Metformin Delayed-Onset Side Effects: What Takes Weeks or Months to Appear

At a glance

  • Best-documented late effect / vitamin B12 decline, subnormal in about 7% over 29-week trials [1]
  • Label B12 testing / every 2 to 3 years, with blood counts annually [1]
  • Kidney monitoring / eGFR before starting and at least yearly [1]
  • eGFR stop point / below 30 mL/min/1.73 m2, contraindicated [1]
  • eGFR 30 to 45 / starting not recommended; reassess benefit and risk if it later falls below 45 [1]
  • Diarrhea, immediate-release trial / 53% vs 12% on placebo; 6% stopped because of it [1]
  • Diarrhea, extended-release trials / 10% vs 3% on placebo; 0.6% stopped [3]
  • Low TSH with treated hypothyroidism / association strongest 90 to 180 days after starting, observational [2]

What Actually Shows Up Late

Metformin's most common adverse reactions are diarrhea, nausea and vomiting, flatulence, asthenia, indigestion, abdominal discomfort and headache, and the label's trial tables give rates, not onset times [1]. The reactions people ask about as delayed are a different group: vitamin B12 status, which falls with continued use [1]; metformin-associated lactic acidosis, a boxed-warning risk the label ties to kidney function, age and other conditions [1]; and low TSH in people already treated for hypothyroidism, which appeared most often within 90 to 180 days of starting metformin in one large observational study [2]. Gut symptoms have their own pattern and are covered further down.

Vitamin B12: The One With a Label Warning

In 29-week trials, approximately 7% of patients had a decrease from normal to subnormal vitamin B12 levels, possibly through interference with B12 absorption from the B12-intrinsic factor complex [1]. The label says the decrease may be associated with anemia but appears to be rapidly reversible with discontinuation of metformin or vitamin B12 supplementation, and it directs that hematologic parameters be measured annually and vitamin B12 at 2 to 3 year intervals [1].

The long-term picture comes from the Diabetes Prevention Program and its follow-up study. Participants took metformin 850 mg twice daily or placebo for 3.2 years; the metformin group then received open-label treatment for another 9 years [4]. Low B12 (203 pg/mL or less) was more common with metformin at 5 years, 4.3% vs 2.3% (p=0.02), and no longer significantly different at 13 years, 7.4% vs 5.4% (p=0.12) [4]. Combined low and borderline-low B12 (298 pg/mL or less) was more common at both points, 19.1% vs 9.5% at 5 years and 20.3% vs 15.6% at 13 years [4]. The odds of deficiency rose 13% for each year of metformin use (OR 1.13, 95% CI 1.06 to 1.20), and the authors suggested routine B12 testing be considered [4]. Anemia was more common in metformin users but did not track B12 status, and neuropathy was more common among metformin users with low B12 [4].

B12 can fall early. A systematic review of 26 papers found significantly lower B12 in metformin users in 10 of 17 observational studies, and a meta-analysis of four trials found a mean reduction of 57 pmol/L after 6 weeks to 3 months of use [5]. At the tissue level, a post hoc analysis of the HOME trial followed 390 insulin-treated patients on metformin or placebo for 52 months and found metformin raised methylmalonic acid, a marker of tissue B12 deficiency, versus placebo (p=0.001). The rise in methylmalonic acid was associated with a worse neuropathy score, while metformin's HbA1c lowering moved the score the other way, leaving a non-significant net effect [6].

Lactic Acidosis: Timing and Risk Factors

The label does not give a set time after starting metformin when lactic acidosis occurs. It identifies renal impairment, certain medicines, contrast procedures, hypoxic states and other factors that increase risk [1].

The boxed warning describes postmarketing cases that resulted in death, hypothermia, hypotension and resistant bradyarrhythmias, with an onset that is often subtle: malaise, myalgias, respiratory distress, somnolence and abdominal pain. The laboratory picture was blood lactate above 5 mmol/L, anion gap acidosis, a raised lactate/pyruvate ratio and metformin plasma levels generally above 5 mcg/mL [1].

The label says the postmarketing cases primarily occurred in patients with significant renal impairment, and that the risk of metformin accumulation rises with the severity of renal impairment because metformin is substantially excreted by the kidney [1]. Named risk factors are renal impairment, certain drugs such as the carbonic anhydrase inhibitor topiramate, age 65 or older, radiologic contrast studies, surgery and other procedures, hypoxic states such as acute congestive heart failure, excessive alcohol intake, hepatic impairment and mitochondrial disease [1].

