ORIGINAL INVESTIGATION Increased Plasma Methylmalonic Acid Level Does Not Predict Clinical Manifestations of Vitamin B12 Deficiency Anne-Mette Hvas, MD; Jørgen Ellegaard, MD; Ebba Nexø, MD Background: The prevalence of vitamin B12 defi- pants, P-MMA levels increased substantially, whereas 44% ciency, defined as an elevated concentration of plasma showed a decrease. Level of P-MMA was significantly but methylmalonic acid (P-MMA), has been estimated to be not strongly associated with levels of plasma cobala- 15% to 44% in the elderly. However, we do not know mins (r=−0.22, P,.001) and plasma total homocyste- whether an increased P-MMA level actually indicates or ine (r=0.37, P,.001). After adjustment for age and sex, predicts a clinical condition in need of treatment. we found no associations between P-MMA concentra- tion and the total symptom score (P=.61), the total Neu- Participants and Methods: In a follow-up study, 432 rological Disability Score (P=.64), or other clinical mani- individuals not treated with vitamin B12 were examined festations related to vitamin B12 deficiency. 1.0 to 3.9 years after initial observation of an increased P-MMA concentration (.0.28 µmol/L). The examina- Conclusions: An increased level of P-MMA did not tion included laboratory tests, a structured interview to predict a further increase with time and clinical mani- disclose symptoms, a food frequency questionnaire, and festations related to vitamin B12 deficiency. We there- a clinical examination including a Neurological Disabil- fore challenge the use of an increased P-MMA concen- ity Score. tration as the only marker for diagnosis of vitamin B12 deficiency. Results: Variation in P-MMA levels over time was high (coefficient of variation, 34%). In only 16% of partici- Arch Intern Med. 2001;161:1534-1541 HE ORIGINAL concept of per- Prevalence estimates of vitamin B12 nicious anemia, defined as deficiency, defined as an elevated P-MMA lack of intrinsic factor, rep- concentration, vary widely. Studies from resents only one possible the United States suggest a prevalence of and rather rare presenta- 15% to 20% among elderly outpatients 12,13 Ttion of vitamin B12 deficiency. Strong in- (P-MMA level .0.37 µmol/L), whereas centives exist to establish accurate diag- European studies suggest a prevalence of nostic tests because of the often diffuse and 39% to 44% among healthy elderly indi- 14,15 nonspecific symptoms of mild vitamin B12 viduals (P-MMA level .0.24 µmol/L) deficiency. Anemia might be absent1,2 and and a prevalence of 24% among free- damage to the nervous system might be living elderly Dutch persons (P-MMA level reversible when treated in time3 but irre- .0.32 µmol/L).16 However, it is now un- versible after delayed diagnosis.4,5 certain to which extent an increase in the Use of the deoxyuridine suppression P-MMA level actually indicates or pre- test has permitted recognition of early and dicts a clinical condition in need of treat- 17,18 mild vitamin B12 deficiency characterized ment, and we still lack consensus about by biochemical dysfunction but lack of clear a gold standard for the diagnosis of vita- 6,7 clinical features of deficiency. However, min B12 deficiency. the test has limited clinical applicability be- In the present study we questioned cause it is cumbersome to perform. Dur- the clinical significance of an increased From the Departments of ing the past 10 years, determination of P-MMA level. The study aims were to Hematology, Aarhus plasma methylmalonic acid (P-MMA) and estimate the long-term trend of an ini- Amtssygehus (Drs Hvas and Ellegaard), and Clinical plasma total homocysteine (P-tHcy) lev- tially elevated P-MMA level in individu- Biochemistry, Aarhus els has been increasingly used. Level of als who did not receive cyanocobalamin Kommunehospital (Dr Nexø), P-MMA has been suggested as a more spe- therapy and to examine the associations Aarhus University Hospital, cific and sensitive marker than levels of between clinical manifestations related 8-11 Aarhus, Denmark. plasma cobalamins. to vitamin B12 deficiency and elevated (REPRINTED) ARCH INTERN MED/ VOL 161, JUNE 25, 2001 WWW.ARCHINTERNMED.COM 1534 ©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 PARTICIPANTS AND METHODS hemoglobin levels were 7.40 to 9.60 mmol/L for women and 8.40 to 10.80 mmol/L for men and for erythrocyte mean STUDY POPULATION cell volume was 85 to 100 fL. Plasma creatinine level was measured using the Jaffe method and a Roche Cobas Inte- From the laboratory information system (Department of gra 700 autoanalyzer (HiCo Creatinine Jaffe method; Boeh- Clinical Biochemistry, Skejby Sygehus, Aarhus University ringer Mannheim GmbH, Mannheim, Germany) (analyti- Hospital, Aarhus, Denmark) we obtained information on cal imprecision ,3%); the reference intervals were 44 to 1754 individuals aged 18 years and older living in the 115 µmol/L (0.5-1.3 mg/dL) for women and 62 to 133 Aarhus municipality (283000 inhabitants) who