Acta Paediatr 81: 870-4. 1992
Efficacy of “high-intensity” blue-light and “standard” daylight phototherapy for non-haemolytic hyperbilirubinaemia KL Tan, G C Lim and KW Boey Departments of Paediatrics. Sociology and Social Work di Psychology, National Universiry of Singapore, Republic of Singapore
Tan KL, Lim GC, Boey KW. Efficacy of “high-intensity” blue-light and “standard” daylight phototherapy for non-haemolytic hyperbilirubinaemia. Acta Pzdiatr 1992;81 :870-4. Stockholm. ISSN 0803-5253 We report our clinical experience with phototherapy in 3802 infants; 3629 were exposed to “standard” daylight phototherapy and 173 to “high-intensity” blue-light phototherapy. High-intensity blue-light phototherapy was twice as effective as standard daylight phototherapy in decreasing bilirubin concentrations. No failures occurred with high-intensity phototherapy compared with an overall failure rate of I .84/1000 with daylight lamps; these cases were transferred to high-intensity phototherapy with prompt response. Rebound after cessation of phototherapy was greater in those exposed to high-intensity blue light with a significantly greater number requiring a second exposure. However, the incidence was still low. No third exposure was required in any infant. Nursing of infants under high-intensity blue light was more difficult and inconvenient as was clinical monitoring. The light also caused more stress on the nursing and medical personnel. However, the infants tolerated both types of phototherapy equally well. High-intensity blue-light phototherapy would seem to be the treatment of choice for infants with rapidly increasing or very high bilirubin levels, as well as in those not responding adequately to daylight phototherapy. 0 High intensity, hyperbilirubinaemia, phototherapy, standard K L Tan, Chief. Department of Neonatology, National University Hospital, 5 Lower Kent Ridge Road, Singapore 0511. Republic of Singapore
Phototherapy is now the accepted method of treating neonatal jaundice, its clinical efficacy being confirmed in many studies (1). Our clinical experience with phototherapy in a large number of infants in a single neonatal unit has already been documented; white daylight lamps were used in that study and found to be highly effective in 3999 infants suffering from non-haemolytic jaundice in the Kandang Kerbau Hospital, Singapore (2). As a result, phototherapy has now replaced exchange transfusion in the management of neonatal hyperbilirubinaemia. Although phototherapy is highly efficacious in the majority of infants presenting with neonatal jaundice, it is also well known that phototherapy using daylight lamps is occasionally not effective, especially in cases of severe or rapidly increasing neonatal jaundice. In such situations, the use of “high-intensity’’ phototherapy to ensure greater effectiveness and a faster rate of decrement in bilirubin levels would be useful; with highintensity phototherapy at “saturation dose” (3) bilirubin levels can be reduced much more rapidly. As a result, we have used high-intensity phototherapy in the treatment of neonatal jaundice in some infants. Such use of high-intensity phototherapy at saturation dose in a large number of infants in a neonatal unit has yet to be documented. Therefore, the purpose of this study was to examine high-intensity phototherapy in a clinical situa-
tion and compare its efficacy and convenience with phototherapy using daylight lamps.
Patients and methods Infants requiring phototherapy were investigated as described previously (4). Birth weights and gestational ages were recorded. The gestational age was obtained by assessing physical and neurological characteristics (9, as well as maternal history; where a discrepancy was observed clinical assessment of gestational age was used. Glucose-6-phosphate dehydrogenase (G6PD) status was screened by a modification of Bernstein’s method (6). “Standard” daylight phototherapy was provided by seven overhead daylight lamps (Philips TL20W/54) which at infant skin level provided an average irradiance of 295 pW/cm2 in the 400-480 nm range, 250 pW/cm2 in the 425-475 nm range and 140 pW/cm2 in the 440-480 nm range and a total light intensity of 4630 Ix. In highintensity phototherapy, special blue lamps (TL20W/52) with an emission spectrum similar to the bilirubin absorption spectrum were used in the same seven overhead lamp set-up together with an additional bank of four lights placed below the infant to provide phototherapy at saturation dose (3). The efficacy of
ACTA PRDIATR 81 (1992)
phototherapy is determined by the total irradiance on the skin surface, independent of the direction(s) of the light (4). With this set-up, the total average irradiance at the infant’s skin level was 1850 pW/cm2, 1690 pW/cm2 and 1070 pW/cm2, respectively, and the total light intensity 1880 Ix. The method of measurement has been described previously (4). The lamps were changed regularly after 2000 h of use, by which time the irradiance was still 80% of the original (7). The infants undergoing phototherapy were clinically well (with no contusions/cephalhaematomata) but presented with non-haemolytic hyperbilirubinaemia as evidenced by the absence of blood group isoimmunization, a negative direct Coombs test, haemoglobin (Hb) concentration > 14 g/dl, packed cell volume (PCV) >42% and a normal haemogram, as well as normal G6PD status. Phototherapy was started when the bilirubin concentration was > 256 pmol/l ( I 5 mg/dl), regardless of the time of the day. In infants with a very rapid onset ofjaundice ( < 24 h postnatal age), phototherapy was started when the bilirubin concentration was > 222pmol/l(I 3 mg/dl) in the first 48 h of life. The infants were exposed, completely