Clinical Question about Diabetes Mellitus

Clinical Question: What is the role of metformin in the treatment of T1DM (type 1 diabetes mellitus), specifically latent autoimmune diabetes of adults (LADA)?

Author: Christine Giesing, PharmD Candidate 2016

Clinical Review: Kelly Cochran, PharmD, BCPS


Diabetes mellitus is characterized by prolonged elevated blood glucose. Type 1 diabetes mellitus (T1DM) evolves as autoimmune destruction of the beta cells in the pancreas prevents secretion of insulin, leading to elevated blood glucose. Elevated blood glucose in type 2 diabetes mellitus (T2DM) arises due to reduced insulin sensitivity in peripheral tissues, leading to impaired glucose absorption from the bloodstream. Sometimes confused with T2DM, latent autoimmune diabetes of adults (LADA) is a type of diabetes that occurs rarely. In this form, patients present as adults like in T2DM, but have circulating antibodies indicative of autoimmune beta cell destruction, as in T1DM. Generally, full beta cell obliteration takes longer in LADA than in T1DM, with most people having severely impaired beta cell function and requiring insulin at 6 years.1


Although the pathogeneses of these disorders differ, the eventual treatment in all cases is exogenous insulin. In T1DM, insulin is necessary for survival once beta cell destruction is complete.  T2DM patients often have a significant period of time when they can control their disease with oral medications. The commonly accepted first-line oral agent for the treatment of T2DM is metformin.2 The mechanism of action of metformin is three-fold: it decreases hepatic glucose production and glucose absorption from the intestine while improving insulin sensitivity in peripheral tissues. The treatment for LADA is less firmly established, but initial treatment with insulin may be beneficial in preserving beta cell function.3 Significant interest in the use of metformin for T1DM as well as LADA has developed in the past 10 years, and research has been conducted to determine the possibility of metformin for other types of diabetes.


To collect relevant literature, PubMed searches with the terms “LADA + metformin” and “Type 1 diabetes + metformin” were conducted. Results included one randomized, double-blind, placebo-controlled study of metformin in addition to insulin in T1DM, one prospective, multicenter analysis of the use of metformin in pediatric T1DM, one review of 9 studies looking at metformin in T1DM, one LADA treatment guideline released by the ADA, one longitudinal study comparing oral agents and insulin in LADA, and a case study of a LADA patient treated with metformin for five years.


The first group of studies addresses the use of metformin in T1DM. Since metformin has the ability to lower glucose release from the liver, it seems to have potential for use for T1DM patients. However, it also impairs absorption of glucose from the intestines and increases peripheral insulin sensitivity, indicating patients with T1DM who do not have insulin resistance may be at risk for hypoglycemia with metformin. The main concern for use of metformin is GI side effects. A 2014 review analyzed 9 trials studying metformin use in T1DM patients. Two of these studies looked at an adolescent population.  Overall, 3 studies showed significant reductions in A1c compared to baseline or placebo. The overall range of A1c change was -1% to +0.4%. However, 7 of the 9 studies reported significant reductions in total daily insulin use of -15.2 to +5.6 units or -0.024 to -0.04 units/kg/day for those studies reporting it. Only two studies revealed significant increases in hypoglycemia with metformin use, although a trend toward increased hypoglycemia occurrences with metformin was observed in all studies. Expected increases in GI side effects were seen in patients treated with metformin, but in general, it was well tolerated. Significant decrease in weight or BMI in patients treated with metformin was seen in 3 studies.4 A single center, randomized, double-blind, placebo-controlled study was done in 2002 comparing the use of metformin to placebo in addition to continuous subcutaneous insulin infusion in 67 patients. While A1c remained unchanged, metformin was associated with significant reductions in insulin requirements. Compared to an increase in daily insulin requirement of 1.7 +/- 8.83 units (2.8 +/- 12.8%) with placebo, patients treated with metformin experienced a decrease in total daily insulin of -4.3 +/- 9.9 units (-7.8 +/- 18%) (P = 0.0043). In addition, basal insulin requirements were statistically significantly reduced in the metformin group. In patients taking placebo, basal insulin dose increased by 1.9 +/- 5.7 units (8.8 +/- 27.1%), while a reduction of -2.6 +/- 3.2 units (-7.9 +/- 23.8%) was seen in patients treated with metformin (P = 0.023). Additionally, significant reductions in total cholesterol (P = 0.04) and LDL cholesterol (P = 0.05) were seen in patients on metformin. Eight patients taking metformin and two patients in the control group reported mild-to-moderate adverse GI effects (P = 0.069).5  A large prospective, multicenter analysis of patients under 21 with T1DM showed that pediatric patients on metformin tended to be older[median (interquartile range)]: [16.1 (14.1 – 17.6) vs. 15.2 (11.5 – 17.5) yr] , female (61.0 vs. 47.2%, p < 0.01) , and have higher BMI-SDS (body mass index standard deviation score) [+2.03 (+1.29 to +2.56) vs. +0.51 (−0.12 to +1.15); p < 0.01]. A follow up, longitudinal study of 285 of the patients showed a marginal improvement in BMI-SDS [−0.01 (−2.01 to +1.40)], but no change in A1c or insulin requirement.6  Contrarily, a 2012 randomized, placebo-controlled trial looking at 42 uncomplicated T1DM patients receiving metformin or placebo in addition to basal/bolus insulin found statistically significant improvements in body weight in the metformin group [−2.27kg (95% confidence interval: −3.99; −0.54); p = 0.012]. Statistically significantly improved flow-mediated dilation (FMD) of the brachial artery [1.32% (0.30; 2.43); p = 0.013], a measure of endothelial function, and 8-iso prostaglandins (PGF2a) [149 pg/mg creatinine (50; 248); p = 0.004], a marker of oxidative stress was also observed in the metformin group.  There was no significant difference in A1c, glycemic variability, or daily insulin dose.7  