had a µmol/L (0.7-1.5 mg/dL) for men. P-MMA level greater than the reference interval (.0.28 µmol/L) between January 1, 1995, and December 31, INTERVIEW AND CLINICAL EXAMINATIONS 1997 (prestudy P-MMA) (Figure 1). Measurement of P-MMA concentration was requested by the physician in A history of present and previous diseases was obtained. charge of the patient because of suspected vitamin B12 Information on symptoms was obtained by structured deficiency. interview. We recorded anemia symptoms (daily fatigue, To exclude individuals who had received cyanoco- palpitations, shortness of breath, and angina on effort), balamin treatment we used a 3-step procedure. From Na- gastrointestinal symptoms (reduced sense of taste, sore tional Health Insurance, Aarhus County, we received in- mouth or tongue, daily reduced appetite, daily nausea, formation on all prescribed cyanocobalamin preparations.19 and daily diarrhea), and neurological symptoms using a For all individuals not excluded by this procedure we asked slightly modified version of the Neurological Symptom their general practitioner about cyanocobalamin prescrip- Score.22 Anemia, gastrointestinal, and Neurological tions. Finally, the initial interview included questions about Symptom Scores were summed to a total symptom score. previous and current treatment with cyanocobalamin. In addition, we recorded current drug use and consump- A total of 571 individuals (33%) had received cyano- tion of alcohol. Dietary vitamin B12 intake was estimated cobalamin treatment and were excluded, and another 28 using part of a validated food frequency question- individuals were excluded because they had participated naire.23-25 in a preceding pilot study. The neurological examination comprised testing for Of 1155 individuals with no report of cyanocobalamin vibration sense, joint position sense, cutaneous sensation, treatment, we included all 336 with prestudy P-MMA levels hyporeflexia, and muscular strength. Vibration sense was of 0.40 µmol/L or greater and took a geographical sample of tested at the medial malleolus, compared with a stimulus 647 individuals from 819 with prestudy P-MMA levels of 0.29 at the processus styloideus ulnae. Joint position sense was to 0.39 µmol/L. tested at the hallux and the index finger. Cutaneous sen- Of the 983 individuals addressed, 49 reported that they sation was tested by pinprick on the pulp of the hallux and had received cyanocobalamin treatment, 21 had died, and the index finger and by light touching of the dorsum of the 1 had emigrated, leaving 912 individuals eligible for fol- foot, the shin, and the forearm. A test for the Romberg sign low-up examination. Of these, 461 individuals (51%) was performed and gait was assessed. “Finger-nose” and volunteered to participate, but 10 did not attend the fol- “heel-knee-shin” tests were performed, as was testing for low-up examination and 19 reported during the interview dysdiadochokinesis. that they had received cyanocobalamin treatment. The fol- We used a slightly modified version of the Neurologi- low-up examinations of the 432 participants were per- cal Disability Score (a summed score of muscle strength, formed between October 7, 1998, and May 31, 1999, 1.0 reflexes, and sensory loss) to quantify the degree of pe- to 3.9 years after the prestudy P-MMA measurement. ripheral neuropathy.22 The Neurological Disability Score The study was approved by the Research Ethics Com- was the sum of 28 item scores, each ranging from 0 (nor- mittee of Aarhus County. Written informed consent was mal) to 4 (high degree of impairment). obtained from all participants. In addition, the examination included assessment of the nutritional state, inspection of the oral cavity, heart and LABORATORY TESTS lung auscultation, blood pressure measurement, and ab- dominal palpation. Levels of P-MMA were measured using stable isotope– All participants were examined by the same investi- dilution capillary gas chromatography–mass spectrometry gator (A.-M.H.), who did not know the laboratory test re- (analytical imprecision ,8%)20; the reference interval was 0.08 sults when the examinations were performed. to 0.28 µmol/L.21 Levels of P-tHcy were measured using an immunologi- STATISTICAL ANALYSIS cal method and Imx (Abbott Laboratories, Abbott Park, Ill) equipment (analytical imprecision ,5%). Plasma was sepa- For analyses of associations among laboratory test results rated from the blood cells within 2 hours. The reference in- we used the t test (independent samples), the x2 test for terval was 5.8 to 11.9 µmol/L. Levels of plasma cobalamins trend, linear regression, the Pearson correlation, and the were determined using an automated chemiluminescence Levene test. To analyze the associations between the bio- system (ACS: Centaur Automated Chemiluminescence chemical markers and the clinical manifestations we used System; Chiron Diagnostics Corporation, East Walpole, Mass) linear and logistic regression.