This was an open-label multi-center randomized controlled trial conducted in four tertiary care neonatal units across India, from November 2007 to July 2008. The study protocol was approved by institutional ethics committees of all the four hospitals and the study was registered with Clinical Trial Registry of India. A written informed consent was obtained from one of the parents before enrolment.

Subjects: Newborn infants born at 35 or more completed weeks of gestation were eligible for enrolment, if they developed hyperbilirubinemia needing phototherapy within first 7 days of life. The decision to start phototherapy was made by bedside physicians on the basis of the age of the baby in hours and STB levels, as per American Academy of Pediatrics guidelines(3). Phototherapy was stopped when two consecutive STB levels, measured 6 hours apart were less than 15 mg/dL. Infants with perinatal asphyxia (Apgar score <4 at 1 minute or <7 at 5 minute), onset of jaundice within 24 h of age, evidence of hemolysis (positive direct Coombs test), rhesus hemolytic disease, culture-positive or clinical sepsis, need for exchange transfusion at the time of enrolment, and major congenital malformations were excluded.

Outcome variables: The duration of phototherapy was the primary outcome. It was calculated by subtracting age at start of phototherapy from age at end of phototherapy in hours. Brief periods of discontinuation of phototherapy for feeding the baby or changing nappy were not excluded while calculating total duration of phototherapy. The secondary outcomes were failure of phototherapy, rate of fall of STB and occurrence of hypothermia. ‘Failure of phototherapy’ was defined as STB rising or becoming more than 20 mg/dL during phototherapy, which required either use of double surface phototherapy or exchange transfusion. ‘Rate of fall of STB’ was calculated by dividing the difference between STB at start and end of phototherapy with duration of phototherapy.

Data collection and monitoring: Clinical monitoring was done for side effects of phototherapy like dehydration and skin rash; and a 4-hourly axillary temperature measurement was done to detect episodes of hypothermia or hyperthermia. STB was measured every 6 to 8 h using bilirubinometers based on direct spectrophotometery (Twin-beam micro-bilimeter, Ginevri Technologie Biomediche; Italy or Unibeam microbilimeter, Ginevri Technologie Biomediche, Italy; or Bil-100, Cosmo Medical, Korea; coefficient of variation 1 to 3%, range of bilirubin measurement 0-30 to 0-40 mg/dL). Apart from daily internal calibration, the bilirubinometers were checked 3-monthly against low and high bilirubin standards supplied by the respective manufacturers. However, bilirubinometers used at different centers were not compared against each other on the same sample in the same laboratory. In one center (PGIMER,Chandigarh), the irradiance of the phototherapy units at the surface of the babies was checked at the level of face, xiphoid and knees by a photoradiometer (Fluoro-lite 451, Minolta/Air Shields, USA) at the initiation of phototherapy, and then once a day for all babies. In the other 3 centers, only periodic checks of irradiance were done for monitoring and their data was not included for calculating the spectral irradiance.

Sample size: In a previous unpublished trial using CFT phototherapy units in non-hemolytic jaundice, the mean duration of phototherapy was 25.3±14 h. To prove, with 80% power and alpha of 0.05, that duration of phototherapy with LED phototherapy unit is not different by more than 6h, we needed to enroll 125 subjects in each group (Power and Precision software ver 2.0, Biostat Inc., USA).

Statistical analysis: Data entry and analysis were done using Epi Info (CDC, Atlanta). Continuous data with normal distribution was analyzed by student t-test and non-normally distributed data by Mann Whitney U test. Categorical data was analyzed by chi-square or Fisher exact test. A P value of <0.05 was taken as significant. Analysis was intention-to-treat. It was decided a priori that LED will be considered equally efficacious if the difference in duration of phototherapy between the two groups is less than 6 hours without any increase in adverse effects.

The study enrollment is depicted in Fig.1. The mean birthweight and gestation of enrolled and excluded neonates were comparable. There were 39, 37, 35 and 31 neonates in LED and 39, 35, 34 and 22 neonates in CFT groups at AIIMS New Delhi, FH Hyderabad, PGIMER Chandigarh and CSMMU Lucknow, respectively. There were 9 exceptions to the protocol. Three neonates were enrolled despite low Apgar score (but had no hypoxic ischemic encephalopathy) and six despite the presence of sepsis. The results with and without these cases were comparable and analysis reported here includes these 9 subjects.

The birthweight, gestation and other demographic variables were similar in the neonates enrolled in LED or CFT groups (Table I). The infant characteristics and laboratory parameters at the start and during phototherapy, which may have impact on the duration of phototherapy were also comparable in the two groups, except for spectral irradiance which was higher in LED group as compared to CFT group (Table II). One neonate in LED group received phenobarbitone while 1 in CFT group received extra intravenous fluids. The STB at the time of stopping phototherapy was similar in the two groups thereby indicating that uniform guidelines were followed.