unclothed with their eyes covered, to continuous phototherapy which was interrupted only for feeding, cleaning and blood sampling. The typeofphototherapy used was largely determined by availability of the phototherapy set-ups, there being more standard set-ups than high-intensity ones; however, the neonatal staff tended to prefer high-intensity phototherapy for those with rapidly increasing or more severejaundice and standard phototherapy for the more usual “standard” cases. Informed consent was obtained in all cases. Capillary blood was sampled at the start of phototherapy and at intervals of 6 or 12 h thereafter; the lights were switched off and the infant removed from the cot for blood sampling. Failure of phototherapy was defined as continued increase in bilirubin concentration on two consecutive determinations beyond the starting bilirubin value. Phototherapy was stopped when bilirubin concentrations had decreased to < 187 pmol/l ( 1 1 mg/dl), the minimum period of exposure being 24 h. At the end of phototherapy, another c‘apillary blood sample was obtained for determination of bilirubin concentration. The duration of phototherapy was recorded. The bilirubin level was monitored daily after phototherapy; the minimum period was two days. H b and PCV were determined one day after treatment. Infants with rebound bilirubin values greater than pre-phototherapy values were again exposed to phototherapy following the same guidelines as the first exposure. The infants were observed for irritability, restlessness, feeding problems or any other unusual behaviour during exposure. The capillary blood samples were labelled and kept in a light-proof box until bilirubin concentration was determined under standard conditions using an A 0 bilirubinometer (American Optical Co, USA) which was checked regularly against known standards. The
Phototherapy for non-haemolyrir hyperbilirubinaemia
87 I
bilirubin concentrations in paired samples of capillary blood were also monitored randomly to determine the reliability and reproducibility of the determinations. Direct-acting bilirubin concentration was also ’determined in random samples as well as in those from babies not responding to phototherapy as described previously (3). The data were compiled and stored in an IBM computer in a fixed-input format and analysed with the statistical package for the social sciences (SPSS-X) (8). Statistical analysis was carried out using the Student’s ttest and chi-square test. Some of the infants in the standard phototherapy group were from a group reported on previously (2).
Results Altogether 3802 infants with non-haemolytic hyperbilirubinaemia were exposed to phototherapy; 3629 were exposed to standard daylight phototherapy and 173 to high-intensity phototherapy provided by the special blue lamps. The infants in each group were divided into three groups (Table 1) according to gestational age and birth weight. During this period (1976 to 1986), 242 821 infants were delivered in the hospital. As most healthy infants and mothers were discharged within 24 to 48 h after delivery, these infants with non-haemolytic hyperbilirubinaemia exposed to phototherapy most probably did not represent the total number as not all infants developing hyperbilirubinaemia were referred back to this hospital. The infants were apparently well and free from any detectable clinical illness at the start of treatment. The majority of the infants were full term and of normal birth weight with a substantial group of small preterm infants (birth weight < 2270 g, gestational age < 37 weeks) and a third group of large preterni infants (birth weight > 2270 g, gestational age < 37 weeks). All the infants remained well during and after exposure. Hb and PCV were comparable at the start and end of exposure in all groups. The normal-sized infants were able to maintain normothermia during phototherapy; those with low birth weight needed extra heat from incandescent lamps (with minimum light irradiance in the range effective for phototherapy (7)) to maintain normothermia. Increased fluid intake was needed to ensure adequate hydration. Nursing of the infants exposed to high-intensity phototherapy was more inconvenient; since no lining of the cot was possible, more frequent cleaning was necessary. As reported previously (9), blue light made clinical monitoring difficult and caused nausea and giddiness, and had to be screened off. The infants, however, tolerated the light and bare plexiglass cot well; no behavioural difference was observed in the infants exposed to white or blue light. Feeds were well tolerated with no vomiting, irritability or restlessness observed in the infants in either group. The concentration of bilirubin in the full-term infants
a
129.28 (4.27) 41.83 (1.23)
51.30 (0.68) 19.14 (0.27) 175.79 (4.27) 56.96 (1.01) 47.52 (1.31) 1.31 (0.04)
265.73 (1.71) 192.20 (1.88) 152.87 (1.54)
293.09 (0.51) 226.06 (1.03) 168.95 (0.85)
305.41 (3.76) 175.79 (3.76) 129.62 (2.74)
265.39 (0.68) 213.24 (0.68) 165.36 (0.51)
100.03 (0.68) 36.86 (0.25) 58.86 (0.48) 0.70 (0.01)
56.39 (0.49) 54.64 (0.47)
56.67 (0.27) 54.46 (0.26)
55.36 (0.79) 51.82 (0.98)
56.56 (0.17) 54.30 (0.17)
124.15 (1.03) 42.17 (0.28) 64.02 (0.81) 0.73 (0.01)
112.86 (2.22) 41.55 (0.71) 49.48 (1.58) 0.98 (0.03)
74.21 (1.88) 27.43 (0.71)
18.29 (0.13) 17.83 (0.13)
18.51 (0.07) 17.74 (0.07)
18.60 (0.21) 17.54 (0.21)
18.47 (0.04) 17.75 (0.04)
66.86 (1.03) 22.70 (0.33)
397 213: 184 1906.92 (15.12) 33.90 (0.12) 4.07 (0.08)
1105 661:444 3108.65 (13.43) 39.12 (0.04) 4.07 (0.03)
141 78:63 31 14.06 (39.60) 39.01 (0.13) 4.25 (0.14)
2879 1664: 1215 3122.01 (8.55) 39.16 (0.03) 3.84 (0.02)
In Singapore, infants weighing > 2270 g are deemed to be of normal birth weight.
