Endocrine Abstracts

Introduction: Hypoaldosteronism is characterized by the development of hyperkalemia, but can also induce hypovolemic hyponatremia. Trimethroprim can cause hypoaldosteronism through mineralocorticoid resistance. That hypoaldosteronism can induce hyponatremia in absence of Addison’s disease has been questioned. We studied the electrolyte disturbances found following initiation of trimetroprim therapy.

Material and methods: Retrospective, analytical. Laboratory tests of 100 consecutive patients receiving trimethroprim in 2017 in a tertiary hospital, after excluding patients with baseline (B) hyperkalemia or hyponatremia. Electrolytes mmol/l. Hyponatremia: serum sodium (SNa) <135; hyperkalemia: serum potassium (SK) >5. Baseline electrolytes were compared to maximum SK and nadir SNa after starting trimethroprim. Results in mean (S.D.), or median (interquartile range).

Results: Age: 71.55 (13.55), 51% males. Bcreatinine: 0.77 mg/dl (0.56–1.13). BSK: 4.14 (0.58). SK rose 0.65 (0.74) (P< 0.0001). SK increment was >0.2 in 76%: 0.92 (0.6) (P<0.0001). 35% patients developed hyperkalemia. 10%: SK>5.5. BSK was higher in patients developing hyperkalemia: 4.32 (0.58), than in those maintaining normokalemia: 4.04 (0.6) (P=0.018). BSNa:138 (2.9) descended to 135.6 (4.5) (P<0.0001). 46% developed hyponatremia: SNa 132.3 (3.8), 6/46 without a SK rise. BSNa in patients presenting hyponatremia was lower: 137.8 (2.8) than in those not: 139.6 (2.76) (P=0.002). 25/35 patients presenting hyperkalemia developed hyponatremia. In these, creatinine rose from 0.82 mg/dl (0.6–1.28) to 0.99 (0.59–1.67) (P=0.038). There was a weak correlation between SK elevation and SNa descent: r=−0,26 (P=0.034). Days of treatment (DOT): 7.5 (5–11.5), with weak positive correlation between DOT and SK increment (r=0.31) (P=0.01), negative correlation for SNa (r=−0.31) (P=0.01). Weak positive correlation between trimethroprim-cumulative dose and SK changes (r=0.23) (P=0.04), negative for SNa (r=−0.27) (P=0.01). Blood gas bicarbonate descended <23 mEq/l in 5/28. No patient had cortisolemia determined. There was no significant difference in SK or SNa changes in 64/100 patients receiving pharmacological doses of prednisone/methyl-prednisolone versus those not. K rise and-or Na descent were not significantly influenced by heparin, ACE inhibitors, ARBs, NSAIDS, beta-blockers, or furosemide therapy, gender, age, the presence of Diabetes Mellitus, urinary tract infection or obstructive uropathy.

Conclusions: Trimethroprim-induced hypoaldosteronism is frequent, potentially causing hyperkalmia and, more often in our series, hyponatremia. In patients presenting both, creatinine levels rose, suggesting hypovolemia onset. Trimethroprim, inducing mineralocorticoid resistance, could also be unmasking subclinical Addison’s Disease. However, the fact that steroid medication in 64% of patients affected neither K rise nor Na descent suggests that these electrolye disturbances should not be attributed to Addison’s Disease.