Comparison of the efficacy of fiberoptic and conventional phototherapy for neonatal hyperbilirubinemia

K. L, Tan, FRCPE, FRACP

From the Department of Paediatrics, National University of Singapore, and the Department of Neonatology, National University Hospital, Singapore, Republic of Singapore

A comparat ive evaluat ion of the ef f icacy of f iberopt ic phototherapy using the Ohmeda Bil iblanket f iberopt ic device, convent ional phototherapy using day- light f luorescent lamps, and a combinat ion of the two forms of phototherapy was made in 165 term healthy infants and 105 preterm infants with hyperbiliru- binemia. In the term infants, the 24-hour decl ine rate for f iberopt ic phototherapy was9.2% _+ 1.6%(mean ___ SEM) versus 21.5% + 1 .8%forday l igh tphoto therapy(p <0.01), and the overal l dec l ine rate was 0.49% ± 0.03%/hr versus 0.70% ± 0.04%/ hr (p <0.001). Combinat ion phototherapy, with a 24-hour dec l ine rate of 29.9% ± 1.0% and an overal l dec l ine rate of 0.97% ± 0.04%/hr, was signif icantly better than dayl ight photofherapy in both respects (p <0.01 and <0.01, respec- tively). The durat ion of exposure for f iberopt ic phototherapy was signif icantly longer than that for dayl ight phototherapy, which in turn was signif icantly longer than that for combinat ion phototherapy. Response to exposure in the preterm infants was signif icantly better than that in the term infants with the respect ive types of phototherapy. The nursing personnel unanimously felt more comfort- ab le with f iberopt ic phototherapy, which did not disturb the swaddled infants as much as convent ional phototherapy. The parents also felt more reassured. Fiberoptic phototherapy proved adequa te in control l ing hyperbi l i rubinemia in preterm infants; in term infants, failures often occurred. Combinat ion photother- apy can be recommended for severe or rapidly increasing jaundice in preterm infants, but its ef f icacy in term infants is uncertain. (J PEDtATR 1994;125:607-12)

New techniques using fiberoptics have enabled photother- apy to be delivered via a flat mat that can be placed directly in contact with the infant's skin. Fiberoptie phototherapy has several advantages: keeping the term infant comfort- ably swaddled during phototherapy, no interruption of ex- posure during feeding, and dispensing with eye pads. A few studies have used the Wallaby phototherapy system (Fi- beroptic Medical Products Inc., Allentown, Pa.); one claimed efficacy superior to that of conventional photother-

Submitted for publication Dec. 23, 1993; accepted April 25, 1994. Reprint requests: K. L. Tan, FRCPE, FRACP, Chief, Department of Neonatology, National University Hospital, 5 Lower Kent Ridge Rd., Singapore 0511, Republic of Singapore. Copyright ® 1994 by Mosby-Year Book, Inc. 0022-3476/94/$3.00 + 0 9/23/56913

apy 1 and another the opposite. 2 A third study demonstrated greatest efficacy with "double" phototherapy when the Wallaby fiberoptic mat was used below and fluorescent lamps above. 3 The numbers involved in those studies were relatively small, especially for specific birth-weight and gestational-age groups. A similar device produced by a dif- ferent company has become available. Our study, involving more infants, compared this relatively new fiberoptic device with our conventional daylight fluorescent lamp units in terms of efficacy and ease of use.

M E T H O D S

Healthy term infants with nonhemolytic hyperbilirubine- mia as previously defined4--normal hemogram, no evidence of blood group isoimmunization, negative result of a direct Coombs test, hemoglobin level >140 gin/L, and hematocrit

607

6 0 8 Tan The Journal of Pediatrics October 1994

>0.40, with exclusion of glucose-6-phosphate dehydroge- nase deficiency (tested by a modification of the Bernstein method)4--were exposed to phototherapy when their bili- rubin concentration was >255 ~mol/L (15 mg/dl), or >222 #mol/L (13 mg/dl) in the first 48 hours of life. The infants were allocated sequentially to three forms of phototherapy: (1) the fiberoptic Biliblanket device (Ohmeda Critical Care, Columbia, Md.), consisting of a halogen lamp with an attached fiberoptic cable containing 2400 optic fibers spread out in a flat mat (the light is transmitted via the fi- bers to the mat, which is placed in direct contact with the skin during phototherapy); (2) seven overhead daylight flu- orescent lamps (Philips TLD 18 W/54; Philips Lighting Co., Somerset, N.J.) arranged in an arc 35 cm above the infant; and (3) a combination of the fiberoptic mat below and seven overhead lamps above the infant. In the first group it was possible to swaddle the infants, with the mat placed against the skin; to ensure maximal efficacy, we used the fiberoptic mat without its sheath and set at maximal power. However, the size of the mat was only 11 × 20 cm and the illuminated part 11 × 13 cm. No eye pads were required. In the other two groups, the infants were exposed completely unclothed, with their eyes covered. Care was taken to ensure that the more active term infants were immediately placed back onto the fiberoptic mats if they rolled off during combina- tion phototherapy.