Duration of phototherapy (h) % Decline in bilirubin concentration/h
%
Decrease in bilirubin concentration at end of phototherapy prnolll
%
Total Sex (M :F) Birth weight (g)” Gestational age (weeks) Age at phototherapy (days) Haemoglobin concentration (g/dl) At the start of phototherapy At the end of phototherapy Packed cell volume (%) At the start of phototherapy At the end of phototherapy Serum bilirubin concentration (pmolll) At the start of phototherapy After 24 h of phototherapy At the end of phototherapy Decrease in bilirubin concentration after 24 h pmol/l
White light
White light (adjusted)
Blue light
185.36 (10.43) 64.21 (2.99) 34.42 (3.62) 2.06 (0.16)
155.78 (9.92) 54.94 (3.21)
287.11 (6.84) 126.20 (8.04) 102.60 (7.35)
55.95 (1.96) 52.18 (1.66)
18.32 (0.47) 17.33 (0.46)
19 7: 12 1891.84 (61.76) 33.84 (0.47) 3.86 (0.21)
Blue light
Small preterm infants
White light (original)
Full-term infants
Table I. Data of infants exposed to phototherapy with decreases in bilirubin concentration in response to phototherapy. Values are mean (SEM).
99.35 (2.39) 36.39 (0.79) 58.23 (1.23) 0.70 (0.02)
54.55 (2.05) 19.71 (0.76)
266.76 (1.71) 212.72 (1.88) 167.41 (1.88)
56.47 (0.46) 55.39 (0.51)
18.45 (0.12) 18.20 (0.17)
353 215: 138 2779.45 (23.88) 34.87 (0.08) 3.89 (0.06)
~
White light
178.52 (9.75) 62.93 (2.12) 43.23 (2.44) 1.54 (0.14)
134.23 (12.14) 47.02 (3.33)
282.83 (10.94) 148.43 (8.72) 104.99 (5.81)
55.29 (3.49) 51.32 (4.15)
19.44 (0.77) 18.33 (0.67)
13 8:s 2713.69 (100.92) 34.92 (0.33) 4.73 (0.42)
~.
Blue light
Large preterm infants
Phototherapy .for non-haemolylic hyperbilirubinaemia
ACTA PRDIATR 81 (1992)
exposed to daylight lamps was significantly lower than that of the full-term infants exposed to high-intensity blue-light phototherapy ( p < 0.001). To ensure comparability of these two groups, the original daylight group of infants was “adjusted” by stratification resulting in the proportion of infants within each 17-pmol/l(I mg/ dl) segment approximating to that of the high-intensity group. A decrease in bilirubin concentration occurred in all three groups of full-term infants (Fig. 1) and was especially marked in the infants exposed to highintensity blue light; by 12 h the bilirubin concentration in the high-intensity group, initially higher than that of the daylight-lamp groups, had decreased to below the concentration of the respective comparable groups exposed to daylight lamps. A similar response was observed in the large preterm and small preterm infants. The rate of bilirubin decrease was assessed in two ways: (1) the 24-h reduction in bilirubin concentration expressed as a proportion of the initial bilirubin concentration, this being a valid method for assessing initial response as all infants were exposed to phototherapy for at least 24 h; and (2) the overall decrease in bilirubin concentration related to the total exposure time and expressed as a “proportionate” decrease per hour (% decline/h). In the full-term infants, the 24-h decrease in bilirubin concentration in the group exposed to highintensity blue-light phototherapy was twice that of the
Fullterm
Large Preterm
a .c 200 n
v
-.-2 z
100
3002
A
Small Preterm
-
f
0
groups exposed to daylight-lamp phototherapy. This difference was highly significant ( p < 0.0001). Similarly, the duration of phototherapy (Table 1) was also significantly shorter ( p < 0.OOOl) and the overall rate of decline in bilirubin concentration for the whole duration of phototherapy was significantly greater ( p < 0.0001); about twice that of the groups exposed to daylight lamps. The same pattern of response was observed in the large preterm infants as well as in the small preterm infants. The response to phototherapy was greatest in the small preterm infants. Both standard daylight phototherapy as well as high-intensity phototherapy resulted in a rate of decline in bilirubin concentration that was significantly greater (p < 0.OOOl) than that of the respective full-term infants exposed to the same type of phototherapy . The failure rate was very low; among the full-term infants, four did not respond adequately to daylight phototherapy and had to be transferred to highintensity phototherapy with a resultant prompt response. One of these four infants was in the adjusted group. Among the small preterms, only one infant did not respond to white light and had to be transferred to high-intensity phototherapy. Two of the large preterm infants failed to respond to white light but responded promptly to high-intensity blue-light phototherapy. The overall failure rate was only 1.84 per 1000. The cause(s) of this failure could not be determined; Hb, postnatal age, bilirubin concentration and rate of bilirubin increase did not differ significantly from those of the responders. There were no failures in the high-intensity group. The rebound after phototherapy was mild with only 30 infants (7.89 per 1000) requiring re-phototherapy; 24 (6.61 per 1000) were from the daylight group compared with 6 (34.68 per 1000) from the high-intensity group, a difference that was highly significant (p < 0.0001). Among the full-term infants, 19 from the original daylight group required a second exposure compared with 4 from the high-intensity group (p < O.OOOl), in the large preterm group, 2 compared with 1 ( p < O.OOl), and in the small preterm group, 3 and 1 ( p > 0.05),respectively. All responded well to the second exposure. No infant required a third exposure. The direct-acting bilirubin concentration in all infants did not exceed 17 pmol/l in all the random samples analysed.
Discussion
200.
100
873
12
24
36
4‘8
Phototherapy (hrs)
80”
;
2
Post (days)
Fig. I . Bilirubin response to daylight and blue-light phototherapy. (Standard errors less than 5 pmol/l are too small to be charted.)
