Increasing antimicrobial resistance has renewed interest in older, less used antimicrobials. Cotrimoxazole shows promise; however, hyperkalaemia and acute kidney injury (AKI) are potential complications. Identifying risk factors for and quantification of these events is required for safe use. This study aimed to evaluate predictors of cotrimoxazole-associated AKI and hyperkalaemia in a clinical setting.
Methods
Patients prescribed cotrimoxazole were identified using electronic healthcare records over 3 years (1 April 2016 to 31 March 2019). Individual risk factors were recognized. Serum creatinine and potassium trends were analysed over the subsequent 21 days. AKI and patients with hyperkalaemia were classified using Kidney Disease Improving Global Outcomes (KDIGO) and laboratory criteria. Univariate and multiple logistic regression analyses were performed.
Results
Among 214 patients prescribed cotrimoxazole, 42 (19.6%, 95% confidence interval (CI) 14.6–25.7) met AKI criteria and 33 (15.4%, 95% CI 11.0–21.1) developed hyperkalaemia. Low baseline estimated glomerular filtration rate (<60 mL/min/1.73 m2, odds ratio (OR) 7.78, 95% CI 3.57–16.13, p < 0.0001) and cardiac disorders (OR 2.40, 95% CI 1.17–4.82, p 0.011) predicted AKI, while low baseline estimated glomerular filtration rate (<60 mL/min/1.73 m2, OR 6.80, 95% CI 3.09–15.06, p < 0.0001) and higher baseline serum potassium (p 0.001) predicted hyperkalaemia. Low-dose cotrimoxazole (<1920 mg/d) was associated with lower AKI and hyperkalaemia risk (p 0.007 and 0.019 respectively). Early (within the first 2–4 days of therapy) serum creatinine changes predicted AKI (OR 3.65, 95% CI 1.73–7.41, p 0.001), and early serum potassium changes predicted hyperkalaemia (>0.6 mmol/L, OR 2.47, 95% CI 1.14–5.27, p 0.0236).
Conclusions
Cotrimoxazole-associated AKI and hyperkalaemia is frequent and dose dependent. Renal function, serum potassium and preexisting cardiac disorders should be evaluated before prescribing cotrimoxazole. Serum creatinine and potassium monitoring within first 2 to 4 days of treatment to identify susceptible patients is recommended, and the lowest effective dose ought to be prescribed.
