Why the parent drug reassures you and the metabolite does the damage
A 47-year-old man on maintenance hemodialysis for diabetic nephropathy has been trying to quit smoking. Three weeks ago his primary care physician started bupropion sustained-release 150 mg daily. He is also on carvedilol 12.5 mg twice daily, insulin, and sevelamer.
He is brought in after a witnessed generalized tonic-clonic seizure at home. He has no seizure history, no alcohol use, and no head trauma. Glucose was 118 at the scene.
Sodium 138 ยท calcium and magnesium normal ยท non-contrast head CT unremarkable. He had dialysed the previous day with a normal run. A send-out bupropion level returns within the usual therapeutic range. BP is 168/96, up from his usual 140s.
When a drug has active metabolites cleared by the kidney, measuring the parent compound tells you very little in renal failure. Bupropion is the clean example, but the principle generalizes โ think of it whenever an assay reassures you and the patient does not.
Honest limitation: the dialysis pharmacokinetic study had only 8 patients and used a single dose, and the authors explicitly called for a multi-dose study. It is thin evidence. It is also the only direct evidence there is, and it points one direction โ which is exactly the situation in which you should say so out loud rather than quote the number as if it were settled.
Andrew Bland, MD, FACP, FAAP
Medical Associates Department of Nephrology ยท University of Illinois College of Medicine at Peoria ยท University of Dubuque PA & DPT Programs ยท Butler College of Osteopathic Medicine
Interactive teaching case ยท Psychotropic & ADHD Medications and Blood Pressure
ยฉ 2026 ยท urinenephrology.org