The study was repeated in preterm infants with nonhe- molytic hyperbilirubinemia in level 2 and level 3 care in the neonatal intensive care unit. The gestational ages of these infants were obtained from the maternal history as well as assessed clinically. 5 All the infants were nursed in open in- cubators and were completely unclothed; the eyes were not covered during fiberoptic phototherapy but were covered for

the other two types of exposure. The irradiance of the fiberoptic device averaged 19.01

#W/cm 2 per nanometer, and that of the seven overhead lamps 6.73 #W/cm 2 per nanometer. Because of the differ- ent spectra of the two types of light, the irradiance of the fiberoptic mat was 867 #W/cm 2 in the 400 to 480 nm range, 437.0 #W/cm 2 in 425 to 475 nm range, 342.0 ~W/cm 2 in the 440 to 480 nm range, and 775.8 ~W/cm 2 in the 440 to 500 nm range; that of the seven overhead lamps was 403.2 /xW/cm 2, 205 #W/cm 2, 106.6 #W/cm 2, and 201.6 t~W/ cm 2, respectively. In the combined setup the total irradiance would be the sum of the irradiance of the other two groups. The fiberoptic irradiance in the green spectrum was sub- stantial (hence the greater values in the spectrum involving the 500 nm band). The measurements were made with a model IL400A radiometer/photometer (International Light Inc., Newburyport, Mass.). Fluid intake was in- creased during phototherapy to offset the increased fluid loss during exposure.

Capillary blood was sampled every 12 hours in the term healthy infants and every 6 hours in the preterm infants to monitor the serum bilirubin response to exposure. The lights were switched off during sampling. The capillary samples were placed in labeled red drinking straws and kept in a light-proof box until the moment of determination under standard conditions with the use of the AO Bilirubinome- ter (American Optical Corp., Southbridge, Mass.), which was calibrated regularly against known standards.

In infants with increasing bilirubin values exceeding the starting value on two consecutive determinations during exposure, direct-acting bilirubin was determined as previ- ously described4; when this value was minimal (<10 #mol/L [0.6 mg/dl ] )phototherapy was deemed to have failed and the infant was transferred to high-intensity pho- totberapy as previously described. 6 Direct-acting bilirubin was also estimated in random samples.

Phototherapy was terminated when bilirubin values had declined to < 185 #mol/L (11 mg/dl) on two successive es- timations, the minimal duration being 24 hours. Bilirubin values were then monitored daily to determine the rebound, for at least 2 days; if rebound bilirubin values increased be- yond those of the prephototherapy values, phototherapy following the same guidelines was repeated. The reaction of the nursing personnel in charge of both the healthy term infants and the preterm infants to the fiberoptic device compared with conventional phototberapy, in terms of ease of use and of infant care, was obtained by interviews. The parents were also interviewed regarding acceptability of the new device. Informed consent from the parent(s) and approval from the director of medical affairs of the hospital were obtained.

The data were statistically evaluated with the Student t test and the chi-square test.

R E S U L T S

Altogether 165 healthy term infants and 105 preterm in- fants (Table I) with nonhemolytic hyperbilirubinemia were studied. All remained well during and after the exposure. In the term infants, phototherapy was effective in decreasing bilirubin levels in all three groups. The response was great- est in the combination group (Figure), followed by the day- light-lamp group. It was least in the fiberoptic group; the duration of phototherapy was significantly longer (Table II) than that for the daylight-lamp and the combination groups. The 24-hour decline rate (expressed as a percentage of the starting bilirubin concentration) and the overall decline rate (decline for the duration of exposure, expressed as a percentage of decline per hour) of the combination group were significantly greater than those of the daylight photo- therapy group, which in turn were significantly greater than those of the fiberoptic group. The failure rate in the

The Journal of Pediatrics Volume 125, Number 4

Tan 6 0 9

Table I. Clinical data of infants studied

White light Fiberoptic mat* Combination phototheropy

Term infants No. (M/F) 55 (33/22) 55 (30/25) 55 (32/23) Birth weight (gm) 3114.2 + 53.4 3122.6 _+ 67.0 3065.9 -+ 54.0 Gestational age (wk) 38.7 +_ 0.2 38.5 + 0.2 38.2 _+ 0.2 Age (days) 3.81 _+ 0.14 3.84 _+ 0.16 3.76 +_ 0.14 Hemoglobin (gm/L)