As demonstrated in an earlier report (2), daylight phototherapy was highly effective in the treatment of non-haemolytic hyperbilirubinaemia. The failure rate was negligible. No factor(s) predictive of failure could be determined. The post-phototherapy rebound was also mild; repeat treatment for those few needing a second exposure was equally effective compared with the first.
874 KL Tan et al. High-intensity phototherapy using double-bank blue lamps with an emission spectrum very similar to the absorption spectrum of bilirubin was much more effective than daylight lamps, resulting in a decrement rate that was about twice that of daylight lamps, not only in full-term but also in small preterm and large preterm infants. The 24-h reduction in bilirubin concentration during phototherapy was expressed as a proportion of the initial bilirubin concentration. This was a reliable method of assessing the initial efficacy of phototherapy as all the infants were exposed to phototherapy for at least 24 h. Problems caused by diurnal variations in bilirubin concentration were effectively avoided using this method of assessment. To assess the overall effectiveness of phototherapy, the “proportionate” decrease in bilirubin concentration during the whole period of exposure, expressed as a percentage decline per hour, was thought to be the most reliable. These two methods of assessment of the efficacy of phototherapy allow the initial and overall efficacy of exposure to be reliably determined and compared among the groups of infants; this was further enhanced by ensuring that the initial bilirubin values of the relevant groups of infants studied were comparable. High-intensity blue-light phototherapy at optimum dose was much more effective than standard daylight phototherapy and many might be tempted to use it as a standard form of treatment for severe neonatal jaundice. However, blue light is not without its disadvantages as reported previously (9). Furthermore, using two banks of light, one above and one below the infant, made nursing of the baby more inconvenient and difficult. The daylight lamps with a more gradual but still adequate efficacy would seem to be the therapy of choice for those cases who d o not need a rapid decline in bilirubin concentration. It would however be wiser to use high-intensity bluelight phototherapy in urgent cases with more severe hyperbilirubinaemia where a rapid decline in bilirubin levels to safer concentrations would be preferred, or in those not responding to white daylight phototherapy. Indeed our blue-light phototherapy has been so effective that even severely haemolytic cases have been treated in this manner with great success (lo), thereby eliminating the need for exchange transfusion; packed cell transfusion for the resulting anaemia was performed with very little morbidity. The anticipated discomfort caused by the hard perspex surface did not arise since the infants tolerated blue-light and daylight phototherapy equally well. The increased efficacy of standard daylight phototherapy in small preterm infants which has been reported previously (2), was also observed with high-intensity phototherapy. The relative increase in surface area as well as skin translucency are obvious enhancing factors. However, the efficacy in large preterm infants was comparable to that of full-term infants.
ACTA PRDIATR 81 (1992)
A more rapid rebound was observed after highintensity phototherapy resulting in a need for rephototherapy that was about five times more frequent than that of the daylight group. This was not evident in an earlier study (3) probably because of the smaller numbers involved; only with vastly larger numbers will this difference become evident. This phenomenon is probably due to the duration of high-intensity phototherapy being significantly shorter, with cessation of phototherapy occurring at a period when the rate of bilirubin accumulation (bilirubin production versus elimination) was still appreciable; at a level still capable of reducing the efficacy of standard daylight phototherapy. The longer duration of standard daylight phototherapy ensured cessation at a later stage when the rate of bilirubin accumulation was probably already declining; hence the lesser rebound and need for a second exposure. Phototherapy would therefore seem to be the preferred method of treatment for all neonates presenting with hyperbilirubinaemia. For infants with especially severe hyperbilirubinaemia, high-intensity blue-light phototherapy should be considered. This should also be the case for those infants not responding to phototherapy with daylight lamps. Our experience has demonstrated such an approach to be successful in the vast majority of patients. Morbidity was low and mortality nil. The non-invasive nature of this form of treatment has much to recommend it as the natural choice for the treatment of neonatal jaundice. Acknowledgements.-The authors thank Dr TS Koh, Medical Director of Kandang Kerbau Hospital, for permission to report these findings, the nurses for their enthusiastic help, LE Lee for technical help, and J Ow for secretarial help.