In 2005, a peer-reviewed paper published by the ADA gave a description of LADA and guidance for treatment. According to this publication, LADA consists of a type 2 phenotype in combination with slowly progressive beta cell failure. The paper indicates insulin therapy is the ideal choice for LADA patients, adding that metformin is probably useful for obese LADA patients.8 A 2011 longitudinal study followed 37 LADA patients assigned to either early insulin or conventional treatment (consisting of diet modifications and either metformin, or in some cases, sulfonylurea, then insulin when necessary). The study was not randomized due to some counts of patient refusal of early insulin. There was no significant difference in C-peptide production between the groups after 12, 24, or 36 months, indicating no difference in endogenous insulin production. However, the conventional group experienced a statistically significant increase in A1c from 7.0% (+/-1.3%) to 7.5% (+/-1.5%) (P = 0.006), while the early insulin group experienced no such increase, going from 7.3% (+/-1.3%) to 7.2% (+/-0.7%) (P = 0.6). No significant hypoglycemic events occurred.9 A study of 103 diabetic patients in which 3 LADA patients were identified showed improved A1c (7% to 6.3% in LADA; 5.9% to 5.5% in T2DM) and decreased insulin release measured by peptide C (6.6 ng/mL to 3.4 ng/mL in LADA; 3.4 ng/mL to 2.7 ng/mL in T2DM) compared T2DM patients after one year of treatment with metformin in addition to insulin. Decreased peptide C in this study was interpreted to mean decreased excess insulin release in response to insulin resistance in LADA patients, as opposed to diminished ability of beta cells to produce insulin.10 Finally, a case report published in 2013 describes the course of disease over 5 years after diagnosis of LADA in a man who was 52 at the time of onset. This man was treated with diet modifications for 6 months after diagnosis; then metformin was added. He remained controlled to an A1c of 7% for all 5 years of treatment with metformin. The patient was slightly overweight with a BMI of 26.3, and had hypercholesterolemia and hypertriglyceridemia, but was otherwise healthy. During these five years, he had no evidence of micro- or macrovascular complications related to diabetes.11



I recommend the use of metformin in addition to insulin for the treatment of T1DM in patients who meet the following criteria:

  1. Have clear evidence of insulin resistance despite lifestyle interventions
  2. Minimize the risk of diabetic ketoacidosis through medical supervision, intensive insulin regimens and self-monitoring of blood glucose and urine ketones (when indicated)
  3. Understand the risk for metabolic acidosis, and frequently evaluate the efficacy of metformin12

for the following reasons:

  • Efficacy: One review including 9 trials demonstrated overall significant improvement in daily insulin required with the use of metformin. Lower daily insulin requirements with metformin were replicated in a randomized controlled trial (P = 0.0043). Additionally, some studies indicated metformin may improve cholesterol (P = 0.04), LDL (P = 0.05), body weight (0.012), FMD (P = 0.013), and PGF2a (P = 0.004).
  • Safety: Metformin use in T1DM appears to increase the risk for minor episodes of hypoglycemia. Metformin is well known to cause adverse GI effects, and is contraindicated in patients with a SCr > 1.5  mg/dl.13
  • Cost: Metformin is generally very low-cost, and appears on the value-list of many pharmacies ($4-5). 14, 15
  • Other considerations: Several studies showed the benefits of metformin might be especially applicable to adolescent and/or obese populations.
  • Strengths/Weaknesses: Strengths of the studies include appropriate treatments and dosages of the treatments, lengths of studies, inclusion/exclusion criteria, some randomization and blinding, accepted measures, and appropriate conclusions drawn. Limitations include no mentions of power and generally small patient populations.

