| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Department of Foods and Nutrition, University of Georgia, Athens (MAJ, NAH, WRB, and JGF); the Centers for Disease Control and Prevention, Atlanta (EWG); and the Department of Medicine, University of Colorado Health Sciences Center, Denver (RHA and SPS).
2 Supported by National Institute on Aging grant AG-09834 (SPS) and the Northeast Georgia Regional and Development Center, Georgia Department of Human Resources and Georgia Agricultural Experiment Station, Hatch GEO00916 (MAJ and JGF). 3 Address reprint requests to SP Stabler, 4200 East 9th Avenue, Box B170, Denver, CO 80262. E-mail: sally.stabler{at}UCHSC.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: The objective was to determine the prevalence of hyperhomocysteinemia and vitamin B-12 deficiency in elderly who were using Title IIIc nutrition services, after folate food fortification in the United States.
Design: Demographic, nutritional, cognitive, routine diagnostic, and serum methylmalonic acid (MMA) and total homocysteine (tHcy) tests were performed in a convenience sample of 103 elderly enrolled in nutrition service programs in rural northeast Georgia. A subgroup (n = 27) was treated with vitamin B-12, 2.5 mg, and a multivitamin with 400 µg folic acid, 2 mg vitamin B-6, and 27 mg ferrous fumarate.
Results: The total cohort included 103 participants (
± SD age: 76.4 ± 8.1; 80% female; 68% white, 32% African American). Vitamin B-12 deficiency (serum vitamin B-12 < 258 pmol/L and MMA > 271 nmol/L) was present in 23%. Mean serum folate was high, 39.3 nmol/L, and no subject had serum folate < 6.8 nmol/L. Mean tHcy was 17.6 ± 7.2 µmol/L in vitamin B-12–deficient subjects and 10.8 ± 3.6 µmol/L in those who were nondeficient. Determinants of high tHcy were vitamin B-12 deficiency, high serum creatinine, and low red blood cell folate. Those with vitamin B-12 deficiency were more likely to have poor cognition (58% compared with 20%, P < 0.001) and anemia (38% compared with 18%, P = 0.042). High-dose oral B-12 therapy lowered mean MMA and tHcy by 49% and 32%, respectively.
Conclusion: Vitamin B-12 deficiency was prevalent and was associated with poor cognition, anemia, and hyperhomocysteinemia.
Key Words: Vitamin B-12 • folate • pepsinogen I • cognition • elderly • homocysteine • methylmalonic acid • anemia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In addition to being a possible risk factor for vascular disease (10), hyperhomocysteinemia may be associated with poor cognitive function, possibly through alterations in vascular function (12–14), and it is associated with depression (15,16). Serum tHcy concentration is influenced primarily by low intakes, poor bioavailability, and poor absorption of folate and vitamin B-12 (17–19). Now that US food prepared with fortified grain flours is widely fortified also with folic acid, it will be important to determine whether the relations between poor cognition and hyperhomocysteinemia and vitamin concentrations still exist.
The major purpose of this study was to determine the prevalence and clinical associations of vitamin B-12 and folate deficiency in elderly referred for nutrition assessment by directors of nutrition service programs (Meals on Wheels) funded by Title IIIc through the Administration on Aging of the US Department of Health and Human Services. This subgroup of the elderly is known to be at risk of nutritional problems (20), but their vitamin B-12 status has not been reported previously. They are receiving nutritional support with either congregate meals or meals delivered to their homes (Meals on Wheels) but not with program-supplied vitamin supplements. We hypothesized that vitamin B-12 deficiency would be 1) highly prevalent, 2) associated with poor diet as well as markers for atrophic gastritis such as serum pepsinogen I, and 3) associated with poor cognition, depression, and anemia. A second major hypothesis was that the food folate fortification would have decreased the prevalence of folate deficiency and hyperhomocysteinemia. We also investigated whether oral vitamin treatment would decrease tHcy and MMA concentrations and improve clinical variables.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
60 y from whom blood samples were collected between September 1997 and September 1998. Because of the high prevalence of vitamin B-12 deficiency discovered in the first 65 subjects (first group), the protocols were amended to add a treatment phase for a new cohort of subjects aged 60 y who were not homebound and not taking supplements containing B vitamins; blood samples were collected from these subjects between April 1999 and September 1999, and these subjects are referred to as the treatment group. They underwent the same data collection as the subjects in the first group, but the treatment group subjects were then offered a treatment phase.
