Configurational photoisomerization of bilirubin in vitro--II. A comparative study of phototherapy fluorescent lamps and lasers

Phototherapy: current methods and future directions

Phototherapy is the most common therapeutic intervention used for the treatment of hyperbilirubinemia. Although it has become a mainstay since its introduction in 1958, a better understanding of the photobiology of bilirubin, characteristics of the phototherapy devices, the efficacy and safety considerations of phototherapy applications, and improvements in spectroradiometers and phototherapy devices are necessary for more predictable and improved clinical practices and outcomes. A step forward in instituting consistent, uniform, and effective use of phototherapy is the recent American Academy of Pediatrics clinical guideline on the management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation, which outlines a clinical strategy for the diagnosis of hyperbilirubinemia and contains direct recommendations for the application of phototherapy. This article reviews the parameters that determine the efficacy of phototherapy, briefly discusses current devices and methods used to deliver phototherapy, and speculates on future directions and studies that are still needed to complement our presently incomplete knowledge of the facets of this common mode of therapy.

A prospective randomized controlled study of phototherapy using blue and blue-green light-emitting devices, and conventional halogen-quartz phototherapy

OBJECTIVE

To determine the efficacy of blue versus blue-green phototherapy using new light sources with narrow luminous spectra. The devices made of high-intensity gallium nitride light-emitting diodes (LEDs) were also compared to conventional halogen-quartz bulbs phototherapy.

DESIGN

Prospective open randomized study.

PARTICIPANTS

A total of 114 jaundiced, but otherwise healthy term infants who met the entry criteria for phototherapy set by the American Academy of Pediatrics' Practice Parameter.

MAIN OUTCOME MEASURES

The duration of phototherapy and the rate of decrease in total serum bilirubin (TSB).

RESULTS

The mean TSB concentrations at initiation and termination of treatment, as well as the duration of phototherapy and the rate of decrease in TSB, were not statistically different in newborns receiving blue LED, blue-green LED or conventional phototherapy. The average rate of decrease in TSB (slope), after adjustment by a linear regression analysis for confounding factors, was -3.61 micromol/hour (95% confidence limits -5.47, -1.75) in the 25 newborns receiving blue LED phototherapy compared with -2.57 micromol/hour (-4.32, -0.82) in the 22 newborns treated with blue-green LED phototherapy and -3.42 micromol/hour (-5.02, -1.81) in the 57 newborns who received conventional phototherapy.

CONCLUSIONS

When using low light irradiance, there was no statistically significant difference in the effectiveness of phototherapy using blue-green LEDs, blue LEDs or conventional halogen-quartz bulbs.

Bilirubin photoisomerization products in serum and urine from a Crigler-Najjar type I patient treated by phototherapy

The relative compositions of the photoisomers of bilirubin-1X alpha (4Z, 15Z-bilirubin) in serum and urine of a patient with Crigler-Najjar type I syndrome treated by phototherapy are reported. High-performance liquid chromatography analysis reveals the presence of high serum levels of the configurational bilirubin photoisomer (4Z,15E-bilirubin) before the beginning of phototherapy (between 12 and 16% of the total bilirubin). The configurational photoisomer value increases during phototherapy with blue fluorescent lamps up to a photoequilibrium of about 25%, similar to that obtained in a bilirubin solution in vitro irradiated by the same lamps. This evidence suggests an inefficient serum excretion of the 4Z,15E-bilirubin. Indeed, its average half-life in serum of the Crigler-Najjar patient is found to be about 8 h. No detectable traces of the bilirubin structural isomer, lumirubin, are found in the serum. On the other hand, lumirubin represents the dominant bilirubin isomer excreted in the urine, as both 15Z and 15E configurations. Smaller amounts of 4Z,15E-bilirubin, 4E,15Z-bilirubin and native 4Z,15Z-bilirubin are observed in urine. The presence in urine of 4Z,15Z-bilirubin is probably due to a fast reversion of the configurational photoisomers to their native form. The half-life of the configurational photoisomers in urine kept at 38 degrees C is found to be of the order of a few minutes. Our study indicates that in Crigler-Najjar type I patients, mechanisms exist to excrete all bilirubin photoisomers. The lumirubin pathway seems to contribute markedly to bilirubin excretion in the urine, as occurs in jaundiced babies under phototherapy. However, the contribution of configurational isomers cannot be neglected.

Light-emitting diodes: a novel light source for phototherapy

High intensity light-emitting diodes (LEDs) are being studied as possible light sources for the phototherapy of hyperbilirubinemic neonates. These power-efficient, low heat-producing light sources have the potential to deliver high intensity light of narrow wavelength band in the blue-green portion of the visible light spectrum, which overlaps the absorption spectrum of bilirubin (BR). We compared the efficacy between single LEDs of different color and then constructed a prototype phototherapy device using 300 blue LEDs. The efficacy of this device was compared with that of conventional phototherapy devices by measuring the in vitro photodegradation of BR in human serum albumin. When blue, blue-green, green, and white LEDs were compared, the blue light was the most effective in degrading BR by 28% of dark control, followed by blue-green (18% of control), and then white light (14% of control). Green light was the least effective (11% of control). The prototype device with three focused arrays, each with 100 blue LEDs, generated greater irradiance (> 200 microW.cm-2.nm-1) than any of the conventional devices tested. It also supported the greatest rate of BR photodegradation. We conclude that light from LEDs should be considered a more effective treatment for hyperbilirubinemia than light from presently used phototherapy devices. Furthermore, the unique characteristics of this light source may make it especially suitable for use in safe and lightweight home phototherapy devices.

