Original Article
Broad-Spectrum Light versus Blue Light for Phototherapy in Neonatal Hyperbilirubinemia: A Randomized Controlled Trial Simone Pratesi, MD1 Sandra Di Fabio, MD2 Shahar Bar, MD1 Carlo Dani, MD1
Cecilia Bresci, MD1
1 Department of Neurosciences, Psychology, Drug Research, and
Children’s Health, University of Florence, Florence, Italy 2 Department of Neonatal Intensive Care Unit, San Salvatore Hospital, L’Aquila, Italy
Cecilia Di Natale, MD2
Address for correspondence Carlo Dani, MD, Division of Neonatology, Careggi University Hospital, University of Florence School of Medicine, Largo Brambilla, 3 Firenze, Italy (e-mail: cdani@unifi.It).
Abstract
Keywords
► neonatal hyperbilirubinemia ► phototherapy ► jaundice ► infant
Phototherapy is standard care for treatment of neonatal hyperbilirubinemia. Our aim was to compare the effectiveness of broad-spectrum light (BSL) to that of blue light emitting diodes (LED) phototherapy for the treatment of jaundiced late preterm and term infants. Infants with gestational age from 35þ0 to 41þ6 weeks of gestation and nonhemolytic hyperbilirubinemia were randomized to treatment with BSL phototherapy or blue LED phototherapy. A total of 20 infants were included in the blue LED phototherapy group and 20 in the BSL phototherapy group. The duration of phototherapy was lower in the BSL than in the blue LED phototherapy group (15.8 4.9 vs. 20.6 6.0 hours; p ¼ 0.009), and infants in the former group had a lower probability (p ¼ 0.015) of remaining in phototherapy than infants in the latter. We concluded that BSL phototherapy is more effective than blue LED phototherapy for the treatment of hyperbilirubinemia in late preterm and term infants. Our data suggest that these results are not due to the different irradiance of the two phototherapy systems, but probably depend on their different peak light emissions.
Hyperbilirubinemia is the most frequent clinical problem neonatologists must deal with during the newborn period because under certain circumstances high levels of total serum bilirubin (TSB) may be potentially toxic to the central nervous system, even in healthy term newborns. Sufficiently elevated levels of bilirubin can lead to bilirubin encephalopathy and subsequently kernicterus, with devastating, permanent neurodevelopmental handicaps.1 Phototherapy is the first-line treatment for neonatal hyperbilirubinemia.2 Its efficacy depends on the absorption of light photons by the bilirubin molecules which converts the toxic Z,Z-bilirubin molecules into more easily excretable compounds (photoisomers) such as the configurational isomers Z,E-bilirubin and E,Z-bilirubin, and the structural isomers Z-lumirubin and E-lumirubin.3 Only light of certain
colors or wavelengths can be absorbed by bilirubin. Since bilirubin is a yellow pigment, blue and green light is best absorbed,4 and when broad-spectrum white light is used for treating only a fraction of the light is acting on the bilirubin.5 Moreover bilirubin does not absorb the different colors of light equally, and blue light at approximately 460 nm is absorbed more readily, while the green light is theoretically less well absorbed.4 However, the effectiveness of phototherapy depends also on the light’s capacity to penetrate the newborn’s skin, and longer wavelength light, such as green light, is expected to penetrate the infant’s skin more deeply.6 In fact, it has been demonstrated that longer than blues wavelength turquoise spectral band is more efficient than the blue spectral band in the clearance of bilirubin.3
received April 30, 2014 accepted after revision October 8, 2014
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1396685. ISSN 0735-1631.
Downloaded by: Florida International University. Copyrighted material.
Am J Perinatol
Pratesi et al.
Thus, we hypothesized that broad-spectrum light (BSL) phototherapy (λ ¼ 420–680 nm) is more effective than narrow-spectrum blue light phototherapy (λ ¼ 400–500 nm) in decreasing TSB concentration in late preterm and term infants. To assess this hypothesis, we performed a randomized, controlled study to compare the effectiveness of BSL phototherapy with that of light-emitting diode (LED) blue light phototherapy in the treatment of neonatal jaundice.