Methods
The participants were assessed either at their home or at the senior center in a 2-h session with rest periods as needed. Because of their advanced age and frailty, participants were not required to fast before blood was collected. Complete blood counts including hemoglobin and mean cell volume (MCV) (automated cell counts), sedimentation rate, and serum ferritin and serum creatinine concentrations were conducted by a local clinical laboratory (SmithKline-Beecham Clinical Laboratories, Atlanta). The normal MCV was defined as 80–100 fL. Anemia was defined as < 120 g/L for women and < 130 g/L for men. Blood samples for the serum folate and vitamin B-12 analyses were frozen at -70 °C in cryogenic vials with minimal air space (Nalgene Brand Products, Rochester, NY) until they were analyzed. Analyses for serum vitamin B-12, red blood cell (RBC) folate, and serum folate were performed at the Centers for Disease Control and Prevention (Atlanta) with a radioassay (Quantaphase II Vitamin B-12/Folate Radioassay; Bio-Rad, Richmond, CA) (21). Ferritin analysis was performed by the Centers for Disease Control and Prevention [Quantimmune 125-I ferritin immunoradiometric assay (IRMA); Bio-Rad, Hercules, CA]. Serum MMA, tHcy, serum 2-methylcitric acid, and cystathionine were analyzed by capillary gas chromatography-mass spectrometry (22–25). The previously determined normal ranges were 73–271 nmol/L for MMA (24), 5.4–13.9 µmol/L for tHcy (6), 44–342 nmol/L for cystathionine (6), and 60–228 nmol/L for 2-methylcitric acid (24). Serum pepsinogen I was measured with the use of a kit (SORIN/Biomedica kit P2560; INCSTAR Corporation, Stillwater, MN). A person was diagnosed with possible iron deficiency if their ferritin was < 45 µg/L (26,27). Folate deficiency was diagnosed if serum folate or RBC folate was < 6.8 nmol/L or 295 nmol/L, respectively [the 10th percentile for their age and sex group in the third National Health and Nutrition Examination Survey (NHANES) III] (28) and if serum homocysteine was > 13.9 µmol/L and MMA was < 271 nmol/L. The definition of vitamin B-12 deficiency was serum B-12 <258 pmol/L, serum MMA >271 nmol/L, and MMA concentration greater than that of the accompanying 2-methylcitric acid (6,24). Serum pepsinogen I was used as an indirect index of atrophic gastritis (29). Subjects were considered to have mild atrophic gastritis if their pepsinogen I concentrations were between 10 and 60 µg/L and to have severe atrophic gastritis if their pepsinogen I was < 10 µg/L (29).
Questions concerning general health, eating patterns, depression, cognition, and activities of daily living (ADL) were read to each subject (or to his or her caregiver), and the answers were recorded by the interviewer. A history of chronic health problems and current medications, including the use of vitamin or vitamin-mineral supplements, was obtained. A 24-h food recall was collected and nutrient intake was estimated by the use of FOOD PROCESSOR software, version 6.20 (Esha Research, Salem, OR). Questions for the Mini-Nutritional Assessment (30) were used to assess the weekly servings of milk and yogurt or cheese (or both) and of meat and poultry or fish (or both). Functional health, or level of physical functioning, was assessed by an ADL inventory provided by the Northeast Georgia Area Agency on Aging. Physical ADL classification is based on 6 items of ADL performance (independent = 0; dependent = 1): transfer, grooming, dressing, eating, bathing, and bladder continence. Instrumental ADL classification is based on 9 items of instrumental performance: managing money, telephoning, preparing meals, laundry, housework, transfer outside the home, following directions of routine and specialized health care, and function in being left alone. Depression was assessed by the 30-item geriatric depression scale (GDS) (31) (normal range: 0–10; possible depression: 11). Cognitive status was assessed by a short orientation memory concentration test (32). This instrument has 6 questions concerning the current month, year, and time of day; phrases to be repeated; and the counting of numbers and months backwards. For each error, a point is assigned (normal or minimal impairment: 0–8; moderate impairment: 9–19; severe impairment: 20). We considered scores 9 to reflect impaired cognition.
Treatment phase
There were 38 elderly recruited into a treatment phase. The enrolled subjects were provided a high-dose (2.5-mg) oral vitamin B-12 supplement (cyanocobalamin; Puritan Pride, Long Island, NY) as well as a multivitamin pill (Women’s Formula; Kroger, Cincinnati) that included 400 µg folic acid, 2 mg vitamin B-6, 400 IU vitamin D, 450 mg calcium, 27 mg ferrous fumarate, and 100% of the recommended daily allowance for vitamin A, vitamin C, vitamin E, vitamin B-12, thiamin, riboflavin, niacin, pantothenic acid, and zinc. Blood samples were obtained at baseline and after 4 mo of supplement use. Compliance was monitored by patient interview and pill count, but only 29 of the original 38 subjects finished the intervention trial because of death and the preference of the patient or the health provider not to participate in the final data collection for blood samples or to participate in the interview to provide health and nutrition information. Further investigations revealed that one subject was receiving parenteral vitamin B-12, and this person was excluded. Another subject did not take the supplements and was excluded.
