Introduction

Type 2 diabetes mellitus (T2DM) is one of the most prevalent and burdensome non-communicable diseases globally, and the pathophysiology of the disease comprises both environmental and genetic factors [1–4]. The aim of treatment of T2DM is to control blood glucose levels in the target and so avoid disease complications affecting the patient’s quality of life and life expectancy [5]. Since its discovery in 1922, metformin has become one of the most prescribed oral medications in the management of T2DM. The mechanisms of action of metformin include reducing glucose neogenesis in the liver and increasing insulin sensitivity in peripheral tissues, such as muscle and the liver [6]. Thus, metformin is very useful for glycemic control, and it helps reduce long-term complications of T2DM [7]. Although metformin is cheap, effective, and widely used, it has certain side effects that can decrease drug tolerance. One of the most neglected side effects of metformin is vitamin B12 deficiency. The precise mechanisms for metformin-induced vitamin B12 deficiency are still not fully understood; however, several potential mechanisms have been proposed, including reduced intrinsic factor secretion, inhibition of intrinsic factor-vitamin B12 complex absorption, changes in bile acid metabolism and reabsorption, and interference in the binding of the intrinsic factor-vitamin B12 complex to the cubilin receptor [8,9]. Long-term vitamin B12 deficiency can have severe consequences, such as anemia, hearing loss, and neuropathy (which is frequently misdiagnosed as diabetic peripheral neuropathy) [10,11].

The first report of metformin-induced vitamin B12 deficiency was published in 1971 [12]. After that, the HOME randomized control trial has undoubtedly confirmed the causative role of metformin in vitamin B12 deficiency [13]. Several meta-analysis studies also showed a strong association between metformin use and vitamin B12 deficiency in diabetic patients [14–17]. Current guidelines recommend annual measurement of vitamin B12 in patients undergoing long-term treatment with metformin [18]. However, the prevalence of metformin-induced vitamin B12 deficiency differs between locations and ethnicities; the rate varies from 4.3% up to 30% [12,19–21]. In Vietnam, there were around 4,000,000 diabetic patients in 2021 [3], and the majority of these patients were treated with metformin. Despite the common use of metformin, no assessment of the rate of vitamin B12 deficiency in Vietnam has been published. Therefore, this study set out to investigate the rate of vitamin B12 deficiency and associated factors in Vietnamese T2DM patients treated with metformin.

Materials and methods

Results

Baseline characteristics of participants

The characteristics of the participants are described in Table 1. The mean age of the studied population was 56.3, the median diabetes duration was 4.8 years, the median duration of metformin use was 48 months, and the median dose of metformin was 1000mg. Overall, the rate of vitamin B12 deficiency was 18.6% (29 out of 156 cases). Folate deficiency and macrocytic anemia were not found in any of the participants. Patients with vitamin B12 deficiency had significantly longer duration of diabetes, longer duration of metformin use, higher metformin dose, lower cholesterol levels, and lower LDL-C levels.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics of the participants.

https://doi.org/10.1371/journal.pone.0302500.t001

Factors associated with vitamin B12 deficiency

Univariate logistic regression analysis showed that factors statistically associated with vitamin B12 deficiency were duration of diabetes, duration of metformin use, metformin dose, and hemoglobin level (Table 2). Long-term use of metformin and metformin dose > median was statistically associated with vitamin B12 deficiency, and ORs were 3.91 and 4.16, respectively. When combining both long-term use of metformin and metformin dose > median dose, the OR increased to 5.25 (95% CI: 2.11–13.15) (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Odd ratios of long-term use of metformin and metformin dose > median for vitamin B12 deficiency.

https://doi.org/10.1371/journal.pone.0302500.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Univariate logistic regression analysis of factors associated with vitamin B12 deficiency.

https://doi.org/10.1371/journal.pone.0302500.t002

After adjusting for covariates using the two models (each model contained four variables) shown in Table 3, it was found that only metformin dose was associated with vitamin B12 deficiency in both models. Each 1mg increase in metformin dose was associated with a 0.2% increase in the risk of vitamin B12 deficiency. Metformin dose > median dose (>1000mg daily) was independently associated with vitamin B12 deficiency with OR = 4.10 (95% CI: 1.62–10.36).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Multivariate logistic regression analysis of factors associated with vitamin B12 deficiency.

https://doi.org/10.1371/journal.pone.0302500.t003

Discussion

The rate of vitamin B12 deficiency in T2DM patients treated with metformin was found in this study to be 18.6%. This result is similar to the results from a Korean study using the same criterion for vitamin B12 deficiency, namely a level of less than 300 pg/mL [22]. The rate, however, was up to 29.1% in a study conducted in the United States in which the criterion for vitamin B12 deficiency was a level of less than 350pg/mL [23]. On the other hand, the rate of vitamin B12 deficiency in patients treated with metformin was only 14.1% in a study in the Netherlands in which the criterion was a level of less than 150 pmol/L [24]. These differences might be explained by differences in measuring methods, ethnicities, or diet. However, in general, the more stringent the criteria used, the lower the detection rate for vitamin B12 deficiency.

Patients with vitamin B12 deficiency had a significantly longer duration of diabetes and longer exposure to metformin, as well as higher daily doses of metformin. These variables were also shown to be statistically associated with vitamin B12 deficiency in univariate logistic regression analysis. The observation that vitamin B12 deficiency is associated with duration of diabetes is similar to the findings of studies conducted in Yongin Severance Hospital, the Madigan Army Medical Center Family Medicine Clinic, and the Diabetic and Endocrine Centre at Al-Noor Specialist Hospital [22,23,25]; however, this association was not found in a study in Karachi, Pakistan [26]. The duration of metformin use and metformin dose are two important factors that have been reported to be associated with vitamin B12 deficiency in several meta-analyses [14,16]. However, these two factors are not always associated with vitamin B12 deficiency, as shown in a study in Saudi Arabia [25]. These inconsistencies can be explained by the fact that there are differences in the inclusion criteria, regions, vitamin supplements, and composition of diet. In this study, the long-term use of metformin (48 months or more) was strongly associated with vitamin B12 deficiency in the univariate analysis, with OR = 3.91. This finding together with recent real-world evidence [27] once again suggests that patients on long-term metformin treatment should be screened for vitamin B12 deficiency.

In the multivariate logistic regression analysis, metformin dose was the only parameter associated with vitamin B12 deficiency in both models. This result emphasizes the important role of metformin dose, not just duration of metformin use, in vitamin B12 deficiency. Despite choosing a low dose threshold of 1000mg, we observed a significant association of high metformin dose (> 1000mg daily) with vitamin B12 deficiency. Several studies have shown similar results, with metformin dose a strong factor associated with vitamin B12 deficiency [17,24,28,29]; thus, screening for this side effect needs to be performed in patients with either long-term use or a high metformin dose.

This study has limitations that need to be discussed. First, methylmalonic acid and homocysteine levels were not measured, and this may lead to the underdiagnosis of early vitamin B12 deficiency. In cellular level, vitamin B12 helps converting homocysteine to methionine, therefore, vitamin B12 deficiency will lead to increased homocysteine level [30]. Furthermore, vitamin B12 promotes the conversion of methylmalonyl-CoA to succinyl-CoA in mitochondria. Vitamin B12 deficiency results in the accumulation of methylmalonyl-CoA which will later convert to methylmalonic acid [30]. The increase of both methylmalonic acid and homocysteine levels reflects vitamin B12 deficiency in tissues and can cause of worsen diabetic peripheral neuropathy [31,32]. Second, participants were not systematically assessed for peripheral neuropathy, an important consequence of vitamin B12 deficiency. Due to the lack of resources, we could not measure methylmalonic acid and homocysteine levels as well as comprehensively examine diabetic peripheral neuropathy. Third, due to the nature of the cross-sectional study, we could not ascertain the causative effect of metformin on vitamin B12 deficiency. Finally, participants were only recruited in a single center, and the rate of vitamin B12 deficiency reported here may not represent the wider population of Vietnamese diabetic patients treated with metformin.

To the best of our knowledge, this is the first study in Vietnam investigating vitamin B12 deficiency in type 2 diabetes patients treated with metformin. We found that 18.6% of patients treated with metformin for at least 6 months had vitamin B12 deficiency. This result will pave the way for further studies with larger sample sizes, multi-center recruitment and, ideally, a longitudinal design to improve understanding of the role of metformin in vitamin B12 deficiency.

Supporting information