#447 - [Journal Club] - 📌 Are we missing UTIs in neonates hospitalized for unexplained jaundice?

In this Journal Club episode, Daphna reviews a retrospective cohort study from Istanbul examining clinical, laboratory, and ultrasound factors associated with UTI in neonates hospitalized for unexplained hyperbilirubinemia. Among 96 term and near-term infants, 31% had culture-proven UTIs, a striking prevalence. Pathological renal ultrasound findings were independently associated with UTI, with affected neonates 4.6 times more likely to have a concurrent infection. Notably, standard laboratory markers including CRP and white blood cell count failed to distinguish UTI-positive from UTI-negative infants. The findings prompt a practical question: should urine culture be part of the routine workup for neonatal hyperbilirubinemia?

----

Renal ultrasonography findings are associated with urinary tract infection in neonates with asymptomatic hyperbilirubinemia. Sarı EE, Salihoğlu Ö.J Perinatol. 2026 Apr 13. doi: 10.1038/s41372-026-02686-x. Online ahead of print.PMID: 41975209

Transcript

[00:00] Ben Courchia, MD: Hello everybody. Welcome back to The Incubator Podcast Journal Club. We're back today for another episode. Daphna, you are presenting today. What's going on?

[00:08] Daphna Barbeau: Well, I'm going to take your primer on neonatal hyperbilirubinemia and talk about this article — which made me go, "what?" It's entitled "Renal Ultrasonography Findings Are Associated with Urinary Tract Infection in Neonates with Asymptomatic Hyperbilirubinemia," published in the Journal of Perinatology. The lead author is Amin Ergul-Sari and the senior author is Ozgul Saluglu. This is coming to us from Istanbul. I'll admit I'm of Turkish descent and have very little Turkish proficiency, so — shame on me.

[01:04] Daphna Barbeau: The objective of this study was to look at clinical, laboratory, and ultrasound factors associated with urinary tract infection (UTI) in neonates hospitalized for unexplained hyperbilirubinemia. The introduction is valuable. As we mentioned, neonatal hyperbilirubinemia is among the most frequent clinical conditions requiring medical attention and even hospital readmissions in the early postnatal period. While the majority of cases are physiologic and typically benign, even indirect hyperbilirubinemia can serve as a sentinel sign of something concerning — especially UTI. The prevalence of UTI in asymptomatic neonates presenting with unexplained jaundice has been reported to range from 6% to as high as 11%. The pathophysiological link is multifactorial, but it's hypothesized that gram-negative bacteria — particularly E. coli — release endotoxins that can lead to cholestasis, resulting in elevated bilirubin levels even in the absence of systemic sepsis.

[02:30] Daphna Barbeau: This was a retrospective observational cohort study conducted at one level three NICU, looking at records of all neonates hospitalized for the management of hyperbilirubinemia between January 2013 and December 2017. They included term and near-term neonates — anyone greater than or equal to 37 weeks gestational age — hospitalized for jaundice requiring phototherapy according to the AAP guidelines at that time, which have since been updated. Patients were included only if they had no identifiable cause for jaundice other than a potential UTI. Asymptomatic hyperbilirubinemia was defined as the absence of fever, feeding intolerance, respiratory distress, lethargy, or hemodynamic instability at admission. Exclusion criteria included hemolytic disease of the newborn, known glucose-6-phosphate dehydrogenase (G6PD) deficiency — which is at even higher rates in that population — major congenital urinary tract anomalies, history of perinatal asphyxia, evidence of systemic sepsis or meningitis, and prior antibiotic use.

[03:20] Daphna Barbeau: A confirmed UTI was defined as: one, a positive urinalysis — presence of leukocyte esterase on urine dipstick and pyuria defined as greater than five white blood cells per high power field on microscopic examination — and two, a positive urine culture with growth of a single uropathogen at a colony count of greater than or equal to 50,000 CFU. Specimens with mixed growth, low colony counts, or lacking associated pyuria were classified as contaminants. All neonates underwent renal and urinary tract ultrasound within the first 48 hours of admission.

[04:00] Daphna Barbeau: A total of 96 neonates were included. Based on culture results, 30 neonates — 31.2% — were diagnosed with culture-proven UTI, while 66 — 68% — had negative cultures. So a third of babies admitted for hyperbilirubinemia had culture-proven UTIs. There was a statistically significant difference in gestational age: neonates in the UTI-positive group were born at a slightly lower gestational age — 37 ± 1.36 weeks versus 37 ± 1.28 weeks. There were no significant differences in birth weight, gender, or mode of delivery. Mean hospital stay was slightly longer in the UTI group, but this was not statistically significant.

