Research Communications in Molecular Pathology and Pharmacology

AUGUST 1995; 89(2): 208-20

PREVENTION OF MYOCARDIAL INFARCTION BY VITAMIN B6

John M. Ellis¹ and Kilmer S. McCully²

Titus County Memorial Hospital,¹ Mt. Pleasant, Texas 75455 Veterans Affairs Medical Center,² Providence, Rhode Island 02908

ABSTRACT

Vitamin B6 is effective in the treatment of carpal tunnel syndrome and related disorders in patients with vitamin B6 deficiency. Hyperhomocysteinemia, a risk factor for atherosclerosis, is associated with deficiencies of vitamin B6, folate, and cobalamin. Patients who were given vitamin B6 for carpal tunnel syndrome and other degenerative diseases were found to have 27% of the risk of developing acute cardiac chest pain or myocardial infarction, compared with patients who had not taken vitamin B6. Among elderly patients of the author (JE) expiring at home, the average age at death from myocardial infarction was 8 years later in those who had taken vitamin B6, compared with those who had not taken vitamin B6. The preventive effect of vitamin B6 on progression of coronary heart disease may be related to increased formation of pyridoxal phosphate, the coenzyme that is required for catabolism of the atherogenic amino acid, homocysteine.

INTRODUCTION

In 1962, the clinical signs and symptoms of human vitamin B6 deficiency, including severe numbness, tingling, pain, and swelling of hands, were found to be responsive to vitamin B6, given as pyridoxine at 50-100 mg/day (Ellis, 1966). Subsequently, bilateral carpal tunnel syndrome (CTS) was found to be associated with vitamin B6 deficiency in pregnancy and in diabetes mellitus (Ellis, 1973). Deficiency of vitamin B₆ in CTS was demonstrated by determination of decreased specific activity of erythrocyte glutamic oxalacetic transaminase (EGOT), a pyridoxal phosphate-dependent enzyme, compared with controls (Folkers et al, 1978; Ellis et al, 1982). Vitamin B6 was given safely to 225 pregnant women in a dose of 50-200 mg/day, improving the edema of pregnancy and bilateral CTS (Ellis, 1987; Ellis and Folkers, 1990). Diabetic women with pregnancy, retinopathy, and CTS are severely deficient in vitamin B6. Doses of 200-300 mg/day prevent macular edema and retinopathy in these cases (Ellis et al, 1991).

During the past 32 years, citizens of Titus County, Texas were encouraged through the news media and by word of mouth to take vitamin B6, as a daily supplement for prevention of CTS, rheumatic diseases, diabetes mellitus, and other related degenerative conditions, including arteriosclerosis (Ellis and Preston, 1988). In 1988, independent pharmacists in Titus County reported selling a monthly total of 59,200 tablets of vitamin B6, each containing 50-100 mg of pyridoxine. When calculated for the population of Titus County (24,000), this volume of sales amounts to 4-8 mg/day per capita during 1988. The RDA for vitamin B6 is 2 mg/ day for adults. Over a period of years, the 32 physicians and surgeons of Titus County reported no cases of toxic neuropathy and only rare cases of photosensitivity to these vitamin B6 supplements.

An important metabolic consequence of clinical vitamin Z, deficiency is hyperhomocysteinemia, which is associated with increased risk of arteriosclerosis (Kang et al, 1992) and myocardial infarction (MI) (Stampfer et al, 1992). Recent studies have shown that hyperhomocysteinemia is attributable in many cases to deficiencies of pyridoxine, folate, and cobalamin in elderly populations (Selhub et al, 1993). For these reasons, a retrospective epidemiological study was initiated to examine whether vitamin B6 supplements that were given to residents of Titus County for CTS, rheumatic disorders, and diabetic retinopathy, prevented MI by counteracting hyperhomocysteinemia secondary to vitamin B6 deficiency in the population at risk.

