consumption of natto.
Additionally, just before consuming chlorella, VK1 blood concentrations were 0.48}0.22 ng/ml, which increased significantly to 1.07}0.81 ng/ml four hours after consumption (p<0.01). There were no significant differences between VK1 blood concentrations before and 24 hours after consumption of chlorella.
The detection threshold of the present MK-4 measurement method is 0.5 ng/ml. When chlorella or 10 or 30 g of natto is consumed, MK-4 was often undetectable just before, and 4, 24, and 48 hours after consumption of the test food. The mean blood concentrations of MK-4 following consumption of the test food were less than 0.5 ng/ml, demonstrating that blood concentrations of MK-4 are unaffected by natto and chlorella.
Figure 3 shows shifts in blood concentrations of MK-7 resulting from consumption of natto or chlorella. The detection threshold of the present MK-7 measurement method is 0.5 ng/ml. Just before consumption of the test food, blood concentrations of MK-7 were undetectable in 75% of the cases. As shown in figure 3, averages were calculated by treating undetectable amounts as 0.5 ng/ml, and then applying statistical analyses. Among subjects who consumed 10 g of natto, MK-7 blood concentrations were 0.57}0.18 ng/ml just before natto intake, which increased significantly to 2.43}0.77 ng/ml after four hours of intake (426% increase as compared to the pre-consumption value, p<0.01), remaining high at 24 and 48 hour intervals with the values of 1.66}0.69 ng/ml (291%) and 1.28(0.46 ng/ml (225%), respectively, (as compared to the pre-consumption value, both p<0.01). Among subjects consuming 30 g of natto, MK-7 blood concentrations were 0.57}0.18 ng/ml just before natto intake, which increased significantly to 7.24}2.92 ng/ml after four hours of intake (1,270% increase as compare to the pre-consumption value, p<0.01) and remaining high 24 and 48 hours later with the values of 4.57}1.82 ng/ml (802%) and 3.00}1.09 ng/ml (526%), respectively, (as compared to the pre-consumption value, both p<0.01).
With the exception of one subject, however, MK-7 blood concentrations fell below the detection threshold at 4, 24, and 48 hours following chlorella consumption, demonstrating that chlorella intake does not increase MK-7 blood concentrations.
and other indices
Thrombotest levels remained constant before and after consumption of any amounts of natto or chlorella. Protein C activities and the blood levels of coagulation factors VII, IX, and X remained constant before and after consumption of natto or chlorella. Table 2 provides the vitamin K content of natto and chlorella, we measured. These findings are similar to those obtained in other studies 13).
We investigated the effects of a single intake of small amounts of natto on vitamin K levels and coagulation activity. We found that MK-7 blood concentrations (vitamin K2) increased four-fold within four hours after consumption of 10 g of natto, and twelve-fold within four hours after consumption of 30 g of natto. Even after forty-eight hours following consumption of 10 or 30 g of natto, MK-7 blood concentrations were still more than twice and more than five times higher, respectively.
Thus, the findings show that the consumption of even small amounts of natto increases vitamin K blood concentrations for a period of over two days.
Certain types of bacteria are known to produce vitamin K in the human intestines. Bacillus natto are Gram-negative bacteria belonging to Bacillus subtilis. Bacillus subtilis produces large amount of vitamin K as compared to the other types of bacteria 13). This may explain, in part, why vitamin K blood concentrations remained to be high over 48 hours after a single consumption of natto.
In clinical studies conducted to investigate the effect of natto on warfarin, Kudo et al. 6) reported that thrombotest levels were stabilized by warfarin anticoagulation therapy, but the level elevated after intake of 100 g of natto. One of their patients was admitted to an emergency room due to a cerebral embolism. Questioned about his eating habits, the patient reported that when his pregnant wife returned home to deliver their baby, he ate more natto than usual, for his convenience.
The results of the present study and previous clinical studies suggest that natto intake must be restricted in patients undergoing warfarin anticoagulation therapy.
Although the chlorella foods used in the present study contained significant quantities of the vitamin K group (VK1), we found only slightly elevated blood concentrations of the vitamin K group four hours after consumption. However, given that a single consumption of chlorella increased levels of VK1 slightly, and that chlorella is generally consumed for extended period, the effects of chlorella intake on warfarin therapy may not be negligible. After conducting a clinical study of the relationship between chlorella foods and warfarin therapy, Ohkawa et al. 14) reported that despite stable thrombotest values of 10% to 20% in patients on warfarin therapy, the values quickly increased by 58% after consumption of chlorella foods (6g/day). Suda 15) also reported that following chlorella consumption, thrombotest values increased by 56% in patients whose thrombotest values had stabilized between 5% and 15%, with levels once again falling below 10% two weeks after discontinuing consumption. Thus, results of both the present study and previous reports suggest that patients on warfarin should avoid chlorella foods.
Levels of thrombotest, protein C, and vitamin K-dependent blood coagulation factors VII, IX, and X were measured to assess the blood coagulation system of healthy individuals. These parameters remained stable following natto or chlorella intake.
The results of this study show that the single consumption of vitamin K does not affect the blood coagulation system of healthy individuals who do not require warfarin.
