It is common practice to withhold metformin prior to cardiac catheterization due to fear of developing lactic acidosis in the setting of contrast-associated acute kidney injury (AKI). The evidence behind this recommendation is currently weak.


We collected 851 articles from PubMed and Embase, of which 3 met our inclusion criteria. Inclusion criteria were age > 18 years, baseline use of long-term metformin with continued or interrupted metformin use in patients during diagnostic or interventional cardiac catheterization. The outcomes studied were differences between post-catheterization and pre-catheterization serum creatinine (SCr) and glomerular filtration rate (GFR). We excluded studies dealing with patients not on long-standing metformin and those in which contrast exposure was through contrast enhanced computerized tomography. We used Hedges’s g with inverse variance method to pool standard mean difference with a random effects model using meta-cont module in CRAN-R software with 95% confidence interval (CI) for statistical significance. Higgins I-squared (I²) statistic was used to evaluate heterogeneity.


. Post-catheterization serum creatinine (Hedges’s g = -0.12 mg/dL; CI = -0.83 to +0.6, p = 0.75, I2 = 95%), post-catheterization GFR (Hedges’s g = +0.18 mL/min; CI = -0.76 to +1.11, p = 0.71, I2 = 97%) and post-catheterization lactate levels (Hedges’s g = +0.03 mg/dL; CI = -0.66 to +0.72, p = 0.75, I2 = 95%) failed to reach statistical significance.


There is no statistically significant difference in SCr or GFR between patients who held metformin prior to cardiac catheterization and those who continued taking the medication.

Figure 1.jpg (93 kB)
Figure 1

Figure 2.jpg (380 kB)
Figure 2

Figure 3.jpg (544 kB)
Visual Abstract

Supplemental files.docx (2855 kB)
Supplemental File

Manuscript Full File.docx (46 kB)
Manuscript with Edits