METHODS

Since Texas covers a wide area from sparsely populated desert to densely populated metropolitan areas, 14 counties in Northeast Texas, including Titus County, each having a relatively uniform population with respect to age of 7802-48,024, were selected for comparison studies of the incidence of MI. Statistics concerning deaths from MI in these 14 counties during the 5 year study period, 1/1/88-12/31/92, are presented in Table I. For comparison, the death rate from MI in the entire State of Texas is given for 1992. The information was obtained from the Texas Department of Health.

All medical records for the study period that were prepared in the emergency room (ER) of Titus County Memorial Hospital (TCMH), the only hospital in the County, were reviewed. Twenty five physicians and surgeons admitted all patients in Titus County with acute cardiac chest pain to the intensive care unit (ICU) of TCMH. In Table II, the numbers of patients with cardiac chest pain and with proven MI are listed for each of 10 physicians with an active hospital practice. The 15 other physicians with limited hospital practices who attended an additional 93 patients during the study period are omitted from the analysis. All physicians who attended patients in the ICO are certified in accordance with the American Heart Association Standards for Basic and Advanced Life Support. Internists RV and MM are additionally qualified as instructors for the American Heart Association Cardiopulmonary Resuscitation and Emergency Cardiac Care Program.

By protocol all patients admitted to the ICU with cardiac chest pain or suspected acute MI had chest X-rays for exclusion of other diseases associated with chest pain. Total serum cholesterol, lactate dehydrogenase (LDH), creatine phosphokinase (CPK) and creatine kinase MB Isoenzyme (CKMB), and 3 additional CKMB studies were determined within the first 24 hours. When serial EGG demonstrated characteristic T wave depression and ST segment elevations, abnormal Q waves, with or without irregular rhythms or bundle branch blocks, in conjunction with elevated CKMB, the clinical diagnosis of acute MI was considered to be proven, and tissue plasminogen activator was given intravenously for thrombolytic therapy. In selected cases, echocardiograms, coronary artery catheterization, and coronary angiography were completed for determination of the degree and type of acute MI and the location and extent of coronary atherosclerosis and occlusion. Further therapy, including coronary angioplasty or bypass grafts, was recommended in some cases.

Among the 13 patients in Table II with cardiac chest pain admitted to the ICU by JE, 7 were diagnosed with acute MI. The clinical histories and vitamin B6 status of these 7 patients are presented in Table III. None of the 13 patients treated by JE in Table II is listed in the deceased group of elderly patients dying at home, as listed in Table IV. All death certificates which list MI as the cause of death, signed by JE in Titus County during the study period, are presented in Table IV. These elderly patients all expired in private homes or in nursing homes, and their clinical records were reviewed to determine prior vitamin B6 therapy. All death certificates which list MI as the cause of death, signed by all other physicians in Titus County, except JE, are presented in Table V. None of these patients had received prescriptions for vitamin B6.

During the study period, the physicians in Table II each treated approximately the same number of patients in their offices, and the general health characteristics of these patients were homogeneous with respect to age, sex, personal habits, and health status. Beginning in 1975, JE prescribed vitamin B6, 50-200 mg/day, for almost all adult patients who came for medical treatment and follow-up care. Compliance in taking the prescribed vitamin B6 was monitored by clinical histories in follow-up visits to JE's office. All of JE's patients in nursing homes were given 50-100 mg/day of vitamin B6. In 1991, RV, the internist who led in admissions of patients to ICU with cardiac chest pain and MI (Table II), was a collaborator in completion of studies of visual acuity and macular edema in patients with diabetes mellitus that were treated with vitamin B6 (Ellis et al, 1991). In 1992, RV began giving vitamin B6, 300 mg/day, to all adult patients with diabetes mellitus, including those admitted to TCMH. The other family physicians and internists in Table II had not prescribed vitamin B in their medical practices. The extent of self administered vitamin B6, among patients of these other physicians, if any, is unknown.

