Birri Lubina 2

Original Articleped_3353 689..693

Management of phototherapy for neonatal hyperbilirubinemia: Is anew radiometer applicable for all wavelengths and light source types?

Toru Kuboi,1 Takashi Kusaka,1 Saneyuki Yasuda,1 Kensuke Okubo,2 Kenichi Isobe2 and Susumu Itoh2

1Maternal Perinatal Center, Faculty of Medicine and 2Department of Pediatrics, Faculty of Medicine, Kagawa University,Kagawa, Japan

Abstract Background: To evaluate the clinical effects of phototherapy for neonatal hyperbilirubinemia, it is necessary to measurethe rate of cyclobilirubin production, which represents the main photochemical pathway of bilirubin metabolism. Sincethe Atom Phototherapy Analyzer can be used to calculate the theoretical relative light energy of irradiance as a meansof assessing the cyclobilirubin production rate for each wavelength spectrum, the clinical effect of phototherapy can beevaluated regardless of the light source type. Using the Atom Phototherapy Analyzer, the correlation between theirradiance of various light sources with different peak wavelengths and the rate of cyclobilirubin production wasinvestigated in vitro. We also investigated the utility of green LED in vitro.Methods: A bilirubin-albumin complex solution was prepared, poured into tubes, and irradiated using various lightsources. All light sources used were bed-type phototherapy devices; that is, green and blue LED and green and bluefluorescence tubes. The concentrations of photoisomers were measured after irradiation and compared with the irradi-ance of the light sources.Results: The irradiance measured by the Atom Phototherapy Analyzer decreased in the following order: blue fluores-cence tube > green LED > blue LED > green fluorescence tube. The cyclobilirubin production rates and irradiance valuesof the light sources were significantly positively correlated (R2 = 0.93, P < 0.05).Conclusion: Our data indicate that the Atom Phototherapy Analyzer can be used to objectively evaluate the effects ofphototherapy using various light sources. Further, the effects of green LED were similar to those of other light sourcesin vitro.

Key words cyclobilirubin production rate, neonatal hyperbilirubinemia, neonatal jaundice, new radiometer, theoretical relative lightenergy.

More than 50 years have passed since the introduction of photo-therapy for neonatal hyperbilirubinemia, but the most efficientphototherapeutic method that induces the minimum adverseeffects is still being investigated. The maximum absorption wave-length of bilirubin is 458 nm, and blue light, which encompassesthis wavelength, is generally used as a light source for photo-therapy. Phototherapy for neonatal hyperbilirubinemia affectsthree different photochemical mechanisms of bilirubin alteration:(i) photooxidation; (ii) photoaddition; and (iii) photoisomeriza-tion. However, the levels of photooxidation, photooxygenation,and photoaddition products were very low in neonatal serum andurine during neonatal jaundice phototherapy.1,2

The photoisomerization induced by phototherapy is con-figurational ((ZE)-/(EZ)-/(EE)-bilirubin) or structural ((EZ)-/(EE)-cyclobilirubin).

The main mechanism responsible for abrogating neonatalhyperbilirubinemia in phototherapy is the production and excre-tion of structural photoisomers in bile and urine.3

The production of cyclobilirubin per unit of light energyvaries among irradiated wavelengths, and it is most efficientlyproduced at 500–520 nm, which belongs to the green wavelengthzone.4 Accordingly, the effects of phototherapy using lightsources with various peak wavelengths can be evaluated in vitroby measuring the cyclobilirubin production rate.5,6

Although the irradiances of light sources have been measuredusing radiometers, such as the Biliblanket Meter (OhmedaMedical, Laurel, MD, USA) and Minolta-Air Shields Fluoro-Litemeter 451 (Minolta-Air Shields, Osaka, Japan), their receiverproperties were adjusted to the bilirubin absorption wavelength,and a marked variation in irradiance was detected amongdifferences light sources. Therefore, it is necessary to be able toevaluate irradiance in order to predict the clinical effects of pho-totherapy, rather than the intensity of a single wavelength.