Statistics
The demographic, nutritional, neuropsychiatric, and chronic illness data were analyzed with SAS software, version 8.2 (SAS Institute Inc, Cary, NC). The biochemical and clinical laboratory variables were analyzed with SPSS Base software, version 10.0 (SPSS Inc, Chicago). Differences in means for continuous variables across 2 categories were evaluated with a t test and Levene’s test for equality of variances. A P value < 0.05 was considered significant. Paired t tests were used to compare the pre- and posttreatment variables in the treatment group. Differences in categorical variables were tested by use of the chi-square statistic. Multiple stepwise logistic regression analyses were used to evaluate the independent effects of factors on tHcy and MMA concentrations, anemia, cognition, and vitamin B-12 deficiency (SAS software, version 8.2; SAS Institute). Spearman’s correlation coefficients were used to test correlations between continuous variables.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Participants had many impairments, and 47% of the first group had a meal delivered to their home. Impaired cognition was frequent in both the first group and the treatment group: 30% and 29.4%, respectively. The GDS indicated possible depression in 18.4% of the entire cohort. Three or more impairments in instrumental ADL were reported in 34% of the cohort, and 3 illnesses or medications reported by 47%. Poor renal function defined by serum creatinine 127 µmol/L was present in 21% and 8% had low serum albumin. The mean (± SD) and actual range of the vitamin concentrations and metabolites are shown in Table 1. The mean serum vitamin B-12 concentration of 332 pmol/L was very similar to the 50th percentile found in NHANES III (28) for all adults aged 70 y (303 pmol/L). However, both the RBC folate (808 nmol/L) and the serum folate (39.3 nmol/L) were much higher than the 50th percentile in NHANES III (501 and 17.0 nmol/L, respectively) (28). The mean serum MMA was 360 nmol/L, which is greater than our previously determined upper level of normal (± 2 SD). The mean tHcy was 12.4 µmol/L, which is within our previously determined normal range.
|
Prevalence of vitamin B-12 deficiency
Of the 103 subjects (combined groups), 24 met the criteria for our definition of vitamin B-12 deficiency, with vitamin B-12 < 258 pmol/L and MMA > 271 nmol/L and MMA > methylcitric acid. The clinical and biochemical variables in those with deficiency and those who appeared to have adequate status are compared in Table 1
.
Vitamin B-12 deficiency was not associated with body mass index, depression, ADL, illnesses, or medications (data not shown). Compared with vitamin B-12–adequate participants, those with deficiency had 63% higher mean concentrations of serum homocysteine (P < 0.001), 26% lower mean RBC folate (P = 0.005), and 28% lower mean serum folate (P = 0.03). There were no age or racial differences between those who were B-12 deficient and those who were not deficient. The mean (± SD) serum pepsinogen I tended to be lower, 83 ± 73 compared with 118 ± 87 (P = 0.082), in those with vitamin B-12 deficiency, and a serum pepsinogen concentration of 60 mg/L was found in a significantly (P = 0.02) smaller proportion of vitamin B-12–deficient subjects than of vitamin B-12–adequate subjects (78% and 54%, respectively). Compared with vitamin B-12–adequate participants, those with vitamin B-12 deficiency were somewhat more likely to be male (16% and 33%, respectively; P = 0.07) and were less likely to consume at least one serving of meat, poultry, or fish daily (69% and 46%, respectively; P = 0.05); less likely to consume at least one serving of milk, yogurt, or cheese daily (75% and 58%, respectively; P = 0.11); and less likely to consume at least 2 servings of meat, poultry, fish, or dairy daily (62% and 33%, respectively; P = 0.02). The use of a multivitamin supplement was not significantly (P = 0.20) lower in the vitamin B-12–deficient group (21%) than in the group without deficiency (35%). There were 87 participants who provided an interviewer-assisted estimate of their food intake for the previous day (intake data from one participant who ate chicken livers on the day of the assessment was eliminated; vitamin B-12 intake for this nondeficient subject was 25 µg/d). For all 87, the mean vitamin B-12 intake was 2.7 ± 2.0 µg/d. Dietary intakes of vitamin B-12 were similar in those who were B-12–deficient and those who had adequate status, and only about one-half of these elderly met the recommended daily allowance ( 2.4 µg/d) for vitamin B-12 regardless of their vitamin B-12 status (see Table 1).
Because it is possible that differences in folate and homocysteine status associated with vitamin B-12 deficiency might be affected by the use of multivitamin supplements by vitamin B-12–deficient persons, the participants who did not use supplements were examined as a subgroup. Among that subgroup, those with vitamin B-12 deficiency (n = 19) had 67% higher mean (± SD) serum tHcy concentrations (17.4 ± 7.5 compared with 11.0 ± 3.7 µmol/L, P = 0.002) and 22% lower mean RBC folate (599 ± 160 compared with 769 ± 344 nmol/L, P = 0.01), tended to have impaired cognition (53% compared with 29%, P = 0.062), and were less likely to consume at least 2 servings of meat, poultry, or fish daily (21% compared with 54%, P = 0.02) than were those who were not vitamin B-12 deficient (n = 51). Thus, the magnitude of the differences in these indexes was similar in the total sample and in the subgroup that did not use multivitamins.