Materials and Methods Study Design The study was performed at the neonatal special care unit of the “Careggi University Hospital” of Florence and “San Salvatore Hospital” of L’Aquila, after approval by the local ethics committees. Caucasian healthy infants were eligible for inclusion in the study if they were born between the ages of 35þ0 and 41þ6 weeks at the time of birth, presented with nonhemolytic hyperbilirubinemia. Exclusion criteria were an Rh hemolytic disease, positive direct Coomb test, glucose-6phosphate dehydrogenase deficiency, hemoglobinopathies, clinical signs of hemolysis, and evidence of hemolysis in peripheral smear,7,8 and congenital infections. Written informed parental consent was obtained before infants were enrolled.
with a phototherapy radiometer (PD300-UV, Ophir Photonics, Jerusalem, Israel) with peak sensitivity at 460 to 490 nm. We placed the BSL phototherapy 28 cm above the infant (suggested mean irradiance ¼ 4.05 µW/cm2), while the blue LED phototherapy was placed 30.5 cm above the infant (suggested mean irradiance ¼ 3.50 µW/cm2), as recommended by the manufacturers. We did not adjust the distance between the lights and the infants to provide a similar irradiance so as to conform our results to normal clinical practice, where the distance is suggested by the manufacturers. During the phototherapy all infants remained in bassinettes, unclothed except for a diaper, and with their eyes covered. The TSB was checked every 6 hours during the phototherapy (until 30 hours after the beginning), and 12 and 24 hours after its discontinuation to evidence a possible rebound. The decision to discontinue treatment was made every 6 hours based on the TSB levels. The daily care (i.e., feeding, monitoring, etc.) of enrolled infants was performed according to local protocols.
Outcomes The primary endpoint of our study was the duration of phototherapy. Prespecified secondary endpoints were changes in the TSB and need of phototherapy at established data points (6, 12, 18, 24, 30 hours of treatment), hourly decrease in TSB, and time needed for TSB < 13 mg/dL.
Randomization
Other Collected Data
Infants were assigned to receive either BSL phototherapy (BiliCrystal Duo.3, Charleroi, Belgium) or blue LED phototherapy (NeoBlu, Natus Medical Inc., San Carlos, CA) in a 1:1 ratio in permuted blocks of variable size. Randomization was stratified according to center and gestational age (35þ0–37þ6 and 38þ041þ6 weeks). Group assignment was contained in sequentially numbered, sealed, and opaque envelopes. The decision to start phototherapy was made by clinicians other than the investigators involved in patient care, and researchers assessing study endpoints were blinded to the nature of the study treatments.
We recorded the following data for each infant: gestational age, birth weight, ABO incompatibility, Apgar score at 5 minutes, cephalohematoma, cord blood TSB, polycythemia (hematocrit > 65%), exclusive breastfeeding, TSB value at beginning and end of phototherapy, hourly increase in TSB before the beginning of phototherapy beginning, hourly decrease in TSB during phototherapy, highest TSB values, number of phototherapy courses, mean weight loss from birth, and length of hospital stay.
Study Intervention The TSB was measured in whole blood from a heel prick sample in a blood gas analyzer9 with a spectrophotometer module (ABL 735, Radiometer, Fiske Street Holliston, MA) in infants who presented with visible jaundice. Hematocrit was measured contemporarily. Entry criteria to phototherapy followed the American Academy of Pediatrics recommendations,2 and the decision to use phototherapy was made by the attending pediatrician. Phototherapy was discontinued in infants when the TSB fell below the level at which phototherapy was initiated or fell below 13 mg/dL in two consecutive measurements.2,10 The BSL phototherapy unit has eight compact fluorescent light tubes emitting light at 420 to 680 nm wavelengths (peak emission at 455 and 524 nm), while the blue LED phototherapy unit has diodes emitting blue light at 400 to 500 nm wavelengths (peak emission at 466 nm). On initiation of phototherapy, irradiance was measured at the clinical setting American Journal of Perinatology
Statistical Analysis On the basis of local data, we hypothesized that the conventional BSL phototherapy might decrease the duration of phototherapy from 24 12 to 19.2 2.4 hours (20%) in comparison with blue LED phototherapy. We calculated that we needed to enroll 19 newborns in each group to detect this difference as statistically significant with 90% power at 0.01 level. Data are reported as mean values and standard deviations, median values and ranges, or rates and percentages. We analyzed continuous variables using Student t-test and categorical variables with Fisher exact test. Moreover, the repeated measures analysis of variance test was used for analyzing intragroup and intergroup changes and decrease of TSB after phototherapy beginning. The Kaplan–Meier estimate was used to compare the probability of remaining in phototherapy during the study period in infants in either the BSL or blue LED phototherapy group. To evaluate whether BSL and blue LED devices can affect the phototherapy duration in our patients, this and the other variables that according to univariate analysis were most
Downloaded by: Florida International University. Copyrighted material.