References 1. Lucey JF. Neonatal jaundice and phototherapy. Pediatr Clin
North Am 1972;19:827-39 2. Tan KL, Boey KW. Efficacy of phototherapy in non-haemolytic hyperbilirubinaemia. BMJ 1986;293:1361-3 3. Tan KL. The pattern of bilirubin response to phototherapy for neonatal hyperbilirubinemia. Pediatr Res 1982;16:6704 4. Tan KL. Comparison of the effectiveness of“sing1edirection” and “double direction” phototherapy for neonatal jaundice. Pediatrics 1975;56:550-3 5. Dubowitz LMS, Dubowitz V, Goldberg C. Clinical assessment of gestational age in the newborn infants. J Pediatr 1970;77:1-10 6. Brown WR, Wong HB. Hyperbilirubinemia and kernicterus in glucose-6-phosphate dehydrogenase deficient infants in Singapore. Pediatrics 1968;41:1055-62 7. Tan KL. Some aspects on management of neonatal jaundice in Singapore. J Singapore Paediatr SOC1978;20:12241 8. Nie NH, Hull CH, Jenkins SG, Stanbienner K, Bent DH. Statistical Package for the Social Sciences. New York: McGraw Hill Book Co, 1975 9. Tan KL. Efficacy of fluorescent daylight, blue and green lamps in the management of non-hemolytic hyperbilirubinemia. J Pediatr 1989i14132-7 10. Tan KL. Light dose-response relationship in phototherapy. In: Rubaltelli FF, Jon G eds Neonatal Jaundice: New Trends in Phototherapy. New York: Plenum Press, 1984:23544 Received Aug. 2, 1991. Accepted Dec. 12, 1991
Efficacy of “high-intensity” blue-light and “standard” daylight phototherapy for non-haemolytic hyperbilirubinaemia KL Tan, G C Lim and KW Boey Departments of Paediatrics. Sociology and Social Work di Psychology, National Universiry of Singapore, Republic of Singapore
Tan KL, Lim GC, Boey KW. Efficacy of “high-intensity” blue-light and “standard” daylight phototherapy for non-haemolytic hyperbilirubinaemia. Acta Pzdiatr 1992;81 :870-4. Stockholm. ISSN 0803-5253 We report our clinical experience with phototherapy in 3802 infants; 3629 were exposed to “standard” daylight phototherapy and 173 to “high-intensity” blue-light phototherapy. High-intensity blue-light phototherapy was twice as effective as standard daylight phototherapy in decreasing bilirubin concentrations. No failures occurred with high-intensity phototherapy compared with an overall failure rate of I .84/1000 with daylight lamps; these cases were transferred to high-intensity phototherapy with prompt response. Rebound after cessation of phototherapy was greater in those exposed to high-intensity blue light with a significantly greater number requiring a second exposure. However, the incidence was still low. No third exposure was required in any infant. Nursing of infants under high-intensity blue light was more difficult and inconvenient as was clinical monitoring. The light also caused more stress on the nursing and medical personnel. However, the infants tolerated both types of phototherapy equally well. High-intensity blue-light phototherapy would seem to be the treatment of choice for infants with rapidly increasing or very high bilirubin levels, as well as in those not responding adequately to daylight phototherapy. 0 High intensity, hyperbilirubinaemia, phototherapy, standard K L Tan, Chief. Department of Neonatology, National University Hospital, 5 Lower Kent Ridge Road, Singapore 0511. Republic of Singapore
Phototherapy is now the accepted method of treating neonatal jaundice, its clinical efficacy being confirmed in many studies (1). Our clinical experience with phototherapy in a large number of infants in a single neonatal unit has already been documented; white daylight lamps were used in that study and found to be highly effective in 3999 infants suffering from non-haemolytic jaundice in the Kandang Kerbau Hospital, Singapore (2). As a result, phototherapy has now replaced exchange transfusion in the management of neonatal hyperbilirubinaemia. Although phototherapy is highly efficacious in the majority of infants presenting with neonatal jaundice, it is also well known that phototherapy using daylight lamps is occasionally not effective, especially in cases of severe or rapidly increasing neonatal jaundice. In such situations, the use of “high-intensity’’ phototherapy to ensure greater effectiveness and a faster rate of decrement in bilirubin levels would be useful; with highintensity phototherapy at “saturation dose” (3) bilirubin levels can be reduced much more rapidly. As a result, we have used high-intensity phototherapy in the treatment of neonatal jaundice in some infants. Such use of high-intensity phototherapy at saturation dose in a large number of infants in a neonatal unit has yet to be documented. Therefore, the purpose of this study was to examine high-intensity phototherapy in a clinical situa-
tion and compare its efficacy and convenience with phototherapy using daylight lamps.
Patients and methods Infants requiring phototherapy were investigated as described previously (4). Birth weights and gestational ages were recorded. The gestational age was obtained by assessing physical and neurological characteristics (9, as well as maternal history; where a discrepancy was observed clinical assessment of gestational age was used. Glucose-6-phosphate dehydrogenase (G6PD) status was screened by a modification of Bernstein’s method (6). “Standard” daylight phototherapy was provided by seven overhead daylight lamps (Philips TL20W/54) which at infant skin level provided an average irradiance of 295 pW/cm2 in the 400-480 nm range, 250 pW/cm2 in the 425-475 nm range and 140 pW/cm2 in the 440-480 nm range and a total light intensity of 4630 Ix. In highintensity phototherapy, special blue lamps (TL20W/52) with an emission spectrum similar to the bilirubin absorption spectrum were used in the same seven overhead lamp set-up together with an additional bank of four lights placed below the infant to provide phototherapy at saturation dose (3). The efficacy of
ACTA PRDIATR 81 (1992)
phototherapy is determined by the total irradiance on the skin surface, independent of the direction(s) of the light (4). With this set-up, the total average irradiance at the infant’s skin level was 1850 pW/cm2, 1690 pW/cm2 and 1070 pW/cm2, respectively, and the total light intensity 1880 Ix. The method of measurement has been described previously (4). The lamps were changed regularly after 2000 h of use, by which time the irradiance was still 80% of the original (7). The infants undergoing phototherapy were clinically well (with no contusions/cephalhaematomata) but presented with non-haemolytic hyperbilirubinaemia as evidenced by the absence of blood group isoimmunization, a negative direct Coombs test, haemoglobin (Hb) concentration > 14 g/dl, packed cell volume (PCV) >42% and a normal haemogram, as well as normal G6PD status. Phototherapy was started when the bilirubin concentration was > 256 pmol/l ( I 5 mg/dl), regardless of the time of the day. In infants with a very rapid onset ofjaundice ( < 24 h postnatal age), phototherapy was started when the bilirubin concentration was > 222pmol/l(I 3 mg/dl) in the first 48 h of life. The infants were exposed, completely unclothed with their eyes covered, to continuous phototherapy which was interrupted only for feeding, cleaning and blood sampling. The typeofphototherapy used was largely determined by availability of the phototherapy set-ups, there being more standard set-ups than high-intensity ones; however, the neonatal staff tended to prefer high-intensity phototherapy for those with rapidly increasing or more severejaundice and standard phototherapy for the more usual “standard” cases. Informed consent was obtained in all cases. Capillary blood was sampled at the start of phototherapy and at intervals of 6 or 12 h thereafter; the lights were switched off and the infant removed from the cot for blood sampling. Failure of phototherapy was defined as continued increase in bilirubin concentration on two consecutive determinations beyond the starting bilirubin value. Phototherapy was stopped when bilirubin concentrations had decreased to < 187 pmol/l ( 1 1 mg/dl), the minimum period of exposure being 24 h. At the end of phototherapy, another c‘apillary blood sample was obtained for determination of bilirubin concentration. The duration of phototherapy was recorded. The bilirubin level was monitored daily after phototherapy; the minimum period was two days. H b and PCV were determined one day after treatment. Infants with rebound bilirubin values greater than pre-phototherapy values were again exposed to phototherapy following the same guidelines as the first exposure. The infants were observed for irritability, restlessness, feeding problems or any other unusual behaviour during exposure. The capillary blood samples were labelled and kept in a light-proof box until bilirubin concentration was determined under standard conditions using an A 0 bilirubinometer (American Optical Co, USA) which was checked regularly against known standards. The
Phototherapy for non-haemolyrir hyperbilirubinaemia
87 I
bilirubin concentrations in paired samples of capillary blood were also monitored randomly to determine the reliability and reproducibility of the determinations. Direct-acting bilirubin concentration was also ’determined in random samples as well as in those from babies not responding to phototherapy as described previously (3). The data were compiled and stored in an IBM computer in a fixed-input format and analysed with the statistical package for the social sciences (SPSS-X) (8). Statistical analysis was carried out using the Student’s ttest and chi-square test. Some of the infants in the standard phototherapy group were from a group reported on previously (2).