Start 185.1 +_ 3.2 182.0 + 2.8 178.5 + 3.0 End 171.3 +_ 2.7 166.8 _+ 2.7 165.9 _+ 3.1

Hematocrit Start 0.56 _+ 0.01 0,56 _+ 0.08 0.55 _+ 0.08 End 0.52 _+ 0.01 0,51 _+ 0.08 0.51 _+ 0.09

Bilirubin (~mol/L) Start 261.5 _+ 2.8 257.0 +_ 3.! 263.9 _+ 3.8 End i57.1 +_ 2.4 168.2 + 3.2 146.9 _+ 2.4

Preterm infants No. (M/F) 35 (20/15) 35 (19/16) 35 (17/18) Birth weight (gm) 1506.2 _+ 62.5 1331.3 _+ 61.4 1388.7 _+ 67.3 Gestational age (wk) 31.7 +_ 0.4 29.9 _+ 0.4 30.5 _+ 0.5 Age (days) 3.66 +_ 0.16 3.76 _+ 1.03 4.03 +_ 0.25 Level 3/level 2 18/17 21/14 25/10 Hemoglobin (gin/L)

Start i69.7 _+ 3.8 163.2 _+ 3.9 164.2 _+ 4.4 End 161.4 _+ 3.9 150.9 +_ 4.7 158.0 _+ 4.3

Hematocrit Start 0.52 _+ 0.01 0.51 _+ 0.01 0.52 +_ 0.02 End 0.50 _+ 0.01 0.46 + 0.02 0.49 _+ 0.01

Bilirubin (#mol/L) Start 256.4 + 3.1 250.2 _+ 3.4 266.3 _+ 3.0 End 158.0 _+ 3.3 159.2 _+ 3.7 137.3 +_ 4.8

Values presented as mean + SEM.

*Biliblanket fiberoptic ma t (Ohmeda) .

fiberoptic group (Table II) was significantly greater (p

<0.001) than that of the combination group. One infant

from the fiberoptic group needed a second exposure, com-

pared with none in the other two groups.

The response of the preterm infants was much better than

that of the term infants (Figure); compared with the

respective term groups (Table II), the duration of exposure

was significantly shorter (p <0.001) and the 24-hour decline

and overall decline rates were significantly greater (p

<0.001). Efficacy of the fiberoptic mat was much improved,

being almost as good as that for the daylight lamps; only the

24-hour decline rate was significantly less than that for the

daylight lamps. The combination group had the best

response; the duration of exposure, 24-hour decline rate,

and overall decline rate were significantly better than those

of the other two groups. There was only one failure of pho-

totherapy in the fiberoptic group and none in the other two

groups. A second exposure was required in nine cases in the

combination group compared with seven in the fiberoptic

group (0.80 > p >0.70) and two in th e daylight-lamp group

(p <0.001); the number needing a second exposure in the

fiberoptic group was significantly greater (p <0.01) than in

the daylight group. Response to the second exPosure was as

good as to the first. Three infants in the preterm combina-

tion group needed a third exposure compared with one in the

fiberoptic group and non e in the daylight-lamp group; these

differences were not significant.

The direct-acting bilirubin value was minimal in all the

samples tested; all the values obtained were <10 #mol /L .

The nurses who cared for the term infants were unani-

mous in their approva!of the fiberoptic mat, considering the

mat as more comfortable, without the relatively bulky con-

ventional phototherapy frame; lack of glare was a POsitive

factor, though glare was thought to be a minor problem.

Cleaning the soiled mats was the only disadvantage, but this

was offset by little need to clean the cots, otherwise required

during conventional phototherapy. The nurses stated that

the major drawback was poor efficacy of the fiberoptic mat.

The infants were immediately comfortable with it, mani-

festing none of the restlessness initially observed with con-

ventional phototherapy. All the parents were more reas-

sured with the infants swaddled and without eye pads. The

nurses who cared for the preterm infants, who were

unclothed even before exposure, thought that easier acces-

6 1 0 Tan The Journal of Pediatrics October 1994

3 0 0 1 Fu I I t e rm In fants

25o- .

