Abstract
RATIONALE: Phototherapy is the mainstay of treatment for neonatal hyperbilirubinaemia. Increasing the dosage or intensity of phototherapy by various means is commonly practised for infants with high levels of serum bilirubin. It is unclear whether these measures of increasing phototherapy dosage are beneficial and safe for neonates. OBJECTIVES: To assess the effects of high-dose versus low-dose phototherapy on bilirubin level and associated clinical outcomes that constitute the major conditions in kernicterus spectrum disorder (KSD), such as acute bilirubin encephalopathy and kernicterus, as well as cerebral palsy and neurodevelopmental disabilities in infants with hyperbilirubinaemia; and to assess the effects of high-dose phototherapy, defined as the use of high levels of measured spectral irradiance greater than 30 μW/cm²/nm over the same bandwidth, as well as low-dose phototherapy, defined as measured levels of spectral irradiance below 30 μW/cm²/nm. SEARCH METHODS: Searches were conducted up to 09 June 2025 in MEDLINE, Embase, CENTRAL, and two trial registries. We checked reference lists of related reviews and included studies. ELIGIBILITY CRITERIA: We included randomised controlled trials (RCTs), quasi-RCTs and clustered RCTs of term (≥ 37 weeks' gestation) and preterm (˂ 37 weeks' gestation) infants with hyperbilirubinaemia, and requiring any type of phototherapy. We compared high- to low-dose phototherapy. High- and low-dose phototherapy are achieved in various ways: by the use of multiple devices; different levels of light intensity (irradiance); or positioning of the devices closer to the skin. We compared these different methods of delivering phototherapy to each other. We specified no exclusion criteria for the infants. We excluded cross-over studies. OUTCOMES: Our critical outcomes were incidence of acute bilirubin encephalopathy and kernicterus, the proportion of infants with moderate or severe cerebral palsy, all-cause mortality before discharge and serum bilirubin levels at the latest periods of measurements, which included 24, 48 and 72 hours after treatment commencement. RISK OF BIAS: We used the Cochrane RoB 2 tool to assess bias in the RCTs. SYNTHESIS METHODS: We synthesised results for each outcome with data using meta-analysis, where possible, with a random-effects model. For outcomes with a high degree of heterogeneity, we explored study characteristics and performed meta-regression. We used GRADE to assess the certainty of evidence for critical and selected important outcomes. INCLUDED STUDIES: We included 41 studies (6927 infants) of preterm, late preterm and term gestations with different birth weights. The phototherapy treatment in the studies lasted from 16 to 60 hours. Thirty-four studies were parallel RCTs, six were quasi-RCTs and one had unclear allocation methods. The studies were published between 1972 and 2025 and conducted in 21 lower-, lower-middle, upper-middle and high-income countries. SYNTHESIS OF RESULTS: Thirty-eight studies (n = 6508) contributed data to the meta-analysis. We identified one ongoing study. A single overall comparison of high-versus low-dose phototherapy was evaluated with various means of achieving high and low dosages. One-third of the studies had a high risk of bias, and another third had some concerns about the randomisation process, while most studies had low risk of bias in all the other domains. There was a very high degree of heterogeneity in several outcomes, including serum bilirubin levels at various time points, which was inadequately explained and resulted in the downgrading of evidence certainty. Evidence certainty was also downgraded for imprecision for the critical outcomes of the incidence of acute bilirubin encephalopathy, proportions with moderate-to-severe cerebral palsy and all-cause mortality before hospital discharge. From a single study, no infants developed acute bilirubin encephalopathy (Not estimable; 1 study, 80 participants; very low-certainty evidence) and no studies assessed kernicterus. It is very uncertain whether high-dose phototherapy reduces moderate or severe cerebral palsy (RR 0.69, 95% CI 0.46 to 1.02; I² = 0%; 2 studies, 1927 participants; very low-certainty evidence). High-dose phototherapy may result in little to no difference in all-cause mortality before hospital discharge (RR 1.05, 95% CI 0.89 to 1.23; I² = 0%; 2 studies, 1985 participants; low-certainty evidence). High-dose phototherapy may result in a small reduction in serum bilirubin (in micromol/L) at 24 and 48 hours after commencement: at 24 hours (MD -22.76, 95% CI -32.02 to -13.50; I² = 97%; 35 studies, 5025 participants; low-certainty evidence); at 48 hours (MD -23.63, 95% CI -40.76 to -6.50; I² = 91%; 8 studies, 776 participants; low-certainty evidence), with very uncertain effects at 72 hours (MD -9.00, 95% CI -19.53 to 1.53; I² not applicable; 1 study, 144 participants; very low-certainty evidence). AUTHORS' CONCLUSIONS: The low- to very low-certainty evidence precluded firm conclusions about the effects of high-dose compared to low-dose phototherapy on acute bilirubin encephalopathy, moderate or severe cerebral palsy and all-cause mortality. No studies evaluated kernicterus. It is unclear how small reductions in serum bilirubin at 12 to 48 hours translate to clinically important benefits. Further RCTs with robust randomisation methods and detailed documentation of population and intervention, with inclusion of patient-important outcomes are needed to improve the certainty of the evidence. FUNDING: This Cochrane review had no dedicated funding. REGISTRATION: Protocol (2001) DOI: 10.1002/14651858.CD0033; Protocol (2020) DOI: 10.1002/14651858.CD003308.pub2.