RESULTS

As shown in Table I, the annual death rate from MI per capita, during the 5 year study period, is higher in Northeast Texas, 150.3/ 100,000, than for the entire State of Texas, 81.4/100,000, during 1992. A possible explanation of this large difference is that many younger people have moved to the metropolitan areas, leaving an older age group in the agricultural areas. Hopkins County, for example, recently ranked third in total milk production among all counties in the USA.

The data in Table I demonstrate that the total incidence of death from MI for all age groups in Titus County was equal to or less than the [cont.]

TABLE I Death rate from MI in 14 Northeast Texas counties.

County	Population (1990>	Age 50-59	MI per 60-69	100,000 70-99+	Total
Anderson	48,024	65.7	282.2	746.9	100.0
Lamar	43,949	135.9	315.0	912.5	160.6
Cherokee	41,049	77.4	178.5	738.6	120.8
Upshur	31,370	63.8	225.9	725.2	107.1
Wood	29,380	63.2	242.5	808.3	164,1
Hopkins	28,804	90.7	246.3	618.8	110.3
Fannin	24,804	186.0	388.1	1,273.8	262.1
TITUS	24,009	106.2	228.7	834.0	126.6
Panola	22,035	186.1	399.0	992.0	172.5
Red River	14,317	192.4	395.3	1,508.0	308.7
Morris	13,200	201.0	264.6	692.3	137.9
Marion	9,984	33.3	250.9	757.2	146.2
Camp	9,904	226.1	317.1	966.4	169.6
Franklin	7,802	145.2	310.4	1,029.1	194.8
TOTAL	348,660	126.6	288.9	893.1	150.3
Texas (1992)	16,986,510	79.4	212.9	785.9	81.5

The number of death certificates showing MI as the cause of death was ascertained for the 5 year period, 1/1/88-12/31/92. The annual death rate per 100,000 for each age group was calculated for the study period. The death rate from Ml for the State of Texas is given for the year 1992.

mortality from MI in 10 of 14 counties of Northeast Texas during the study period, 1988-1992. The death rate from MI in Titus County, 126.6/100,000, is 15.8% lower than the death rate from MI for all Northeast Texas counties, 150.3/100,000, during the study period. In the 60-69 age group, the incidence of death from MI in Titus County was lower than the mortality from MI in 12 of 14 counties of Northeast Texas. The death rate from MI in Titus County in the 60-69 age group, 228.7/100,000, is 20.8% lower than the death rate from MI in the same age group for all Northeast Texas counties, 288.9/100,000. The genetic, nutritional, or environmental factors that may explain the differences in death rates from MI between the Northeast Texas counties and the entire State of Texas for the various age groups cannot be determined from the data in this study. The data in Table II show that, among those physicians with active hospital practices admitting patients to the ICU with cardiac chest pain, the fewest patients were admitted by JE. During the study period, JE admitted only 13 patients, compared with the other 9 physicians, who admit-

TABLE II

Patients admitted with cardiac chest pain and with acute MI by attending physicians during the 5 year study period.

Physician	Chest pain	Sex M	F	Age M	F	Acute MI	Sex M	F	Age M	F
EM	20	10	10	63	63	9	5	4	64	61
GC	27	17	10	65	74	10	6	4	64	75
GT	30	13	17	66	72	11	5	6	64	73
JIJ*	22	14	8	70	73	10**	6	4	70	72
JE	13	7	6	70	74	7	5	2	66	73
LM	73	40	33	67	70	20	12	8	68	64
MM	61	38	23	61	70	24	13	11	65	71
OR	39	24	15	50	57	9	6	3	52	64
PM	55	33	22	64	66	12	8	4	67	65
RV	121	78	43	63	66	31	19	12	58	68
Total	461	274	187	64	69	143	85	58	64	69

The number of patients with acute cardiac chest pain admitted to the ICU of TCMH for evaluation and the number of patients with a clinical diagnosis of MI are listed for each physician for the 5 year study period. The sex ratios and average ages are given for patients with cardiac chest pain and with acute MI treated by each physician. All physicians are family practitioners, except for MM and RV, who are Board certified internists. John Ellis [had] 4 patients on B6.