Results Altogether 3802 infants with non-haemolytic hyperbilirubinaemia were exposed to phototherapy; 3629 were exposed to standard daylight phototherapy and 173 to high-intensity phototherapy provided by the special blue lamps. The infants in each group were divided into three groups (Table 1) according to gestational age and birth weight. During this period (1976 to 1986), 242 821 infants were delivered in the hospital. As most healthy infants and mothers were discharged within 24 to 48 h after delivery, these infants with non-haemolytic hyperbilirubinaemia exposed to phototherapy most probably did not represent the total number as not all infants developing hyperbilirubinaemia were referred back to this hospital. The infants were apparently well and free from any detectable clinical illness at the start of treatment. The majority of the infants were full term and of normal birth weight with a substantial group of small preterm infants (birth weight < 2270 g, gestational age < 37 weeks) and a third group of large preterni infants (birth weight > 2270 g, gestational age < 37 weeks). All the infants remained well during and after exposure. Hb and PCV were comparable at the start and end of exposure in all groups. The normal-sized infants were able to maintain normothermia during phototherapy; those with low birth weight needed extra heat from incandescent lamps (with minimum light irradiance in the range effective for phototherapy (7)) to maintain normothermia. Increased fluid intake was needed to ensure adequate hydration. Nursing of the infants exposed to high-intensity phototherapy was more inconvenient; since no lining of the cot was possible, more frequent cleaning was necessary. As reported previously (9), blue light made clinical monitoring difficult and caused nausea and giddiness, and had to be screened off. The infants, however, tolerated the light and bare plexiglass cot well; no behavioural difference was observed in the infants exposed to white or blue light. Feeds were well tolerated with no vomiting, irritability or restlessness observed in the infants in either group. The concentration of bilirubin in the full-term infants
a
129.28 (4.27) 41.83 (1.23)
51.30 (0.68) 19.14 (0.27) 175.79 (4.27) 56.96 (1.01) 47.52 (1.31) 1.31 (0.04)
265.73 (1.71) 192.20 (1.88) 152.87 (1.54)
293.09 (0.51) 226.06 (1.03) 168.95 (0.85)
305.41 (3.76) 175.79 (3.76) 129.62 (2.74)
265.39 (0.68) 213.24 (0.68) 165.36 (0.51)
100.03 (0.68) 36.86 (0.25) 58.86 (0.48) 0.70 (0.01)
56.39 (0.49) 54.64 (0.47)
56.67 (0.27) 54.46 (0.26)
55.36 (0.79) 51.82 (0.98)
56.56 (0.17) 54.30 (0.17)
124.15 (1.03) 42.17 (0.28) 64.02 (0.81) 0.73 (0.01)
112.86 (2.22) 41.55 (0.71) 49.48 (1.58) 0.98 (0.03)
74.21 (1.88) 27.43 (0.71)
18.29 (0.13) 17.83 (0.13)
18.51 (0.07) 17.74 (0.07)
18.60 (0.21) 17.54 (0.21)
18.47 (0.04) 17.75 (0.04)
66.86 (1.03) 22.70 (0.33)
397 213: 184 1906.92 (15.12) 33.90 (0.12) 4.07 (0.08)
1105 661:444 3108.65 (13.43) 39.12 (0.04) 4.07 (0.03)
141 78:63 31 14.06 (39.60) 39.01 (0.13) 4.25 (0.14)
2879 1664: 1215 3122.01 (8.55) 39.16 (0.03) 3.84 (0.02)
In Singapore, infants weighing > 2270 g are deemed to be of normal birth weight.