Vitamin B-12 deficiency was frequent in the men; 33% of the B-12–deficient subjects were men, and 4 of the 5 highest MMA values (3130, 2390, 1680, and 1380 nmol/L) were found in the men, even though the studied group as a whole was only 20% male. The percentage of men in the two subgroups (vitamin B-12–deficient and –nondeficient) was not statistically significant (P = 0.07), but, compared with the women, the men did have significantly higher mean (± SD) serum MMA (672 ± 800 and 281 ± 213 nmol/L, respectively; P = 0.038) and significantly higher tHcy (15.8 ± 7.2 and 11.5 ± 4.6 µmol/L, respectively; P = 0.015). The serum creatinine was also significantly higher (P = 0.003) in men than in women (117 ± 40 and 99 ± 32 mmol/L, respectively). However, serum vitamin B-12, RBC folate, and serum folate did not differ significantly between men and women.
The serum MMA was > 1000 nmol/L in 5 (4.8%) of the total cohort. There were 22 (21.4%) subjects who had a serum MMA >376 nmol/L (> 3 SD above the mean in healthy persons), but 5 of the 22 did not meet the definition for vitamin B-12 deficiency because their serum B-12 concentration was > 258 pmol/L, and 3 of them had mild to moderately elevated serum creatinine. However, 2 subjects had normal renal status. The hemoglobin was low (< 120 g/L) in 11 of these 22 subjects with serum MMA >376 nmol/L. However, only one nonanemic subject had an elevated MCV that was suggestive of a megaloblastic process.
Vitamin B-12 deficiency and cognition
The vitamin B-12–deficient subjects were almost 3 times as likely to have cognitive impairment as the subjects without B-12 deficiency (58% and 20%, respectively; P < 0.001). The mean MMA was 599 ± 729 and 262 ± 160 nmol/L, respectively (P = 0.020), and the mean tHcy was 15.3 ± 7.0 and 11.2 ± 4.4 µmol/L, respectively (P = 0.006), in those from the entire cohort of 103 subjects who had impaired cognition and in those who did not. Both the serum folate and the RBC folate were also lower in those who had poor cognition than in those who did not, although the serum vitamin B-12 and the serum creatinine did not differ significantly. Impaired cognition was significantly more frequent in men than in women (63% and 22%, P < 0.001).
Vitamin B-12 deficiency and anemia
Anemia was defined as hemoglobin < 130 g/L in men and < 120 g/L in women. Anemia was very prevalent, and 23 subjects (22%) met the definition. The anemia was severe (< 85 g/L) in 3 subjects. The mean and median hemoglobin were 107 and 108 g/L in the anemic subjects.
The anemic subjects had higher mean (± SD) serum MMA (595 ± 663 compared with 293 ± 313, P = 0.044), tHcy (16.6 ± 7.1 compared with 11.2 ± 4.2, P = 0.002), cystathionine (497 ± 263 compared with 278 ± 131, P = 0.001), and methylcitric acid (262 ± 148 compared with 168 ± 71, P = 0.007) than did the nonanemic subjects. Serum vitamin B-12, folate, and RBC folate did not differ significantly between the anemic and the nonanemic subjects. The MCV was significantly lower (84.9 ± 8.4 compared with 91.8 ± 4.4 fl, P = 0.001) in the anemic subjects than in the nonanemic subjects, as was pepsinogen I (80 ± 52 compared with 118 ± 90 mg/L, P = 0.020). The creatinine tended to be higher (118 ± 45 compared with 100 ± 27, P = 0.082) and the serum ferritin tended to be lower (54 ± 43 compared with 127 ± 203, P = 0.098) in the anemic subjects than in the nonanemic subjects. The vitamin B-12– deficient subjects were anemic twice as often as the vitamin B-12–nondeficient subjects: 38% and 18%, respectively. There were 9 anemic vitamin B-12–deficient subjects, none of whom had an elevated MCV, and, in fact, their mean value, 85 ± 8.3 fl, was surprisingly low. Two vitamin B-12–deficient anemic subjects were frankly microcytic, including the subject with the most severe anemia in the entire cohort. In the anemic vitamin B-12–deficient subjects, the mean (± SD) values for pepsinogen I, 43 ± 40 µg/L (median: 26), and ferritin, 38 ± 31 µg/L (median: 31) were both quite low, which suggests coexisting iron deficiency and atrophic gastritis. The mean serum creatinine was 117 µmol/L, and 3 of the anemic vitamin B-12–deficient subjects had elevated serum creatinine concentrations of 127, 135, and 225 µmol/L.
Microcytosis was present in 6 of the 23 anemic subjects, and 5 of these 6 had low serum ferritin, which suggests a diagnosis of iron deficiency anemia. The mean (± SD) serum ferritin was 54 ± 43 µg/L in the anemic subjects and 127 ± 203 µg/L in the nonanemic subjects (P = 0.098), and pepsinogen I tended to be lower in the former, 80 ± 52 µg/L, than in the latter, 118 ± 91 µg/L (P = 0.060). Serum ferritin was significantly directly correlated with the hemoglobin and MCV (Spearman’s r = 0.32 and 0.30, respectively).