In the present study, we investigated the utility of a new lightmeter (the Atom Phototherapy Analyzer) and a green LED pho-totherapy light (LF-111) in vitro.

Correspondence: Toru Kuboi, MD, Maternal Perinatal Center, Facultyof Medicine, Kagawa University, 1750-1 Ikenobe, Miki, Kita-gun,Kagawa 761-0793, Japan. Email: [email protected]

Received 20 June 2010; revised 9 February 2011; accepted 14February 2011.

Pediatrics International (2011) 53, 689–693 doi: 10.1111/j.1442-200X.2011.03353.x

© 2011 The AuthorsPediatrics International © 2011 Japan Pediatric Society

Methods

To evaluate the efficacy of phototherapy in vitro, light was used toirradiate bilirubin-human serum albumin complex solution, andthe production of bilirubin photoisomers was studied. TheLF-111, which utilizes a green LED light (Toitu, Tokyo, Japan);the neoBLUE cozy (Natus, San Carlos, CA, USA), which utilizesa blue LED light; the Bilibed (Medela, Baar, Switzerland), whichutilizes a blue-white light (BW FL20S, 20 W; Toshiba, Tokyo,Japan); and the Lightbed (Toitu, Tokyo, Japan), which utilizes agreen light (GR FL20S, 18 W; Toitu, Tokyo, Japan), were used asbed-type phototherapy devices. The light source spectrum ofeach device was measured using an MCPD-100 (Union Giken,Osaka, Japan). The mat pad and cloth attached to each lightsource were the same as those used in the clinical setting. Thesubsequent experiments were carried out in a dark room. A2 g/dL-solution of human serum albumin (HSA) (Sigma, StLouis, MO, USA) was prepared by dissolving HSA in 0.1 Mphosphate buffer (pH 7.4). Then, bilirubin (Tokyo Kasei, Tokyo,Japan) was dissolved in 0.1 N sodium hydroxide and mixed withHSA solution to produce 10 mg/dL bilirubin-HSA complex solu-tion. Then, 1 mL of bilirubin-HSA complex solution was pouredinto a Pyrex tube. One tube was placed on a centrally locatedstirrer in each phototherapy unit. Light was irradiated from thebase of the bed, and samples were taken at 0, 5, 10, and 15 minafter irradiation. The mean light energy at the light irradiationsurface was measured using the Atom Phototherapy Analyzer.The measurements were performed three times for each photo-therapy device, and the mean values were calculated. Theconcentrations of bilirubin photoisomers in the samples weremeasured using the high-performance liquid chromatography(HPLC) method, as described by Itoh et al.7 In brief, a gradientelution technique was employed using acetonitrile-0.01 M phos-phate buffer (pH 5.5)-dimethyl formamide (5:30:65 v/v/v) as aprimary eluent and acetonitrile-0.01 M phosphate buffer (pH5.5)-dimethyl formamide (20:15:65 v/v/v) as a secondary eluent.Each sample was mixed with an equal volume of dimethyl sul-

foxide and acetonitrile and then centrifuged. Then, 25 mL of thesupernatant was injected into an HPLC column. The flow ratewas 1 mL/min, and the measured wavelength was 445 nm. HPLCanalysis of the bilirubin/HSA complex solution after irradiationshowed that the peaks for (ZZ)-bilirubin IIIa, IXa, and XIIIa andtheir photoisomers were well separated. We evaluated the rate ofcyclobilirubin production. The concentration of cyclobilirubinwas calculated as the sum of the (EE)-cyclobilirubin IXa and(EZ)-cyclobilirubin IXa concentrations.