Broad-Spectrum Light Phototherapy
Broad-Spectrum Light Phototherapy
Pratesi et al.
60 Screened for eligibility 20 Not eligible 1 Congenital malformation 1 Rh hemolytic disease 5 Positive Coomb test 7 Consent not given 6 Other reasons
40 Randomized
20 Included in the intention-to-treat analysis
20 Allocated to the blu LED phototherapy group 20 Received allocated treatment
20 Included in the intention-to-treat analysis
Fig. 1 The CONSORT flow diagram.
probably related to the duration of phototherapy and markedly different between the groups (p < 0.10) were included in multiple linear regression analysis to assess their independent role. Effect estimates are expressed as relative risk (RR) with profile likelihood-based 95% confidence limits (Cl). The reported RRs represent the effect of the TSB at phototherapy beginning or > 16 mg/dL, and gestational age < or 37 weeks. These values were decided because they represent the median value in our patients (TSB), or arbitrarily (gestational age).
Results The numbers of infants deemed eligible for the study and the numbers randomly assigned to receive blue LED or BSL phototherapy are shown in ►Fig. 1. We studied 40 infants,
20 of whom were included in the blue LED phototherapy group and 20 in the BSL phototherapy group, whose clinical characteristics are shown in ►Table 1. The average irradiance at the infant’s abdomen was 1.64 μW/cm2 in the BSL phototherapy group and 3.2 μW/cm2 in the blue LED phototherapy group. Infants in the BSL phototherapy group needed shorter treatment than infants in the blue LED phototherapy group (15.8 4.9 vs. 20.6 6.0 hours; p ¼ 0.009); the percentage of infants who still required phototherapy after 12 (45 vs. 80%; p ¼ 0.048), and 18 (15 vs. 50%; p ¼ 0.020) hours of treatment was lower in the BSL than in the blue LED phototherapy group. Changes (p < 0.0001) and decrease (BSL group: p ¼ 0.039; blue Led group: p ¼ 0.009) of TSB after phototherapy beginning were statistically significant in both the group. Patients in the BSL phototherapy group had a lower
Table 1 Clinical characteristics of studied infants (mean [SD], or median and [range], or rate and [%]) Broad-spectrum light phototherapy (N ¼ 20)
Blue LED phototherapy (N ¼ 20)
p-Value
38 2
39 1
1.000
6 (30)
1 (5)
0.096
Birth weight (g)
2,950 630
3,170 500
0.229
ABO incompatibility
8 (40)
7 (35)
1.000
Gestational age (wk) 37 wk of gestation
Apgar score at 5 min
9 (9–10)
9 (9–10)
1.000
Cephalohematoma
1 (5)
0
1.000
Cord TSB (mg/dL)
1.1 0.5
1.3 0.4
0.171
Hematocrit > 65%
6 (30)
5 (25)
1.000
Exclusive breastfeeding
9 (45)
9 (45)
1.000
Weight loss (%)
6.6 3.1
6.2 2.7
0.666
Length of hospital stay (d)
4.1 2.7
3.8 2.1
0.697
Abbreviations: LED, light emitting diode; TSB, total serum bilirubin. American Journal of Perinatology
Downloaded by: Florida International University. Copyrighted material.
20 Allocated to the BSL phototherapy group 20 Received allocated treatment
Broad-Spectrum Light Phototherapy
Pratesi et al.
