Propensity score matching yielded patients taking metformin plus insulin and 12 taking metformin plus sulfonylurea. Patients took metformin for a median of 14 interquartile range [IQR] 5—30 months, and the median glycated hemoglobin level was 8.

In the group who added insulin, first hypoglycemia events occurred, and first events occurred in the group who added sulfonylurea For recurrent hypoglycemia, there were events in the insulin group and events in the sulfonylurea group In separate competing risk analyses, the adjusted HR for hypoglycemia was 1.

Interpretation: Among patients using metformin who could use either insulin or sulfonylurea, the addition of insulin was associated with a higher risk of hypoglycemia than the addition of sulfonylurea.

This finding should be considered by patients and clinicians when discussing the risks and benefits of adding insulin versus a sulfonylurea. Hypoglycemia remains one of the most common medication-related adverse events among patients with diabetes and a leading cause of hospital admissions and emergency department visits.

Sulfonylurea is easier to initiate, but insulin dose can be modified in response to daily variation in food intake, exercise or other variables that cause fluctuations in glucose values. Within the Veterans Health Administration clinical practice guideline, both the combination of metformin plus sulfonylurea or the use of bedtime insulin combined with metformin are considered acceptable based on level I evidence.

We recently reported that intensification of metformin with insulin compared with sulfonylurea was associated with an increased risk of all-cause mortality among veterans with diabetes. We assembled a retrospective cohort of patients whose data were contained in the Veterans Health Administration database.

For patients enrolled in Medicare or Medicaid, we obtained data on enrolment, claims files and prescriptions. The institutional review boards of Vanderbilt University and the Tennessee Valley Healthcare System approved the study.

The study population comprised veterans aged 18 years and older who received regular care at least once every days for at least 2 years Appendix 1, Supplemental Figure 1, available at www. Metformin initiators were patients who filled a metformin prescription from Oct. These patients were eligible for inclusion in the treatment intensification cohort on the date that they filled a prescription for either insulin or sulfonylurea intensification prescription.

We restricted the cohort to patients who were adherent to metformin by excluding patients with no metformin available on the date of their intensification prescription or the prior days. Follow-up began days after the intensification prescription to distinguish patients who continued intensified therapy from those who switched to either insulin or sulfonylurea monotherapy.

We excluded patients receiving hospice care or dialysis at the time of intensification. Follow-up continued through the st day without medical contact inpatient, outpatient or pharmacy use ; or nonpersistence, defined as 90 days without metformin or addition of a third antidiabetic drug; an outcome; death; or study end Sept.

The outcome was a composite of 3 types of hypoglycemia event: hospital admission for hypoglycemia, emergency department visit for hypoglycemia, or outpatient blood glucose level less than 3. The outcome was adapted from prior studies evaluating hypoglycemia risks associated with medication use.

Hospital admission for hypoglycemia was defined as a primary discharge diagnosis of hypoglycemia or poisoning by insulin or other antidiabetic agents International Classification of Diseases, 9th Revision [Clinical Modification] code An emergency department visit for hypoglycemia was defined as an emergency department visit with a compatible coded diagnosis: xx, For example, an outpatient blood glucose level of 2.

We also evaluated the risk of recurrent hypoglycemia events all hypoglycemia events, not just the first one during study follow-up. Study covariates were collected in the days before intensification and included age, sex, race white, black or other , fiscal year, indicators of health care use hospital admission, months from hospital admission to intensification, nursing home use, number of outpatient visits, Medicare or Medicaid use in past year , physiologic variables body mass index, blood pressure, HbA 1C level, low-density lipoprotein level, presence of proteinuria and creatinine, which was used to calculate estimated glomerular filtration rate [eGFR, using the Chronic Kidney Disease Epidemiology Collaboration formula] , 33 duration of monotherapy before intensification of treatment diabetes duration , selected medications, smoking and presence of comorbidities Appendix 1, Supplemental Table 1.

For missing covariates, we conducted multiple imputation using the Markov Chain Monte Carlo method and a noninformative Jeffreys prior. The primary analysis investigated the risk of hypoglycemia first events and recurrent events all events in propensity score—matched cohorts.

The propensity score modelled the probability of use of metformin plus insulin, given covariates and service network of care.

Because of size differences between the 2 groups, observations of metformin plus insulin were matched to observations of metformin plus sulfonylurea using a optimal matching algorithm 35 , 36 Appendix 1, Supplemental Table 2 and Supplemental Figure 2.

We used Cox proportional hazards models to compare outcomes for metformin plus insulin versus metformin plus sulfonylurea reference in the matched cohorts adjusted for covariates with robust standard errors to account for multiple events within patients.

Our second analysis used a competing risk model, in which hypoglycemia competes with the risk of death. The evaluation of recurrent hypoglycemic events used the mean cumulative count method in the presence of a competing risk. We conducted sensitivity analyses to determine the robustness of our findings.

First, we used the intensification regimen to define drug exposure and ignored any subsequent changes in regimens or the day refill requirement persistent exposure not required. Lack of censoring for nonpersistence increased follow-up and events.

Second, in the cohort construction, we were interested in long-term outcomes, cardiovascular disease and death; therefore, follow-up began days post—treatment intensification to minimize exposure misclassification.

To evaluate early effects of intensification, we performed an alternate analysis using a new-user design, in which follow-up began at intensification and continued through the first days. Finally, we stratified the reference group by sulfonylurea type and assessed the effect of insulin compared with either glyburide or glipizide and glimepiride.

Analyses were conducted using R 41 , 42 and SAS version 9. After propensity score matching, our study included 14 patients; patients were taking metformin plus insulin, and 12 were taking metformin plus sulfonylurea.

Characteristics between the matched groups were similar Table 1. Reasons for censoring were treatment change The median follow-up prior to censoring or reaching the first outcome was 1. The median follow-up when considering recurrent events was 1.

This yielded Event rates were We evaluated median HbA 1C levels at the start of follow-up and over time. HbA 1C levels declined to a median of 7. HbA 1C levels remained similar between the two groups over follow-up Figure 2. The separate analysis of the risk of first hypoglycemia accounting for the competing risk of death showed similar results adjusted HR 1.

Risk of hypoglycemic event among patients taking metformin plus insulin versus metformin plus sulfonylurea in the propensity score—matched cohort.

Cumulative incidence plots for risk of first hypoglycemic event A and mean cumulative counts for recurrent hypoglycemic event B among patients taking metformin plus insulin and patients taking metformin plus sulfonylurea.

In sensitivity analyses in which patients remained in their original exposure group persistent exposure was not required , there were first hypoglycemia events among patients who intensified their treatment with insulin and first events among those who intensified their treatment with sulfonylurea, yielding When evaluating the risk of recurrent events, there were and events, yielding In the primary analyses, follow-up began 6 months before treatment intensification to ensure that patients were intensifying rather than switching therapies.

In an alternate analysis that evaluated the first 6 months following medication initiation, there were insulin initiators and 62 sulfonylurea initiators. Some of these patients switched to single-agent insulin or sulfonylurea, or discontinued their medication use.

After propensity score matching, there were and 23 patients Appendix 1, Supplemental Table 4. Those who initiated insulin versus sulfonylurea had Recurrent event rates were Sensitivity analyses that varied the outcome definition were consistent with our main findings Appendix 1, Supplemental Table 5 and Supplemental Figure 3.

Subgroups stratified by age, eGFR or hospital admission in the 90 days before treatment intensification demonstrated consistent results with no evidence of effect modification.

Results were also consistent when patients taking glyburide or glipizide and glimepiride were used as reference Figure 3 ; Appendix 1, Supplemental Table 6. Adjusted hazard ratio for risk of first hypoglycemic event and recurrent hypoglycemic event among subgroups. Values above 1. In a national cohort of veterans with diabetes intensifying their metformin treatment, we found that among those who could use either medication, the addition of insulin versus sulfonylurea to metformin was associated with an increased risk of hypoglycemia.

Our group recently reported that adding insulin versus sulfonylurea to metformin was associated with an increased risk of all-cause mortality in this same cohort adjusted HR 1. By design i. Results from this study would allow those patients and their providers to make informed decisions about drug choice for intensification.

The high rate of documented hypoglycemia also serves as a reminder to individualize HbA 1C targets, with higher targets for older patients and those with factors that increase hypoglycemia risk.