-ted an average of 50 patients each with cardiac chest pain during the same period. Therefore, among JE's patients, almost all of whom had received vitamin B6, the risk of admission for chest pain was only 26% of the risk of admission of the patients of the other physicians, none of whom had received vitamin B6. Similarly, the risk of proven MI among JE's patients who took vitamin B6, 4 patients, is only 27% of the risk of MI among the patients of the other 9 physicians. Each of these physicians treated an average of 15 patients with proven MI during the study

period, and none of these patients had taken vitamin B6. The sex ratios and average ages of onset of symptoms are similar among all groups of patients treated by the physicians in Table II, except for OR, whose patients with cardiac chest pain and males with MI are considerably younger than the patients of the other physicians. JE's patients are among the oldest of those in Table II.

A question might arise concerning the total number of patients attended by JE at this period of his medical practice, 30 years after his beginning clinical research with vitamin B6. Did the limited number of patients admitted with cardiac chest pain and with proven MI reflect a downward trend in his medical practice during the study period? Analy-

TABLE III Patients with proven MI admitted to ICU by JE during 1988-1992.

Age  Sex Date   Sequence of clinical conditions                  Vitamin B6

74   M   1988   Insulin dependent diabetes diagnosed in 1983     100 mg/day
                Coronary artery bypass surgery on 3/88.          1978-1988
                Postoperative staphylocaccal sepsis and          (10 years) 
                congestive heart failure, died 1988.

61   F   1989   History of heavy smoking. Non-insulin dependent   NONE
                diabetes diagnosed in   1975. Reperfused by
                intravenous streptokinase following MI with
                excellent myocardial salvage. Cardiac
                catheterization revealed stenosis of
                coronary arteries, right proximal 98% and
                left circumflex 20%.  Received chemotherapy
                for lung cancer on 3/94.

73   M   1990   Acute MI was caused by stress by threat of attack  100-300 mg/day
                by an intruder. Reperfused by intravenous tissue   1962-1994
                plasminogen activator. Cardiac catheterization     (32 years)
                showed decreased perfusion of lateral and inferior
                myocardium. All coronary arteries were widely
                patent by angiography. Stress test 9/90 showed
                normal ejection fraction at rest, 50% and 70% at
                peak exercise. Physical condition excellent
                at age 77, 1994.

60   M   1990   Recovered from acute MI, died in ER, 1992.         NONE

57   M   1991   Insulin dependent diabetes diagnosed in            NONE
                1979. Coronary artery bypass graft surgery
                in 1984. Ambulatory and employed in 1992.
                Died of recurrent MI, 1993.

67   M   1992   1992 Recovered from acute MI. Living at home       50-100 mg/day
                with no complaints, 1994.                            1990-1992
                                                                     (2 years)

86    F    1992  Blood pressure 200/120 in 1988. Recovered from      100 mg/day
                 MI and returned home. Required digitalization       1986-1992
                 for congestive heart failure. Free of pain and      (6 years)
                 ambulatory, age 88, 1994.

-sis of JE's records shows that an average of 3,259 patients were treated per year during the study period, amounting to 13.6% of the total population of Titus County (24,000). Furthermore, JE had personal contacts with a large number of families and would have known about any cases of heart disease among his patients during the study period.

A detailed clinical description of what happened to a total of 7 patients with MI attended by JE is reported in Table III. During 5 years of active practice, JE admitted only 7 patients to ICU with MI, 3 of whom had diabetes mellitus and 3 of whom had not taken therapeutic doses of vitamin B6, prior to admission for MI. Two of those who had not taken vitamin B6 subsequently died within 3 years. The average age of the 3 patients who had taken no vitamin B6 prior to the onset of MI, 59.3 years, is approximately 15 years younger than the average age of the 4 patients who had taken therapeutic doses of vitamin B6, 75 years (Table III). Of the 4 patients who took therapeutic doses of vitamin B6, 2-28 years prior to developing MI, 3 survived to return home and were alive, 2-4 years after recovery from the MI. One of these 3 survivors was shown to have widely patent coronary arteries by angiography. During the study period, JE admitted only one patient for MI who had taken therapeutic doses of vitamin B6 for several years and who did not return to good health. That patient survived the acute phase of the MI but died of staphylococcal septicemia 5 months after surgery for coronary artery bypass grafts.

