1. Subjects
Preterm neonates were enrolled at the neonatal intensive care unit at Nara Medical University Hospital, Japan, between October 2018 and May 2021. Selected were ≤ 33 weeks of gestational age and had uncomplicated hyperbilirubinemia. Neonates eligible for this study were subjected to single-color PT; they were randomly divided into two groups for either blue LED exposure (neoBlue in High Mode, Natus Medical, San Carlos, CA) or green fluorescent lamp (FL) exposure (FL20S-BG-NU (20 W ×5), PIT-220TLR, Atom Medical Corp., Tokyo, Japan). Of the 106 neonates, 38 showed resistant hyperbilirubinemia after single PT and had to be switched to dual PT (simultaneous exposure to both the blue LED and green FL). Seven neonates did not provide both pre-PT and post-PT urine samples. As a result, a total of 61 neonates were included in this study. Decisions regarding the initiation and termination of PT were made based on the standard clinical criteria [20].
2. Phototherapy devices and light intensities
Spectral intensities were determined by integrating each spectrum as a function of wavelength measured using an HR4000 spectrometer (Ocean Optics, Dunedin, FL) calibrated with Excel spreadsheet software (Microsoft Excel for Mac, ver. 16.68, Microsoft Co., Redmond, WA). Light intensity was measured at a distance of 30 cm in both devices. As light is most transmissive in the range of 400 to 520 nm in the epidermis [21], the light intensity was examined over four sub-ranges; 400–430, 430–460, 460–490, and 490–520 nm (shown in Fig. 1). The spectral intensities of the wavelength bands were integrated (µW/cm2) in an Excel spreadsheet and divided by 30 nm to obtain the average light intensity.
3. Blood samples
Blood samples were collected from patients by pricking the heel into sodium heparinized micro-hematocrit capillary tubes (Becton, Dickinson and Company, Franklin Lakes, NJ). Serum was immediately separated by centrifugation for 5 min at 11,800×g (KUBOTA3220, Kubota Corp., Tokyo, Japan) for measurement of total serum bilirubin (TSB) using 2-wavelength (455 nm, 575 nm) spectrophotometry (BL-300, TOITU Co., Tokyo, Japan). Blood collection was executed within 2 h of the initiation of PT for measurement of pre-PT TSB or > 2h after PT initiation for mid-PT TSB.
4. Urine samples
Urine samples were collected by placing several small cotton balls near the external genitalia in a diaper. After removal, the cotton balls were immediately squeezed to transfer the urine into a micro-tube (ST-0150R, INA, OPTICA Co., Ltd, Osaka, Japan), and the sample was frozen at − 80°C until use. Urine samples were centrifuged for 10 min at 2,600×g (KUBOTA 2800, Kubota Corp., Tokyo, Japan) at 4°C, and the supernatants were subjected to PUZZLU assay. Urinary creatinine was measured using an enzymatic assay kit (sarcosine oxidase-peroxidase) (Serotec Co., Ltd., Sapporo, Japan). Urine collection was performed three times everyday from postnatal day 0 (P0) to P7.
5. Bilirubin (ZZ-BR) solution
Reference bilirubin (ZZ-BR) solutions were prepared to generate a standard curve for the measurement of urinary UnaG-binding bilirubin (UUB). A total of 2 mg of bilirubin (98%) (FUJIFILM Wako Pure Chemical Co., Osaka, Japan) was dissolved in 2 mL of 0.1 mol/L NaOH, which was immediately neutralized with 1 mL of 0.1 mol/L phosphoric acid and then mixed with 7 mL of human serum albumin (HSA; albumin 5% I.V. 5 g/100 mL, Japan Blood Products Organization, Tokyo, Japan). The concentration of bilirubin in the preparation was measured using a BL-300 2-wavelength spectrophotometer (TOITU Co., Ltd., Tokyo, Japan) (455 nm, 575 nm). The bilirubin solution was diluted with phosphate-buffered saline (0.1 mol/L, pH 7.2) to desired concentrations. The above reagents, except HSA, were purchased from FUJIFILM Wako Pure Chemical Co. (Osaka, Japan).
6. UnaG mixture
UnaG is a fluorescent protein isolated from Japanese eel muscle [3]. Its apo-form, apoUnaG, was prepared as a purified bacterially expressed recombinant protein. The “UnaG mixture” contained apoUnaG, HSA, and ascorbic acid at a ratio of 1:1:1 (v/v/v), in which their final concentrations were 5 µmol/L, 0.01%, and 0.1%, respectively [4]. L (+)-ascorbic acid was purchased from FUJIFILM Wako Pure Chemical Co. and added to reduce photo-oxidation of bilirubin and its derivatives.
7. PUZZLU
A black microplate (Microtest™ 96-well assay plate, black, flat bottom, BD Biosciences, Franklin Lakes, NJ) was prepared, and 50 µL of each sample (reference bilirubin or urine) was added to 150 µL of UnaG mixture. The resulting 200-µL reaction mixture was pipetted into a well of the microplate. The fluorescence intensity of each well was measured spectrophotometrically at 37°C with fluorescence filters for excitation and emission at 498 and 527 nm, respectively. For calibration curve generation, the bilirubin solution was diluted with PBS to concentrations of 0, 1, 2, 3, 4, and 5 µmol/L and analyzed as described above. Before light exposure, the calibration curve was drawn as UnaG-bound bilirubin (shown in Fig. 2).
Thawed urine samples were exposed to blue-light LED irradiation (P4630-blue LED unit; wavelength range 420–520 nm, with a peak emission at 450 nm, 10.5 mW/cm2 at a distance of 4 cm above the microplate, Ushio Inc., Tokyo, Japan). Samples were assayed in duplicate every 15 min (until 90 min) or 30 min (over 90 min). The UUB concentration was defined as the highest fluorescence intensity during the exposure time, as determined from the calibration curve. Urine sampled before PT was undiluted, whereas that during PT was diluted 5-fold. Final UUB levels were corrected for urinary creatinine and expressed as UUB/Cr (µg/mg Cr).
8. Calculations
The rate of TSB reduction was expressed as 1 − (post-TSB/pre-TSB) × 100/ (PT exposure time) (%/h). Light in the range of 400 to 520 nm entering the epidermis is mostly transmitted [21]. Therefore, the TSB reduction rate was divided by the average light intensity from 400–520 nm to express the therapeutic efficiency of the PT device (%/h/µW/cm2/nm). Similarly, maximum UUB/Cr was also divided by the average light intensity from 400–520 nm to express the excretion efficiency of the phototherapy device (µg/mg Cr/µW/cm2/nm).
9. Statistical analysis
Linear regression analyses and Mann-Whitney U tests were performed using StatFlex software, ver. 6 (Artech Co., Osaka, Japan).