Serum creatinine was 127 µmol/L in 7 of the anemic subjects, which suggests a role for renal insufficiency in the pathogenesis of the anemia seen. Subjects who had low ferritin (< 45 µg/L) also had lower hemoglobin, vitamin B-12, and MCV (data not shown).
Two of the microcytic anemic subjects also met the criteria for diagnosis of vitamin B-12 deficiency and had serum MMA values of 1547 and 754 nmol/L. One value for ferritin was not available, so 22 anemic subjects could be analyzed for combinations of vitamin B-12 deficiency, iron deficiency, and renal insufficiency, all of which could be contributors to anemia. Ferritin was low in 11 subjects, 4 of whom also met our criteria for diagnosis of vitamin B-12 deficiency. Serum creatinine was the only abnormality in 2 subjects, and 2 others had abnormalities in ferritin, vitamin B-12, and creatinine. Vitamin B-12 deficiency was present in 10 of the anemic subjects and 2 also had elevated creatinine; in the others, vitamin B-12 deficiency was combined with low serum ferritin as noted above. The serum creatinine was elevated in 7 of the anemic subjects and combined with the other deficiencies as already noted in all except 2 subjects. Elevated serum MMA and borderline high tHcy, without elevated creatinine or low ferritin, was found in 2 additional subjects who may very well have vitamin B-12 deficiency, except that their serum B-12 concentration was above our arbitrary cutoff. In summary, multiple abnormalities in vitamin B-12, iron, and renal status coexisted in many of the anemic subjects.
Prevalence of folate deficiency
The subjects in this investigation were consuming folate-fortified food during the period of blood sample collection. As a result, folate deficiency as previously defined had disappeared (33). Mean serum folate in this cohort was 39.3 nmol/L, a concentration that was between the 85th and 90th percentiles in subjects of all racial groups aged 70 y, as reported in NHANES III (28). The lowest serum folate was 10.8 nmol/L, which was near the 25th percentile in NHANES III, and the mean RBC folate was 808 nmol/L, which was between the 75th and 85th percentile in NHANES III (28). The 8 subjects with RBC folate levels below the 50th percentile in NHANES III for the corresponding age group (< 477 nmol/L) are characterized in Table 2. Subject 2 had the lowest serum folate concentration, at 10.8 nmol/L, but tHcy was not elevated. Four of these 8 subjects (subjects 1, 4, 5, and 7) met our criteria for a diagnosis of vitamin B-12 deficiency. Subject 7, who took multivitamin supplements, showed an interesting discrepancy between the serum folate, which was high at 53.1 nmol/L, and the other values: the RBC folate was the seventh highest in the study, the MCV was the highest, and the tHcy was the fourth highest. It seems likely that the RBC folate was relatively low because of vitamin B-12 deficiency in this subject, rather than because of poor folate nutritional status. The other 7 subjects described in Table 2 had serum folate concentrations considerably below the mean for the entire cohort and did not ingest multivitamins; thus, they may have had relatively impaired folate status. However subject 3 also had a tHcy concentration considerably below the mean. Out of the 103 subjects, 12 had serum folate < 17.0 nmol/L (the 50th percentile in NHANES III, 28; data not shown), and we diagnosed vitamin B-12 deficiency in 5 of those 12 subjects. The tHcy was elevated in 3 of the vitamin B-12–deficient subjects, but in only 1 of the 7 nondeficient subjects. None of these subjects had macrocytosis. Multivitamins were used by only one of these subjects, who also had vitamin B-12 deficiency. The corresponding RBC folate ranged from 304 to 1150 nmol/L in the subjects. Cognition was impaired in 5 of the 12 subjects.
|
). All of the subjects with markedly elevated tHcy (>3 SD above the normal mean) also had elevations of MMA, and in 90% of this group, the MMA was also > 3 SD above the mean. Only 5 subjects who had a tHcy concentration between 13.9 and17.9 µmol/L, which is between 2 and 3 SD above the normal mean, did not also have elevated MMA; 2 of those 5 had elevated serum creatinine (207 and 153 µmol/L), and 1 of them was taking multivitamins, as was reflected in his serum folate concentration, 47.7 nmol/L. The other 3 of these 5 subjects were not ingesting multivitamins and had normal serum creatinine and serum folate concentrations ranging from 20.8 to 52 nmol/L.
|
Atrophic gastritis and vitamin B-12 deficiency
The pepsinogen I concentration was < 10 µg/L in 2 subjects and < 60 µg/L in 28 (27%) of the 103 subjects. Serum MMA was elevated in 16 of the 28 with evidence for atrophic gastritis by pepsinogen I. Those with the highest serum MMA usually had among the lowest values for pepsinogen I. Mean serum MMA was more than twice as high, 624 ± 747 nmol/L, in those with low pepsinogen than in those with normal values, 263 ± 134 nmol/L (P = 0.017). There were no significant differences in pepsinogen I concentrations when analyzed by sex, race, or low ferritin.