Duration of phototherapy (h) % Decline in bilirubin concentration/h
%
Decrease in bilirubin concentration at end of phototherapy prnolll
%
Total Sex (M :F) Birth weight (g)” Gestational age (weeks) Age at phototherapy (days) Haemoglobin concentration (g/dl) At the start of phototherapy At the end of phototherapy Packed cell volume (%) At the start of phototherapy At the end of phototherapy Serum bilirubin concentration (pmolll) At the start of phototherapy After 24 h of phototherapy At the end of phototherapy Decrease in bilirubin concentration after 24 h pmol/l
White light
White light (adjusted)
Blue light
185.36 (10.43) 64.21 (2.99) 34.42 (3.62) 2.06 (0.16)
155.78 (9.92) 54.94 (3.21)
287.11 (6.84) 126.20 (8.04) 102.60 (7.35)
55.95 (1.96) 52.18 (1.66)
18.32 (0.47) 17.33 (0.46)
19 7: 12 1891.84 (61.76) 33.84 (0.47) 3.86 (0.21)
Blue light
Small preterm infants
White light (original)
Full-term infants
Table I. Data of infants exposed to phototherapy with decreases in bilirubin concentration in response to phototherapy. Values are mean (SEM).
99.35 (2.39) 36.39 (0.79) 58.23 (1.23) 0.70 (0.02)
54.55 (2.05) 19.71 (0.76)
266.76 (1.71) 212.72 (1.88) 167.41 (1.88)
56.47 (0.46) 55.39 (0.51)
18.45 (0.12) 18.20 (0.17)
353 215: 138 2779.45 (23.88) 34.87 (0.08) 3.89 (0.06)
~
White light
178.52 (9.75) 62.93 (2.12) 43.23 (2.44) 1.54 (0.14)
134.23 (12.14) 47.02 (3.33)
282.83 (10.94) 148.43 (8.72) 104.99 (5.81)
55.29 (3.49) 51.32 (4.15)
19.44 (0.77) 18.33 (0.67)
13 8:s 2713.69 (100.92) 34.92 (0.33) 4.73 (0.42)
~.
Blue light
Large preterm infants
Phototherapy .for non-haemolylic hyperbilirubinaemia
ACTA PRDIATR 81 (1992)
exposed to daylight lamps was significantly lower than that of the full-term infants exposed to high-intensity blue-light phototherapy ( p < 0.001). To ensure comparability of these two groups, the original daylight group of infants was “adjusted” by stratification resulting in the proportion of infants within each 17-pmol/l(I mg/ dl) segment approximating to that of the high-intensity group. A decrease in bilirubin concentration occurred in all three groups of full-term infants (Fig. 1) and was especially marked in the infants exposed to highintensity blue light; by 12 h the bilirubin concentration in the high-intensity group, initially higher than that of the daylight-lamp groups, had decreased to below the concentration of the respective comparable groups exposed to daylight lamps. A similar response was observed in the large preterm and small preterm infants. The rate of bilirubin decrease was assessed in two ways: (1) the 24-h reduction in bilirubin concentration expressed as a proportion of the initial bilirubin concentration, this being a valid method for assessing initial response as all infants were exposed to phototherapy for at least 24 h; and (2) the overall decrease in bilirubin concentration related to the total exposure time and expressed as a “proportionate” decrease per hour (% decline/h). In the full-term infants, the 24-h decrease in bilirubin concentration in the group exposed to highintensity blue-light phototherapy was twice that of the
Fullterm
Large Preterm
a .c 200 n
v
-.-2 z
100
3002
A
Small Preterm
-
f
0
groups exposed to daylight-lamp phototherapy. This difference was highly significant ( p < 0.0001). Similarly, the duration of phototherapy (Table 1) was also significantly shorter ( p < 0.OOOl) and the overall rate of decline in bilirubin concentration for the whole duration of phototherapy was significantly greater ( p < 0.0001); about twice that of the groups exposed to daylight lamps. The same pattern of response was observed in the large preterm infants as well as in the small preterm infants. The response to phototherapy was greatest in the small preterm infants. Both standard daylight phototherapy as well as high-intensity phototherapy resulted in a rate of decline in bilirubin concentration that was significantly greater (p < 0.OOOl) than that of the respective full-term infants exposed to the same type of phototherapy . The failure rate was very low; among the full-term infants, four did not respond adequately to daylight phototherapy and had to be transferred to highintensity phototherapy with a resultant prompt response. One of these four infants was in the adjusted group. Among the small preterms, only one infant did not respond to white light and had to be transferred to high-intensity phototherapy. Two of the large preterm infants failed to respond to white light but responded promptly to high-intensity blue-light phototherapy. The overall failure rate was only 1.84 per 1000. The cause(s) of this failure could not be determined; Hb, postnatal age, bilirubin concentration and rate of bilirubin increase did not differ significantly from those of the responders. There were no failures in the high-intensity group. The rebound after phototherapy was mild with only 30 infants (7.89 per 1000) requiring re-phototherapy; 24 (6.61 per 1000) were from the daylight group compared with 6 (34.68 per 1000) from the high-intensity group, a difference that was highly significant (p < 0.0001). Among the full-term infants, 19 from the original daylight group required a second exposure compared with 4 from the high-intensity group (p < O.OOOl), in the large preterm group, 2 compared with 1 ( p < O.OOl), and in the small preterm group, 3 and 1 ( p > 0.05),respectively. All responded well to the second exposure. No infant required a third exposure. The direct-acting bilirubin concentration in all infants did not exceed 17 pmol/l in all the random samples analysed.
Discussion
200.
100
873
12
24
36
4‘8
Phototherapy (hrs)
80”
;
2
Post (days)
Fig. I . Bilirubin response to daylight and blue-light phototherapy. (Standard errors less than 5 pmol/l are too small to be charted.)
As demonstrated in an earlier report (2), daylight phototherapy was highly effective in the treatment of non-haemolytic hyperbilirubinaemia. The failure rate was negligible. No factor(s) predictive of failure could be determined. The post-phototherapy rebound was also mild; repeat treatment for those few needing a second exposure was equally effective compared with the first.