A systematic review and meta-analysis of glyburide versus other secretagogues and insulin on hypoglycemia risk included 21 randomized trials and patients followed from 4 weeks to 10 years. Consistent with the meta-analytic results, our estimate of hypoglycemia appeared higher when insulin was compared with glipizide and glimepiride versus glyburide.

There was no difference in HbA 1C levels over follow-up between those who intensified with insulin versus sulfonylurea Figure 2 , which suggests that the outcome differences were not due to large between-group differences in glycemic targets.

A systematic review that included 5 trials of intensive versus conventional glucose control reported a twofold risk of hypoglycemia in patients who received intensive treatment intensification. With outcomes adjudication, 42 of deaths were adjudicated to be probably, possibly or definitely related to hypoglycemia 27 intensive v.

In a follow-up to the Action in Diabetes and Vascular Disease ADVANCE trial, Zoungas and colleagues 7 noted that, of 11 patients who completed a median follow-up of 5 years, 2. The proportion who died was However, that secondary analysis did not determine whether hypoglycemia was causal or a marker of illness severity.

In our cohort, patients with reduced renal function, those aged 65 years and older and those with a recent hospital admission had high rates of hypoglycemia, with rates including recurrent hypoglycemia of 4 to 5 per person-years of follow-up.

Some have advocated using insulin without dose reduction for patients with chronic kidney disease stages 3 to 5. HbA 1C targets should be individualized to account for age, comorbidities and life expectancy. Our study has several limitations. First, although we used multiple strategies to address confounding by indication and disease severity, including selecting patients adherent to metformin at intensification, matching propensity scores and adjusting covariates, residual confounding from difficult to measure factors, such as patient frailty, remains possible.

Second, we did not account for medication doses or concurrent medication use that may increase hypoglycemia risk. Third, veterans may not receive all their care or medications in veteran facilities, 24 , 25 which may result in missing outcomes or medications, which we partially addressed through supplementation with Medicare and Medicaid data.

We also chose to combine substantial hypoglycemia with low outpatient glucose measurements, which may be symptomatic or asymptomatic. We did not quantify the patient-level impact of those low glucose values. Fourth, to reduce exposure misclassification, we started follow-up 6 months after intensification.

Although this approach excluded the initial study regimen exposure period, a separate sensitivity analysis evaluated hypoglycemia in the first 6 months following regimen intensification with similar results. Finally, our population reflects a typical veteran population; therefore, caution is warranted when extrapolating our findings to other settings.

In our study, the addition of insulin to metformin monotherapy was associated with a higher risk of hypoglycemia than the addition of sulfonylurea.

This finding, in combination with our previous finding of increased risk of death, should be considered by patients and clinicians when discussing the risks and benefits of adding insulin versus a sulfonylurea. This result also calls into question the apparent equivalence of these choices in current guidelines.

Competing interests: Christianne Roumie, Adriana Hung and Tom Elasy report grants from Veterans Affairs Clinical Science Research and Development. Jea Young Min reports grants from the National Center for Advancing Translational Sciences. Marie Griffin reports grants from the Agency for Healthcare Research and Quality and MedImmune.

No other competing interests were declared. Contributors: Christianne Roumie, Robert Greevy, Carlos Grijalva, Adriana Hung, Tom Elasy and Marie Griffin designed the study. Call your doctor or call if taking in sugar does not help, or the hypo symptoms come back. Make sure your friends and family know about your diabetes and the symptoms of low blood sugar levels so they can recognise a hypo if it happens.

In rare cases, it's possible to have a serious allergic reaction anaphylaxis to metformin. You or the person who's unwell may also have a rash that's swollen, raised, itchy, blistered or peeling. These can be signs of a serious allergic reaction and may need immediate treatment in hospital.

Taking metformin can cause vitamin B12 deficiency if you take it for a long time. This can make you feel very tired, breathless and faint, so your doctor may check the vitamin B12 level in your blood.

If your vitamin B12 levels become too low, vitamin B12 supplements will help. These are not all the side effects of metformin. For a full list, see the leaflet inside your medicine packet. Page last reviewed: 24 March Next review due: 24 March Home Medicines A to Z Metformin Back to Metformin.

Side effects of metformin. Common side effects These common side effects of metformin happen in more than 1 in people. There are things you can do to help cope with them: Feeling sick nausea Take metformin with food to reduce the chances of feeling sick.

If you're in a lot of pain, speak to your pharmacist or doctor. Try chewing sugar-free gum. Immediate action required: Call now if:.

your lips, mouth, throat or tongue suddenly become swollen you're breathing very fast or struggling to breathe you may become very wheezy or feel like you're choking or gasping for air your throat feels tight or you're struggling to swallow your skin, tongue or lips turn blue, grey or pale if you have black or brown skin, this may be easier to see on the palms of your hands or soles of your feet you suddenly become very confused, drowsy or dizzy someone faints and cannot be woken up a child is limp, floppy or not responding like they normally do their head may fall to the side, backwards or forwards, or they may find it difficult to lift their head or focus on your face You or the person who's unwell may also have a rash that's swollen, raised, itchy, blistered or peeling.

Maintenance of normal blood glucose concentrations Enlarge image Close. Maintenance of normal blood glucose concentrations Maintenance of normal blood glucose concentrations in individuals with prediabetes during treatment with metformin.

Related Content. An emerging connection between circadian rhythm disruption and type 2 diabetes mellitus. Medical Professionals Metformin revisited. Show the heart some love! Give Today.

Help us advance cardiovascular medicine. Find a doctor. Explore careers. Sign up for free e-newsletters. About Mayo Clinic. About this Site. Contact Us. Health Information Policy. Media Requests. News Network. Price Transparency. Medical Professionals. Clinical Trials. Mayo Clinic Alumni Association.

Refer a Patient. Executive Health Program. International Business Collaborations. Supplier Information. Admissions Requirements. Degree Programs.

Research Faculty.

This content does not have an English version. This content does not have an Arabic version. Metformin revisited. April 11, Chemical structure for metformin Enlarge image Close. Chemical structure for metformin Chemical structure for metformin 1,1-dimethylbiguanide; C4H11N5.

Maintenance of normal blood glucose concentrations Enlarge image Close. Maintenance of normal blood glucose concentrations Maintenance of normal blood glucose concentrations in individuals with prediabetes during treatment with metformin. Related Content. An emerging connection between circadian rhythm disruption and type 2 diabetes mellitus.

Medical Professionals Metformin revisited. Show the heart some love! Give Today. Help us advance cardiovascular medicine. Find a doctor. Explore careers. Sign up for free e-newsletters.

About Mayo Clinic. About this Site. Contact Us. Health Information Policy. Media Requests. News Network. Price Transparency.

Medical Professionals. Clinical Trials. Mayo Clinic Alumni Association. Refer a Patient. A second cannula was inserted into an antecubital vein of the contralateral arm. Both cannulas were connected to long thin tubes that enabled blood sampling and adjustment of the rate of dextrose infusion from an adjacent room without being noticed by the subject.

After a min baseline period starting at h, plasma glucose was lowered to a plateau of 2. Ninety minutes after the start of insulin infusion h , plasma glucose was further decreased to a plateau of 2.

Plasma glucose levels were measured Glucose Analyzer, Beckman Coulter, Inc. At h, insulin infusion was stopped, and plasma glucose levels were returned to normal by dextrose infusion.

At h, insulin infusion was started again at a rate of 2. During this hypoglycemic clamp, a plasma glucose plateau of 2. After the clamp, the subject went home. Eating was allowed until h.

On d 16 of the treatment period subjects again reported to the research unit at h. This third clamp was performed in the same way as the first hypoglycemic clamp starting at h.

During the baseline period and the first min of the first and third hypoglycemic clamps, blood samples were drawn every 15 min for determination of insulin, epinephrine, norepinephrine, ACTH, cortisol, GH, and glucagon in plasma and serum, respectively.

The hormone concentrations were measured as previously described Serum IGF-I concentrations were measured by ELISA Active IGF-I, Diagnostics Systems Laboratories, Inc. A semiquantitative symptom questionnaire was presented every 15 min during the hypoglycemic clamps.

Subjects rated from 0 none to 9 severe each of the following symptoms: tremor, hunger, palpitations, sweating, irritability, anxiety, dizziness, blurred vision, difficulty in thinking, tingling, and faintness.