Principle behind the Atom Phototherapy Analyzer

Irradiance represents light energy. Cyclobilirubin produces irra-diance at 400–520 nm, and the light energy of a light source inthis spectrum can be determined from the relative light energydistribution using a spectrum measurement device (Fig. 1).However, cyclobilirubin production is dependent on the wave-length, and cyclobilirubin production per unit energy variesamong wavelengths.4 Thus, the 400–520 nm spectrum of therelative light energy distribution was divided into three 40 nmwavelength zones, and the light energy of each zone (area inFig. 1) was multiplied by its cyclobilirubin production per unitenergy. The total of the three zones was defined as the “theoreti-cal relative energy = irradiance” because this corresponds to thelight energy involved in cyclobilirubin production, allowing theeffect of phototherapy to be judged regardless of the peak wave-length of the light source.5

This can be explained as follows. The relative light energy ata given wavelength is designated as El. The light energy distri-bution at 400–520 nm measured using the MPCD-100 wasdivided into 40 nm regions (A-C), and the area of each regionwas determined: E400–440 = A, E440–480 = B, and E480–520 = C. Theresultant values were then multiplied by Kl as shown below,and their total was defined as the “theoretical relative light

energy E K d= ⋅∫ λ λ λ400

520”.

Wavelength (nm)

A B C

Rel

ativ

e lig

ht e

nerg

y

Fig. 1 Relative light energy distribution.

690 T Kuboi et al.


Kl was determined by setting an identical light energy for allwavelengths and setting the amount of (EZ)-cyclobilirubin pro-duced by 80-minute irradiation at 400–440 nm to 1 as follows:E400–440 = 1.0, E440–480 = 1.9, and E480–520 = 2.8.

Accordingly, “theoretical relative light energy” = 1.0 ¥ A +1.9 ¥ B + 2.8 ¥ C.

Statistical analysis

Statistical analyses were performed with Microsoft Excel(Microsoft, Redmond, WA, USA) and Statcel 2 (OMS publish-ing, Saitama, Japan). Correlations were regarded as significantwhen P < 0.05.

Results

The light source spectrum of each device is shown in Figure 2.The mean light energy on the light irradiation surface as mea-sured by the Atom Phototherapy Analyzer was 156, 97.0, 189,and 41.1 mW/cm2 per nm for the LF-111, neoBLUE cozy,Bilibed, and Lightbed, respectively. Figure 3 shows the cyclobi-lirubin production (mean 1 SD) with the duration of irradiation

Fig. 2 Emission spectra of phototherapy devices. The spectra of (a) Bilibed (a blue-white fluorescent light), (b) LF-111 (a green LED light),(c) neoBLUE cozy (a blue LED light), and (d) Lightbed (a green fluorescent light).

Fig. 3 Cyclobilirubin production during irradiation (mean 1 SD).(�) Bilibed; ( ) LF-111; (�) neoBLUE cozy; (�) Lightbed.

New radiometer for phototherapy 691


by light source. The cyclobilirubin production measured every5 min increased mostly linearly regardless of the light source.The mean production every 15-min irradiation with the lightsources decreased in the order of bilibed (4.7 mg/dL), LF-111(3.3 mg/dL), neoBLUE cozy (2.7 mg/dL), and Lightbed (1.5 mg/dL). Figure 4 shows the correlation between the irradiance mea-sured using the Atom Phototherapy Analyzer and rate ofcyclobilirubin production. Irradiation by each light source wassignificantly correlated with the rate of cyclobilirubin production(R2 = 0.93, P = 0.006).

Discussion

New light sources for efficient phototherapy for neonatal hyper-bilirubinemia, such as LED, have recently been developed anddemonstrated to be effective in many clinical reports.8,9 To evalu-ate the effects of phototherapy using different light source wave-lengths and types, measurement of the rate of cyclobilirubinproduction is necessary, but it has only been investigated in a fewreports.10,11 The Atom Phototherapy Analyzer does not merelymeasure the irradiance, but rather estimates cyclobilirubin pro-duction based on the theoretical relative light energy withoutHPLC measurement. In this study, the rate of cyclobilirubinproduction was significantly correlated with irradiance, as mea-sured by the Atom Phototherapy Analyzer, suggesting that theclinical effects of various phototherapy lights can be measuredobjectively and simply using this device. In very-low-birth-weight neonates with a low bilirubin excretion capacity, whencyclobilirubin production exceeds its excretion, the risk of bronzebaby syndrome increases.12 In addition to being able to evaluatethe effects of phototherapy, this new device may serve as anobjective index for selecting a light source that does not lead tothe excessive production of cyclobilirubin, thereby reducing therisk of toxicity.