[05:35] Daphna Barbeau: Neonates in the UTI group were significantly older at presentation — 8.4 ± 5 days versus 4.8 ± 2 days in the non-UTI group. In a subgroup analysis, late-onset presentation at greater than or equal to seven days of life was significantly associated with a higher risk of UTI — I thought that was one of the most impressive findings. Laboratory values at admission did not distinguish the groups: total bilirubin, direct bilirubin, white blood cell counts, and C-reactive protein (CRP) levels showed no statistically significant differences. The most frequent uropathogens were Klebsiella and E. coli. All neonates in the UTI group had significant pyuria, while neonates in the control group had normal urinalysis findings.

[06:30] Daphna Barbeau: Pathological renal ultrasound findings were significantly more frequent in the UTI group — 30% (9 babies) — compared to controls — 9% (6 babies). Among UTI-positive neonates with pathological urinary tract findings, the most common anomaly was pelvicaliectasis followed by hydronephrosis. In contrast, all pathological findings in the control group were isolated pelvicaliectasis — no cases of hydronephrosis. The authors highlight that this likely indicates that mild dilation is common, but hydronephrosis in this cohort was exclusively clustered within the UTI-positive group. In a multivariate model including gestational age, gender, and ultrasound findings, pathological renal ultrasound findings were independently associated with UTI — neonates with structural anomalies were 4.6 times more likely to have a concomitant UTI.

[07:48] Daphna Barbeau: Their conclusions are that the study demonstrated a high prevalence of UTI — 31% — among neonates hospitalized for unexplained hyperbilirubinemia, and identifies renal ultrasonographic abnormalities as the factor most strongly associated with UTI. And again, this was a group of babies sick enough to need hospitalization for phototherapy — so a sicker group than the general population. But I think it should give us pause. Sometimes we use the white blood cell count to try to rule out infection. In our practice, we are sending blood cultures for babies admitted with high bilirubin levels.

[09:09] Ben Courchia, MD: Hold on — I'm going to push back. I don't think we order blood cultures on everybody admitted for hyperbilirubinemia.

[09:22] Daphna Barbeau: I do, especially if they need to come back to the NICU with our new bilirubin thresholds. I don't put them on antibiotics, but I send the culture.

[09:30] Ben Courchia, MD: We're going to get a lot of emails about that.

[09:42] Daphna Barbeau: Listen, if somebody can show me the global policy on obtaining blood cultures, I'm happy to follow it.

[09:50] Ben Courchia, MD: My point here is that that's the real discussion — a baby comes back for hyperbilirubinemia, do you get a blood culture or not? Maybe you should just get a urine culture.

[10:04] Daphna Barbeau: Yeah, that's really what this says to me — UTIs are much more likely to be the problem than sepsis in a well-appearing baby. How many blood cultures have we had come back positive in recent memory? None.

[10:20] Ben Courchia, MD: No — you'd get a positive blood culture on a kid with urosepsis, but that kid would be really sick. So it might be much higher yield to just get a urine culture. This is what makes the data interesting. I tend to dismiss retrospective studies unless the effect size is impressive — and this one is small, fewer than 100 patients. But that finding from Figure 1, 30% pathological renal ultrasound findings in the UTI group versus 9% in controls — that's not close. And it reflects a lot of our clinical reality. I also love that they included CRPs, since we send a lot of them in our unit. And they were essentially normal — 0.28 versus 0.31. No help whatsoever. Same with the white blood cell counts — 14 versus 13, nothing to write home about, especially if obtained by heel stick. And the direct bilirubin wasn't meaningfully different either.

[11:37] Daphna Barbeau: Yeah. We tend to think of direct hyperbilirubinemia as the trigger to look for UTI, but indirect? We're less worried. So which baby are we not worried about? The baby that lost weight — we say, "oh, probably not eating enough." But maybe they're not eating enough because they don't feel well. Which baby can we look at and say this is definitely physiologic? It gives me pause.

[12:07] Ben Courchia, MD: That reminds me — my first real ear-pulling as an attending was related to this. I admitted two babies in a row with hyperbilirubinemia and got blood cultures on both — a practice I'd picked up in fellowship. The medical director at the time sent an email to the entire group, in a very non-accusatory way, saying that the practice for babies admitted with hyperbilirubinemia was not to get blood cultures. You know the kind — the email goes to everyone, but you know it's for you. I went to him afterwards and said, "I think that was for me." He said, "Yep, that was for you."

[13:16] Daphna Barbeau: I remember those emails. And for anyone listening — Dr. Doron, wonderful medical director. He never made things punitive. But you still knew. You always knew it was you.

[13:36] Ben Courchia, MD: Anyway. All right, I'll see you tomorrow.

[13:43] Daphna Barbeau: Sounds good.