Quantum yield and skin filtering effects on the formation rate of lumirubin

Photocyclization of bilirubin to lumirubin in the skin of jaundiced infants exposed to blue-green light irradiation is considered to be the most important process for bilirubin elimination from the organism. The quantum yield phi LR of the bilirubin-->lumirubin photoreaction has been recently measured and found to vary with the excitation wavelength, with a peak at about 520 nm. The quantum yield phi ZE for the strongly competing reversible configurational photoisomerization of bilirubin has also been recently shown to be wavelength dependent and to decrease significantly in the long-wavelength part of the absorption band of bilirubin. These new data are taken into account to model the bilirubin photochemistry in vivo by using a simplified skin optical model based on the Kubelka-Munk theory. The rate kappa LR of formation of lumirubin has been evaluated for the case of a four-layer skin and for monochromatic and narrow-band coloured fluorescent lamps. The effects of long-wavelength increase in phi LR, decrease in phi ZE and skin optical losses all combine to shift significantly the optimal rate of formation of lumirubin towards the green. These results suggest that a significant improvement in phototherapy might be obtained with the introduction of new lamps emitting in the blue-green spectral region between 490 and 510 nm.

Evaluation of the quantum yield for E-->Z isomerization of bilirubin bound to human serum albumin. Evidence of internal conversion processes competing with configurational photoisomerization

The quantum yield (phi zz) for the E--> photoisomerization of bilirubin (BR) bound to human serum albumin (HSA) under laser light irradiation has been evaluated with absorbance spectroscopy implemented by the high-performance liquid chromatography technique. The value at 458 nm is about 0.2, four times smaller than the value previously reported in the literature. As for the Z-->E configurational and the structural photoisomerization processes, the quantum yield varies with excitation wavelength, from 0.23 in the blue to 0.16 in the green. The sum of the quantum yields of the Z<-->E configurational photoisomerization reactions is 0.2-0.3 in the blue-green spectral region, thus giving direct evidence of the existence of internal conversion processes of the BR-HSA complex which compete with configurational photoisomerization. An evaluation of the quantum yields of the Z<-->E reactions for filtered broad-band light excitation in the 390-530 nm spectral region from data already published in the literature is also reported. Good agreement with our data of the quantum yields phi ZZ and phi ZE is found, despite the diversity of the experimental procedures in the two cases.

New trends in photobiology (invited review). Recent advances in bilirubin photophysics

The high photosensibility of bilirubin represents the main reason why neonatal jaundice is treated with light. The relaxation processes of excited bilirubin appear to be governed by complex mechanisms which are not fully known. At room temperature, radiationless channels, such as Z----E configurational photoisomerizations and internal conversions, contribute mostly to the deactivation of singlet excited bilirubin. Exciton coupling processes between the two pyrromethenone chromophores constituting the molecule may play a crucial role in bilirubin photophysics.

UV excitable fluorescence of lumirubin

Solutions of bilirubin in different solvents show negligible fluorescence when they are excited with UV light. After irradiation, blue-violet fluorescence (400-450 nm) is observed, whose intensity and peak position depend on the irradiation time. The isolation of pure lumirubin led us to attribute this fluorescence specifically to this photoisomer, even if small contributions from other photoproducts cannot be ruled out. The emission peak of the pure lumirubin in aqueous solution (phosphate buffer) is at 415 nm while the main excitation peak is at 315 nm. Finally, an interpretation of the observed fluorescence is proposed on the basis of the similarity of the present results with previously reported experimental data.

Green light phototherapy in newborn infants with ABO hemolytic disease

The efficacy of fluorescent green light phototherapy was compared with that of blue light phototherapy in the treatment of full-term infants with hemolytic disease and jaundice caused by ABO incompatibility. The efficacy of the treatment was expressed as actual (milligrams per hour) and quantum (milligrams per hour per square centimeter per megawatt) efficiency, taking into account the differential emission of energy from the green versus the blue fluorescent tubes. No statistically significant difference in the rate of serum bilirubin photodegradation was found between the two groups after treatment for 84.6 +/- 14.1 hours versus 81.5 +/- 14.2 hours with the green and the blue phototherapy, respectively. These results, coupled with the known effects of the blue light on the genetic apparatus of mammalian cells, support the application of the green light phototherapy for the treatment of neonatal hyperbilirubinemia caused by ABO incompatibility.

Phototherapy for neonatal jaundice: clinical equivalence of fluorescent green and "special" blue lamps

The relative efficacy of fluorescent green (Sylvania F20T12/G) and "special" blue (Westinghouse F20T12/BB) lamps in the phototherapy of jaundiced neonates was investigated. Two groups of low birth weight infants with a mean gestational age of 35 weeks and mean birth weight of 1930 gm, who developed hyperbilirubinemia within the first 5 days of life, were given green or blue lamp phototherapy under the same irradiation conditions. No statistically significant difference in plasma bilirubin concentrations was found between the two groups after 24 or 48 hours of treatment. Because recent measurements indicate that green lamps are much less efficient than special blue lamps for the production of Z, E isomers of bilirubin in vitro and in vivo, the clinical equivalence of these two types of lamps seems to support the hypothesis that production of structural photoisomers of bilirubin is the main mechanism of phototherapy in humans. Therefore, fluorescent green lamps provide an alternative to special blue lamps for treatment of neonatal hyperbilirubinemia.