Table 2 Primary and secondary outcomes (mean [SD], or rate and [%]) Broad-spectrum light phototherapy (N ¼ 20)
Blue LED phototherapy (N ¼ 20)
p-Value
15.8 4.9
20.6 6.0
0.009
6 h of phototherapy
13.1 1.7
14.0 2.9
0.205
12 h of phototherapy
11.7 2.1
12.7 2.5
0.201
18 h of phototherapy
9.8 1.6
12.0 1.6
Broad-Spectrum Light versus Blue Light for Phototherapy in Neonatal Hyperbilirubinemia: A Randomized Controlled Trial Simone Pratesi, MD1 Sandra Di Fabio, MD2 Shahar Bar, MD1 Carlo Dani, MD1
Cecilia Bresci, MD1
1 Department of Neurosciences, Psychology, Drug Research, and
Children’s Health, University of Florence, Florence, Italy 2 Department of Neonatal Intensive Care Unit, San Salvatore Hospital, L’Aquila, Italy
Cecilia Di Natale, MD2
Address for correspondence Carlo Dani, MD, Division of Neonatology, Careggi University Hospital, University of Florence School of Medicine, Largo Brambilla, 3 Firenze, Italy (e-mail: cdani@unifi.It).
Abstract
Keywords
► neonatal hyperbilirubinemia ► phototherapy ► jaundice ► infant
Phototherapy is standard care for treatment of neonatal hyperbilirubinemia. Our aim was to compare the effectiveness of broad-spectrum light (BSL) to that of blue light emitting diodes (LED) phototherapy for the treatment of jaundiced late preterm and term infants. Infants with gestational age from 35þ0 to 41þ6 weeks of gestation and nonhemolytic hyperbilirubinemia were randomized to treatment with BSL phototherapy or blue LED phototherapy. A total of 20 infants were included in the blue LED phototherapy group and 20 in the BSL phototherapy group. The duration of phototherapy was lower in the BSL than in the blue LED phototherapy group (15.8 4.9 vs. 20.6 6.0 hours; p ¼ 0.009), and infants in the former group had a lower probability (p ¼ 0.015) of remaining in phototherapy than infants in the latter. We concluded that BSL phototherapy is more effective than blue LED phototherapy for the treatment of hyperbilirubinemia in late preterm and term infants. Our data suggest that these results are not due to the different irradiance of the two phototherapy systems, but probably depend on their different peak light emissions.
Hyperbilirubinemia is the most frequent clinical problem neonatologists must deal with during the newborn period because under certain circumstances high levels of total serum bilirubin (TSB) may be potentially toxic to the central nervous system, even in healthy term newborns. Sufficiently elevated levels of bilirubin can lead to bilirubin encephalopathy and subsequently kernicterus, with devastating, permanent neurodevelopmental handicaps.1 Phototherapy is the first-line treatment for neonatal hyperbilirubinemia.2 Its efficacy depends on the absorption of light photons by the bilirubin molecules which converts the toxic Z,Z-bilirubin molecules into more easily excretable compounds (photoisomers) such as the configurational isomers Z,E-bilirubin and E,Z-bilirubin, and the structural isomers Z-lumirubin and E-lumirubin.3 Only light of certain
colors or wavelengths can be absorbed by bilirubin. Since bilirubin is a yellow pigment, blue and green light is best absorbed,4 and when broad-spectrum white light is used for treating only a fraction of the light is acting on the bilirubin.5 Moreover bilirubin does not absorb the different colors of light equally, and blue light at approximately 460 nm is absorbed more readily, while the green light is theoretically less well absorbed.4 However, the effectiveness of phototherapy depends also on the light’s capacity to penetrate the newborn’s skin, and longer wavelength light, such as green light, is expected to penetrate the infant’s skin more deeply.6 In fact, it has been demonstrated that longer than blues wavelength turquoise spectral band is more efficient than the blue spectral band in the clearance of bilirubin.3
received April 30, 2014 accepted after revision October 8, 2014
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1396685. ISSN 0735-1631.
Downloaded by: Florida International University. Copyrighted material.
Am J Perinatol
Pratesi et al.
Thus, we hypothesized that broad-spectrum light (BSL) phototherapy (λ ¼ 420–680 nm) is more effective than narrow-spectrum blue light phototherapy (λ ¼ 400–500 nm) in decreasing TSB concentration in late preterm and term infants. To assess this hypothesis, we performed a randomized, controlled study to compare the effectiveness of BSL phototherapy with that of light-emitting diode (LED) blue light phototherapy in the treatment of neonatal jaundice.
