LETTER
Year : 2014  |  Volume : 60  |  Issue : 1  |  Page : 95-96

Authors' reply

Correspondence Address:
A K Singh
Department of Medicine, Rajendra Institute of Medical Sciences, Ranchi, Jharkhand
India

Source of Support: None, Conflict of Interest: None

Check

1. The power calculation was based on previous studies done by Reinstatler et al., ^[2] Liu et al., ^[3] and Wile et al. ^[4] Our outcome variable for power calculation defined B ₁₂ deficiency as B ₁₂ level < 220 pg/ml, as has been done by most of these studies.. The values are 16.2, ^[2] 52, ^[3] and 31% ^[4] respectively in the metformin arms of these studies. We thus chose a middle value of expected vitamin B ₁₂ level deficiency percentage to be 25%. Even in the study ^[1] quoted by the reader, 16.2% of the patients using metformin had borderline B ₁₂ deficiency < 220 pg/ml, while 5.8% (quoted in the abstract of the study by the reader ^[1] had severe B ₁₂ deficiency (B ₁₂ < 148 pg/ml). In our clinical setting and indeed in most of the studies around the world, a B ₁₂ level of between 200 - 220 pg/ml has been used as a marker of B ₁₂ deficiency. In our study, possible B ₁₂ deficiency and severe deficiency were seen in 21.4% and 7.1% of the patients, respectively. This proportion of B ₁₂ deficiency observed in our study was in agreement with studies across the world, including the National Health and Nutrition Examination Survey (NHANES) ^[2] referenced by the reader ^[1] and us, where this figure is 16.8 and 5.8%, respectively. Based on the studies ^[2],[3],[4] referenced above, we believe that valid assumptions for power calculation were made in our study leading to an appropriate sample size.
2. Even in the metformin-taking group in our study, 71.5% patients had normal vitamin B ₁₂ levels. Only 21.4% had possible B ₁₂ deficiency and 5.7% had severe B ₁₂ deficiency. As a majority of this group population had above-normal B ₁₂ levels, the mean B ₁₂ level was normal despite this group having a higher proportion of patients with B ₁₂ deficiency than the non-metformin group. The mean level of B ₁₂ in non-vegetarians was higher than vegetarians. Please refer erratum on page (92). Thus, "Vitamin B ₁₂ levels were significantly higher in the non-vegetarian population than in the vegetarian population (529.7 ± 162.3 versus 450 ± 170, p = 0.048, 95% CI 4.41-164.75)." The chi-square test was used to compare the proportion of -vegetarians and not the absolute numbers, as inferred by the reader. ^[1] As the metformin group had a higher number of patients (n = 84) than the non-metformin group (n = 52), therefore, despite having a higher number of -vegetarians (12 versus 6), the difference in proportion was statistically insignificant (12/84 versus 6/52, difference = 8.30%, 95% CI = -1.59 to 18.19, chi-square = 1.477, p = 0.2243).
3. Low B ₁₂ levels were observed in all patients with neuropathy. There was negative correlation between the B ₁₂ levels and neuropathy scores. As a majority of the patients in the metformin group had normal B ₁₂ levels, the mean B ₁₂ levels in this group was normal despite having a higher proportion of B ₁₂ deficiency. Thus, a higher proportion of patients with B ₁₂ deficiency in the metformin group accounts for a higher proportion of neuropathy observed in this group, as compared to the non-metformin group.
4. As has rightly been pointed out by the reader, ^[1] further studies should be done to evaluate which modes of B12 supplements (intramuscular or oral) are suitable to correct this deficiency.

:: References

1.	Shah V. Association of B12 deficiency and clinical neuropathy with metformin use in type 2 diabetes patients. Jr Postgrad Med 2014;60:95.
2.	Reinstatler L, Qi YP, Williamson RS, Garn JV, Oakley GP Jr. Association of biochemical B12 deficiency with metformin therapy and vitamin B12 supplements: The national health and nutrition examination survey, 1999-2006. Diabetes Care 2012;35:327-33.
3.	Liu KW, Dai DL, Ho W, Lau E, Woo J. Metformin-associated vitamin B12 deficiency in the elderly. Asian J Gerontol Geriatr 2011;6:82-7.
4.	Wile DJ, Toth C. Association of metformin, elevated homocysteine, and methylmalonic acid levels and clinically worsened diabetic peripheral neuropathy. Diabetes Care 2010;33:156-61.