TABLE IV Death certificates signed as MI by JE during 1988-1992.

Age	Sex	Date	Vitamin B6
88	F	1988	100 mg/day for 16 years
73	F	1988	100 mg/day for 45 days
78	M	1988	NONE
94	M	1988	100 mg/day for 1 year
80	F	1989	100 mg/day for 18 months
74	F	1990	50 mg/day for 3 years
82	F	1990	100 mg/day for 3 years
80	M	1990	NONE
83	F	1991	50 mg/day for 14 years
92	F	1991	50 mg/day for 17 years
76	M	1991	100 mg/day for 6 months
83	F	1992	100 mg/day for 6 years

Summary: Average age at death

8 patients taking vitamin B6, 1-17 years 84.5 years
4 patients taking vitamin B6, 0-1 years 76.7 years

The data in Table IV show that, among elderly patients treated by JE, who died of MI at home or in nursing homes, the average age at death of the patients who had taken vitamin B6 was 84.5 years. Among those who did not take vitamin B6 or who had taken vitamin B6 for less than one year, the average age at death was 76.7 years. The data in Table V show that, among patients of other physicians dying at home or in the ER of TCMH, none of whom had taken vitamin B6, the average age at death was 74 years.

TABLE V

Death certificates signed as MI by all physicians except JE during 1988-1992.

Physician	MI	Sex M	Sex F	Average age	Vitamin B6
EM	8	6	2	72	NONE
GC	7	2	5	84	NONE
GT	5	2	3	86	NONE
JB	3	2	1	82	NONE
LM	4	1	3	88	NONE
MM	24	13	11	74	NONE
OR	3	2	1	67	NONE
PM	4	3	1	67	NONE
RV	17	8	9	76	NONE
ER*	23	16	7	67	NONE
Total	98	55	43	74	NONE
Summary :					Average age at death
Total of 98					74
23 deceased in ER					67
*TCMH ER physicians

for all patients. The average age of those patients who died of MI in the ER was 67 years.

DISCUSSION

The data in Table I show that the death rate for all age groups from MI in Titus County is lower by 15.8% than the death rate for all 14 Northeast Texas counties. For the 60-69 age group, the difference is 20.8%. A conceivable explanation of these differences could be the increased consumption of vitamin B6 by residents of Titus County, amounting to 4-8 mg/ day per capita over a period of years. The RDA for vitamin B6 is 2 mg/day for adults, and the average resident of Titus County, therefore, consumed 2 to 4 times the RDA for vitamin B6 prior to and during the study period. This analysis would not, however, explain why 4 other Northeast Texas counties, Anderson, Cherokee, Upshur, and Fannin, had even lower death rates from MI than Titus County. Moreover, the distribution of consumption of vitamin B6 supplements is uneven among Titus County residents, since most of the consumption is attributable to patients of JE. The consumption of vitamin supplements other than vitamin B6 is presumed to be similar among residents of all Northeast Texas counties, but no data are available on this point. The listing of MI as the cause of death on death certificates is of questionable accuracy unless the diagnosis had been proven before death by clinical study or after death by autopsy. Therefore, the data in Table I are of undetermined significance with regard to whether vitamin B6 supplements may have altered the risk of death from Ml among residents of Titus County, compared with residents of other Northeast Texas counties.