Consistent with the categories used by previous investigators 22 , the first six symptoms were considered autonomic, and the other five were considered neuroglycopenic.

All values are presented as the mean ± sem. Statistical analysis was based on ANOVA for repeated measures, including the factors drug for metformin vs. placebo treatment and ante-hypo for the effects of antecedent hypoglycemia first vs. third hypoglycemia. To compare the hormonal and symptomatic responses between the treatment conditions and the first and third hypoglycemic clamps, areas under the curve with reference to the baseline values were calculated.

For pairwise comparisons, paired t test was applied. Plasma glucose concentrations on both first and third hypoglycemic clamps were very similar between the metformin and placebo conditions Fig.

Likewise, the decrease in plasma glucose during the third and third hypoglycemic clamps was almost identical.

Serum insulin concentrations as well as the rate of dextrose infusion required to control plasma glucose did not differ between the metformin and placebo conditions for all hypoglycemic clamps data not shown.

On d 15, before starting the clamp experiments, baseline levels of serum insulin and plasma glucose were significantly lower in the metformin than in the placebo condition[ insulin, The responses of epinephrine, norepinephrine, ACTH, cortisol, glucagon, and GH to the first and third hypoglycemias are shown in Fig.

Compared with the first hypoglycemic clamp, counterregulatory responses of these hormones were attenuated during the third hypoglycemic clamp. However, metformin had no effect on the responses of epinephrine, norepinephrine, ACTH, cortisol, and glucagon Table 1. Serum glucagon levels were distinctly higher in the metformin than in the placebo condition throughout the experiments Fig.

before and during the hypoglycemic clamps. However, the hypoglycemic response of glucagon itself was not changed by metformin Table 1. Metformin slightly enhanced the GH response to hypoglycemia Fig.

Supplementary determinations of serum IGF-I revealed significantly lower levels in the metformin than the placebo condition at baseline Mean ± sem plasma or serum concentrations of epinephrine, norepinephrine, ACTH, cortisol, glucagon, and GH during the first left and third right hypoglycemic clamp.

Mean ± SEM for hormonal and subjective responses AUC min to the first and third hypoglycemias. Ante-hypo, Effects of antecedent hypoglycemia first vs.

third hypoglycemia ; Drug, effects of metformin vs. placebo ; Drug × ante-hypo, interaction between metformin and antecedent hypoglycemia.

Subjects experienced distinctly more autonomic and neuroglycopenic symptoms during the first than during the third hypoglycemic clamp Fig. Metformin slightly increased awareness of neuroglycopenic symptoms, but had no influence on autonomic symptoms Table 1.

Mean ± sem autonomic and neuroglycopenic symptom scores during the first left and third right hypoglycemic clamps. The novel finding reported here is that metformin does not adversely affect hormonal and symptomatic responses to hypoglycemia.

The present data confirm previous results 20 — 25 showing an attenuated hormonal and symptomatic response to hypoglycemia after antecedent hypoglycemia. However, metformin does not interact with the effect of antecedent hypoglycemia on subsequent counterregulation. Animal studies 13 — 17 have indicated a sympathoinhibitory effect of metformin with possible relevance for hypoglycemic counterregulation.

Activation of the sympathoadrenal system essentially contributes to the normalization of glucose levels and also to awareness of hypoglycemia by promoting typical hypoglycemic symptoms such as sweating, tremor, and palpitations often referred to as autonomic symptoms 18 , However, the present data indicate that metformin in healthy men neither inhibits the sympathoadrenal nor impairs the symptomatic, especially autonomic, response to hypoglycemia.

It should be pointed out that the metformin dose studied here, i. Although it appears unlikely that a higher dose of metformin will adversely affect hypoglycemic hormonal and symptomatic responses to hypoglycemia, it should be emphasized that this possibility cannot be excluded by the present data.

Metformin enhanced the GH response to hypoglycemia. This effect is probably a consequence of a tonic reducing influence of metformin on IGF-I concentration, as IGF-I is an important inhibitory feedback signal to the pituitary in the control of GH release A decreased feedback on the pituitary caused by lower IGF-I levels will increase the responsiveness of GH release on stimuli such as hypoglycemia.

However, although remarkable and new, this finding appears to be of minor clinical relevance with regard to hypoglycemic counterregulation, where GH does not play a central role Serum glucagon concentrations were higher in the metformin than in the placebo condition throughout the clamps, indicating that metformin increased glucagon levels regardless of the glycemic state.

There may be several explanations for this finding. Metformin could increase the glucagon concentration via its reducing influence on insulin and IGF-I release. Both insulin 30 , 31 and IGF-I 28 directly suppress pancreatic glucagon secretion.

Alternatively, metformin could increase circulating glucagon levels simply by impairing glucagon degradation.

In the context of hypoglycemic counterregulation, it is important to note that in the present study the response of glucagon to hypoglycemia was unaffected by metformin. However, it should be pointed out that the present data do not exclude an adverse effect of metformin on glucose recovery after hypoglycemia.

There is some evidence that metformin decreases hepatic glucose output by inhibiting the effects of glucagon on hepatic glucose metabolism 32 , Metformin slightly, but significantly, decreased the fasting plasma glucose concentration, most likely as a consequence of diminished endogenous glucose production 8.

Reduced insulin secretion and increased glucagon secretion probably represent physiological responses to decreased plasma glucose concentration, preventing a further decline in plasma glucose.

The fact that the decrease in plasma glucose concentration during metformin treatment was small suggests that these mechanisms that prevent hypoglycemia, i. low insulin and high glucagon levels, were operative and effective.

One may speculate that in advanced type 2 diabetes these mechanisms may be less operative, which could explain the occurrence of hypoglycemic episodes during metformin monotherapy in diabetic patients 4 , This view is supported by the finding of unchanged glucagon levels during metformin therapy in type 2 diabetic patients The present results were obtained in healthy subjects, which limits conclusions with regard to the conditions in diabetic patients.

Healthy subjects were used because they show a distinctly lower variability in physiological responses to hypoglycemia than diabetic patients. In diabetic patients, the magnitude of hormonal and symptomatic responses to hypoglycemia depends on many factors, such as previous glycemic control 35 , 36 , recent hypoglycemic episodes 22 , 23 — 25 , and prevalence of neuropathy 37 — These factors are difficult to sufficiently control experimentally to allow clear-cut conclusions with regard to the effects of metformin on hormonal and symptomatic responses to hypoglycemia.

On the other hand, there is no evidence to date against the assumption that the influence of metformin on the clinically most important variables, i. sympathoadrenal and symptomatic responses to hypoglycemia, will not differ between healthy and diabetic subjects.

From a clinical point of view, the important finding here is that metformin does not impair the sympathoadrenal or symptomatic response to hypoglycemia, thereby leaving the awareness of hypoglycemia unaffected.

Previous studies 40 — 42 have clearly indicated that hypoglycemia unawareness is an important risk factor for the experience of severe hypoglycemic episodes. The present data lend themselves to predict that metformin in combination with insulin therapy will not increase the frequency and severity of hypoglycemic episodes.

Recent observations, in fact, suggest that adding metformin to insulin therapy decreases rather than increases the frequency of symptomatic and biochemical hypoglycemic episodes in patients with type 2 diabetes 9.

We thank Christiane Zinke for her expert and invaluable laboratory assistance and Anja Otterbein for her organizational work. Diabetes Control and Complications Trial Research Group Adverse events and their association with treatment regimens in the diabetes control and complications trial.

Diabetes Care 18 : — UK Prospective Diabetes Study UKPDS Group Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes UKPDS Lancet : — Veneman TF , Erkelens DW Clinical review hypoglycemia unawareness in noninsulin-dependent diabetes mellitus.

J Clin Endocrinol Metab 82 : — Google Scholar. UK Prospective Diabetes Study UKPDS Group Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes UKPDS DeFronzo RA , Goodman AM Efficacy of metformin in patients with non-insulin-dependent diabetes mellitus.

N Engl J Med : — Fontbonne A , Charles MA , Juhan-Vague I , et al. BIGPRO Study Group. Diabetes Care 19 : — Charles MA , Morange P , Eschwege E , Andre P , Vague P , Juhan-Vague I Effect of weight change and metformin on fibrinolysis and the von Willebrand factor in obese nondiabetic subjects: the BIGPRO1 Study.