Predictors of abnormal values
The tHcy was highly correlated to MMA and to the other metabolites, as shown in Table 3. Stepwise logistic regression analyses were performed to determine the major predictors of high serum tHcy, high serum MMA, anemia, and poor cognition. The potential predictors were age (< 80 y compared with 80 y), sex, race (white or African American), vitamin B-12 status (normal or deficient), MMA (
271 compared with > 271 nmol/L), tHcy ( 13.9 compared with > 13.9 µmol/L), RBC folate (< 580 compared with 580 nmol/L, which is the lower 25th percentile for this sample), serum creatinine (< 127 compared with 127 µmol/L), serum pepsinogen I (< 60 compared with 60 mg/L), and the use of a multivitamin supplement (yes or no). Forward logistic regression analyses were used with the criteria for entry into a model and inclusion in the model, because the variable was statistically significant at P 0.05.
|
30 mm/hr; P < 0.0001; OR: 15.6; 95% CI: 4.0, 61.0). The predictors of poor cognition score ( 9, see ref 32) were vitamin B-12 deficiency (P < 0.01; OR: 4.7; 95% CI: 1.7, 13.5) and male sex (P = 0.004; OR: 5.3; 95% CI: 1.7, 16.7). Age and use of a multivitamin mineral supplement did not enter into any of these models. A stepwise forward logistic regression analysis also was conducted to determine the major predictors of vitamin B-12 deficiency related to diet, including not consuming at least 2 servings daily of animal foods, not meeting the recommended dietary allowance from diet alone (< 2.4 µg/d), nonuse of a multivitamin supplement, and low serum pepsinogen (< 60 µg/L). The major predictors of vitamin B-12 deficiency were not eating at least 2 servings of meat, poultry, fish, or dairy foods daily (P < 0.02; OR: 3.45; 95% CI: 1.3, 9.3) and low serum pepsinogen (< 60 µg/L; P < 0.03; OR: 3.14; 95% CI: 1.1, 8.6). Dietary intake of vitamin B-12, use of a multivitamin supplement, age, sex, and race were not related to vitamin B-12 deficiency in this regression analysis.
Effects of vitamin treatment in the treatment group
There were 38 elderly in the treatment group who were offered intensive vitamin therapy with high-dose (2.5 mg) oral vitamin B-12 and a multivitamin that included 400 µg folic acid, 2 mg vitamin B-6, and 27 mg ferrous fumarate. However, only 29 finished the intervention trial and 2 were excluded afterward: 1 because of parenteral B-12 treatment and 1 because none of the supplement was taken. Compliance monitoring by pill count showed that average compliance was 70% (range, 20–90%). The mean serum metabolites and other variables before and after the 4 mo of oral vitamin therapy are shown in Table 4. Although the mean serum vitamin B-12, serum folate, and RBC folate were already quite high in these subjects before vitamin therapy, those measures increased markedly after 4 mo of vitamin therapy. There was a 32% fall in the mean (± SD) tHcy, to 7.1 ± 2.4 µmol/L, and a 49% drop in the MMA to 130 ± 50 µmol/L. The MMA was corrected to normal in 100% of these subjects, and the tHcy was elevated after treatment in only 1 subject, who had a mild elevation in serum creatinine and who had only 20% compliance according to a pill count. The mean ferritin values and the number of subjects with low ferritin values did not change after treatment. The mean MCV was normal before treatment and increased slightly with treatment, and the mean hemoglobin was normal before therapy but actually decreased after therapy. The percentage of subjects who were anemic increased significantly. The hemoglobin fell > 10 g/L in 11 subjects, 3 of whom had renal insufficiency and 2 of whom had combined iron and B-12 deficiency. The mean GDS score increased slightly after therapy, but the proportion of those with a score suggesting depression ( 11) decreased from 18.5 to 11.1%. There was not a significant change in the orientation memory concentration score after treatment. However, 2 of the 3 vitamin B-12–deficient subjects who completed the trial had a 5-point decrease in their posttreatment score, which suggests a trend for improved cognition (1 subject had scores of 19 at baseline and 14 after intervention; the other subject scores of 25 at baseline and 20 after intervention). The third vitamin B-12–deficient subject who completed the intervention trial had normal cognition at baseline: scores of 6 at baseline and 6 after intervention).