The data in Table II show that 74% fewer patients of JE's practice, 13 patients, were admitted with cardiac chest pain during the study period, compared with the average number, 50 patients, admitted by the other 9 physicians. Almost all of JE's patients had taken vitamin B6, 50-200 mg/day, for periods of months to years, and none of the patients of the other 9 physicians had taken vitamin B6. Only 4 of JE's patients with proven MI had taken vitamin B6 (Table III), compared with an average of 15 patients with Ml per physician, among the group which did not take vitamin B6. The risk of MI among JE's patients who took vitamin B6 during the study period, therefore, is only 27% of the risk of MI among those patients who had taken no vitamin B6. Since the patient populations treated by JE and by each of the other 9 physicians in Table II are virtually identical, the reduced risks of both cardiac chest pain and MI in JE's patients are attributed to a preventive effect of vitamin B6 on the progression of coronary heart disease.

The apparent 73% reduction in risk of Ml from vitamin B6 therapy among JE's patients could conceivably be related to other factors. One such factor, a possible reduction in the scope of JE's practice, is eliminated by consideration of the annual number of patients treated by JE, 3,259, showing a very active practice during the study period. Moreover, JE examined and treated all of his patients with cardiac chest pain and MI, only referring selected cases to internists or cardiologists for specialized therapy after initial treatment. Therefore, the likelihood that all cardiac cases among JE's patients were treated by JE is the same as the likelihood that all cardiac cases among the other physicians' patients were treated by those physicians. Although exact figures are not available, the total numbers of patients treated by each of the other 9 physicians in Table II are approximately the same as the number treated by JE.

Another conceivable explanation for the reduced risk of MI among JE's patients is that they are derived from a younger and healthier population, compared with the patients of the other 9 physicians in Table II. The information in Tables II and IV, however, indicates that many of JE's patients were elderly and in nursing homes. The distribution of the ages and health of the patients in Table II is believed to be essentially identical to JE's patients. The similar sex ratios and age of onset of symptoms among the various groups, except for patients of OR, support this assumption.

A detailed analysis of other known risk factors for coronary heart disease, including diabetes mellitus, hypertension, obesity, dietary practice, tobacco abuse, family history, and hyperlipemia, is beyond the scope of the present retrospective study. However, these important risk factors are believed to be essentially identical among the groups of patients at risk for MI treated by the physicians in Table II. The relation of known risk factors to prevention of MI by vitamin B6 should be addressed in a properly designed prospective prevention trial.

The analysis of the elderly patients of JE in Table IV who died of MI at home or in nursing homes shows that the average age at death of the patients taking vitamin B6 for periods of 1-17 years, 84.5 years, is 8 years later than the average age at death of patients who did not take vitamin B6 or who took it for less than one year, 76.7 years. The clinical Information in Table III shows that among JE's patients with MI, 3 of the 7 took no vitamin B6. Among the other 4 patients with MI who took vitamin B6, 3 made rapid and satisfactory recoveries, and the fourth died of post-operative septicemia. The average age of the 3 patients with MI taking no vitamin B6, 59.3 years, is more than 15 years younger than the average age of the 4 patients with HI who took vitamin B6, 75 years. These data suggest that vitamin B6 therapy delayed the age of onset of fatal MI in elderly patients dying at home (Table IV) and in patients admitted to the ICU with proven MI (Table III).

The analysis of all other patients, except those of JE (Table IV), who died of MI in Titua County during the study period at home or in the ER shows that the average age at death was 74 years for the entire group and 67 years for those dying in the ER, as shown in Table V. None of these patients had taken vitamin B6, and the average age at death was 10-17 years younger than the patients in Table IV who had

taken vitamin B6. Therefore, these data show that death from MI occurred 10-17 years younger among patients who took no vitamin B6 (Table V), compared with those who took vitamin B6 for 1-17 years (Table IV).

The first report that vitamin B6 may be involved in the pathogenesis of human arteriosclerosis was the observation of vascular lesions with pathological features resembling human atherosclerosis in monkeys with chronic, experimental pyridoxine deficiency (Rinehart and Green-berg, 1951). The key observation that related hyperhomocysteinemia to pyridoxine deficiency and arteriosclerosis was the discovery of arteriosclerotic lesions in children with inherited enzymatic disorders of homocysteine metabolism (McCully, 1969). The homocysteine theory of arteriosclerosis was developed to explain the experimental, pathophysiological, clinical, and epidemiological characteristics of atherosclerosis (McCully, 1983, 993). In that analysis, the declining incidence of mortality from MI in the USA since 1965 was related to increased consumption of synthetic pyridoxine (McCully, 1983).

The findings of the present study suggest that consumption of vitamin B6, a precursor of pyridoxal phosphate, reduces the risk of MI in a susceptible population. Pyridoxal phosphate is the coenzyme for cystathionine synthase, the enzyme that converts the atherogenic amino acid, homocysteine, to cystathionine, a non-atherogenic amino acid. The metabolites of cystathionine, including cysteine, cysteine sulfinic acid, taurine, and sulfate, are also non-atherogenic. Vitamin B6 reduces the risk of thromboembolism in those patients with homozygous deficiency of cystathionine synthase and homocystinuria who respond to pyridoxine therapy (Mudd et al, 1985). Pyridoxine therapy and a low animal protein diet, measures that counteract hyperhomocysteinemia, improve clinical symptoms of angina, increase exercise tolerance, improve ECG abnormalities, and improve glucose tolerance in patients with coronary heart disease (Suzman, 1973).

The blood homocysteine, cholesterol, and LDL levels of patients with atherosclerosis and hyperlipemia are reduced by pyridoxine, folic acid, cobalamin, chollne, riboflavin, and troxerutin (Rang et al, 1992; Olszewski et al, 1989). Moderately elevated blood homocysteine levels are reduced by a diet containing abundant vitamin B6, folic acid and cobalamin and by these vitamins taken as supplements (Ubbink et al, 1993).

In a recent study, vitamin B6-deficient rats were found to have normal fasting homocysteine levels, but the animals develop dramatic, transient, hyperhomocysteinemia 1-5 hours after methionine loading (Miller et al, 1994). In contrast, folate deficient rats have elevated fasting homocysteine levels, and no increase in plasma homocysteine occurs after methionine loading. This experimental finding indicates that in a future prospective human prevention trial, the response of plasma homocysteine to vitamin B6 needs to be assessed by study of plasma homocysteine before and after a methionine load.

In a prospective, human prevention trial, known risk factors for coronary heart disease, such as diabetes mellitus, hypertension, obesity, dietary practice, smoking, family history, and hyperlipemia, need to be analyzed and compared with evidence of both fasting and post-prandial hyperhomocysteinemia in treated and control groups. The quantity of dietary and supplemental vitamin B consumed by the treated group needs to be verified and compared with the untreated group and correlated with the incidence of fasting and post-methionine hyperhomocysteinemia. The occurrence of cardiac chest pain, MI, and other cardiovascular events needs to be documented by strict clinical and pathological criteria among treated and control groups in a prospective design. In this way, conclusive evidence can be sought for determining whether vitamin B6 prevents MI by counteracting hyperhomocysteinemia.

ACKNOWLEDGMENT

We thank Martin A. Weinstock, M.D., Ph.D. for his helpful comments,

REFERENCES

Ellis, J.M. (1966).: The Doctor Who Looked at Hands. Arco Publ. Co., New York.

Ellis,J.M. (1973).: Vitamin B6 the Doctor's Report. Harper and Row, New York.

Ellis, J.M. (1987).: Treatment of carpal tunnel syndrome with vitamin B6. South. Med. J. 80:882-884.

Ellis, J.M. and Folkers, K. (1990).: Clinical aspects of treatment of carpal tunnel syndrome with vitamin B6. Ann. New York Acad. Sci.

585:302-320.

Ellis, J.M., Folkers, K., Levy, M., Shizukuishi, S., Lewandowski, J., Nishii, S., Schubert, H., and Ulrich, R. (1982),: Response of vitamin B6 deficiency and the carpal tunnel syndrome to pyridoxine. Proc. Natl. Acad. Sci. USA 79:7494-7498.

