Intravenous Contrast in Patients With Diabetes on Metformin

New Common Sense Guidelines

Arti D. Shah, MD; Cody McHargue, BA; Judy Yee, MD; Robert J. Rushakoff, MD


Endocr Pract. 2016;22(4):502-505. 

In This Article


Metformin remains the cornerstone of treatment for type 2 diabetes mellitus (DM). Historically, as intravenous contrast has been thought to predispose patients with DM, especially those with underlying renal impairment, to acute renal failure and thus increasing the risk of lactic acidosis in those on metformin therapy,[1] the Food and Drug Administration (FDA) mandates the cessation of metformin therapy for 48 hours after the administration of contrast and that renal function be re-evaluated prior to resuming the drug.[2] However, this recommendation is supported by minimal data,[3] and as we discuss below, rarely followed. Additionally, the incidence of contrast-induced nephropathy in patients with DM on metformin is low (0% in patients without prior renal failure and up to 4.7% in patients with renal failure).[4–6]

In their 2013 guidelines, the American College of Radiology (ACR) continued to recommend that metformin be held for 48 hours postprocedure, but in most instances, a repeat creatinine (Cr) was not required. These specific recommendations were: (1) those with normal renal function and no known comorbidities do not need Cr checked after receiving contrast; (2) in those with multiple comorbidities but with normal renal function, metformin should be discontinued at the time the patient receives contrast, held for 48 hours, and renal function should be reassessed prior to restarting metformin; and (3) in those with renal dysfunction, metformin should be stopped at the time they receive contrast and renal function should be monitored until it is safe to resume metformin.[7]

In 2015, these guidelines were updated to remove restrictions on pre- and postcontrast metformin use ( Table 1 ).[8] How did we get to this point and do these new guidelines make sense?

Historically, given metformin's renal excretion, caution was advised when prescribing it to those with or at high risk for renal insufficiency. The FDA package insert states that metformin is contraindicated in males with a Cr ≥1.5 mg/dL, in females with a Cr ≥1.4 mg/dL and anyone with abnormal creatinine clearance.[2] While metformin accumulation in the setting of renal insufficiency may cause life-threatening lactic acidosis, systematic clinical reviews have found that metformin per se is not associated with an increased risk of lactic acidosis or increased lactate levels, even in those with kidney disease (estimated glomerular filtration rate [eGFR] 30–60 mL/min/1.73 m2).[9,10] Metformin levels remain within the therapeutic range in these patients.[10] In addition, when lactic acidosis is found in patients taking metformin, the drug is not necessarily responsible. These patients may have an underlying medical condition such as sepsis that is causing the lactic acidosis.[10] Furthermore, those on metformin with an eGFR of 30 to 60 mL/min/1.73 m2 have decreased mortality compared to those who are not on metformin.[11] Therefore, there has been a move towards lowering the eGFR cutoff for metformin use from 60 to 30 mL/min/1.73 m2 (10,12,13). In 2012, the American Diabetes Association and the European Association for the Study of Diabetes released a position statement endorsing the United Kingdom's National Institute for Health and Care Excellence guidelines that recommend using an eGFR cutoff of 30 mL/min/1.73 m2 below which metformin is contraindicated and a dose reduction at 45 mL/min/1.73 m2 (14).

Given concerns about metformin-induced lactic acidosis in the setting of contrast-induced nephropathy, the FDA and the 2013 guidelines from the ACR[7] stated that metformin should be held and renal function re-evaluated prior to resuming metformin 48 hours following the procedure, even in those with baseline normal renal function. Anecdotally, our colleagues and we have observed that patients rarely have their Cr repeated before restarting metformin. Prior to the new 2015 guidelines, we surveyed radiologists and internists throughout the U.S. attending a series of continuing medical education courses. Of 30 radiologists, only 4 (13%) almost always or always ordered a Cr to be drawn after administering intravenous (IV) contrast. When asked about the postcontrast instructions that they give, only 6 (20%) stated that they tell patients to resume metformin only after having Cr re-checked. Of 417 internists, 23% tell patients to restart metformin immediately, 24% said metformin should be restarted in 2 days, 47% said Cr should be checked in 2 days and metformin restarted if Cr is normal, and 5% said the radiologist takes care of this.

