Trimethoprim-induced hyperkalemia in burn patients treated with intravenous or oral trimethoprim sulfamethoxazole for methicillin-resistant Staphylococcus aureus and other infections: nature or nurture?

Trimethoprim is well known to cause rashes; however, what is not commonly known is that it causes sudden and profound hyperkalemia in 10 to 20% of treated patients. The uniqueness of burn patients begs the question whether changes known to occur in these patients might also increase this trimethoprim effect. After institutional review board approval, a retrospective study evaluated 224 patients with thermal injury who had been treated with trimethoprim sulfamethoxazole (TMP-SMX), 24 of whom had underlying renal impairment (creatinine clearances <50 ml/min) and were excluded, leaving 200 patients for analysis. Three definitions of drug-induced hyperkalemia were used: 1) a ≥ 1 mEq/L rise, 2) a >0.8 mEq rise in potassium in <24 hours warranting early discontinuation of TMP-SMX, and 3) "marked" hyperkalemia defined as serum potassium of ≥ 5.5 mEq/L within 48 hours. A potassium level before trimethoprim exposure (TxK) and after TxK were collected retrospectively. Demographic data were analyzed with Student's t-test and trimethoprim dose alone, demonstrating a significant difference. Analysis of 200 patients exposed to trimethoprim demonstrated an elevation of potassium (first definition) in 31 patients (15.5%), a rapid change in serum potassium in two patients (second definition), and marked hyperkalemia (>5.5 mEq/L) in 13 patients (6.5%). Hyperkalemia never occurred in 166 of 200 patients (82%; before TxK, 3.9 ± 0.4; after TxK, 4.3 ± 0.5 mEq/L). Change in serum potassium among patients with hyperkalemia was 4.0 ± 0.5 mEq/L before TxK and 5.3 ± 0.7 mEq/L after TxK. Twelve published hyperkalemia risk factors were reviewed in these 200 patients and only history of hypertension and need for intubation was more common in those with hyperkalemia. A nearly 20% incidence of hyperkalemia and 6% serious hyperkalemia in burn patients is consistent with reports in patients without burn injury. These data also suggest that the metabolic and hormonal changes associated with burn injury do not increase further the genetically predisposed hyperkalemia resulting from exposure to trimethoprim. These data suggest patients treated with TMP-SMX should have routine serum potassium monitoring before discharge.