Sensitivity of the Total B12 Test Compared to Methylmalonic Acid - A Systematic Review

Problem

The total vitamin B12 test (tB12) is the standard test to determine whether someone is vitamin B12 (B12) deficient. It is used for symptomatic patients, e.g. with anaemia or tiredness, and to rule out treatable causes of dementia and other neurological disease. Incidence of B12 deficiency is increasing with the popularity of vegan lifestyle and use of nitrous oxide as a recreational drug. Several studies indicate tB12 is a poor test for deficiency but a systematic review of its sensitivity and specificity has not been done before. We aim to assess the evidence for sensitivity, specificity, and diagnostic yield of tB12 for the diagnosis of B12 deficiency, compared to methylmalonic acid (MMA).

Approach

A systematic search of six online databases and two trial registries was conducted for the terms ‘methylmalonic’ and either ‘B12’ or ‘holotranscobalamin’. Eligible publications were to provide data on sensitivity and specificity, or allow calculation of diagnostic 2x2 tables.

Findings

After title and abstract review, 1208 publications were deemed eligible, of 729 full text was available. These have been reviewed. Twenty-two studies met criteria. After quality assessment, 13 studies were deemed to be of sufficient quality to allow data extraction and analysis. Even at cut-off levels for tB12 significantly higher than used in clinical practice (< 271 pg/mL), compared to MMA > 400 nmol/L, tB12 had a sensitivity of 8-48.6%, and specificity 83.1-98.4%. False negative results, i.e. potential missed diagnoses, occurred in 9.3-36.8% of five study populations of older people. The highest of these occurred in a study that included participants living in a nursing home.

Consequences

The tB12 test has good specificity but low sensitivity even at cut-off levels higher than used in clinical practice. Many diagnoses of eminently treatable deficiency may be missed. This confirms Hooper et al's finding that many patients experience diagnostic delay of years, and occasionally decades (2014). Older people, because of their high prevalence of B12 deficiency and multiple co-morbidities obscuring the clinical picture, are particularly at risk of being missed. High levels of clinical suspicion, access to more sensitive tests, and a symptom score to rate the effect of trial of treatment are needed to improve diagnosis of B12 deficiency in General Practice.