It’s “time to get rid of this age-old practice,” said Sripal Bangalore, who noted that patients are happier when not fasting.
Though it remains a common practice, requiring patients to have nothing by mouth (NPO) for several hours before percutaneous cardiovascular procedures doesn’t appear to have any advantages over allowing a no-fasting approach, according to a meta-analysis of randomized trials.
Satisfaction scores were much higher when patients were allowed to eat, in fact, with no increase in intraprocedural adverse events, lead author M. Haisum Maqsood, MD (Houston Methodist Hospital, TX), and colleagues report in aresearch letterpublished in the March 10, 2025, issue of JACC: Cardiovascular Interventions.
The findings “suggest that no fasting is a reasonable option and should be adopted in the majority of patients undergoing percutaneous cardiovascular procedures,” they write, adding, however, that the approach “must be used with caution in patients with gastroenteritis, gastroparesis, and other diseases that increase the risk for nausea, vomiting, or aspiration.”
A2021 scientific statementfrom the American Heart Association, led by Sripal Bangalore, MD (NYU Langone Health, New York, NY), highlighted the weak evidence supporting NPO protocols before procedures that require conscious sedation and pointed to emerging trial evidence—from theCHOW NOWstudy, for instance—questioning the practice. Since then, additional trials, includingCALORI, have further shown that routine fasting may not be needed.
Bangalore, senior author of the new meta-analysis, identified eight RCTs—six that enrolled patients undergoing cardiac catheterization (angiography or PCI) and two that included patients undergoing TAVI, ablation for arrhythmia, or placement of a cardiac implantable electronic device. Overall, there were 3,131 patients (mean age 68 years; 68% men) randomized to fasting or no fasting before the procedures. The median duration of fasting was 3.11 hours in the no-fasting group and 11.9 hours in the fasting group.
Allowing patients to eat before the procedure did not significantly influence the likelihood of intraprocedural adverse safety events (OR 0.77; 95% CI 0.59-1.01), aspiration pneumonia (OR 1.38; 95% CI 0.36-5.24), nausea or vomiting (OR 1.16; 95% CI 0.67-2.02), hypoglycemia (OR 0.78; 95% CI 0.45-1.35), or contrast-induced nephropathy (OR 1.91; 95% CI 0.93-3.92). In general, the findings were similar across procedure types.
Patients were more satisfied, however, when they were not required to fast before the procedure, with a mean difference in satisfaction score of 1.11 points (on a scale of 1 to 5).
It was a source of frustration, so we now feel that patients and physicians alike are much happier. Sripal Bangalore
It’s “time to get rid of the age-old practice” of fasting prior to cath lab procedures, Bangalore said. “There is no disadvantage to letting patients eat. There is no risk of adverse events. But at the same time, patient satisfaction is much better.”
After removing the NPO rule at his center 4 to 5 months ago, satisfaction improved not only among the patients, but also among the members of the care team, said Bangalore. “It was a source of frustration, so we now feel that patients and physicians alike are much happier.”
There is momentum toward getting rid of NPO rules at other centers, too, but it will take some time and effort to raise awareness of the data showing that the practice isn’t necessary on a routine basis, he said, cautioning that this does not apply to patients who require general anesthesia. “But for most coronary procedures, you don’t need it,” Bangalore added.
The authors acknowledge that the analysis is limited by the inclusion of small-scale RCTs conducted in specific geographic areas, raising questions about generalizability, and by heterogeneity in satisfaction scores.