Materials and Methods Study Design The study was performed at the neonatal special care unit of the “Careggi University Hospital” of Florence and “San Salvatore Hospital” of L’Aquila, after approval by the local ethics committees. Caucasian healthy infants were eligible for inclusion in the study if they were born between the ages of 35þ0 and 41þ6 weeks at the time of birth, presented with nonhemolytic hyperbilirubinemia. Exclusion criteria were an Rh hemolytic disease, positive direct Coomb test, glucose-6phosphate dehydrogenase deficiency, hemoglobinopathies, clinical signs of hemolysis, and evidence of hemolysis in peripheral smear,7,8 and congenital infections. Written informed parental consent was obtained before infants were enrolled.
with a phototherapy radiometer (PD300-UV, Ophir Photonics, Jerusalem, Israel) with peak sensitivity at 460 to 490 nm. We placed the BSL phototherapy 28 cm above the infant (suggested mean irradiance ¼ 4.05 µW/cm2), while the blue LED phototherapy was placed 30.5 cm above the infant (suggested mean irradiance ¼ 3.50 µW/cm2), as recommended by the manufacturers. We did not adjust the distance between the lights and the infants to provide a similar irradiance so as to conform our results to normal clinical practice, where the distance is suggested by the manufacturers. During the phototherapy all infants remained in bassinettes, unclothed except for a diaper, and with their eyes covered. The TSB was checked every 6 hours during the phototherapy (until 30 hours after the beginning), and 12 and 24 hours after its discontinuation to evidence a possible rebound. The decision to discontinue treatment was made every 6 hours based on the TSB levels. The daily care (i.e., feeding, monitoring, etc.) of enrolled infants was performed according to local protocols.
Outcomes The primary endpoint of our study was the duration of phototherapy. Prespecified secondary endpoints were changes in the TSB and need of phototherapy at established data points (6, 12, 18, 24, 30 hours of treatment), hourly decrease in TSB, and time needed for TSB < 13 mg/dL.
Randomization
Other Collected Data
Infants were assigned to receive either BSL phototherapy (BiliCrystal Duo.3, Charleroi, Belgium) or blue LED phototherapy (NeoBlu, Natus Medical Inc., San Carlos, CA) in a 1:1 ratio in permuted blocks of variable size. Randomization was stratified according to center and gestational age (35þ0–37þ6 and 38þ041þ6 weeks). Group assignment was contained in sequentially numbered, sealed, and opaque envelopes. The decision to start phototherapy was made by clinicians other than the investigators involved in patient care, and researchers assessing study endpoints were blinded to the nature of the study treatments.
We recorded the following data for each infant: gestational age, birth weight, ABO incompatibility, Apgar score at 5 minutes, cephalohematoma, cord blood TSB, polycythemia (hematocrit > 65%), exclusive breastfeeding, TSB value at beginning and end of phototherapy, hourly increase in TSB before the beginning of phototherapy beginning, hourly decrease in TSB during phototherapy, highest TSB values, number of phototherapy courses, mean weight loss from birth, and length of hospital stay.
Study Intervention The TSB was measured in whole blood from a heel prick sample in a blood gas analyzer9 with a spectrophotometer module (ABL 735, Radiometer, Fiske Street Holliston, MA) in infants who presented with visible jaundice. Hematocrit was measured contemporarily. Entry criteria to phototherapy followed the American Academy of Pediatrics recommendations,2 and the decision to use phototherapy was made by the attending pediatrician. Phototherapy was discontinued in infants when the TSB fell below the level at which phototherapy was initiated or fell below 13 mg/dL in two consecutive measurements.2,10 The BSL phototherapy unit has eight compact fluorescent light tubes emitting light at 420 to 680 nm wavelengths (peak emission at 455 and 524 nm), while the blue LED phototherapy unit has diodes emitting blue light at 400 to 500 nm wavelengths (peak emission at 466 nm). On initiation of phototherapy, irradiance was measured at the clinical setting American Journal of Perinatology
Statistical Analysis On the basis of local data, we hypothesized that the conventional BSL phototherapy might decrease the duration of phototherapy from 24 12 to 19.2 2.4 hours (20%) in comparison with blue LED phototherapy. We calculated that we needed to enroll 19 newborns in each group to detect this difference as statistically significant with 90% power at 0.01 level. Data are reported as mean values and standard deviations, median values and ranges, or rates and percentages. We analyzed continuous variables using Student t-test and categorical variables with Fisher exact test. Moreover, the repeated measures analysis of variance test was used for analyzing intragroup and intergroup changes and decrease of TSB after phototherapy beginning. The Kaplan–Meier estimate was used to compare the probability of remaining in phototherapy during the study period in infants in either the BSL or blue LED phototherapy group. To evaluate whether BSL and blue LED devices can affect the phototherapy duration in our patients, this and the other variables that according to univariate analysis were most
Downloaded by: Florida International University. Copyrighted material.
