Cobalamin-responsive disorders in the ambulatory care setting: unreliability of cobalamin, methylmalonic acid, and homocysteine testing.

Early recognition of cobalamin (Cbl)-responsive disorders in the ambulatory care setting is essential to prevent irreversible neurologic deficits. However, diagnostic algorithms using Cbl, methylmalonic acid (MMA), and homocysteine (HCys) measurements reflect studies in academic centers, and their negative predictive values have not been established. Thus, records of 456 ambulatory patients evaluated for Cbl deficiency at a staff model HMO were reviewed. Pretherapy Cbl, MMA, and HCys values in individual patients varied by 23%, 23%, and 17%, respectively, over 2 to 6 weeks. Hematologic or neurologic responses to pharmacologic doses of Cbl occurred in 37 of the 95 evaluable patients. In these patients, pretherapy Cbl, MMA, and HCys values were normal in 54%, 23%, and 50%, respectively. If therapy had been restricted to symptomatic patients with both low or intermediate Cbl levels and increased metabolite values, 63% of responders would not have been treated. Twenty-five patients did not respond to treatment, including 5 of 11 patients (45%) with low Cbl, 22 of 49 patients (45%) with high MMA, and 13 of 30 patients (43%) with high HCys values. It is concluded that Cbl, MMA, and HCys levels fluctuate with time and neither predict nor preclude the presence of Cbl-responsive hematologic or neurologic disorders.