I recommend the use of metformin in conjunction with insulin in LADA patients who display evidence of insulin resistance and still have some beta cell function, and have been educated on the risk of and treatment procedure for hypoglycemia for the following reasons:

  • Efficacy: a longitudinal study showed no difference in C-peptide between insulin and metformin in LADA patients; however, patients taking metformin monotherapy experienced an increase in A1c (P = 0.006), indicating metformin alone may not provide enough control in all LADA patients. Another study showed that compared to T2DM patients, those with LADA treated with metformin plus insulin had greater improvements in A1c, but more diminished insulin release after one year. A case study showed metformin alone was efficacious in treating one patient’s LADA for five years.
  • Safety: One study reported no major hypoglycemic events. The case report noted no adverse effects from metformin, and no evidence of micro- or macrovascular complications related to diabetes during treatment with metformin. Metformin is well known to cause adverse GI effects, and is contraindicated in patients with a SCr > 1.5  mg/dl.13
  • Cost: Metformin is generally very low-cost, and appears on the value-list of many pharmacies ($4-5). 14, 15
  • Other considerations: Metformin may be especially useful in obese LADA patients.  
  • Strengths/Weaknesses: Strengths of the studies include appropriate treatments and dosages of the treatments, lengths of studies, inclusion/exclusion criteria, some randomization and blinding, and accepted measures. Limitations include no mentions of power and generally small patient populations.



  1. Littorin B, Sundkvist G, Hagopian W, et al. Islet cell and glutamic acid decarboxylase antibod- ies present at diagnosis of diabetes predict the need for insulin treatment: a cohort study in young adults whose disease was initially labeled as type 2 or unclassifiable diabetes. Diabetes Care. 1999; 22: 409–412
  2. American Diabetes Association. Standards of medical care in diabetes – 2015. Clin Diabetes. 2015; 33(2): 97-111 doi: 10.2337/diaclin.33.2.97
  3. Cernea S, Buzzetti R, Pozzilli P. β-Cell Protection and Therapy for Latent Autoimmune Diabetes in Adults. Diabetes Care. 2009; 32(Suppl 2): S246-S252
  4. DeGeeter M, Williamson B. Alternative Agents in Type 1 Diabetes in Addition to Insulin Therapy: Metformin, Alpha-Glucosidase Inhibitors, Pioglitazone, GLP-1 Agonists, DPP-IV Inhibitors, and SGLT-2 Inhibitors. J Pharm Pract. 2014. doi:10.1177/0897190014549837
  5. Meyer L, Bohme P, Delbachian I, et al. The Benefits of Metformin Therapy During Continuous Subcutaneous Insulin Infusion Treatment of Type 1 Diabetic Patients. Diabetes Care. 2002; 25(12): 2153–2158
  6. Konrad K, Datz N, Engelsberger I, et al. Current use of metformin in addition to insulin in pediatric patients with type 1 diabetes mellitus: an analysis based on a large diabetes registry in Germany and Austria. Pediatr Diabetes 2014; doi: 10.1111/pedi.12203
  7. Pitocco D, Zaccardi F, Tarzia P, et al. Metformin improves endothelial function in type 1 diabetic subjects: a pilot, placebo-controlled randomized study. Diabetes Obes Metab. 2013; 15(5): 427-31; doi: 10.1111/dom.12041
  8. Stenstrom G, Gottsater A, Bakhtadze E, et al. Latent autoimmune diabetes in adults: definition, prevalence, beta-cell function, and treatment. Diabetes. 2005; 54 Suppl 2: S68-72
  9. Thunander M, Thorgeirsson H, Petersson C, et al. Beta cell function and metabolic control in latent autoimmune diabetes in adults with early insulin versus conventional treatment: a 3-year follow up. Eur J Endocrinol. 2011; 164(2) 239-45 doi: 10.1530/EJE-10-0901
  10. Peralta L, Bonilla C, Ybern G, et al. Detection of LADA-type diabetes in overweight diabetic patients. Is treatment with metformin suitable? Aten Primaria. 2007; 39(3): 133-7
  11. Hirata T, Shimada A, Maruyama T, et al. Slowly progressive type 1 diabetes treated with metformin for five years after onset. Intern Med. 2013; 52(23): 2635-7
  12. Faichney J, Tate P. Metformin in Type 1 Diabetes: Is this a good or bad idea? Diabetes Care. 2003; 26(5): 1655
  13. Metformin [package insert]. Princeton, NJ; Merck; Revised January 2009
  14. Walmart/Sam’s Club: $4 Prescription Program. Published June 5, 2008. Accessed August 25, 2015.
  15. Walgreens Prescription Savings Club: Value-Priced Medication List. Published 2012. Updated December 19, 2014. Accessed August 24, 2015.