Biguanides and the Prevention of the Risk of Obesity. Diabetes Care 21 : — Bailey CJ , Turner RC Metformin. Yki-Jarvinen H , Ryysy L , Nikkila K , Tulokas T , Vanamo R , Heikkila M Comparison of bedtime insulin regimens in patients with type 2 diabetes mellitus.

Biguanides are antihyperglycemic agents, not hypoglycemic agents; they promote euglycemia but alone are unlikely to cause hypoglycemia. Biguanides are used both as monotherapy and in combination with other oral hypoglycemic agents. They can exacerbate hypoglycemia caused by other types of antidiabetic medicines [ 3,4 ].

The major toxicity from acute or chronic biguanide use is lactic acidosis. The high rate of severe lactic acidosis from phenformin led to the withdrawal of this drug from the US market in , although it remains available in several countries. Metformin is the principal biguanide in clinical use.

The management of metformin toxicity is reviewed here. A summary table to facilitate emergent management is provided table 1. General issues relating to hypoglycemia, the therapeutic use of biguanides, and the general clinical management of drug intoxication are presented separately.

Why UpToDate? Product Editorial Subscription Options Subscribe Sign in. Learn how UpToDate can help you.

It should be emphasized to people with type 2 diabetes that healthy behaviour interventions and weight loss can lead to withdrawal of antihyperglycemic medication and even remission of type 2 diabetes in some cases 2. The Look AHEAD Action for Health in Diabetes trial showed that an intensive healthy behaviour intervention resulted in a significantly greater weight loss and likelihood of diabetes remission after 1 year compared to standard care, with the greatest benefit seen in persons with new-onset type 2 diabetes Antihyperglycemic therapy with metformin may also be initiated at diagnosis, depending on the current and target glycated hemoglobin A1C.

A1C targets may be higher up to 8. S42 for recommendations. It should be emphasized to people with type 2 diabetes that reductions in A1C levels are associated with better outcomes even if recommended glycemic targets cannot be reached, and inability to achieve A1C target should not be considered a treatment failure 3,4.

If the A1C level at diagnosis is less than 1. If healthy behaviour interventions are insufficient to achieve target A1C levels within 3 months, they should be combined with antihyperglycemic medications. In the face of significant hyperglycemia i. People who have evidence of metabolic decompensation e.

Insulin may later be tapered or discontinued once stability is achieved. In general, A1C will decrease by about 0. By and large, the higher the baseline A1C, the greater the A1C reduction seen for each given agent.

The maximum effect of noninsulin antihyperglycemic agent monotherapy is observed by 3 to 6 months 5,6. Evidence indicates that initial combination of metformin with another agent is associated with an additional mean 0.

The initial use of combinations of submaximal doses of antihyperglycemic agents produces more rapid and improved glycemic control and fewer side effects compared to monotherapy at maximal doses 13— Table 1 lists all the available classes of antihyperglycemic therapies.

These include insulin and noninsulin therapies. Unless contraindicated, metformin should be the initial pharmacotherapy in people with type 2 diabetes. The recommendation to use metformin as the initial agent in most people is based on its efficacy in lowering A1C, its relatively mild side effect profile, long-term safety track record, affordability, negligible risk of hypoglycemia and lack of weight gain.

Compared to sulfonylureas, metformin monotherapy has comparable A1C-lowering effects, but better glycemic durability 18 , a lower risk of hypoglycemia 19 , less weight gain 19,20 and lower CV risk Metformin is associated with less weight gain than thiazolidinediones 21 , and has better A1C lowering and weight loss than DPP-4 inhibitors The demonstrated CV benefit of metformin monotherapy in newly diagnosed participants who were overweight in the UKPDS trial 17 is also cited as a reason to select metformin as first-line treatment, although other evidence from a meta-analysis of metformin trials has been equivocal on this matter 21, Metformin should be started at a low dose and gradually increased over several weeks to minimize the risk of gastrointestinal side effects.

If metformin is contraindicated or if initial combination therapy is required, then a second agent should be chosen based on individual patient characteristics and the efficacy and safety profile of other agents see Table 1 and Figure 2. DPP-4 inhibitors, GLP-1 receptor agonists or SGLT2 inhibitors should be considered over other antihyperglycemic agents as they are associated with less hypoglycemia and weight gain 19,23—27 , provided there are no contraindications and no barriers to affordability or access.

Insulin may be used at diagnosis in individuals with marked hyperglycemia and can also be used temporarily during illness, pregnancy, stress or for a medical procedure or surgery. The use of intensive insulin therapy may lead to partial recovery of beta cell function when used in people with metabolic decompensation, and studies suggest that early insulin treatment may induce remission in people with newly diagnosed type 2 diabetes 28,29— Trials of this approach are ongoing.

The natural history of type 2 diabetes is that of ongoing beta cell function decline, so blood glucose BG levels often increase over time even with excellent adherence to healthy behaviours and therapeutic regimens Treatment must be responsive as therapeutic requirements may increase with longer duration of disease.

If A1C target is not achieved or maintained with current pharmacotherapy, treatment intensification is often required.

A review of potential precipitants of increasing A1C e. infection, ischemia and medication adherence should first be conducted, and current therapy may need to be modified if there are significant barriers to adherence. Healthy behaviour interventions, including nutritional therapy and physical activity, should continue to be optimized while pharmacotherapy is being intensified.

Metformin should be continued with other agents unless contraindicated. In general, when combining antihyperglycemic agents with or without insulin, classes of agents that have different mechanisms of action should be used.

sulfonylureas and meglitinides or DPP-4 inhibitors and GLP-1 receptor agonists is currently untested, may be less effective at improving glycemia and is not recommended at this time. Table 1 identifies the mechanism of action for all classes of antihyperglycemic agents to aid the reader in avoiding the selection of agents with overlapping mechanisms.

A1C , glycated hemoglobin; CHF , congestive heart failure; CV , cardiovascular; CVD , cardiovascular disease; DKA , diabetic ketoacidosis; eGFR , estimated glomerular filtration rate; HHS , hyperosmolar hyperglycemic state.

Figure 1 continued Management of hyperglycemia in type 2 diabetes. In deciding upon which agent to add after metformin, there must be consideration of both short-term effects on glycemic control and long-term effects on clinical complications.

While intensive glycemic control with a variety of agents is associated with a reduction in microvascular complications 3 and possibly CV complications 34 see Targets for Glycemic Control chapter, p. S42 , Table 1 highlights agent-specific effects on CV or microvascular complications e.

CKD based on trials where glycemic differences between treatment arms were minimized. The effect of exogenous insulin on the risk of CV complications has been shown to be neutral 35, There was a neutral effect on CV outcomes and cancer, and a slight increase in hypoglycemia and weight 36, Earlier trials evaluated effects of thiazolidinediones on CV events.

Meta-analyses of smaller studies suggested possible higher risk of myocardial infarction MI with rosiglitazone 38,39 ; however, CV events were not significantly increased in a larger randomized clinical trial 40, Conversely, the evidence for pioglitazone suggests a possible reduced risk of CV events, but the primary CV outcome was neutral 42, While these agents have comparable glucose-lowering effects to other drugs, the edema, weight gain, risk of congestive heart failure CHF 44 , increased risk of fractures 45,46 and inconsistent data regarding MI risk with rosiglitazone 38—40 and bladder cancer risk with pioglitazone significantly limit the clinical utility of this drug class 47, Based on controversies regarding rosiglitazone, in , the United States Food and Drug Administration FDA required that all new antidiabetic therapies undergo evaluation for CV safety at the time of approval.

Subsequently, several industry-sponsored placebo-controlled trials were initiated to evaluate CV outcomes of drugs from 3 newer classes: DPP-4 inhibitors, GLP-1 receptor agonists and SGLT2 inhibitors see Table 2.

Trial durations are from 1. Therefore, findings from these trials are directly relevant to people with established type 2 diabetes and clinical CV disease or multiple risk factors. Studies have not evaluated whether findings are generalizable to people with new-onset type 2 diabetes or those at average or lower CV risk.

Three DPP-4 inhibitor trials have been completed Table 2. None have shown inferiority or superiority compared to placebo for the risk of major CV events 49, There was a non-statistically significant increase in hospitalizations for CHF with alogliptin in the Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care EXAMINE trial 49 and there is limited experience treating people with a history of CHF with linagliptin; therefore, these agents should be used with caution in that setting.

