Co-trimoxazole induced hyperkalemia and potassium monitoring in hospitalized patients.

Background Co-trimoxazole is an antibiotic combination used for the treatment of Pneumocystis jirovecii pneumonia, amongst others. Co-trimoxazole is known to increase serum potassium. For this reason, Dutch guidelines advise serum potassium monitoring in high-risk patients. Objective This study aimed to determine average serum potassium rise after administration of intravenous co-trimoxazole in hospitalized patients, compared to intravenous ceftriaxone. This study also aimed to determine adherence to Dutch guidelines by measuring the incidence of serum potassium monitoring in these patients. Setting Five departments of the Canisius Wilhelmina Hospital, a teaching hospital in Nijmegen, the Netherlands. Method Data was collected and compared from patients that received intravenous co-trimoxazole (n = 66) and intravenous ceftriaxone (n = 132) in the period of November 2008-November 2017. For each patient using co-trimoxazole, two patients using ceftriaxone were included in a paired fashion. Baseline and follow-up potassium were collected, if available. Additionally, it was tested if serum potassium was measured around the initiation of antibiotic therapy. Main outcome measure Changes in serum potassium where obtainable in 30 patients using cotrimoxazole and 40 patients using ceftriaxone. When compared to ceftriaxone, administration of intravenous co-trimoxazole was associated with a significant mean increase in serum potassium (+ 0.55 mmol/l, 95% CI 0.29-0.80, p < 0.001). After correction for confounders (baseline potassium, estimated glomerular filtration rate 30 to < 60, the presence of haematological malignancies and the usage of corticosteroids), this effect shrunk noticeably, but remained significant (+ 0.28 mmol/l, 95% CI 0.03-0.53, p = 0.031). Results The incidence of hyperkalemia at follow-up was 20% in the cotrimoxazole group, compared to 5% in the ceftriaxone group. Despite this, serum potassium was often not measured in patients using intravenous cotrimoxazole, being 76% at baseline and 55% in the period of 48-120 h after antibiotic therapy initiation, compared to 87% and 34% in the ceftriaxone group respectively. Conclusion Adherence to Dutch guidelines was poor as serum potassium monitoring was often not performed. As intravenous co-trimoxazole usage is associated with a significant increase in mean serum potassium, monitoring is strongly recommended.