Evaluating the correlation between serum vitamin B12 levels and various haematologic indices among metformin-treated type 2 diabetic patients: A prospective analytical study

Abstract

Background: Metformin-induced vitamin B12 deficiency state or metformin-induced hypocobalaminemia is gradually becoming an epidemic among diabetic patients on moderate-to-high doses of metformin or those diabetic patients on metformin for a long period of time. The potential effect of chronic metformin pharmacotherapy to cause vitamin B12 deficiency with abnormalities in haematologic indices and central/peripheral neuropathy has been widely reported. Long-term usage of metformin has been reported to be associated with intestinal malabsorption of vitamin B12 culminating in vitamin B12 deficiency with likely associated haematologic abnormalities (including macro-ovalocytic anaemia and immune dysfunctioning due to hypersegmentation of polymorphonuclear leukocytes), central/peripheral neuropathy and manifestation of biochemical derangements such as elevated homocysteine and methyl malonate levels. Aim: This study aimed to determine the correlation between serum vitamin B12 levels and various haematologic indices among metformin-treated type 2 diabetic patients in a clinical practice setting with the rational purpose of alleviating/preventing the associated derangements. Materials and Methods: This was a case-control, prospective, analytical, observational study of 200 adult participants (100 per group) attending the Endocrinology Out-patients Clinic of Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria. For each participant, serum vitamin B12 level was determined using a vitamin B12 immunoassay technique, while the corresponding complete blood count was done using PCE-210N autohaematology analyser. Data were presented using tables and figures. Chi-square test was used to compare categorical variables, Student t-test was used in comparing means of continuous variables, while Pearson’s correlation study was done to determine the existence of any statistically significant correlation(s) between the serum vitamin B12 levels and various haematologic indices among the participants. Results: Approximately 41% versus 20% of the metformin-treated and metformin-naive diabetic patients, respectively, had frank vitamin B12 deficiency. There was a statistical difference between the total serum vitamin B12 levels in male and female diabetic patients with p = 0.048. Also, statistically significant differences existed with respect to mean corpuscular volume (MCV), mean corpuscular haemoglobin and total white blood cells count among the metformin-treated and metformin-naive diabetic patients. Furthermore, a statistically significant weak positive correlation existed between pack cell volume (PCV) and serum vitamin B12 level ( r = +0.148, p = 0.037), but a statistically significant weak negative correlation existed between MCV and serum vitamin B12 level ( r = −0.245, p = 0.0001). In addition, the test for associations between the serum vitamin B12 categorization status or metformin exposure status and the peripheral neuropathy components assessment revealed that there were statistically significant associations between the serum vitamin B12 categorization status or metformin exposure status versus pain sense ( p < 0.0001 or <0.001), vibration sense ( p < 0.0001 or <0.001) and light touch sense ( p < 0.0001 or <0.001) among the participants. Conclusion: In this study, statistically significant weak positive and weak negative correlations existed between serum vitamin B12 level versus PCV, and serum vitamin B12 level versus MCV, respectively. The peripheral neuropathy components assessment revealed that there were statistically significant associations between the serum vitamin B12 categorization status or metformin exposure status versus pain sense, vibration sense and light touch sense among the participants.