Eight Years Out, a Positive Culture, and a Habit the Trials Don't Support
An established transplant recipient's routine surveillance culture comes back positive with no symptoms at all, and the reflex to treat a single, irreplaceable kidney runs directly against three separate randomized trials that found nothing to treat.
Marisol T., a 45-year-old dental hygienist, received a deceased-donor kidney transplant eight years ago after lupus nephritis put her on dialysis in her thirties, and has spent every one of those eight years treating the graft the way she treats her own patients' teeth: something worth protecting from the smallest problem before it becomes a large one, checked and rechecked on a schedule rather than waited out. Her lupus itself has been quiet since transplant, with no flares and no need to modify her baseline immunosuppression for disease activity. Her creatinine has held steady at 1.1 for four years running, she has had one biopsy-proven rejection episode early on that resolved fully with pulse steroids and left no lasting mark on her graft function, and she has never had a single symptomatic urinary tract infection since transplant — a record she attributes, not unreasonably, to how quickly she has always treated anything her quarterly labs turned up, positive culture or otherwise.
This quarter's routine urine culture, drawn at a visit where she reports feeling entirely normal — no dysuria, no frequency, no flank tenderness, no fever, no change in her usual voiding pattern — grew E. coli at 100,000 CFU/mL. She is more than two months past transplant by a factor of nearly fifty, well outside the early post-operative window where a stent or indwelling catheter might independently explain a positive culture on its own, and she has no known anatomic abnormality, no reflux on prior imaging, no stone disease, nothing to distinguish her urinary tract from a transplant recipient with an entirely unremarkable course. On paper she is not an edge case at all: she sits inside the actual enrollment window of Origuen and colleagues' 2016 trial — more than two months post-transplant, with no indwelling ureteral stent or catheter at the time in question — the same window three separate randomized trials have already tested this precise question in, at three different centers, years apart, and all three found the same thing.
Quarterly visit, a positive culture and no symptoms
She has one kidney and it isn't hers by birth. A five-day course of nitrofurantoin costs her almost nothing, and if there's even a modest chance it prevents a pyelonephritis in a graft with zero reserve to absorb one, that trade looks obvious to me.
I recognize the trial data exists. My hesitation isn't that I doubt the numbers — it's that a graft failure from an untreated infection is a much larger, much less reversible event than anything a short antibiotic course could plausibly cost her.
Origuen's 2016 trial, Sabe's 2019 trial, and Rao and colleagues' 2023 pooled meta-analysis of five RCTs and 566 patients all found the same thing: no reduction in symptomatic UTI or pyelonephritis from treating asymptomatic bacteriuria beyond the early post-transplant window, and no difference in graft loss or graft function between the treated and untreated groups in any of them.
The 'zero reserve, trivial cost' framing treats the antibiotic as free. It isn't — the same meta-analysis found a relative risk of 1.51 toward drug-resistant organisms in the treated arm. It didn't reach statistical significance, but the direction is consistent with what you'd expect from repeated, unnecessary antibiotic exposure in someone who will need effective antibiotics again.
This isn't a case of extending a general finding to an individual and hoping it holds. Origuen's trial specifically enrolled kidney transplant recipients more than two months out, excluding only those with an indwelling ureteral stent or catheter at randomization — that is her, on every actual enrollment criterion, not an approximation of her.
When a patient sits this precisely inside a trial's own population, the null result isn't a population-level statistic being applied to an individual case — it's the individual case the trial was built to answer.
Agreed: no antibiotic today. The culture is documented and will not be repeated reflexively; she'll be asked to return promptly if any symptom — fever, flank pain, dysuria — actually develops, at which point the calculus changes entirely and treatment starts without hesitation.
The primary care physician's underlying worry — that a single-kidney patient can't afford to be treated the same as someone with two — wasn't argued away so much as answered with a more specific fact: the trials that produced this null result already included patients exactly like her. Everyone left the visit agreeing on the plan; not everyone left equally at ease with how little room that leaves for erring toward caution the next time the culture comes back positive.