B12 deficiency is a common, reversible cause of macrocytic anemia and neurological symptoms. Suspected B12 deficiency can be evaluated both directly and indirectly using a variety of assays. The serum cobalamin level, despite highly variable sensitivity and specificity, is often the sole test relied upon to diagnose B12 deficiency, despite being influenced by many common medical conditions. B12 levels tend to fall late in deficiency, making it less useful in detecting acute fluctuations in body stores. In addition, assay methodology has proven problematic, with high rates of falsely normal cobalamin levels using newer chemi-luminescent technologies. Most assays also tend to measure total serum cobalamin, notwithstanding the fact that 80% of cobalamin circulates in biochemically inert form. These factors make it difficult to establish thresholds for "normal" serum cobalamin. B12 status exist along a continuous spectrum ranging from subclinically low vitamin concentrations, as observed in vegans who maintain a system of enterohepatic circulation, to fulminant deficiency with severe clinical signs and symptoms. Serum cobalamin <200 pg/mL is a threshold commonly used to delineate true deficiency, though such low levels are infrequently observed.

Methylmalonic acid (MMA), a molecular intermediate in a unique metabolic pathway requiring cobalamin as a cofactor, can be also be used to assess B12 status. MMA reflects tissue availability of biochemically active cobalamin rather than total cobalamin, with an excellent sensitivity. Fluctuations in MMA occur rapidly, and are detectable in the setting of subtle neurologic, psychiatric or hematologic signs and symptoms when corresponding serum cobalamin levels may remain normal. False positive MMA can be seen in renal dysfunction, though typically to a far milder degree than true deficiency.

To explore the operating characteristics of these serum tests in detecting clinical B12 deficiency, we retrospectively identified all MMAs measured at our institution over the 2015 calendar year and compared any elevated values with corresponding serum cobalamin levels drawn within the same week. 34 of 42 (81%) elevated MMAs were associated with a serum cobalamin level within our laboratory's reference range, and six (14%) of these were actually greaterthan the upper limit of normal. Acknowledging the limited size of our data set, this translates to a 19% sensitivity of serum cobalamin for detecting elevations in MMA and, by extrapolation, detecting clinical B12 deficiency. This sensitivity is far lower than that commonly reported in the literature.

Despite the superior test characteristics of MMA, serum cobalamin is often the first and only test performed to evaluate B12 status due to "economic" reasons or force of habit. If only the cobalamin level were relied upon, many patients would go untreated for a curable disease. While the cost difference of serum cobalamin and MMA assays at our hospital ($7.00 and $18.00, respectively) is not negligible, the time and expense of repeated cobalamin measurements or other testing necessary to accurately diagnose B12 deficiency is arguably greater. The mass of accumulated data shows that serum cobalamin is an insensitive assay for B12 deficiency and should be abandoned. MMA is superior for detecting diminished functional B12 stores; increased utilization of this test will result in more accurate and cost-efficient diagnosis of true B12 deficiency.


Taylor:Baxalta/Shire: Consultancy, Research Funding; Novo Nordisk: Research Funding; Kedrion: Research Funding; CSL Behring: Consultancy, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract