Survey studies consistently show that 7–30% of older subjects with normal serum Cbl values have elevated levels of the Cbl-dependent metabolites, MMA and HCys, even when renal function is normal. Although this finding may reflect insensitivity of the measurement of total serum Cbl to early Cbl depletion, the possibility of Cbl “resistance” with “functional” Cbl deficiency should also be considered. I recently reported that Cbl, MMA and HCys levels are not predictive of clinical responses to Cbl therapy (

Blood 2005;105, 2005
). Interestingly, 8 of the 30 neurologic responders had DM and 5 of these DM pts had elevated MMA values despite normal serum Cbl levels. DM is common in the elderly and produces neurologic changes similar to those of Cbl deficiency. Thus, a retrospective review of Cbl, MMA and HCys levels in patients evaluated for Cbl deficiency during a 10 year period at a staff model HMO was performed with particular attention to the presence of DM. Since both MMA and HCys increase in renal disease, men with serum creatinine >1.5 mg/dl and women with serum creatinine >1.3 mg/dl were excluded from analysis. MMA was measured in 406 subjects, including 49 pts with DM and 357 non-diabetic pts (Non-DM). HCys levels were considered evaluable in 223 non-alcoholic pts (30 DM pts and 193 Non-DM pts) all of whom had normal serum or red cell folate values and normal thyroid function. When serum Cbl levels were ≤ 300 pg/ml, high MMA values (>250 nmol/l) occurred with similar frequency in DM pts (7 of 16; 44%) and in Non-DM pts (93 of 153; 61%)(χ 2=0.305; NS). In contrast, when Cbl values were >300 pg/ml, MMA was significantly more likely to be increased in DM pts (20 of 33; 61%) than in Non-DM pts (70 of 204; 34%)(χ 2=7.238; p<0.01). Indeed, even when serum Cbl levels were >600 pg/ml, MMA values were increased in 6 of 8 DM pts (75%) and in only 10 of 41 Non-DM pts (24%)(χ2=5.813;p<0.02). Moreover, of the 70 Non-DM pts with normal Cbl and increased MMA values, 43 had fasting blood glucose studies performed within 6 months of evaluation and 15 of them had increased values (≥ 100 mg/dl). Thus, DM or impaired fasting glucose was present in at least 35 of the 90 pts (39%) with normal Cbl and increased MMA values. Hgb A1C was measured within 4 mo of evaluation in 38 of the 49 DM pts. MMA was increased in 15 of the 21 DM pts with A1C values >7.0% (71%) and in only 7 of 17 pts with A1C values ≤ 7.0% (41%)(χ2=3.936; p<0.05), suggesting that poorly controlled DM pts were more likely to have high MMA levels. Although metformin use in DM can impair Cbl absorption, only 6 of the 27 DM pts with high MMA values were on metformin therapy (22%). In Non-DM pts with normal Cbl levels, MMA was increased in 31 of 108 pts ≤ 60 years old (29%) and in 29 of 57 patient >70 years old (53%)(χ 2=7.395; p<0.01). This difference remained significant when only those Non-DM pts with serum creatinine ≤ 1.1 mg/dl were considered. High HCys values (>12.1 μmol/l) were also more frequent in DM pts (9 of 22; 40%) than in Non-DM pts (29 of 115; 25%) when serum Cbl was >300 pg/ml, but this difference was not statistically significant (χ2= 2.452; p=0.12). It is concluded that 1) DM is associated with “functional” Cbl deficiency and a role for pharmacologic doses of Cbl in the therapy of diabetic neuropathy should be further explored; and 2) age is also a determinant of MMA in Non-DM pts with normal serum Cbl levels and normal renal function suggesting that other age-related disorders can impair Cbl utilization.

Topics:
cobalamin, diabetes mellitus, homocysteine, methylmalonic acid, brachial plexus neuritis, creatinine tests, serum, metformin, diabetic neuropathies, fasting blood glucose measurement, geriatric disorders

Corresponding author

2005, The American Society of Hematology
2005
Sign in via your Institution