Current guidelines for correcting serum potassium levels in patients who present with acute MI are out of date and aim for targets that are too high, raising the risk of in-hospital mortality, according to a report in the Jan. 11 issue of JAMA.
In a retrospective study of 38,689 cases of MI in a nationally representative database, there was a U-shaped relationship between serum potassium level at admission an in-hospital mortality. The lowest mortality occurred in patients with potassium levels of 3.5-4.5 mEq/L, with higher mortality in those with potassium levels of 4.5 mEq/L or higher as well as those with levels less than 3.5 mEq/L.
Current practice guidelines recommend raising "low" potassium to 4.0-5.0 mEq/L, "and some experts even advise a higher range of 4.5-5.5 mEq/L," said Dr. Abhinav Goyal of the schools of medicine and public health at Emory University, Atlanta, and his associates.
"These guidelines are based on small, older studies that focused only on [preventing] ventricular arrhythmias (and not mortality) and were conducted before the routine use of beta-blockers, reperfusion therapy, and early invasive management in acute MI patients. Our data suggest that ... potassium levels of greater than 4.5 mEq/L are associated with increased mortality and probably should be avoided," they noted.
Dr. Goyal and his colleagues performed their study because measuring and repleting potassium to prevent arrhythmia in patients who present with acute MI is an entrenched practice, despite "the lack of current, adequately powered studies that define the optimal range of serum potassium levels with respect to mortality and other important clinical outcomes." They used information from the Cerner Corporation’s Health Facts database, which covers nearly 400,000 consecutive MI patients who presented to 67 U.S. hospitals representing all geographic regions of the country in 2000-2008.
A total of 2,679 of these study subjects (6.9%) died during hospitalization.
Compared with the reference group (potassium level of 3.5-4.0 mEq/L), in which in-hospital mortality was 4.8%, patients with levels of 4.0-4.5 had comparable mortality (5.0%).
In contrast, mortality was twice as great (10.0%) for patients with potassium levels of 4.5-5.0 mEq/L, and was even greater at higher levels, the investigators said (JAMA 2012;307:157-64).
"Our findings suggest that overly aggressive repletion of potassium levels (which is often automated through the implementation of hospital order sets) may not be advisable in patients with acute MI ... as potassium levels of at least 4.5 mEq/L are associated with harm," they noted.
As expected, mortality also was greater for patients with low potassium levels below 3.5 mEq/L.
These associations persisted in further analyses that adjusted for potentially confounding factors such as patient age, sex, and glomerular filtration rate at admission. They also remained robust regardless of whether patients did or did not receive potassium supplementation during hospitalization, and remained robust when the analysis was restricted only to patients who survived past the first 24 hours.
The investigators cautioned that their findings apply only to patients with acute MI and should not be extrapolated to those with other cardiac conditions, including heart failure.
This study was supported in part by Saint Luke’s Mid America Heart Institute, Kansas City, Mo. No conflicts of interest were reported.