Vesicoureteral Reflux Prophylaxis: A Low-Grade Infant and a High-Grade Child with Bladder-Bowel Dysfunction
Two children with vesicoureteral reflux, evaluated the same week, sit on opposite ends of the trial evidence: a three-month-old with grade II reflux found before any infection, and a four-year-old with grade IV reflux, a second febrile UTI, and documented bladder-bowel dysfunction.
Wren A., a three-month-old girl, was found to have left-sided hydronephrosis on her mother's twenty-week anatomy scan, and a postnatal renal ultrasound at two weeks confirmed mild pelviectasis without evidence of obstruction. She was born at thirty-nine weeks after an uncomplicated pregnancy, feeds and grows well, and has never had a fever or a urine culture drawn. A voiding cystourethrogram at six weeks, ordered to complete the antenatal-hydronephrosis workup rather than in response to any illness, showed grade II vesicoureteral reflux on the left and a normal right kidney. Her mother is due back at her nursing job in six weeks, which has made the question of daily antibiotics — who gives it, whether daycare staff will be asked to — practically as pressing as the pharmacology.
Grade II sits at the milder end of the five-grade reflux scale, and that number carries more weight here than it might in an older child: every major placebo-controlled trial that has actually shown a prophylaxis benefit — PREDICT, in 292 infants randomized to trimethoprim-based regimens or no treatment, found a first UTI in 21.2% of the prophylaxis group against 35.6% untreated (hazard ratio 0.55, 95% CI 0.35–0.86) — enrolled only grade III, IV, or V reflux, and required a first UTI, not a scan finding, as the outcome being prevented. Wren has neither the grade nor, since she has never been infected, anything yet to prevent a recurrence of. The number needed to treat in that trial, even restricted to the population it actually studied, was 7 over two years; nothing in the trial's own inclusion criteria says anything about a child like her at all.
Deciding before any infection has happened
I'd start trimethoprim-sulfamethoxazole now, once daily, and reassess with a repeat VCUG around her first birthday. Grade II reflux resolves spontaneously in a majority of infants by then, but “majority” still leaves a real minority carrying reflux — and unlike an older child we can monitor by symptom, an infant's first pyelonephritis often presents as nothing more specific than a fever, by which point some scarring may already be underway.
I recognize PREDICT didn't enroll children at her grade — that's exactly why I'm not citing it as proof prophylaxis works for her; I'm citing the cost side instead: at this age, TMP-SMX once daily is well tolerated, and the downside of being wrong by treating is far smaller than the downside of being wrong by not.
You're right that the cost of one daily antibiotic dose in an otherwise healthy infant is small in any single case — but “small and repeated across every incidentally-found low-grade reflux we see” is exactly how resistant organisms accumulate in a population, and PREDICT itself found a shift toward less-susceptible organisms in its treated arm even though the trial enrolled children at a HIGHER grade than Wren.
Calling PREDICT's own numbers “the cost side” undersells what the trial showed: it wasn't just neutral on scarring at grade III–V, it also documented a real resistance cost from precisely the intervention you're proposing on weaker evidence. If prophylaxis's downside showed up even where the trial found a modest UTI benefit, I don't see why we'd expect a better trade in a child outside that population altogether.
Neither of you is wrong about the evidence — there genuinely isn't a trial that answers this for grade II reflux, in either direction. But “no trial answers it” doesn't have to resolve to a flat yes or no held for years; it can resolve to a bounded trial of observation with a hard revisit date. Watch for fever without antibiotics until the twelve-month re-imaging, culture promptly if she does spike one, and let that scan — not the guess we're making today — decide whether the conversation about daily antibiotics even needs to happen again.
The family and the urology team settled on a twelve-month follow-up VCUG rather than a decision now, with instructions to bring Wren in for a urine culture before any antibiotic is given for a fever in the meantime, not after.
Not settled: whether a fever between now and then, with a positive culture, should convert the plan to prophylaxis immediately or wait for the scheduled scan regardless. The pediatrician's read is that a real infection would answer the underlying question on its own; the urologist thinks a single febrile UTI in an infant already known to reflux is reason enough to stop waiting.
Camille O., a four-year-old girl, has had two febrile urinary tract infections in the past eight months, the second of which prompted the renal ultrasound and voiding cystourethrogram that found grade IV reflux on the right. She is the middle of three sisters and, by her parents' account, treats bathroom breaks as an interruption to whatever game is already in progress — she has been observed crossing her legs and dancing in place rather than stop playing, voids only two or three times some days, and has a documented history of hard, infrequent stools going back to before her toilet training was even complete. A bladder-bowel dysfunction screening questionnaire administered at her urology visit scored above threshold on both the voiding and stooling domains.
Grade IV reflux and a documented second febrile infection would place Camille inside the population RIVUR actually enrolled — 607 children with grade I–IV reflux after a first or second UTI — where prophylaxis cut recurrence roughly in half (13% versus 24%, hazard ratio 0.5). But RIVUR's own later subgroup analysis found that headline number undersold children who look like her specifically: among the 12% of RIVUR's cohort with bladder-bowel dysfunction at enrollment, the hazard ratio for recurrence fell to 0.21, and a subsequent reanalysis using penalized regression put the number needed to treat for that subgroup at 4, against roughly 10 for the trial as a whole. Whatever the reflux grade contributes, her holding and her constipation appear to be doing at least as much of the actual work.
What the reflux number leaves out
Start prophylaxis now and treat the bladder-bowel dysfunction alongside it, not sequentially. Camille meets RIVUR's own entry criteria almost exactly — grade IV reflux, second febrile UTI — and the subgroup work on that same trial found children with bladder-bowel dysfunction got MORE benefit from prophylaxis, not less, with the hazard ratio dropping to 0.21 against the trial's overall 0.5.
I'd treat this as an unusually strong case for prophylaxis, not a marginal one — the trial evidence and her own clinical picture happen to point the same direction.
I don't dispute the subgroup numbers — you're right that they favor her specifically, more clearly than most reflux cases we see. But that same analysis is a secondary, hypothesis-generating reanalysis of RIVUR, not a prospective trial of prophylaxis-plus-BBD-treatment versus prophylaxis alone, and the trial's own primary authors have been explicit that bladder-bowel dysfunction predicts who benefits FROM treating it, as much as who benefits from antibiotics riding alongside.
Starting prophylaxis today without a real bowel regimen in place risks crediting the antibiotic for whatever improvement actually comes from fixing her constipation and voiding habits — the two are entangled in her case in a way the subgroup analysis can describe but can't separate.
Then do both, deliberately, so we can actually tell them apart later: start the daily antibiotic now given how clearly her picture matches the highest-benefit subgroup, and start a structured bowel regimen — scheduled toileting, osmotic laxative for the constipation — on the same visit rather than waiting to see if antibiotics alone are enough. If her voiding and stooling normalize and infections stop, we'll have a real answer at her recheck about how much of the credit belongs to which intervention, instead of guessing.
Camille started trimethoprim-sulfamethoxazole prophylaxis and a scheduled-voiding-plus-osmotic-laxative bowel program on the same visit, with both started together rather than staggered, and a follow-up in three months to reassess bladder-bowel symptoms directly rather than infer them from infection recurrence alone.
Not fully resolved: if she reaches three months infection-free with her bowel and voiding symptoms also resolved, whether that will read as evidence prophylaxis worked or as evidence it was never separately necessary once the BBD was treated. The nephrologist expects the recheck to leave that question about as open as it is today; the urologist expects the subgroup data to hold up regardless of how neatly the two threads untangle.