|
Nine subjects did not complete the entire 4-mo treatment and phlebotomy phase. Unfortunately, 5 of these subjects met our diagnosis of vitamin B-12 deficiency; their serum MMA ranged from 367 to 754 µmol/L and the tHcy ranged from 9.8 to 24.6 µmol/L. Serum pepsinogen I was low in 3 of the vitamin B-12–deficient subjects, ranging from 13 to 35 µg/L. Anemia was present in 3, and 2 of them also had definite iron deficiency with microcytosis. The median creatinine was 117 µmol/L in this group, and the mean tended to be higher than in those who completed the study. Reasons given for stopping treatment in 3 cases were skin rash, which was thought possibly to be related to treatment, possibly supplement-induced gastrointestinal distress, and development of multiple health problems (1 person each).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The serum vitamin B-12 and the MMA values were major determinants of the tHcy, whereas the serum folate or the intake of multivitamins was not. There were only 5 subjects with elevated tHcy who did not have corresponding increases in MMA, and 2 of these subjects had some degree of renal insufficiency, which can be associated with elevated tHcy (23). The mean tHcy in those without B-12 deficiency was actually quite low for this age group, 10.8 µmol/L (see Table 1). Our findings suggest that elevations in serum tHcy in the elderly are associated with vitamin B-12–deficient status more than with folate status at the present time in the United States. This possibility is in agreement with a recent report in a younger cohort with coronary artery disease (36). Our findings have important clinical implications and suggest that patients with elevated tHcy should either be investigated for vitamin B-12 deficiency or treated empirically with high-dose oral or parenteral vitamin B-12 therapy. High-dose (2.5 mg) oral vitamin B-12 therapy combined with a standard multivitamin supplement containing 400 µg folic acid was extremely effective in lowering the mean MMA and tHcy to the normal range in this cohort of elderly, who frequently were chronically ill. This oral combination was as effective as an intensive parenteral vitamin combination in similar-aged populations (7) and intensive oral treatment (37). Despite the advanced age and poor clinical status of these elderly subjects, an extremely low mean tHcy (7.1 µmol/L) was achieved after 4 mo of therapy. This mean tHcy concentration is much lower than that reported recently (10 µmol/L) in similar-aged patients who received only folic acid (400 µg) and vitamin B-6 (1.6 mg) for 3 mo without supplemental vitamin B-12 (38). Likewise, there was only a small decrease in mean tHcy from 12.3 to 10.9 µmol/L with high-dose folic acid (5 mg/d) in patients with coronary artery disease from Canada who had been studied during implementation and after food folate fortification in that country (39). As in the previous study (38), the disappointing response probably reflects the baseline adequate folate status of the subjects and the lack of treatment with vitamin B-12.
The B-12–deficient subjects in our investigation were almost 3 times as likely to have impaired cognition as were those who were not deficient. Both MMA and tHcy were associated with poor cognition, and values were higher in those with impaired cognition. It has been known for many years that vitamin B-12 deficiency causes neurologic disorders including myelopathy and neuropathy, as well as disorders of mood, orientation, and memory (40). However, investigations have shown that elevated tHcy is associated with poor performance on tests measuring cognition (12–15). Elevated tHcy has been found to be a stronger predictor than elevated MMA for cognition in some populations with dementia (41) and for improvement after intensive vitamin B-12 and folate therapy (42). We were not able to tell which metabolite was more correlated with poor cognition because of the extremely strong correlation of the two metabolites. The cognitive test scores in 2 treated B-12–deficient subjects improved by 5 points in this study. We previously found that depression was twice as common in vitamin B-12–deficient patients (16). The number of patients with an elevated GDS score decreased after treatment in this study, but not significantly so. Our data show that treatment of dementias or affective disorders in subjects with hyperhomocysteinemia should include correction of vitamin B-12 status as well as of folate status.
When the food folate fortification program was implemented, it was recognized that the elderly, who have the highest incidence of vitamin B-12 deficiency, would also have the highest risk of "masking" anemia, because of vitamin B-12 deficiency. We found that only one nonanemic vitamin B-12–deficient subject had an elevated MCV, which may be due to the high folate status of these subjects. However, coexisting iron deficiency was quite common in our vitamin B-12–deficient subjects: 2 actually had frank microcytic anemia and others had low-normal MCV with low serum ferritin. We previously found microcytic anemia in elderly women with vitamin B-12 deficiency (6) and in subjects who had undergone gastric surgery (43). Many of the subjects in the latter study had low pepsinogen I coexisting with low ferritin and vitamin B-12 deficiency, which led to the speculation that both vitamin B-12 and iron deficiency were related to chronic atrophic gastritis. However, chronic gastrointestinal blood loss is thought to be the most common cause of iron deficiency in older adults, and it is not known whether atrophic gastritis can cause clinically significant impairments of iron absorption (44). It is also possible that this group of elderly did not ingest enough heme-containing animal protein, which is the source of iron that is absorbed best. It is likely that the iron deficiency in these subjects may be multifactorial, resulting from poor absorption, poor diet, and possibly blood loss. In any case, vitamin B-12 deficiency should be investigated in anemic elderly regardless of the MCV value. The prevalent iron deficiency probably played a role in preventing any improvement in anemia in those we treated. The multivitamin contained ferrous fumarate but no iron–deficient subject achieved an improved status. The subjects with chronic renal failure and anemia also did not respond to vitamin therapy with an increase in hemoglobin. Our study provides evidence that anemia in elderly with many chronic illnesses is likely to be multifactorial in origin, and correcting only one aspect, such as vitamin B-12 deficiency, may not result in a rise in the hemoglobin. Detailed investigations of iron, vitamin B-12 status, and renal status with erythropoeitin measurements and assessments of chronic inflammatory disease should be performed in this population before assuming whether an intervention was or will be successful.