The guidelines proposed by the American Academy of Pedi-atrics in 2004 specify the use of a light source that is capable of

producing irradiation at a level greater than 30 mW/cm2/nm,which highlights the importance of measuring irradiance.13

However, the guidelines only specify values for irradiation at430–490 nm using blue-white light and do not describe inten-sive phototherapy with green light, which has a much lowerirradiance value. Blue-white light has been reported to be themost effective for treating neonatal hyperbilirubinemia becauseits peak wavelength is close to the bilirubin absorption wave-length. The studies performed by us and Ennever et al. revealedthat the production and excretion of cyclobilirubin were themain pathways of the photochemical metabolism of bilirubin inphototherapy for neonatal hyperbilirubinemia, and the mosteffective wavelength for producing (EZ)-cyclobilirubin wasfound to be 500–520 nm using green light.3,4,14,15 The irradiance,that is, the rate of cyclobilirubin production of bilibed,decreased in the order of LF-111, neoBLUE cozy, and Light-bed. Cyclobilirubin is most efficiently produced in the 500–520 nm wavelength range, but this does not mean that nocyclobilirubin is produced in the 400–500 nm range (seeFig. 1). As shown in Figure 2, the peak wavelength variedamong the light sources, but the spectra were present in the400–520-nm range. This may have been the reason forthe finding that the rate of cyclobilirubin production in thepresence of green LED light, which showed a high cyclobiliru-bin production efficiency, was the second highest after that ofbilibed.

The wavelength of blue-light has been reported to causeadverse events, such as light-induced DNA damage and reactiveoxygen production against an endogenous photosensitizer(vitamin B2).16,17 Increases in DNA chain cleavage, the sisterchromatid exchange rate, and the rate of the intracellular uptakeof nucleic acid thymidine labels and a decrease in the cell sur-vival rate have been reported as the cytotoxic effects of blue-white light.18,19 The latent toxicity caused by phototherapy forneonatal hyperbilirubinemia can not be separated from the clini-cal effects of light sources until the photochemical mechanism bywhich bilirubin is broken down during phototherapy has beenfully elucidated. However, green LED light does not producelatent toxicity during phototherapy, and the effects of green LEDlight, as evaluated in vitro by measuring the cyclobilirubin pro-duction rate, were similar to those of other bed-type photothera-pies. If green light with a peak limited to the effective wavelengthrange can be prepared using LED, and irradiance, that is, the rateof cyclobilirubin production, of all types of light sources can besimply and objectively evaluated using the Atom PhototherapyAnalyzer, green light may be described in the guidelines in thefuture.

Conclusion

Our data indicated that irradiance with different wavelengths andtypes of light source, as measured using the Atom PhototherapyAnalyzer, were significantly positively correlated with the rate ofcyclobilirubin production, which represents the effects of photo-therapy. The effects of green LED light, as evaluated in vitro bymeasuring the cyclobilirubin production rate, were similar tothose of other bed-type phototherapies.

Fig. 4 Correlation between the irradiance measured by the AtomPhototherapy Analyzer and the rate of cyclobilirubin production. (�)Bilibed; ( ) LF-111; (�) neoBLUE cozy; (�) Lightbed.

692 T Kuboi et al.


References

1 Lightner DA, Linnane WP 3rd, Ahlfors CE. Bilirubin photooxida-tion products in the urine of jaundice neonates receiving photo-therapy. Pediatr. Res. 1984; 18: 696–700.

2 Jori G, Rossi E, Rubaltelli FF. Phototherapy-induced covalentbinding of bilirubin to serum albumin. Pediatr. Res. 1980; 14:1363–6.