A Cochrane review pooling 347 trials and cohort studies found no cases of lactic acidosis in 70,490 patient-years of metformin use, with a calculated upper limit for the true incidence of 4.3 cases per 100,000 patient-years against 5.4 in 55,451 patient-years of non-metformin treatment [7]. The metformin label separately reports postmarketing cases, including fatal ones [1].

The label's safeguards are built around kidney function. It asks for an eGFR before starting and at least yearly, more often in people at risk of renal impairment; it contraindicates metformin below 30 mL/min/1.73 m2, does not recommend starting at 30 to 45, says to assess benefit and risk of continuing if eGFR later falls below 45, and says to discontinue below 30 [1]. For iodinated contrast, the label says to stop metformin at the time of or before the procedure in patients with eGFR between 30 and 60, a history of hepatic impairment, alcoholism or heart failure, or intra-arterial contrast, then recheck eGFR 48 hours later and restart only if renal function is stable [1]. Excessive alcohol is named because alcohol potentiates metformin's effect on lactate metabolism [1]. Carbonic anhydrase inhibitors lower serum bicarbonate and raise lactic acidosis risk, so the label advises more frequent monitoring [1]. Drugs that interfere with renal tubular transport — ranolazine, vandetanib, dolutegravir and cimetidine — can raise metformin exposure, and the label advises weighing benefits and risks [1]. When lactic acidosis is suspected, the label says to discontinue metformin and treat in hospital, and recommends prompt hemodialysis [1].

Gastrointestinal Effects: More Common With Metformin, Declining Over Time

In a US trial of immediate-release metformin up to 2,550 mg a day, diarrhea occurred in 53% of 141 patients versus 12% of 145 on placebo, nausea or vomiting in 26% vs 8%, flatulence in 12% vs 6%, asthenia in 9% vs 6%, indigestion in 7% vs 4%, abdominal discomfort in 6% vs 5% and headache in 6% vs 5% [1]. Diarrhea led to discontinuation in 6% [1]. In placebo-controlled trials of extended-release metformin, diarrhea occurred in 10% of 781 patients vs 3% on placebo and nausea or vomiting in 7% vs 2%, with 0.6% stopping because of diarrhea [3]. The label cautions that reaction rates from different trials cannot be directly compared [1].

The longest view comes from the Diabetes Prevention Program and its follow-up: gastrointestinal symptoms were more common with metformin than placebo and declined over time, and no significant safety issues were identified across the double-blind period and 7 to 8 years of open-label follow-up [8]. The label's titration schedule starts at 500 mg twice daily or 850 mg once daily with meals and increases by 500 mg weekly or 850 mg every 2 weeks [1].

Low TSH in People Already on Thyroid Treatment

Fournier and colleagues used a UK primary care database to compare metformin monotherapy with sulfonylurea monotherapy. The study included 5,689 patients with treated hypothyroidism and 59,937 with normal thyroid function [2]. Among those with treated hypothyroidism, metformin was associated with a 55% higher risk of a low TSH level below 0.4 mIU/L (adjusted HR 1.55, 95% CI 1.09 to 2.20), with the highest risk in the 90 to 180 days after starting (adjusted HR 2.30, 95% CI 1.00 to 5.29) [2]. No association appeared in euthyroid patients (adjusted HR 0.97, 95% CI 0.69 to 1.36), and the authors said the clinical consequences need further investigation [2]. This is an observed association in one population, not a demonstrated cause, and it was seen in people already treated for hypothyroidism.

Postmarketing Liver Injury

Cholestatic, hepatocellular and mixed hepatocellular liver injury have been reported in postmarketing use of metformin, and because these reports come voluntarily from a population of uncertain size, the label says frequency cannot be reliably estimated [1]. Hepatic impairment is also one of the named lactic acidosis risk factors, and the label says metformin is not recommended in patients with hepatic impairment [1].

How Timing Differs by Effect

The B12 testing interval of every 2 to 3 years is the closest thing the label has to a schedule for a delayed effect [1]. The other effects follow different patterns: gut symptoms were more common with metformin and declined over time, lactic acidosis depends on kidney function, liver function and circulation rather than on time alone, and low TSH was an association seen over months in one group of patients with treated hypothyroidism.