Moreover, a secondary analysis of the data suggested a possibly higher relative risk of unstable angina and all-cause mortality with saxagliptin in those under 65 years The significance of these findings is unclear and further studies are needed.

The GLP-1 receptor agonist, lixisenatide, was also shown to be non-inferior to placebo after a median 2. Figure 2 Antihyperglycemic medications and renal function. Based on product monograph precautions.

CKD, chronic kidney disease; CV , cardiovascular; GFR , glomerular filtration rate; TZD , thiazolidinedione. Three approved and one unapproved antihyperglycemic agent, thus far, have shown benefit in reducing major CV outcomes in individuals with clinical CVD, the SGLT2 inhibitors empagliflozin 53 and canagliflozin 54 , and the GLP-1 receptor agonists liraglutide 55 and semaglutide Those treated with empagliflozin had significantly fewer CV events CV death, nonfatal MI, nonfatal stroke compared to placebo-treated participants after a median 3.

In a secondary analysis, empagliflozin was associated with a significant reduction in hospitalizations for CHF 4. Recent meta-analyses of SGLT2 inhibitors confirmed a significant benefit of this class of agents on major CV outcomes, which was largely driven by EMPA-REG OUTCOME results 58— The CANagliflozin cardioVascular Assessment Study CANVAS program, which integrated findings from 2 placebo-controlled trials CANVAS and CANVAS-R , evaluated the CV effects of canagliflozin The trials enrolled 10, participants 4, in CANVAS and 5, in CANVAS-R with type 2 diabetes mean duration Over a median follow up of 2.

There were no statistical differences in the individual components of the composite outcome. There was a reduction in hospitalization for heart failure and in several adverse renal outcomes; however, these were considered exploratory outcomes due to pre-specified rules of evidence hierarchy.

While one-third of participants did not have CVD, a significant decrease in the primary endpoint was only found in those with CVD. Therefore, as with other CV outcome trials, these results largely apply to people with type 2 diabetes requiring add-on antihyperglycemic therapy who have established clinical CVD.

Canagliflozin was also associated with an increase in fracture rates HR 1. Importantly, canagliflozin was associated with doubling in the risk of lower extremity amputation HR 1. This risk was strongest in participants with a prior amputation.

Canagliflozin should, therefore, be avoided in people with a prior amputation, as the harms appear to be greater than the benefits in that population.

The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results LEADER trial enrolled 9, participants with longstanding type 2 diabetes median duration Over a median follow up of 3.

Therefore results are most applicable to people with type 2 diabetes with clinical CVD requiring add-on antihyperglycemic therapy. The Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes SUSTAIN-6 enrolled 3, participants with a mean duration of type 2 diabetes of After a median follow up of 2.

There was, however, a higher rate of diabetic retinopathy complications in the semaglutide group compared to placebo group 3. It is unclear at this time if there is a direct effect of semaglutide or other explanations for this unexpected difference in retinopathy complication rates, although the risk appeared greatest in individuals with pre-existing retinopathy and rapid lowering of A1C.

All 4 trials reported lower rates of kidney disease progression in the treated groups compared to placebo 53,55, It should also be noted that the majority of people in these trials had pre-existing CVD and required add-on antihyperglycemic therapy. In addition, because these were placebo-controlled trials, no conclusions can be made about how the cardioprotective properties of empagliflozin, canagliflozin, liraglutide and semaglutide compare to those of other agents.

CV outcome trials for other agents are expected to be completed by ; therefore, based on evidence to date, a GLP-1 receptor agonist or SGLT2 inhibitor with demonstrated CV outcome benefit should be considered as initial add-on therapy for people with pre-existing type 2 diabetes and clinical CV disease who have not achieved target A1C on existing treatment to reduce CV risk.

A careful review of the methods and findings from these trials was conducted by an independent committee. While primary analyses results were similar for canagliflozin, empagliflozin and liraglutide, it was concluded that the strength of evidence for CV benefit was weaker for canagliflozin than for the other agents.

This conclusion was based on three factors. First, in an interim analysis of the CANVAS study for medication approval necessitated unblinding of study data. A decision was then made to combine this study with the CANVAS-R study, presumably to provide greater power for CV outcomes.

The interim unblinding and protocol revision were viewed as potential threats to internal validity, thereby weakening the strength of evidence for benefit. Second, while canagliflozin was associated with a significant decrease in the composite MACE outcome, there was no significant benefit on individual outcomes, such as all-cause or CV mortality.

Third, the findings of increased risk of fractures and amputations with canagliflozin treatment in the context of a noninferiority design where the comparator is placebo was particularly concerning, indicating that harms may outweigh benefits.

For these reasons, the committee decided that the uncertainty regarding benefits should be acknowledged with a lower grade of recommendation for canagliflozin than for other agents with demonstrated CV benefit.

In the absence of evidence for long-term clinical benefit, agents effective at A1C lowering should be considered in terms of both the degree of baseline hyperglycemia needing correction, and any heightened concerns regarding hypoglycemia e.

elderly people or those with renal or hepatic dysfunction see Diabetes in Older People chapter, p. While most medications added to metformin lower A1C to a similar extent, insulin and insulin secretagogues are associated with higher rates of hypoglycemia than other agents 21,23,24, In those who are stable, other agent-specific advantages and disadvantages should be weighed as treatment is individualized to best suit the patient's needs and preferences.

Each of the agents listed in Table 1 and Figure 1 has advantages and disadvantages to consider. Figure 2 illustrates the basis on which agent selection is influenced by renal function as dictated by product monograph precautions. Recent meta-analyses have summarized head-to-head comparisons of metformin-based combinations 19,24,62, Combinations of metformin with a sulfonylurea, a thiazolidinedione TZD , an SGLT2 inhibitor and a DPP-4 inhibitor have comparable A1C-lowering effects 19,24,62—66 , while the combination of metformin with a GLP-1 receptor agonist reduced A1C more than combination with a DPP-4 inhibitor.

TZDs, insulin and sulfonylureas are associated with the most weight gain 1. Hypoglycemia risk is also lower with TZDs, DPP-4 inhibitors, SGLT2 inhibitors and GLP-1 receptor agonists compared to sulfonylureas and insulin 19,24,62—65,67, Network meta-analyses that indirectly compared the net benefits of second- and third-line treatment options have found similar results 21,23,24,69— Evidence on comparative effectiveness of acarbose and orlistat is limited, although they are associated with a low risk of hypoglycemia and weight gain.

The safety of incretin agents, SGLT2 inhibitors and TZDs in pregnancy is unknown; therefore, these agents should be avoided or discontinued in women who are pregnant or planning a pregnancy see Diabetes and Pregnancy chapter, p.

If a sulfonylurea is added to metformin, gliclazide should be considered as first choice as it is associated with a lower risk of hypoglycemia 67,72 , CV events and mortality relative to other sulfonylureas Glimepiride is also associated with a lower risk of CV events and mortality 73 , but has a similar rate of hypoglycemia 67,72 compared to other sulfonylureas.

For people already taking metformin and a sulfonylurea, the addition of either a DPP-4 inhibitor, a GLP-1 receptor agonist or SGLT2 inhibitor may be considered as they are associated with effective A1C lowering with less hypoglycemia than insulin or TZDs 21,69,70,74,75 ; GLP-1 receptor agonists and SGLT2 inhibitors are also associated with weight loss 70,71 see Weight Management in Diabetes chapter, p.

For instance, the combination of a DPP-4 inhibitor or a GLP-1 receptor agonist and an SGLT2 inhibitor added to metformin has been shown to be as safe and more efficacious at lowering A1C after 24 weeks than either agent alone 76, SGLT2 inhibitors and GLP-1 receptor agonists added to metformin have also been shown to reduce systolic BP compared to metformin alone, and add-on of SGLT2 inhibitors reduce systolic BP more than add-on of sulfonylureas or DPP-4 inhibitors A combination of noninsulin antihyperglycemic agents and insulin often effectively controls glucose levels.

Insulin treatment includes long-acting or intermediate-acting insulin analogue injections once or twice daily for basal glycemic control, and bolus injections at mealtimes for prandial glycemic control.