Table I



Description of Study Population

The duration of phototherapy was normally distributed in the LED group, but was skewed positively in the CFT group (P <0.001; D’Agostino and Balanger test for normality) due to 7 cases in this group needing phototherapy for more than 60 hours (Fig. 2). The median duration of phototherapy was comparable in the two groups [26 h, (IQR: 22-36) versus 25 h, (IQR: 22-36); P=0.44]. At any time point, a similar proportion of neonates were under phototherapy in the two groups (Fig. 3), (log-rank test, P=0.38).

Fig. 2 Duration of phototherapy.

Fig. 3 Survival analysis: Duration of phototherapy.

Two hundred and seventy two healthy term and late preterm neonates with non-hemolytic hyperbili-rubinemia were randomized to receive LED or CFT phototherapy, across four neonatal units. The efficacy of both types of phototherapy devices in terms of duration of phototherapy, rate of fall of STB, incidence of ‘failure of phototherapy’ and need for exchange transfusion was similar in this trial.

Since we were testing a newly introduced LED system, we enrolled a relatively ‘low-risk’ population of healthy neonates with non-hemolytic jaundice. Whether LED units will be similarly effective in hemolytic jaundice is not known. The efficacy of a phototherapy system is influenced by initial STB levels, body surface area exposed and spectral irradiance(3). We kept LED light source at a relatively large distance from the body surface to match the distance achievable with CFT photo-therapy. As LED light sources do not produce much infrared light, they can be brought much closer to the baby with potential increase in efficacy, without danger of hyperthermia or burns.

The side effects like hypothermia and hyperthermia were rare and comparable in the two groups. This may be partly because the enrolled neonates were treated in temperature controlled environments with regular monitoring of body temperature. Since LEDs do not produce much heat, hypothermia may be a problem when used in small and sick babies, and in environments without temperature control. In such situations, a closer monitoring and external heat source may be required.

In different centers, STB was measured by three different bilirubinometer models from two manu-facturers. It was not feasible to compare the bilirubinometers used at different centers against each other by running the same sample. However, all centers followed regular internal calibrations and periodic checking against known standards. The irradiance could not be measured regularly at all centers. However, that should not affect the results since a strategy of keeping ‘similar distance’ for the two devices was being followed rather than ‘similar irradiance’.

Although phototherapy has been used for the treatment of neonatal hyperbilirubinemia for more than four decades, the most efficacious method with least side effects is yet to be developed. There is a need to conduct actual cost-effectiveness studies using different phototherapy devices. Studies are also required to compare different types of LED devices. For better comparison, future studies should record not only the distance and irradiance but also the body surface area covered by them. The effects of LEDs on conversion of bilirubin to various bilirubin isomers also needs to be studied in vitro and in vivo.

Contributors: PK, AKD, GKM, SM and DC designed the study. KN, SS, VS and SNS recruited the subjects and collected the data. PK, AKD, GKM, PG and SM monitored the patient recruitment and data collection. DC and PK analyzed data and wrote the manuscript with inputs from AKD, GKM and SM. All authors reviewed the final manuscript and made the decision to submit the manuscript for publication.

Financial assistance: None. The prototype LED phototherapy units at all sites were provided by Srichakra Scientifics, Hyderabad, free of cost. CFL unit at AIIMS, New Delhi, was provided by Phoenix Medical Systems, Chennai, free of cost.

Conflict of interest: None stated.

1. National Neonatology Forum of India. National Neonatal Perinatal Database Network. Report 2002-2003. New Delhi: 2004.

2. Ennever JF. Blue light, green light, white light, more light: treatment of neonatal jaundice. Clin Perinatol 1990; 17: 467-481.

3. American Academy of Pediatrics. Subcommittee on Hyperbilrubinemia. Management of hyper-bilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004; 114: 297-316.

4. Seidman DS, Moise J, Ergaz Z , Laor A, Vreman HJ, Stevenson DK , et al. A new blue light-emitting phototherapy device: a prospective randomized controlled study. J Pediatr 2000; 136: 771-774.

5. Vreman HJ, Wong RJ, Stevenson DK, Route RK, Reader SD, Fejer MM, et al. Light-emitting diodes: a novel light source for phototherapy. Pediatr Res 1998; 44: 804-809.

6. Martins BM, de Carvalho M, Moreira ME, Lopes JM. Efficacy of new microprocessed photo- therapy system with five high intensity light emitting diodes (Super LED). J Pediatr (Rio J) 2007; 83: 253-258.

7. Maisels MJ, Kring EA, DeRidder J. Randomized controlled trial of light-emitting diode photo-therapy. J Perinatol 2007; 27: 565-567.

8. Seidman DS, Moise J, Ergaz Z, Laor A, Vreman HJ, Stevenson DK, et al. A prospective randomized controlled study of phototherapy using blue and blue-green light-emitting devices, and conventio-nal halogen-quartz phototherapy. J Perinatol 2003; 23: 123-127.

9. Chang YS, Hwang JH, Kwon HN, Choi CW, Ko SY, Park WS, et al. In vitro and in vivo efficacy of new blue light emitting diode phototherapy compared to conventional halogen quartz photo-therapy for neonatal jaundice. J Korean Med Sci 2005; 20: 61-64.

10. Research Randomizer. Available from: URL: http://randomizer.org/form.htm. Accessed October 24, 2007.