This confusion is understandable. A systematic review by Goergen et al found that there was a lack of agreement among 5 different previous guidelines (published between 1999 and 2008) regarding the need to stop metformin after administration of IV contrast, the risk of lactic acidosis in patients with normal renal function before IV contrast, the recommended method of measuring renal function, and the values used to define abnormal renal function.[3] The evidence that was the basis for these guidelines consisted of observational studies including case reports and series.[3] Furthermore, a literature review by McCartney et al showed that 17 of the identified 18 cases of IV contrast-related metformin-induced lactic acidosis were in patients with baseline renal dysfunction, and the authors concluded the risk of metformin-induced lactic acidosis after contrast to be negligible.[4] Many cases of reported metformin-associated lactic acidosis are not due to metformin accumulation but rather are due to associated hypoxic conditions such as heart failure and end-stage liver failure.[15]

Parra et al studied 97 veterans who underwent a total of 111 radiologic procedures over 27 months.[6] This patient population had a high prevalence of comorbid disease, and 30% had an eGFR <60 mL/min/1.73 m2. There was no significant difference in pre- and postcontrast Cr. However, 4 patients developed contrast-induced nephropathy defined as at least a 25% increase in serum Cr, but none developed lactic acidosis. No defining characteristics were noted in these four patients. They had an estimated Cr clearance of 60 to 70 mL/min/1.73 m2, and their renal function returned to baseline within 2 weeks.[6] More recently, Gómez-Herrero et al studied 98 diabetic patients on metformin who were undergoing contrast-enhanced computed tomography (CT) scans and were enrolled in a lactic acidosis prevention protocol defined as stopping metformin 2 days prior to contrast if renal function was abnormal prior to the scan (Cr >1.2 mg/dL) and holding metformin in all patients for 2 days after the scan until renal function was re-evaluated.[5] No patients with an eGFR >60 mL/min/1.73 m2 developed contrast-induced nephropathy. In patients with an eGFR <60 mL/min/1.73 m2, 1 out of 21 (4.7%) developed contrast-induced nephropathy. They concluded that the risk of developing contrast-induced nephropathy in patients with baseline normal renal function is minimal and that recommendations to stop metformin in patients receiving contrast may be restricted to patients with abnormal renal function.

Prior to the new guideline publication, we began looking at our patients taking metformin who were undergoing contrast procedures both prospectively (at the University of California, San Francisco) and retrospectively (at the San Francisco Veterans Affairs Medical Center). Protocol details, patient demographics, and characteristics are outlined in an online supplement ( In our retrospective cohort of 130 veterans taking metformin for DM management, IV contrast (Omnipaque or Visipaque) and precontrast renal function did not predict a decline in postcontrast renal function. Similarly to the above previous studies, we found that patients with precontrast eGFR >60 mL/min/1.73 m2 did not have any deterioration in renal function. Furthermore, there was no significant deterioration in renal function in patients with a precontrast eGFR <60 mL/min/1.73 m2. However, 4 of 63 patients with precontrast eGFR <60 developed acute kidney injury (AKI) that resolved with conservative treatment. All 4 of these patients were on an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker and had a diagnosis of hypertension. Whether it is these specific characteristics that increased the risk of AKI in these patients will need further investigation. Interestingly, even accounting for the change in Cr in these patients, the eGFR levels would not have been an absolute contraindication to continued metformin use.

In our recent prospective look at 40 patients with DM who were being treated with metformin, no significant change in serum creatinine was observed (baseline range from 0.7 to 1.2 mg/dL). Notably, an additional 15 patients were recruited, but they did not obtain a postcontrast creatinine measurement despite receiving reminders.

From our experience and as indicated from our survey of radiologists and internists, a postcontrast Cr check is rarely ordered prior to restarting metformin. The most common practice is for the patient to be told to hold their metformin for 2 days after contrast administration and then restart the medication. If the Cr level is not routinely being checked and changes in Cr are rare and clinically insignificant, why discontinue metformin at all? As we and others have shown, there are no significant postcontrast increases in Cr in patients on metformin. In the rare patient who has reduced renal function to begin with, it is possible to see a temporary Cr increase, but those levels were not a contraindication to continuing metformin.