Broad-Spectrum Light Phototherapy
Broad-Spectrum Light Phototherapy
Pratesi et al.
60 Screened for eligibility 20 Not eligible 1 Congenital malformation 1 Rh hemolytic disease 5 Positive Coomb test 7 Consent not given 6 Other reasons
40 Randomized
20 Included in the intention-to-treat analysis
20 Allocated to the blu LED phototherapy group 20 Received allocated treatment
20 Included in the intention-to-treat analysis
Fig. 1 The CONSORT flow diagram.
probably related to the duration of phototherapy and markedly different between the groups (p < 0.10) were included in multiple linear regression analysis to assess their independent role. Effect estimates are expressed as relative risk (RR) with profile likelihood-based 95% confidence limits (Cl). The reported RRs represent the effect of the TSB at phototherapy beginning or > 16 mg/dL, and gestational age < or 37 weeks. These values were decided because they represent the median value in our patients (TSB), or arbitrarily (gestational age).
Results The numbers of infants deemed eligible for the study and the numbers randomly assigned to receive blue LED or BSL phototherapy are shown in ►Fig. 1. We studied 40 infants,
20 of whom were included in the blue LED phototherapy group and 20 in the BSL phototherapy group, whose clinical characteristics are shown in ►Table 1. The average irradiance at the infant’s abdomen was 1.64 μW/cm2 in the BSL phototherapy group and 3.2 μW/cm2 in the blue LED phototherapy group. Infants in the BSL phototherapy group needed shorter treatment than infants in the blue LED phototherapy group (15.8 4.9 vs. 20.6 6.0 hours; p ¼ 0.009); the percentage of infants who still required phototherapy after 12 (45 vs. 80%; p ¼ 0.048), and 18 (15 vs. 50%; p ¼ 0.020) hours of treatment was lower in the BSL than in the blue LED phototherapy group. Changes (p < 0.0001) and decrease (BSL group: p ¼ 0.039; blue Led group: p ¼ 0.009) of TSB after phototherapy beginning were statistically significant in both the group. Patients in the BSL phototherapy group had a lower
Table 1 Clinical characteristics of studied infants (mean [SD], or median and [range], or rate and [%]) Broad-spectrum light phototherapy (N ¼ 20)
Blue LED phototherapy (N ¼ 20)
p-Value
38 2
39 1
1.000
6 (30)
1 (5)
0.096
Birth weight (g)
2,950 630
3,170 500
0.229
ABO incompatibility
8 (40)
7 (35)
1.000
Gestational age (wk) 37 wk of gestation
Apgar score at 5 min
9 (9–10)
9 (9–10)
1.000
Cephalohematoma
1 (5)
0
1.000
Cord TSB (mg/dL)
1.1 0.5
1.3 0.4
0.171
Hematocrit > 65%
6 (30)
5 (25)
1.000
Exclusive breastfeeding
9 (45)
9 (45)
1.000
Weight loss (%)
6.6 3.1
6.2 2.7
0.666
Length of hospital stay (d)
4.1 2.7
3.8 2.1
0.697
Abbreviations: LED, light emitting diode; TSB, total serum bilirubin. American Journal of Perinatology
Downloaded by: Florida International University. Copyrighted material.
20 Allocated to the BSL phototherapy group 20 Received allocated treatment
Broad-Spectrum Light Phototherapy
Pratesi et al.
Table 2 Primary and secondary outcomes (mean [SD], or rate and [%]) Broad-spectrum light phototherapy (N ¼ 20)
Blue LED phototherapy (N ¼ 20)
p-Value
15.8 4.9
20.6 6.0
0.009
6 h of phototherapy
13.1 1.7
14.0 2.9
0.205
12 h of phototherapy
11.7 2.1
12.7 2.5
0.201
18 h of phototherapy
9.8 1.6
12.0 1.6