3 Onishi S, Isobe K, Itoh S et al. Metabolism of bilirubin and itsphotoisomers in newborn infants during phototherapy. J. Biochem.1986; 100: 789–95.

4 Itoh S, Onishi S, Isobe K, Manabe M, Yamakawa T. Wavelengthdependence of the geometric and structural photoisomerization ofbilirubin bound to human serum albumin. Biol. Neonate. 1987; 51:10–7.

5 Yaguchi Y, Itoh S, Isobe K, Onishi S. Phototherapy for neonatalhyperbilirubinemia comparison of various forms of phototherapylight. J. Jpn. Soc. Premature Newborn Med. 1997; 9: 17–26.

6 Okada H, Abe T, Etoh Y et al. In vitro production of bilirubinphotoisomers by light irradiation using neoBLUE. Pediatr. Int.2007; 49: 318–21.

7 Itoh S, Isobe K, Onishi S. Accurate and sensitive high-performanceliquid chromatographic method for geometrical and structural pho-toisomers of bilirubin IXa using the relative molar absorptivityvalues. J. Chromatogr. A 1999; 848: 169–77.

8 Seidman DS, Moise J, Ergaz Z et al. A new blue light-emittingphototherapy device: A prospective randomized controlled study.J. Pediatr. 2000; 136: 771–4.

9 Seidman DS, Moise J, Ergaz Z et al. A prospective randomizedcontrolled study of phototherapy using blue and blue-green light-emitting devices, and conventional halogen-quartz phototherapy. J.Perinatol. 2003; 23: 123–7.

10 Yasuda S, Itoh S, Imai T, Isobe K, Onishi S. Cyclobilirubin for-mation by in vitro photoirradiation with neonatal phototherapylight. Pediatr. Int. 2001; 43: 270–75.

11 Verman HJ, Wong RJ, Murdock JR, Stevenson DK. Standardizedbench method for evaluating the efficacy of phototherapy devices.Acta Paediatrica. 2008; 97: 308–16.

12 Okada H, Masuya K, Yasuda S et al. Developmental changes inserum half-life of (EZ)-cyclobilirubin. Early Hum. Dev. 2005; 81:619–22.

13 AAP Clinical Practice Guideline, Subcommittee on Hyperbiliru-binemia. Management of hyperbilirubinemia in the newborninfant 35 or more weeks of gestation. Pediatrics 2004; 114: 297–316.

14 Ennever JF, Knox I, Denne SC, Speck WT. Phototherapy forneonatal jaundice: In vivo clearance of bilirubin photoproducts.Pediatr. Res. 1985; 19: 205–8.

15 Ennever JF, Dresing TJ. Quantum yields for the cyclization andconfigurational isomerization of 4E,15Z-bilirubin. Photochem.Photobiol. 1991; 53: 25–32.

16 Sanvordeker DR, Kostenbauder HB. Mechanism for riboflavinenhancement of bilirubin photodecomposition in vitro. J. Pharm.Sci. 1974; 63: 404–8.

17 Ando M, Isobe K, Itoh S, Onishi S. Pathophysiological signifi-cance of neonatal jaundice in defense system against active oxygenin the neonatal period: A comparison of bilirubin photochemicalreactions of blue-white light and green light in the presence offlavin mononucleotide. J. Jpn. Soc. Premature Newborn Med.1990; 2: 96–103.

18 Speck WT, Rosenkranz HS. Base substitution mutations inducedin Salmonella strains by visible light (450 nm). Photochem. Pho-tobiol. 1975; 21: 369–71.

19 Sideris EG, Papageorgiou GC, Charalampous SC et al. A spectrumresponse study on single strand DNA breaks, sister chromatidexchanges, and lethality induced by phototherapy light. Pediatr.Res. 1981; 15: 1019–23.

New radiometer for phototherapy 693


Copyright of Pediatrics International is the property of Wiley-Blackwell and its content may not be copied or

emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.

However, users may print, download, or email articles for individual use.

Documents

Birri Lubina 2