Frequently asked questions

Do metformin side effects get worse the longer you take it?
Not in a straight line. In the Diabetes Prevention Program, gastrointestinal symptoms were more common with metformin than placebo and declined over time [8]. Vitamin B12 moves the other way: the odds of deficiency rose 13% for each year of metformin use in the follow-up study [4].
How quickly can metformin lower vitamin B12?
The label reports subnormal B12 in about 7% of patients in trials of 29-week duration [1]. A meta-analysis pooling four trials found a mean fall of 57 pmol/L after 6 weeks to 3 months of metformin [5]. In the Diabetes Prevention Program follow-up, low B12 was 4.3% vs 2.3% at 5 years and 7.4% vs 5.4% at 13 years [4].
Can lactic acidosis appear months or years after starting metformin?
The label does not give a set time after starting metformin when lactic acidosis occurs. It ties the risk to kidney function and other factors: metformin accumulation and lactic acidosis become more likely as renal impairment worsens, and the postmarketing cases primarily occurred in patients with significant renal impairment [1]. The label asks for an eGFR before starting and at least yearly, and contraindicates metformin below 30 mL/min/1.73 m2 [1].
Does metformin change thyroid test results?
In a UK database study, metformin monotherapy was associated with a 55% higher risk of a low TSH level in patients with treated hypothyroidism (adjusted HR 1.55, 95% CI 1.09 to 2.20), with the highest risk 90 to 180 days after starting; no association was found in euthyroid patients [2].
Does metformin cause liver damage?
Cholestatic, hepatocellular and mixed hepatocellular liver injury have been reported after metformin reached the market, and the label says frequency cannot be reliably estimated from voluntary reports [1]. Hepatic impairment is listed as a lactic acidosis risk factor, and the label says metformin is not recommended in patients with hepatic impairment [1].

References

  1. U.S. National Library of Medicine, DailyMed. Metformin hydrochloride tablets (Granules, immediate-release) prescribing information. Metformin hydrochloride tablets (Granules, immediate-release) label
  2. Fournier JP, Yin H, Yu OH, Azoulay L. Metformin and low levels of thyroid-stimulating hormone in patients with type 2 diabetes mellitus. CMAJ. 2014;186(15):1138-45. Metformin and low levels of thyroid-stimulating hormone in patients with type 2 diabetes mellitus
  3. U.S. National Library of Medicine, DailyMed. Metformin hydrochloride extended-release tablets (Sun) prescribing information. Metformin hydrochloride extended-release tablets (Sun) label
  4. Aroda VR, Edelstein SL, Goldberg RB, Knowler WC, Marcovina SM, Orchard TJ, et al. Long-term Metformin Use and Vitamin B12 Deficiency in the Diabetes Prevention Program Outcomes Study. J Clin Endocrinol Metab. 2016;101(4):1754-61. Long-term Metformin Use and Vitamin B12 Deficiency in the Diabetes Prevention Program Outcomes Study
  5. Chapman LE, Darling AL, Brown JE. Association between metformin and vitamin B12 deficiency in patients with type 2 diabetes: A systematic review and meta-analysis. Diabetes Metab. 2016;42(5):316-327. Association between metformin and vitamin B12 deficiency in patients with type 2 diabetes: A systematic review and meta-analysis
  6. Out M, Kooy A, Lehert P, Schalkwijk CA, Stehouwer CDA. Long-term treatment with metformin in type 2 diabetes and methylmalonic acid: Post hoc analysis of a randomized controlled 4.3year trial. J Diabetes Complications. 2018;32(2):171-178. Long-term treatment with metformin in type 2 diabetes and methylmalonic acid: Post hoc analysis of a randomized controlled 4.3year trial
  7. Salpeter SR, Greyber E, Pasternak GA, Salpeter EE. Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus. Cochrane Database Syst Rev. 2010;2010(4):CD002967. Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus
  8. Diabetes Prevention Program Research Group. Long-term safety, tolerability, and weight loss associated with metformin in the Diabetes Prevention Program Outcomes Study. Diabetes Care. 2012;35(4):731-7. Long-term safety, tolerability, and weight loss associated with metformin in the Diabetes Prevention Program Outcomes Study
Evidence overview for Metformin Delayed-Onset Side Effects: What Takes Weeks or Months to Appear