874 KL Tan et al. High-intensity phototherapy using double-bank blue lamps with an emission spectrum very similar to the absorption spectrum of bilirubin was much more effective than daylight lamps, resulting in a decrement rate that was about twice that of daylight lamps, not only in full-term but also in small preterm and large preterm infants. The 24-h reduction in bilirubin concentration during phototherapy was expressed as a proportion of the initial bilirubin concentration. This was a reliable method of assessing the initial efficacy of phototherapy as all the infants were exposed to phototherapy for at least 24 h. Problems caused by diurnal variations in bilirubin concentration were effectively avoided using this method of assessment. To assess the overall effectiveness of phototherapy, the “proportionate” decrease in bilirubin concentration during the whole period of exposure, expressed as a percentage decline per hour, was thought to be the most reliable. These two methods of assessment of the efficacy of phototherapy allow the initial and overall efficacy of exposure to be reliably determined and compared among the groups of infants; this was further enhanced by ensuring that the initial bilirubin values of the relevant groups of infants studied were comparable. High-intensity blue-light phototherapy at optimum dose was much more effective than standard daylight phototherapy and many might be tempted to use it as a standard form of treatment for severe neonatal jaundice. However, blue light is not without its disadvantages as reported previously (9). Furthermore, using two banks of light, one above and one below the infant, made nursing of the baby more inconvenient and difficult. The daylight lamps with a more gradual but still adequate efficacy would seem to be the therapy of choice for those cases who d o not need a rapid decline in bilirubin concentration. It would however be wiser to use high-intensity bluelight phototherapy in urgent cases with more severe hyperbilirubinaemia where a rapid decline in bilirubin levels to safer concentrations would be preferred, or in those not responding to white daylight phototherapy. Indeed our blue-light phototherapy has been so effective that even severely haemolytic cases have been treated in this manner with great success (lo), thereby eliminating the need for exchange transfusion; packed cell transfusion for the resulting anaemia was performed with very little morbidity. The anticipated discomfort caused by the hard perspex surface did not arise since the infants tolerated blue-light and daylight phototherapy equally well. The increased efficacy of standard daylight phototherapy in small preterm infants which has been reported previously (2), was also observed with high-intensity phototherapy. The relative increase in surface area as well as skin translucency are obvious enhancing factors. However, the efficacy in large preterm infants was comparable to that of full-term infants.
ACTA PRDIATR 81 (1992)
A more rapid rebound was observed after highintensity phototherapy resulting in a need for rephototherapy that was about five times more frequent than that of the daylight group. This was not evident in an earlier study (3) probably because of the smaller numbers involved; only with vastly larger numbers will this difference become evident. This phenomenon is probably due to the duration of high-intensity phototherapy being significantly shorter, with cessation of phototherapy occurring at a period when the rate of bilirubin accumulation (bilirubin production versus elimination) was still appreciable; at a level still capable of reducing the efficacy of standard daylight phototherapy. The longer duration of standard daylight phototherapy ensured cessation at a later stage when the rate of bilirubin accumulation was probably already declining; hence the lesser rebound and need for a second exposure. Phototherapy would therefore seem to be the preferred method of treatment for all neonates presenting with hyperbilirubinaemia. For infants with especially severe hyperbilirubinaemia, high-intensity blue-light phototherapy should be considered. This should also be the case for those infants not responding to phototherapy with daylight lamps. Our experience has demonstrated such an approach to be successful in the vast majority of patients. Morbidity was low and mortality nil. The non-invasive nature of this form of treatment has much to recommend it as the natural choice for the treatment of neonatal jaundice. Acknowledgements.-The authors thank Dr TS Koh, Medical Director of Kandang Kerbau Hospital, for permission to report these findings, the nurses for their enthusiastic help, LE Lee for technical help, and J Ow for secretarial help.
References 1. Lucey JF. Neonatal jaundice and phototherapy. Pediatr Clin
North Am 1972;19:827-39 2. Tan KL, Boey KW. Efficacy of phototherapy in non-haemolytic hyperbilirubinaemia. BMJ 1986;293:1361-3 3. Tan KL. The pattern of bilirubin response to phototherapy for neonatal hyperbilirubinemia. Pediatr Res 1982;16:6704 4. Tan KL. Comparison of the effectiveness of“sing1edirection” and “double direction” phototherapy for neonatal jaundice. Pediatrics 1975;56:550-3 5. Dubowitz LMS, Dubowitz V, Goldberg C. Clinical assessment of gestational age in the newborn infants. J Pediatr 1970;77:1-10 6. Brown WR, Wong HB. Hyperbilirubinemia and kernicterus in glucose-6-phosphate dehydrogenase deficient infants in Singapore. Pediatrics 1968;41:1055-62 7. Tan KL. Some aspects on management of neonatal jaundice in Singapore. J Singapore Paediatr SOC1978;20:12241 8. Nie NH, Hull CH, Jenkins SG, Stanbienner K, Bent DH. Statistical Package for the Social Sciences. New York: McGraw Hill Book Co, 1975 9. Tan KL. Efficacy of fluorescent daylight, blue and green lamps in the management of non-hemolytic hyperbilirubinemia. J Pediatr 1989i14132-7 10. Tan KL. Light dose-response relationship in phototherapy. In: Rubaltelli FF, Jon G eds Neonatal Jaundice: New Trends in Phototherapy. New York: Plenum Press, 1984:23544 Received Aug. 2, 1991. Accepted Dec. 12, 1991