A high dose of oral vitamin B-12 (2.5 mg) along with folate and B6 was effective in lowering MMA and tHcy in this investigation. We previously showed that an oral dose of 2 mg vitamin B-12 daily was as effective as standard intramuscular injections in lowering metabolites and correcting clinical abnormalities in patients who had much more severe B-12 deficiency than was seen in this present investigation (45). It has been known for many decades that patients with pernicious anemia can be treated with high-dose oral vitamin B-12, although the minimum dose necessary is not well delineated (46). Atrophic gastritis was present in about 50% of our vitamin B-12–deficient subjects, which implies some degree of malabsorption of oral vitamin B-12. This may explain why the use of multivitamin supplements at screening was not associated with lower tHcy or MMA concentrations in this cohort. However, we did not quantitate the amounts of vitamin B-12 in the multivitamin supplements that our subjects were ingesting. Our previous studies showed that only multivitamins with higher quantities of vitamin B-12 (> 33–50 µg) will prevent elevated MMA (4,6).
A limitation of our study is that the cohort was referral- rather than population-based. We may have studied the most poorly nourished elderly in these 7 northeast Georgia counties, because they were considered by the center directors to be at risk. Therefore, it is all the more remarkable that their mean serum and RBC folate concentrations were in the high-normal range. Our treatment phase was limited by the poor compliance with our protocol on the part of the subjects with vitamin B-12 deficiency.
Our data may be important for public health in the United States in light of a recent cost- effectiveness analysis that predicted that lowering tHcy concentrations by the use of a combination of 1 mg folic acid and 0.5 mg vitamin B-12 could result in 310 000 fewer deaths if all patients with known coronary heart disease were treated (47). Because high-dose oral vitamin B-12 treatment was very effective in achieving a low mean tHcy value in these chronically ill elderly subjects, smaller doses such as 0.5 mg vitamin B-12 should be studied for efficacy.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. C. Pflipsen, R. C. Oh, A. Saguil, D. A. Seehusen, and R. Topolski The Prevalence of Vitamin B12 Deficiency in Patients with Type 2 Diabetes: A Cross-Sectional Study J Am Board Fam Med, September 1, 2009; 22(5): 528 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hoey, J. Strain, and H. McNulty Studies of biomarker responses to intervention with vitamin B-12: a systematic review of randomized controlled trials Am. J. Clinical Nutrition, June 1, 2009; 89(6): 1981S - 1996S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Ganji and M. R. Kafai Demographic, Lifestyle, and Health Characteristics and Serum B Vitamin Status Are Determinants of Plasma Total Homocysteine Concentration in the Post-Folic Acid Fortification Period, 1999-2004 J. Nutr., February 1, 2009; 139(2): 345 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. McCracken, P. Hudson, R. Ellis, A. McCaddon, and the Medical Research Council Cognitive Function an Methylmalonic acid and cognitive function in the Medical Research Council Cognitive Function and Ageing Study Am. J. Clinical Nutrition, December 1, 2006; 84(6): 1406 - 1411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P Stabler, R. H Allen, E. T Dolce, and M. A. Johnson Elevated serum S-adenosylhomocysteine in cobalamin-deficient elderly and response to treatment Am. J. Clinical Nutrition, December 1, 2006; 84(6): 1422 - 1429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Solomon Cobalamin-responsive disorders in the ambulatory care setting: unreliability of cobalamin, methylmalonic acid, and homocysteine testing Blood, February 1, 2005; 105(3): 978 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. McLean, P. F. Jacques, J. Selhub, K. L. Tucker, E. J. Samelson, K. E. Broe, M. T. Hannan, L. A. Cupples, and D. P. Kiel Homocysteine as a Predictive Factor for Hip Fracture in Older Persons N. Engl. J. Med., May 13, 2004; 350(20): 2042 - 2049. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Refsum, A. D. Smith, P. M. Ueland, E. Nexo, R. Clarke, J. McPartlin, C. Johnston, F. Engbaek, J. Schneede, C. McPartlin, et al. Facts and Recommendations about Total Homocysteine Determinations: An Expert Opinion Clin. Chem., January 1, 2004; 50(1): 3 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Obeid, H. Schorr, R. Eckert, and W. Herrmann Vitamin B12 Status in the Elderly as Judged by Available Biochemical Markers Clin. Chem., January 1, 2004; 50(1): 238 - 241. [Full Text] [PDF]
|
|
|
|
|
|
S. P Stabler Vitamins, homocysteine, and cognition Am. J. Clinical Nutrition, September 1, 2003; 78(3): 359 - 360. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Carmel, R. Green, D. S. Rosenblatt, and D. Watkins Update on Cobalamin, Folate, and Homocysteine Hematology, January 1, 2003; 2003(1): 62 - 81. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