Ellis, J.M., Folkers, K., Minadeo, M., VanBuskirk, R., Li-Jun, X., and Tamagawa, H. (1991).: A deficiency of vitamin B6 is a plausible molecular basis of the retinopathy of patients with diabetes mellitus. Biochem. Biophys. Res, Commun. 179:615-619.

Ellis, J.M. and Preston, H. (1988).: Free of Pain. Revised Edition Update, Vitamin B6 Deficiency U.S.A., the Cause of Soft Tissue Rheumatism. Southwest Publishing Co.,Dallas. Texas, pps 167-168.

Folkers, K., Ellis, J.M., Watanabe, T., Saji, S., and Kaji, M. (1978).: Biochemical evidence for a deficiency of vitamin B6 in the carpal tunnel syndrome based on a cross-over clinical study, Proc. Natl. Acad. Sci. DSA 75.:3410-3412.

Kang, S-S., Wong, P.W.K., and Malinow, M.R. (1992).: Hyperhomocysteinemia as a risk factor for occlusive vascular disease. Annu. Rev. Nutr. 12:279-298.

McCully, K.S. (1969).: Vascular pathology of homocysteinemia: implications for the pathogenesis of arteriosclerosis. Am. J. Pathol. 66:111-128.

McCully, K.S. (1983).: Homocysteine theory of arteriosclerosis: development and current status. In: Atherosclerosis Reviews, Vol. 11. Edited by A.M. Gotto, Jr. and R. Paoletti. Raven Press, New York, pps 157-246.

McCully, K.S. (1993).: Chemical pathology of homocysteine. I Athero-genesis. Ann. Clin. Lab. Sci. 23:477-493.

Miller, J.W., Nadeau, M.R., Smith, D., and Selhub, J. (1994).: Vitamin B6 deficiency vs folate deficiency: comparison of response to meth-iSnine loading in rats. Am. J. Clin. Nutr. 59:1033-1039.

Mudd, S.H., Skovby, F., Levy, H.L., Pettigrew, K.D., Wilcken, B., Pyeritz, R.E., Andria, G., Boers, G.H.J., Bromberg, I.L., Cerone, R., Fowler, B., Grobe, H., Schmidt, H., and Schweitzer, L. (1985).: The natural history of homocystinuria due to cystathionine beta synthase deficiency. Am. J. Hum. Genet. 37:1-31.

Olsaewski, A.J., Szoatak, W.B., Bialkowska, M., Runicki, S., and McCully, K.S. (1989).: Reduction of plasma lipid and homocysteine levels by pyridoxine, folate, cobalamin, choline, r.'boflavin, and troxerutln in atherosclerosis. Atherosclerosis 75:1-9.

Rinehart, J.F. and Greenberg, L.D. (1951).: Pathogenesis of experimental arteriosclerosis in pyridoxine deficiency. With notes on similarities to human arteriosclerosis. Arch. Pathol. 51:12-18.

Selhub, J., Jacques, P.F., Wilson, P.W.F., Rush, D., and Rosenberg, I.H. (1993).: Vitamin status and intake as primary determinants of homocysteinemia in an elderly population. J. Am. Med. Assoc. 270; 2693-2698.

Stampfer, M.J., Malinow, M.R., Willett, W.C., Newcomer, L.M., Upson, B., Ullmann, D., Tishler, P.V., and Hennekens, C.H. (1992).: A pros-spec tive study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. J. Am. Med. Assoc. 268:877-881.

Suzman, M.M. (1973).: Effect of pyridoxine and low animal protein diet in coronary heart disease. Circulation 48:Suppl IV:254.

Ubbink, J.B., VanderMerwe, A., Vermaak, W.J.H., and Delport, R. (1993).: Homocysteinemia and the response to vitamin supplements. Clin. In-vestig. 71:993-998.