We concluded from the data described that, following consumption of natto, vitamin K2 blood concentrations (MK-7) were markedly elevated for over 48 hours. Following consumption of chlorella, vitamin K1 blood concentrations increased slightly, returning to pre-consumption levels within 24 hours. The intake of even small amounts of natto should be restricted in patients on warfarin anticoagulation therapy. Whereas, a single instance of eating food containing significant quantities of the vitamin K group does not appear to affect the blood coagulation system of healthy individuals.
The authors would like to thank the following individuals for assistance in carrying out the present study: Minoru Tanaka, Nakahora Masayoshi and Makoto Shimotsuura, of the Tokyo Medical University Hachioji Medical Center Clinical test Laboratories; Atsuko Fukumoto, of the Department of Nutrition, Tokyo Medical University Hachioji Medical Center; Yasuo Dohiguchi, Naoki Magario, Masahiko Saitoh, and Arata Furubayashi of Eisai K.K.; Toshihiko Hirano, of the Tokyo University of Pharmacy and Life Science Clinical Pharmacology Laboratories.
Shimodaira, H., et al.: Q&A for the use of warfarin: Patient compliance
regarding food containing vitamin K, including natto.
Clinical Pharmacy No. 48: 79-83 (1996) (in Japanese)
6) Kudo, T., et al.: Warfarin antagonism of natto during anticoagulation therapy. Igaku No Ayumi (Progress in Medicine), 104: 36-38 (1978) (in Japanese)
7) Kudo, T.: Warfarin antagonism of natto and increase in serum vitamin K by intake of natto. Artery, 17: 189-201 (1990)
8) Kaneki, M., et al.: Significance of serum vitamin K2 concentrations in osteoporosis, significant geographic area differences, and the effects of natto intake. A study of their relationship to bone fractures in the cervical region of the femur. Nihon Kotsutaisha Gakkai Zasshi (Journal of the Japanese Bone Metabolism Association), 13: 39 (1995) (in Japanese)
9) Steyn, J. M., et al.: Reversed-phase high-performance liquid chromatographic method for the determination of warfarin from biological fluids in the low-nanogram range. J. Chromatogr. 387: 254-260 (1986)
10) Owren, P. A.: Thrombotest, a new method for controlling anticoagulant therapy. Lancet 2: 752-754 (1959)
11) Takahashi, K., et al.: Measurement of protein C activity using snake venom. Comparison with protein C antigen in patients with congenital protein C deficiency, disseminated intravascular coagulation syndrome, thrombotic thrombocytopenic purpura, diabetes or hepatic dysfunction, or those on warfarin. Ketsueki To Myakukan (Blood and Vessels), 18: 548-556 (1987) (in Japanese)
12) Dati, F., et al.: Multicenter evaluation of a chromogenic substrate method for photometric determination of prothrombin time. Thromb. Haemost. 58: 856-865 (1987)
13) Iwanaga, S., et al.: Vitamin K: New developments in the fields of medicine and biology. Medical Journal Inc. Tokyo (1994) (in Japanese)
14) Ohkawa, S., et al.: Warfarin therapy and chlorella (health food). Abstract from the 59th Kinki District Japan Neurology Association (1993) (in Japanese)
15) Suda, M.: Chlorella weakens the anticoagulation action of warfarin. Nihon Naika Gakkai Zasshi (Journal of the Japanese Internal Medicine Association), 82 (2): 137 (1993) (in Japanese)
Table 2. Contents of vitamin K in natto and chlorella foods
The amount of vitamin K in the food
products used in the present study:
Natto contained significant quantities of Menaquinone-7 (vitamin K2),while chlorella foods contained significant quantities of K1.
|Menaquinone-4 (vitamin K2) @||16}2||53}1|
|Menaquinone-7 (vitamin K2)||10,893}1,112||834}95|
(5) mean(SD 28.5(5.1 58.2(14.8
(6) Group A
(7) 8AM, 10 g of natto
(8) Two-week wash-out period
(9) Pre-consumption blood collection
A/ (In B group, 30 g, then 10 g of natto was consumed).
Figure 1. Natto intake and blood collection protocol
The subjects were divided into two groups: Group A (n=7) and Group B (n=6). According to the cross-over method, 10 or 30 g of natto was consumed once, and four blood samples were collected at four different points in time following consumption.
Note: pre-consumption blood collection (a blood sample was collected just before natto intake)
4hr blood collection, 24-hr blood collection, 48-hr blood collection: A blood sample was collected at 4, 24, and 48 hours following consumption.
(10) 30 g of natto (n=13)
(11) 10 g of natto (n=13)
(12) Chlorella (n=6)
Wilcoxon test. Comparison with pre-consumption values
(11) Just before consumption
Figure 2. Shifts in VK1 brought about by natto or chlorella intake
Although VK1 blood concentrations significantly increased four hours after chlorella consumption, no significant differences could be confirmed more than 24 hours after consumption.
Figure 3. Shifts in MK-7 brought about by natto or chlorella intake
When 10 or 30 g of natto was consumed, MK-7 blood concentrations increased significantly four hours after consumption, remaining significantly elevated 48 hours after consumption.
Table 2. Contents of vitamin K in natto and chlorella foods
The amount of vitamin K in the food products used in the present study: Natto contained significant quantities of MK-7 (vitamin K2), while chlorella foods contained significant quantities of K1.
(14) Chlorella foods
(15) 1. Vitamin K1
2. MK-4 (vitamin K?)
3. MK-7 (vitamin K2)