Adding insulin to noninsulin antihyperglycemic agent s may result in better glycemic control with a smaller dose of insulin 78 , and may induce less weight gain and less hypoglycemia than that seen when non-insulin antihyperglycemic agents are stopped and insulin is used alone 79, A single injection of an intermediate-acting NPH 81 or long-acting insulin analogue insulin glargine U, insulin glargine U, insulin detemir or insulin degludec 82—84 may be added.

The addition of bedtime insulin to metformin therapy leads to less weight gain than insulin plus a sulfonylurea or twice-daily NPH insulin When insulin is used in type 2 diabetes, the insulin regimen should be tailored to achieve good metabolic control while trying to avoid hypoglycemia.

With intensive glycemic control, there is an increased risk of hypoglycemia, but this risk is lower in people with type 2 diabetes than in those with type 1 diabetes.

The mode of insulin administration continuous subcutaneous infusion vs. injections , the number of insulin injections 1 to 4 per day and the timing of injections may vary depending on each individual's situation As type 2 diabetes progresses, insulin requirements will likely increase and higher doses of basal insulin intermediate-acting or long-acting analogues may be needed.

DPP-4 inhibitors, GLP-1 receptor agonists and SGLT2 inhibitors have been shown to be efficacious at further lowering glucose levels when combined with insulin therapy 87— A meta-analysis determined that the addition of a GLP-1 receptor agonist to basal insulin regimens results in greater A1C reduction, more weight loss and less hypoglycemia compared to the addition of bolus insulin A GLP-1 receptor agonist should, therefore, be considered before bolus insulin as add-on therapy in people on basal insulin with or without other agents who require antihyperglycemic treatment intensification if there are not barriers to affordability or access.

If glycemic control is suboptimal on treatment regimens that include basal insulin with other agents, bolus insulin at mealtimes short- or rapid-acting analogues may be added.

Generally, once bolus insulin is introduced into a treatment regimen, either as a separate mealtime bolus or as part of a premixed containing regimen, insulin secretagogues, such as sulfonylureas and meglitinides, should be discontinued.

Concomitant therapy with metformin and, if applicable, a GLP-1 receptor agonist, DPP-4 inhibitor or SGLT2 inhibitor should be continued with regimens containing bolus insulin unless contraindicated, to allow for improved glycemic control with less risk of weight gain and hypoglycemia The reduction in A1C achieved with insulin therapy depends on the dose and number of injections per day A meta-analysis of 12 articles compared basal-bolus and biphasic insulin regimens, and found that both approaches are equally efficacious at lowering A1C, with comparable effects on hypoglycemia risk and weight—although basal-bolus regimens were modestly more efficacious in people with type 2 diabetes already on insulin Bolus insulin should be initiated using a stepwise approach starting with 1 injection at the largest meal and additional mealtime injections at 3-month intervals if needed , as it was shown to be as efficacious at A1C lowering as a full basal-bolus regimen, and is associated with less hypoglycemia and greater patient satisfaction after 1 year Lower rates of hypoglycemia have been observed in some studies of individuals with type 2 diabetes treated with rapid-acting insulin analogues insulin aspart, insulin lispro, insulin glulisine compared to those treated with short-acting regular insulin — Use of long-acting basal insulin analogues insulin detemir, insulin glargine, insulin degludec in those already on antihyperglycemic agents reduces the relative risk of symptomatic and nocturnal hypoglycemia compared to treatment with NPH insulin 83,,— Meta-analyses indicate a relative reduction of 0.

NPH Insulin degludec has been associated with lower rates of overall and nocturnal hypoglycemia compared to glargine U 82,84, After 32 weeks of treatment, insulin degludec was associated with a significantly lower rate of the primary endpoint of overall symptomatic hypoglycemic episodes rate ratio 0.

The proportions of patients with hypoglycemic episodes were 9. The Trial Comparing Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events DEVOTE randomized patients with type 2 diabetes at high risk of CV disease to insulin degludec or glargine U, and found no difference in the primary outcome of CV events but a significant decrease in severe hypoglycemia with degludec 4.

There is also some evidence of lower hypoglycemia rates with glargine U compared to glargine U and may also be considered over glargine U if reducing hypoglycemia is a priority Efficacy and rates of hypoglycemia are similar between glargine U and detemir Aside from effects of some antihyperglycemic agents on the occurrence of hypoglycemia and weight, there are adverse effects unique to each agent Table 1.

Gastrointestinal side effects are more common with metformin, alpha glucosidase inhibitors, GLP-1 receptor agonists and orlistat than with other agents. Metformin can cause diarrhea, which tends to resolve over time and is minimized with starting at a low dose and subsequent slow titration of the dosage.

Extended-release metformin can also be used to improve tolerability in individuals experiencing gastrointestinal side effects with immediate-release metformin — Metformin is also associated with an approximate 2-fold increased incidence of vitamin B12 deficiency — , and vitamin B12 levels should be measured periodically in people taking metformin or with signs or symptoms of deficiency such as impaired proprioception or peripheral neuropathy.

GLP-1 receptor agonists and, less commonly, DPP-4 inhibitors can cause nausea and GLP-1 receptor agonists can also cause diarrhea. A meta-analysis comparing the risk of congestive heart failure between antihyperglycemic therapies found an increased risk with TZDs and DPP-4 inhibitors driven by higher risk with saxagliptin 44 , although another meta-analysis and a large observational study of over one million participants failed to find an increased risk of heart failure with DPP-4 inhibitors compared to other agents.

Reports of acute pancreatitis have been noted with DPP-4 inhibitors and GLP-1 receptor agonists. A small significant increase in pancreatitis but not pancreatic cancer was seen with DPP4-inhibitors in a meta-analysis of 3 large randomized controlled trials of over 20, participants However, a recent large Canadian observational study of over 1.

SGLT2 inhibitors are associated with a 3- to 4-fold increased risk of genital mycotic infections 19,69,95 , as well as higher rates of urinary tract infections, volume depletion, rare acute kidney injury and rare DKA , Canagliflozin treatment is associated with an increased risk of fractures 54, and a twofold increased risk of amputations In a retrospective analysis, empagliflozin was not associated with an increased risk of amputations in the EMPA-REG trial There is evidence of a higher risk of bladder cancer with pioglitazone in some studies 47,48 but not others — , and some reports of increased bladder cancer risk with dapagliflozin GLP-1 receptor agonists have been shown to promote the development of pancreatic and medullary thyroid cancer in rodents, but an increased risk has not been seen in humans Semaglutide was associated with a higher risk of retinopathy in SUSTAIN-6 see above Earlier epidemiological evidence suggesting a possible link between insulin glargine and cancer has not been substantiated in review of clinical trial data for either glargine or detemir 36,, Insulin glargine U may be considered over insulin glargine U to reduce overall and nocturnal hypoglycemia [Grade C, Level 3 ].

A1C , glycated hemoglobin; BG , blood glucose; BP , blood pressure; CHF , congestive heart failure; CHD , coronary heart disease; CI , confidence interval; CV , cardiovascular; CVD , cardiovascular disease; DKA , diabetic ketoacidosis; HR , hazard ratio; MI ; myocardial infarct; NPH , neutral protamine Hagedorn; TZD , thiazolidinedione.

Appendix 9. Examples of Insulin Initiation and Titration Regimens in People With Type 2 Diabetes. Literature Review Flow Diagram for Chapter Pharmacologic Glycemic Management of Type 2 Diabetes in Adults.

From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group P referred R eporting I tems for S ystematic Reviews and M eta- A nalyses: The PRISMA Statement. PLoS Med 6 6 : e pmed For more information, visit www. Goldenberg reports personal fees from Abbott, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Merck, Novo Nordisk, Sanofi, and Servier, outside the submitted work.

MacCallum reports personal fees from Janssen and Novo Nordisk, outside the submitted work. No other author has anything to disclose. All content on guidelines. ca, CPG Apps and in our online store remains exactly the same.

For questions, contact communications diabetes. Become a Member Order Resources Home About Contact DONATE. Next Previous. Key Messages Recommendations Figures Full Text References. Chapter Headings Introduction Treatment Regimens Effects of Antihyperglycemic Agents on Microvascular and Cardiovascular Complications Effects of Antihyperglycemic Agents on Glycemic Control and Other Short-Term Outcomes Insulin Treatment in Type 2 Diabetes Adverse Effects Other Relevant Guidelines Relevant Appendices Author Disclosures.

