Answers are grounded in this paper and cite their sources.
I read with interest the recent annotation ( Chanarin & Metz, 1997) noting limitation on commonly used methods for cobalamin (Cbl) deficiency diagnosis. There may be benefits in using the random spot urinary methylmalonic (MMA) assay by gas chromatography mass spectrometry (GC/MS). The advantages of using the urinary MMA test by GC/MS are greater sensitivity and specificity, functional tissue Cbl assessment, and non-invasive and convenient testing. Fig 1 illustrates a plot of urinary MMA levels versus serum Cbl results for 35/36 subjects identified after screening 809 elderly individuals for elevated urinary MMA ( Norman & Morrison, 1993). Of the 35 individuals depicted, 28 had concurrent urinary MMA and serum Cbl tests. The other seven had a follow-up serum Cbl test: four with low serum Cbl, two with low-normal serum Cbl (212, 242 pg/ml), and one with normal serum Cbl (406 pg/ml). Two subjects, upon follow-up testing, had normal MMA; oral Cbl supplements could not be ruled out. As previously described, all subjects (n = 16) treated with Cbl injections showed normal MMA levels with follow-up testing. High MMA results with normal serum Cbl values represent functional Cbl deficiency at the tissue level. Other investigators have also reported studies finding elevated serum MMA levels in non-anaemic elderly individuals having normal serum Cbl levels ( Yao et al, 1992 ; Pennypacker et al, 1992 ; Lindenbaum et al, 1994 ; Naurath et al, 1995 ). 80–900 pg/ml; ▪, normal range 200–900 pg/ml. More recent studies demonstrating the high sensitivity of the urinary MMA assay involved a total (strict) vegetarian group ( Crane et al, 1997 ). Serum Cbl was < 200 pg/ml in 8/9 (the remaining one was 331), serum MMA was elevated in 5/8 (all except one normalized with Cbl therapy) and urinary MMA was elevated in 7/8 (all normalized). Combining serum Cbl and urinary MMA tests, all nine subjects showed evidence of Cbl deficiency. The urinary MMA assay by GC/MS is highly specific for Cbl deficiency. The serum MMA assay may give falsely high results in conditions of renal insufficiency, thyroid disease, pregnancy, small bowel bacterial overgrowth, haemoconcentration, and for unexplained reasons (reviewed in Norman & Cronin, 1996). As noted ( Chanarin & Metz, 1997), the specificity of the serum MMA assay has not been adequately evaluated. The greater specificity of the urinary MMA assay over the serum MMA test could be related to the normalization of urine MMA to urine creatinine and urine MMA being 40 times more concentrated than serum MMA, thereby impeding falsely-high values. Problems with false positive results have not been reported for the urinary MMA test ( Matchar et al, 1987 ; Norman & Morrison, 1993; McCann et al, 1996 ). Upon re-evaluation of the data from the previously mentioned screening study using the current normal value for urinary MMA (< 4.0), 6.7% (54/809) of individuals tested had elevated urinary MMA. A major advantage of the urinary MMA test over the serum Cbl assay is that MMA levels are directly related to a Cbl-dependent metabolic pathway giving a measurement of Cbl function at the tissue level. Analogously, when assessing automobile velocity would you rather rely upon a speedometer or a petrol gauge? The serum Cbl assay has very poor specificity. The presence of low serum Cbl levels without tissue deficiency may be partially explained by the fact that < 20% of the total serum Cbl is bound to transcobalamin II, the carrier protein needed for delivery of Cbl to marrow and nerve cells ( Norman & Morrison, 1993). One study determined the positive predictive value of a low Cbl level to be 22% ( Matchar et al, 1994 ) and another evaluation of patients with low serum Cbl found only 22% (109/504) with clinically important disease ( Cooper et al, 1986 ). The first study had a 5-year follow-up period to ensure low results were not incipient deficiency. The poor specificity for the serum Cbl in no way compares with the 99% specificity reported for the urinary MMA test ( Matchar et al, 1987 ).
No takes yet. Share an insight, caveat, or question.
Eric J. Norman (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: