Using Potassium to Prevent AF After CABG? Don’t Do That, TIGHT-K Indicates

LONDON, England—The common practice of administering potassium to keep levels up during CABG with the aim of preventing postoperative atrial fibrillation (AF) doesn’t seem to be doing much good after all, the randomized TIGHT K trial shows.

A more relaxed approach of administering potassium only when levels fell below the normal range was noninferior to keeping levels in the upper end of normal during surgery, Benjamin O’Brien, MD, PhD (Charité Universitätsmedizin Berlin, Germany), reported here at the European Society of Cardiology Congress 2024.

Moreover, the effort to maintain higher potassium levels came at an extra cost that averaged $111.89 per patient. Safety was similar with either approach.

“There’s a widespread practice of routinely and proactively, or even aggressively, supplementing potassium to achieve a high-normal level after heart bypass surgery, and we believe that [based] on the findings of our trial, that can now be abandoned,” O’Brien said at a press conference. “Abandoning this will decrease patient risk from an unnecessary intervention and reduce healthcare costs.”

The findings were published simultaneously online in JAMA.

Changing an Engrained Practice

AF is the most common adverse event after cardiac surgery—seen in about 30% of patients after CABG, for instance—and is associated with increases in early and late morbidity and mortality, longer hospital stays, and higher costs, O’Brien noted.

There are many interventions used to prevent postoperative AF, although there is limited evidence to support any of them, resulting in much variability in practice.

One widely used approach is to aggressively administer potassium during surgery to keep serum levels in the high-normal range, with a prior worldwide practice survey from O’Brien and colleagues showing that use of the strategy varied across regions. For example, about two-thirds of European centers said they had protocols for tight control in place, whereas only about one-third of US centers did.

O’Brien and his colleagues put the practice to the test in TIGHT K, a noninferiority trial conducted at 23 centers in the United Kingdom and Germany.

Investigators enrolled randomized 1,667 patients (mean age 65 years; 15% women) who had no history of AF, atrial flutter, or atrial tachyarrhythmia and were scheduled to undergo isolated CABG to relaxed or tight control of potassium during the operation. In the relaxed arm, potassium was administered only when the serum concentration fell below 3.6 mEq/L into the low range. In the tight arm, potassium was given whenever the serum concentration fell below 4.5 mEq/L to keep it in the upper end of the normal range.

The median number of potassium administrations was zero in the relaxed arm and seven in the tight arm, leading to a significant separation in mean serum concentration between groups.

The primary endpoint of clinically detected and electrographically confirmed new-onset AF in the first 120 hours after CABG or until hospital discharge occurred at a rate of 27.8% with relaxed potassium control and 26.2% with tight control. The adjusted between group difference was 1.7% (95% CI -2.6%-5.9%), establishing noninferiority of the relaxed versus tight approach since the upper bound of the CI was below 10%. Results were similar in both the intention-to-treat and per-protocol analyses.

Patients wore ambulatory heart rhythm monitors (AHRMs) as well, and there were no differences between groups in secondary endpoints looking at AF episodes detected by AHRMs alone or episodes detected either with the monitors or clinically. Other dysrhythmias, like ventricular fibrillation or tachycardia, occurred at similar rates in the two groups.

In-hospital mortality (four deaths in each group) and median length of stay in critical care (2 days) or the hospital (6 days) did not differ between the relaxed and tight control groups.

The only significant difference came in the cost of purchasing and administering potassium, which was four times higher with tight control.

Commenting for TCTMD, cardiothoracic surgeon Torsten Doenst, MD, PhD (Universitätsklinikum Jena, Germany), said this topic is “not groundbreaking but important.”

Allowing patients to be managed according to the relaxed approach “clears out an old paradigm that maintenance of high potassium levels is important,” Doenst said via email, highlighting the added benefit of reducing costs. “I would like to congratulate the authors because they questioned this paradigm and tested a pathophysiology-driven hypothesis. We need more of that in cardiac surgery.”

Likewise, Faiez Zannad, MD, PhD (Université de Lorraine, Nancy, France), said during a discussion following O’Brien’s presentation that the investigators “must be complimented in shaking up this well-established practice with hard evidence. . . . You have reached a very important conclusion which hopefully will change practice.”