Neonatal Jaundice Treatment Versus Recommendations: The Challenge of Treatment Without Rapid Diagnostic Capability

Background: Neonatal jaundice (NNJ) is a leading cause of death in the early neonatal period, disproportionately affecting newborns in sub-Saharan Africa. In a setting without access to rapid assessment via transcutaneous bilirubin meter, we sought to determine how closely the diagnosis and treatment of neonatal jaundice at an urban district hospital aligned with retrospective assessment and treatment recommendations using the BiliApp (based on the UK NICE Guideline CG98). Methods: This retrospective chart review study aimed to identify: 1) What percent of admissions within 7 days of birth to the Mother and Baby Unit (MBU) at our study site were admitted for a primary diagnosis of neonatal jaundice, and what characterized those admissions? 2) How did treatment provided compare to the recommendations of the United Kingdom NICE Guideline CG98 via the “BiliApp”? and 3) Among those with jaundice, what factors were associated with an increased likelihood of severity indicative of the need for blood exchange therapy? The charts of all neonates admitted to the MBU at Suntreso Government Hospital (SGH), in Kumasi, Ghana, in 2020 were reviewed by trained research assistants. Data were collected regarding demographics, reason for admission, diagnostic markers (e.g. serum bilirubin level), treatments performed in the hospital and outcome. Data were analyzed using Stata 18.0. Results: There were 1,059 newborns admitted to the MBU in 2020 at less than 8 days of age. 179 (16.9%) were admitted with a primary diagnosis of neonatal jaundice. According to the BiliApp, 29.4% (N=50) of newborns admitted for jaundice had bilirubin levels that were normal or below the phototherapy threshold for their gestational age on admission; 25.3% (N=43) were at or near the threshold for phototherapy; 21.2% (N=36) were above the phototherapy threshold; and 24.1% (N=41) were above the blood exchange therapy threshold. The BiliApp recommended no treatment for 21.2% (N=36) of newborns, repeated assessment for 33.6% (N=57), phototherapy for 21.2% (N=36), and exchange therapy for 24.1% (N=41). By comparison, 8.2% (N=14) of neonates admitted for jaundice received no treatment, 77.8% (N=133) received phototherapy only, and 14.0% (N=24) received both phototherapy and exchange therapy. Absent sufficient data on G6PD status and parent/newborn blood type to include in the analysis, the biggest risk factors for a BiliApp recommendation of exchange therapy included serum bilirubin level (OR 1.01, p< 0.001) and gestational age (OR 0.51, p< 0.001), even after controlling for breastfeeding and male sex of the newborn. Conclusions: Without access to rapid assessment tools, many providers in low-resource settings are put in a position to presumptively treat newborns suspected of having jaundice, rather than waiting for serum lab tests to return. Given the cost of transcutaneous bilirubin meters, additional options for rapid diagnostic testing are warranted.