Key Messages Healthy behaviour interventions should be initiated in people newly diagnosed with type 2 diabetes. In the absence of metabolic decompensation, metformin should be the initial agent of choice in people with newly diagnosed type 2 diabetes, unless contraindicated.

In people with clinical cardiovascular CV disease in whom A1C targets are not achieved with existing pharmacotherapy, an antihyperglycemic agent with demonstrated CV outcome benefit should be added to antihyperglycemic therapy to reduce CV risk.

In people receiving an antihyperglycemic regimen containing insulin, in whom glycemic targets are not achieved, the addition of a GLP-1 receptor agonist, DPP-4 inhibitor or SGLT2 inhibitor may be considered before adding or intensifying prandial insulin therapy to improve glycemic control with less weight gain and comparable or lower hypoglycemia risk.

Key Messages for People with Diabetes Some people who have type 2 diabetes can achieve their target blood glucose levels with nutrition guidance and physical activity alone, but most also need glucose-lowering medications.

The decision about which medications are best for you depends on many factors, including your blood glucose level, symptoms, other health problems you have and affordability of medications.

Your health-care provider may even combine medications that act differently on your body to help you control your blood glucose. Glucose-lowering medications for type 2 diabetes include: First-line glucose-lowering medication: Metformin: Metformin is generally the first choice for people with type 2 diabetes because of its safety, low cost and possible heart benefits.

It works by making your body respond better to insulin so that your body uses insulin more effectively. Metformin also lowers glucose production from the liver. Nausea and diarrhea are possible side effects and usually go away within 1 to 2 weeks as your body gets used to the medicine.

It is associated with a low risk of hypoglycemia and does not cause weight gain. If metformin and healthy behaviour changes are not enough to control your blood glucose level, other medications can be added. Second-line glucose-lowering medication: DPP-4 inhibitors: These medications work to lower blood glucose by increasing insulin levels after meals and lowering glucagon levels a hormone that raises blood glucose.

They do not cause weight gain and are associated with a low risk of hypoglycemia. GLP-1 receptor agonists: These injectable medications act when blood glucose increases after eating.

They increase insulin levels, which helps lower blood glucose and lower glucagon levels a hormone that raises blood glucose. They also slow digestion and reduce appetite.

Possible side effects include nausea, which usually goes away with time. They are associated with weight loss and a low risk of hypoglycemia. SGLT2 inhibitors: These medications work by eliminating glucose into the urine. Side effects may include genital yeast infections, urinary tract infections, increased urination and low blood pressure.

Insulin secretagogues meglitinides, sulfonylureas : These medications help your pancreas release more insulin. Possible side effects include hypoglycemia and weight gain. Thiazolidinediones: Like metformin, these medications make the body's tissues more sensitive to insulin.

Side effects include weight gain and an increased risk of heart failure and fractures. Insulin therapy: Some people who have type 2 diabetes need insulin therapy as well. Depending on your needs, your health-care provider may prescribe a mixture of insulin types to use throughout the day and night.

Often, people with type 2 diabetes start insulin use with 1 injection of long-acting insulin at night. Discuss the pros and cons of different treatment plans with your healthcare provider.

Together, you can decide which medication is best for you after considering many factors, including costs and other aspects of your health.

Christian Meier, MD ; Christian Meier, MD. Stephan Krähenbühl, MD, PHD ; Stephan Krähenbühl, MD, PHD. Susan S. Jick, DSC ; Susan S. Jick, DSC. Christoph R.

Meier, PHD, MSC Christoph R. Meier, PHD, MSC. Corresponding author: Christoph R. Meier, meierch uhbs. Diabetes Care ;31 11 — Article history Received:. Get Permissions. toolbar search Search Dropdown Menu. toolbar search search input Search input auto suggest.

Table 1— Oral antidiabetes drug use in seven case subjects with lactic acidosis. Case no. Use of oral antidiabetes drugs. Approximate duration of metformin use prior to diagnosis months. Time between last metformin prescription and diagnosis days.

Relevant comorbidities and outcome died or survived. View Large. Case subjects. Control subjects. P value. Cases subjects. Controls subjects. Insulin no use 1, Exposure to antidiabetes drugs. Current No No This study was funded by an unconditional grant from Merck SA, Lyon, France.

UK Prospective Diabetes Study UKPDS Group: Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes UKPDS Tahrani AA, Varughese GI, Scarpello JH, Hanna FW: Metformin, heart failure, and lactic acidosis: is metformin absolutely contraindicated?

Brown JB, Pedula K, Barzilay J, Herson MK, Latare P: Lactic acidosis rates in type 2 diabetes. Diabetes Care. Stang M, Wysowski DK, Butler-Jones D: Incidence of lactic acidosis in metformin users. Salpeter S, Greyber E, Pasternak G, Salpeter E: Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus.

Cochrane Database Syst Rev. Lalau JD, Race JM: Lactic acidosis in metformin-treated patients. Prognostic value of arterial lactate levels and plasma metformin concentrations. Drug Saf. Holstein A, Egberts EH: Risk of hypoglycemia with oral antidiabetic agents in patients with type 2 diabetes.

Exp Clin Endocrinol Diabetes. Lalau JD, Mourlhon C, Bergeret A, Lacroix C: Consequences of metformin intoxication. Bolen S, Feldman L, Vassy J, Wilson L, Yeh HC, Marinopoulos S, Wiley C, Selvin E, Wilson R, Bass EB, Brancati FL: Systematic review: comparative effectiveness and safety of oral medications for type 2 diabetes mellitus.

Ann Intern Med. UK Prospective Diabetes Study UKPDS Group: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes UKPDS van Staa T, Abenhaim L, Monette J: Rates of hypoglycemia in users of sulfonylureas.

J Clin Epidemiol. Garcia Rodriguez LA, Perez Gutthann S: Use of the UK General Practice Research Database for pharmacoepidemiology. Br J Clin Pharmacol. Wood L, Martinez C: The general practice research database: role in pharmacovigilance.

Jick H, Jick SS, Derby LE: Validation of information recorded on general practitioner based computerised data resource in the United Kingdom. Jick SS, Kaye JA, Vasilakis-Scaramozza C, Garcia Rodriguez LA, Ruigomez A, Meier CR, Schlienger RG, Black C, Jick H: Validity of the general practice research database.

Meier C, Kraenzlin ME, Bodmer M, Jick SS, Jick H, Meier CR: Use of thiazolidinediones and fracture risk. Arch Intern Med. Salpeter SR, Greyber E, Pasternak GA, Salpeter EE: Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus: systematic review and meta-analysis.

Misbin RI, Green L, Stadel BV, Gueriguian JL, Gubbi A, Fleming GA: Lactic acidosis in patients with diabetes treated with metformin. N Engl J Med. Runge S, Mayerle J, Warnke C, Robinson D, Roser M, Felix SB, Friesecke S: Metformin-associated lactic acidosis in patients with renal impairment solely due to drug accumulation?

Diabetes Obes Metab. Rachmani R, Slavachevski I, Levi Z, Zadok B, Kedar Y, Ravid M: Metformin in patients with type 2 diabetes mellitus: reconsideration of traditional contraindications.

Eur J Intern Med. Jennings AM, Wilson RM, Ward JD: Symptomatic hypoglycemia in NIDDM patients treated with oral hypoglycemic agents.

Miller CD, Phillips LS, Ziemer DC, Gallina DL, Cook CB, El-Kebbi IM: Hypoglycemia in patients with type 2 diabetes mellitus. Hirai FE, Moss SE, Klein BE, Klein R: Severe hypoglycemia and smoking in a long-term type 1 diabetic population: Wisconsin Epidemiologic Study of Diabetic Retinopathy.

DIABETES CARE. View Metrics. Email alerts Article Activity Alert. Online Ahead of Print Alert. This content does not have an Arabic version. Metformin revisited. April 11, Chemical structure for metformin Enlarge image Close.

Chemical structure for metformin Chemical structure for metformin 1,1-dimethylbiguanide; C4H11N5. Maintenance of normal blood glucose concentrations Enlarge image Close.

Maintenance of normal blood glucose concentrations Maintenance of normal blood glucose concentrations in individuals with prediabetes during treatment with metformin. Related Content. An emerging connection between circadian rhythm disruption and type 2 diabetes mellitus.

Medical Professionals Metformin revisited. Show the heart some love! Give Today. Help us advance cardiovascular medicine. Find a doctor.

