The Snowball Effect of Low-Value Care

Why Admit for Phototherapy?

At the time of admission, the infant’s TSB was 1.2 mg/dL below the phototherapy threshold recommended by the 2004 AAP hyperbilirubinemia guideline.² Although both the guideline and BiliTool specify that thresholds are approximations and that phototherapy can be provided at lower TSB levels,³ there were several clinical factors that, in conjunction with the guideline, could have supported not admitting this infant. He was feeding well with demonstrated weight gain since discharge, and the rate of rise of his TSB was slowing, dropping from 0.13 mg/dL per hour between 51 and 111 hours to 0.07 mg/dL per hour between 111 and 129 hours. These factors, combined with the family’s demonstrated reliability for follow-up, could have kept the infant safely out of the hospital. Although it is noted the admission history that the patient’s grandmother, a health care worker, strongly desired admission given the closeness of the bilirubin to the threshold for phototherapy, the extent of the conversation around risks and benefits of admission is unclear.

New literature since the publication of the AAP hyperbilirubinemia guideline suggests that even significantly elevated bilirubin levels are less harmful than previously thought,^4–6 whereas phototherapy itself is being investigated for potential physiologic harms, including damage to DNA and links to childhood cancer.^7–9 Admission for phototherapy may also disrupt breastfeeding and maternal-child bonding.^10–12 Furthermore, based on work by Newman et al,¹³ for a male infant that was 38 weeks’ gestation and >72 hours old with a TSB 2 mg/dL to <1 mg/dL below the AAP phototherapy threshold, the number needed to treat to prevent 1 infant from reaching the AAP’s exchange transfusion threshold is 1840. Importantly, this is not the number needed to treat to prevent the true outcome of concern, chronic bilirubin encephalopathy, which is likely much higher, because the estimated threshold of TSB for causing chronic bilirubin encephalopathy is at least 5 mg/dL above the exchange transfusion threshold.⁵

Here, the preference of at least 1 family member weighed significantly into the decision to admit for phototherapy. Shared decision-making includes exploring patient and family preferences around 2 or more medically reasonable options, after clinicians provide explicit information about benefits, risks, and costs of each option.^14–16 Given the current literature, if an infant does not meet the published threshold for phototherapy, it seems questionable as to whether the option of admission should be offered to the family at all. However, when a discussion around the need for phototherapy does occur, sharing emerging data on how conservative current thresholds are, as well as the possible harms of phototherapy, may help prevent unnecessary admissions, such as in this case.

Why Perform a Partial Sepsis Workup?

The AAP recommended workup for an infant admitted to the hospital for phototherapy is a blood type, Coomb’s, complete blood count with smear, and a direct bilirubin, unless the history and/or presentation suggest sepsis.² This patient had normal vital signs and was well appearing yet underwent a fairly extensive workup, with the admitting physician citing factors in the history and physical that do not indicate an increased risk for a serious bacterial infection. The nonpurulent eye discharge, without additional signs of conjunctivitis and the mother’s negative gonococcal and Chlamydia screen, is unlikely to represent ophthalmia neonatorum, which in isolation would be a rare presentation of sepsis anyway.^17–20 Furthermore, the elevated direct bilirubin cited as possibly being secondary to a UTI, was not actually elevated per the AAP guideline, which defines an abnormal direct bilirubin as >1.0 mg/dL if the TSB is ≤5 mg/dL, or >20% of a TSB >5 mg/dL.² Therefore, there was no indication for additional workup beyond routine hemolysis laboratories.

How Should the Urine Culture Results Be Interpreted?

The 2011 AAP clinical practice guideline for febrile UTIs states that the diagnosis of UTI “is made on the basis of the presence of both pyuria and at least 50 000 colonies per mL of a single uropathogenic organism.”²¹ It describes that “organisms such as Lactobacillus spp., coagulase-negative staphylococci, and Corynebacterium spp. are not considered clinically relevant urine isolates for otherwise healthy, 2- to 24-month old children.” Although this patient is a neonate and does not technically fall within the guideline’s purview, through our literature review we could find no reported cases of S epidermidis causing true community-acquired UTI in the neonatal period. In fact, there are limited case reports of S epidermidis causing community-acquired UTI in the entire pediatric age range.^22,23

There is an argument to be made for reevaluating the infant after the urine culture results were received. He was an uncircumcised neonate, with 5 to 10 WBC/hpf on a urinalysis that grew >50 000 CFUs/mL of a single species. According to Schroeder et al,²⁴ pyuria >3 WBC/hpf may have a sensitivity and specificity of up to 96% and 91%, respectively, in infants <3 months old, leading to a notable positive likelihood ratio of 10.7. However, based on the guidelines discussed previously, the urine studies were not indicated, and the organism was not a uropathogen. Because the infant continued to be well appearing on follow-up without having received previous antibiotics, the initial urine culture results almost certainly represented either contamination or asymptomatic bacteriuria, for which there is no evidence to justify a repeat catheterization.