Metformin Side Effects: Potentially Permanent Side Effects Explained

Which metformin side effects can be permanent?

Most of what people notice on metformin is temporary. 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, and nausea or vomiting in 26% versus 8% [2]. The label starts immediate-release metformin at 500 mg twice daily or 850 mg once daily with meals and increases the dose gradually [2]. Diarrhea led to discontinuation in 6% of metformin patients in that trial [2].

Two harms can outlast the treatment. The first is nerve damage from a vitamin B12 deficiency that is not diagnosed. The second is lactic acidosis, which the label carries as a boxed warning because postmarketing cases have resulted in death, hypothermia, hypotension and resistant bradyarrhythmias [2].

Vitamin B12: the realistic route to lasting nerve damage

How often does metformin lower B12?

The label states that in 29-week clinical trials, previously normal serum B12 fell to subnormal levels in about 7% of patients, that the drop may be associated with anemia, and that it appears to be rapidly reversible with discontinuation of metformin or with B12 supplementation [2]. It directs prescribers to measure hematologic parameters annually and vitamin B12 every 2 to 3 years, and to manage any abnormalities [2].

Long-term metformin use is associated with B12 deficiency. In the Diabetes Prevention Program Outcomes Study, low B12 (203 pg/mL or less) was found in 4.3% of metformin users versus 2.3% of placebo participants at 5 years (P=0.02), and in 7.4% versus 5.4% at 13 years, a difference that was no longer statistically significant [5]. Combined low and borderline-low B12 (298 pg/mL or less) was more common with metformin at both points: 19.1% versus 9.5% at 5 years and 20.3% versus 15.6% at 13 years [5]. Each additional year of metformin use was associated with 13% higher odds of B12 deficiency in this analysis (OR 1.13, 95% CI 1.06 to 1.20) [5].

In a Dutch randomized placebo-controlled trial of 390 patients with type 2 diabetes on insulin, metformin 850 mg three times daily for 4.3 years lowered B12 by a mean of 19% (95% CI -24% to -14%) and raised the absolute risk of deficiency by 7.2 percentage points, a number needed to harm of 13.8 over 4.3 years [6]. A pooled analysis of 29 studies covering 8,089 patients found deficiency more common with metformin (OR 2.45, 95% CI 1.74 to 3.44) and average B12 roughly 65.8 pmol/L lower [7].

Why B12 falls

The label attributes the drop to possible interference with B12 absorption from the B12-intrinsic factor complex [2].

When a low level leaves damage behind

The label says the B12 decrease appears to be rapidly reversible with B12 supplementation or stopping metformin [2]. The lasting harm is neurological, and it can happen when the deficiency is missed. A 2010 case report described a patient on long-term metformin whose B12 deficiency presented as peripheral neuropathy; the author wrote that failing to diagnose the cause allows central or peripheral neuronal damage to progress, which can be arrested but not reversed with B12 replacement [3]. In the Diabetes Prevention Program Outcomes Study, neuropathy was more common among metformin users who had low B12 [5]. A single case report does not give a rate [3].

Lactic acidosis: rare, and concentrated in kidney impairment

The boxed warning describes metformin-associated lactic acidosis as subtle in onset, with malaise, muscle pain, breathing difficulty, sleepiness and abdominal pain. The laboratory picture is blood lactate above 5 mmol/L, anion gap acidosis and metformin plasma levels generally above 5 mcg/mL [2]. Risk factors named in the label are renal impairment, certain interacting drugs such as carbonic anhydrase inhibitors, 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 [2]. The label states that postmarketing cases primarily occurred in patients with significant renal impairment [2].

How rare is it? Pooled data from 347 comparative trials and cohort studies found no cases of fatal or nonfatal lactic acidosis in 70,490 patient-years of metformin use; using Poisson statistics, the upper limit for the true incidence was 4.3 cases per 100,000 patient-years for metformin and 5.4 for non-metformin treatment [8]. A systematic review put the incidence at roughly 3 to 10 per 100,000 person-years, generally indistinguishable from the background rate in people with diabetes [9].

The risk is not spread evenly across kidney function. In a community cohort of 75,413 patients with diabetes, metformin use was not associated with hospitalization for acidosis overall (adjusted HR 0.98, 95% CI 0.89 to 1.08), nor at eGFR 45 to 59 (HR 1.16, 95% CI 0.95 to 1.41) or 30 to 44 (HR 1.09, 95% CI 0.83 to 1.44). Below an eGFR of 30 the association appeared (adjusted HR 2.07, 95% CI 1.33 to 3.22) [4]. In a Taiwanese cohort of patients with stage 5 chronic kidney disease, metformin users matched to non-users had higher all-cause mortality (53% versus 41%; adjusted HR 1.35, 95% CI 1.20 to 1.51), while the difference in metabolic acidosis events was not statistically significant [10].

Metformin-associated lactic acidosis has caused deaths [2]. The label also says hemodialysis has often resulted in reversal of symptoms and recovery, and recommends prompt hemodialysis when lactic acidosis is diagnosed or strongly suspected [2].

Gastrointestinal side effects often lessen over time

In the extended-release trials, 781 patients on metformin extended-release were compared with 195 on placebo: diarrhea 10% versus 3%, nausea or vomiting 7% versus 2%, and diarrhea led to discontinuation in 0.6% [11]. In the Diabetes Prevention Program and its follow-up, gastrointestinal symptoms were more common with metformin than placebo and declined over time [1].

A retrospective chart review of 471 patients (310 on extended-release, 158 on immediate-release) found the frequency of any gastrointestinal adverse event similar between the two formulations, even though about 25% of the extended-release group had been switched because of prior gastrointestinal side effects [12].

Kidney and liver effects: what the label says

The label reports liver injury after metformin use but does not establish whether that injury is permanent [2]. It limits metformin use when kidney function is reduced because the drug can accumulate and increase lactic acidosis risk [2]. eGFR is checked before starting and at least yearly; metformin is contraindicated below 30 mL/min/1.73 m2, starting is not recommended at 30 to 45, and if eGFR later falls below 45 the benefit and risk of continuing are reassessed [2]. In patients with an eGFR of 30 to 60, a history of liver disease, alcoholism or heart failure, or intra-arterial contrast, the label says to stop metformin at the time of or before the procedure, recheck eGFR 48 hours later and restart only if renal function is stable [2].

The label lists cholestatic, hepatocellular and mixed hepatocellular liver injury among postmarketing reactions. No frequency is given for those liver reports because postmarketing reactions are reported voluntarily from a population of uncertain size [2]. Separately, because hepatic impairment has been linked to lactic acidosis cases, the label says to avoid metformin in patients with clinical or laboratory evidence of hepatic disease [2]. Separately, a meta-analysis of 9 randomized trials in 417 patients with non-alcoholic fatty liver disease found metformin lowered ALT by 8.12 U/L and AST by 4.52 U/L but did not improve steatosis, inflammation, ballooning or fibrosis [13].

Weight and anemia

During the double-blind Diabetes Prevention Program, body weight fell 2.06% with metformin versus 0.02% with placebo, and over 7 to 8 years of open-label follow-up weight loss remained greater with metformin (2.0% versus 0.2%) [1].

Metformin can lower B12, and the label notes that decrease may be associated with anemia [2]. In the Diabetes Prevention Program, hemoglobin and hematocrit were slightly lower with metformin, with the change occurring in the first year and no further change over time [1]. In the follow-up B12 analysis, anemia was more common with metformin but did not track B12 status [5].