Explore careers. Sign up for free e-newsletters. About Mayo Clinic. About this Site. Contact Us. Health Information Policy. Media Requests. News Network. Price Transparency. Medical Professionals. Clinical Trials. Mayo Clinic Alumni Association. Refer a Patient. Executive Health Program.

Metformin is the principal biguanide in clinical use. The management of metformin toxicity is reviewed here. A summary table to facilitate emergent management is provided table 1.

General issues relating to hypoglycemia, the therapeutic use of biguanides, and the general clinical management of drug intoxication are presented separately. Why UpToDate? Product Editorial Subscription Options Subscribe Sign in. Learn how UpToDate can help you.

Select the option that best describes you. View Topic. Font Size Small Normal Large. Metformin poisoning. Formulary drug information for this topic. UK Prospective Diabetes Study Group U. prospective diabetes study Diabetes 44 : — Andersen D , Haugan K , Sorensen AM , Christensen S , Petersen JS Cardiovascular actions of chronic intracerebroventricular administration of metformin in normotensive rats.

Pharmacol Toxicol 81 : 7 — Muntzel MS , Abe A , Petersen JS Effects of adrenergic, cholinergic and ganglionic blockade on acute depressor responses to metformin in spontaneously hypertensive rats.

J Pharmacol Exp Ther : — Petersen JS , Liu W , Kapusta DR , Varner KJ Metformin inhibits ganglionic neurotransmission in renal nerves.

Hypertension 29 : — Petersen JS , DiBona GF Acute sympathoinhibitory actions of metformin in spontaneously hypertensive rats. Hypertension 27 : — Petersen JS , DiBona GF Effects of central metformin administration on responses to air-jet stress and on arterial baroreflex function in spontaneously hypertensive rats.

J Hypertens 15 : — Gerich JE , Mokan M , Veneman T , Korytkowski M , Mitrakou A Hypoglycemia unawareness.

Endocr Rev 12 : — Towler DA , Havlin CE , Craft S , Cryer P Mechanism of awareness of hypoglycemia. Diabetes 42 : — Fruehwald-Schultes B , Born J , Kern W , Peters A , Fehm HL Adaptation of cognitive function to hypoglycemia in healthy men. Diabetes Care 23 : — Fruehwald-Schultes B , Kern W , Deininger E , et al.

J Clin Endocrinol Metab 84 : — Veneman T , Mitrakou A , Mokan M , Cryer PE , Gerich J Induction of hypoglycemia unawareness by asymptomatic nocturnal hypoglycemia. Heller SR , Cryer PE Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after 1 episode of hypoglycemia in nondiabetic humans.

Diabetes 40 : — Davis SN , Shavers C , Mosqueda-Garcia R , Costa F Effects of differing antecedent hypoglycemia on subsequent counterregulation in normal humans.

Diabetes 46 : — Davis MR , Shamoon H Counterregulatory adaptation to recurrent hypoglycemia in normal humans. J Clin Endocrinol Metab 73 : — Davis SN , Shavers C , Costa F , Mosqueda Garcia R Role of cortisol in the pathogenesis of deficient counterregulation after antecedent hypoglycemia in normal humans.

J Clin Invest 98 : — Peters A , Rohloff F , Kerner W Preserved counterregulatory hormone release and symptoms after short term hypoglycemic episodes in normal men. J Clin Endocrinol Metab 80 : — Le Roith D Seminars in medicine of the Beth Israel Deaconess Medical Center. Insulin-like growth factors.

Boyle PJ , Cryer PE Growth hormone, cortisol, or both are involved in defense against, but are not critical to recovery from, hypoglycemia. Am J Physiol : E — E Havel PJ , Taborsky GJ The contribution of the autonomic nervous system to changes of glucagon and insulin secretion during hypoglycemic stress.

Endocr Rev 10 : — Liu D , Moberg E , Kollind M , Lins PE , Adamson U A high concentration of circulating insulin suppresses the glucagon response to hypoglycemia in normal man.

J Clin Endocrinol Metab 23 : — Alengrin F , Grossi G , Canivet B , Dolais-Kitabgi J Inhibitory effects of metformin on insulin and glucagon action in rat hepatocytes involve post-receptor alterations. Diabet Metab 13 : — Yu B , Pugazhenthi S , Khandelwal RL Effects of metformin on glucose and glucagon regulated gluconeogenesis in cultured normal and diabetic hepatocytes.

Biochem Pharmacol 48 : — Stumvoll M , Nurjhan N , Perriello G , Dailey G , Gerich JE Metabolic effects of metformin in non-insulin-dependent diabetes mellitus. Korzon-Burakowska A , Hopkins D , Matyka K , et al. Meyer C , Grossmann R , Mitrakou A , et al.

Bottini P , Boschetti E , Pampanelli S , et al. Evidence for a nonselective defect. Fanelli C , Pampanelli S , Lalli C , et al. Effects on hypoglycemia awareness and couterregulation. Mokan M , Mitrakou A , Veneman T , et al. Diabetes Care 17 : — Muhlhauser I , Overmann H , Bender R , Bott U , Berger M Risk factors of severe hypoglycaemia in adult patients with Type I diabetes—a prospective population based study.

Diabetologia 41 : — Gonder-Frederick L , Cox D , Kovatchev B , Schlundt D , Clarke W A biopsychobehavioral model of risk of severe hypoglycemia. Diabetes Care 20 : — Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide.

Sign In or Create an Account. Endocrine Society Journals. Advanced Search. Search Menu. Article Navigation.

Close mobile search navigation Article Navigation. Volume Article Contents Subjects and Methods. Journal Article. Metformin Does Not Adversely Affect Hormonal and Symptomatic Responses to Recurrent Hypoglycemia.

Bernd Fruehwald-Schultes , Bernd Fruehwald-Schultes. Oxford Academic. Werner Kern. Kerstin M. Stefan Sopke. Barbara Toschek. Jan Born. Horst L. Achim Peters. PDF Split View Views. Cite Cite Bernd Fruehwald-Schultes, Werner Kern, Kerstin M. Select Format Select format.

ris Mendeley, Papers, Zotero. enw EndNote. bibtex BibTex. txt Medlars, RefWorks Download citation. Permissions Icon Permissions. Figure 1. Open in new tab Download slide. Figure 2. Table 1. P value. Drug × ante-hypo. Open in new tab. Figure 3. Adverse events and their association with treatment regimens in the diabetes control and complications trial.

Search ADS. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes UKPDS Clinical review hypoglycemia unawareness in noninsulin-dependent diabetes mellitus.

Google Scholar Crossref. In rare cases, it's possible to have a serious allergic reaction anaphylaxis to metformin. You or the person who's unwell may also have a rash that's swollen, raised, itchy, blistered or peeling. These can be signs of a serious allergic reaction and may need immediate treatment in hospital.

Taking metformin can cause vitamin B12 deficiency if you take it for a long time. This can make you feel very tired, breathless and faint, so your doctor may check the vitamin B12 level in your blood.

If your vitamin B12 levels become too low, vitamin B12 supplements will help. These are not all the side effects of metformin. For a full list, see the leaflet inside your medicine packet. Page last reviewed: 24 March Next review due: 24 March Home Medicines A to Z Metformin Back to Metformin.

Side effects of metformin. Common side effects These common side effects of metformin happen in more than 1 in people. There are things you can do to help cope with them: Feeling sick nausea Take metformin with food to reduce the chances of feeling sick.

If you're in a lot of pain, speak to your pharmacist or doctor. Try chewing sugar-free gum. Immediate action required: Call now if:. your lips, mouth, throat or tongue suddenly become swollen you're breathing very fast or struggling to breathe you may become very wheezy or feel like you're choking or gasping for air your throat feels tight or you're struggling to swallow your skin, tongue or lips turn blue, grey or pale if you have black or brown skin, this may be easier to see on the palms of your hands or soles of your feet you suddenly become very confused, drowsy or dizzy someone faints and cannot be woken up a child is limp, floppy or not responding like they normally do their head may fall to the side, backwards or forwards, or they may find it difficult to lift their head or focus on your face You or the person who's unwell may also have a rash that's swollen, raised, itchy, blistered or peeling.

Information: You can report any suspected side effect using the Yellow Card safety scheme. Visit Yellow Card for further information.