2 0 0 - ,~. " ~ o . . . - .

e.- • " 0 ..,, ' - . . " - . - . . _ . .

e, 1 5 o - . . . . . . . - . . . . . . . . . . . . . . . -

1 0 0 -

0 3 0 0 -

2 5 0

~ 2 0 0

;~ 1 5 0

' ~ 0

. . . . . . . / / ,

Pre te rm In fants

~ , . ,o- ~" "~ ~ . ~ O.. ~*'"

• __.__.-.-.---"'" N__.__.__.__.-.-.

~ - End photo mean =~ s . e . m .

0 B I I I b l s n k e !

1 O 0 • Daylight lamps 0 / j ® C o m b i n e d

' ' i ~Tr~ ~ I

1 0 112 2~1 3t6 4'8 610 7'2 814 96 Poatpho.t 2 Duration of phototherapy (hrs) (days)

s.e.m. < 5 Funol/I too small to be charted

Figure. Decline in bilirubin concentration in response to the three different forms of phototherapy.

sibility and handling of the infant were important advan- tages; obstruction posed by the conventional phototherapy stand was eliminated.

D I S C U S S I O N

The efficacy of fiberoptic phototherapy was distinctly less than that of conventional phototherapy in the term infants, even though the fiberoptic irradiance per unit area was greater than that of the conventional setup; the reason could be the small size of the mat, which resulted in a relatively small area of exposure (with reduced total light dosage), in

comparison with the much greater area of exposure during conventional phototherapy. In addition, much of the spec- trum was in the green region, which is less effective than that of blue l ight] The interruptions in exposure for feeding of the infants under conventional or combination exposure did not appear to interfere with efficacy; this was not unex- pected, because short interruptions after a period of about 3 hours of phototherapy do not affect efficacy. 8

All three forms of phototherapy were more effective in the low birth weight preterm infants, as previously observed, 4 probably because the infants with the relatively greater

The Journal o f Pediatrics Tan 6 1 1 Volume 125, Number 4

Table II. Response to photo therapy

Therapy group

Term infants

Response

Preterm Infants

p Response ~-~ p

Duration (fir)

Daily (24-hr) decline rate (%)

Overall decline rate (%/hr)

Failures

Fiberoptic* 86.6 _+ 5.1 53.5 + 5.8 <0.001 0.2 > >0.1

Daylight 63.7 + 3.6 43.5 _+ 3.7 <0.001 <0.001

Combination 46.4 _+ 1.5 28.1 _+ 1.2

Fiberoptic 9.2 _+ 1.6 17.9 _+ 2.5 <0.01 <0.01

Daylight 21.5 _+ 1.8 28.1 _+ 2.3 <0.001 <0.001

Combination 29.9 + 1.0 43.5 _+ 2.4

Fiberoptic 0.49 _+ 0.03 0.75 _+ 0.10 <0.001 <0.05

Daylight 0.70 + 0.04 1.10 _+ 0.10 <0.01 <0.001

Combination 0.97 + 0.04 1.84 + 0.09

Fiberoptic 9 of 55 1 of 35 <0.05 0.3 > > 0.2

Daylight 4 of 55 0 of 35 <0.05

Combination 0 of 55 0 of 35

Values presented as mean --- SEM. *Biliblanket fiberoptic mat (Ohmeda).

surface area and a much th inner skin were exposed to a

proport ionately increased light dosage and bet ter l ight pen-

etration. This was especially marked with the fiberoptic de-

vice because the same size b lanket was able to cover a much

greater proportion of the body surface of the small infant;

the durat ion of exposure was comparable to tha t of the

dayl ight lamps, though the ra te of decline in serum biliru-

bin was still slower.

The best result was obtained by combinat ion exposure, as

expected; l ight dosage was significantly increased in this

a r rangement , being equal to the sum of the two forms of

phototherapy. The ra te of response, however, was less than

the sum of the other two, which was not unexpected given

the pa t tern of response of hyperbi l i rubinemia to increasing

irradiance. 6 This type of photo therapy was not more diffi-

cult to adminis ter than conventional phototherapy, though

care had to be taken to place the te rm infant back on the

fiberoptic ma t immedia te ly should he roll off. However, it

appeared to be less effective than our own double-bank blue

light photo therapy 9 used at the optimal dosel°,ll; the effi-

cacy of the fiberoptic ma t in te rm infants with severe or

rapidly increasing jaundice, especially of a hemolytic

nature, is uncer ta in but worth a trial, in case the double-

bank blue l ight setup is not available. However, in pre te rm

infants this double photo therapy proved much more effec-

tive and should adequately control severe or rapidly in-

creasing hyperbi l i rubinemia; our high-intensi ty setup would

be difficult to use with an infant in an incubator .

Our fiberoptic ma t was not compared with the Wal laby

system. Our s tudy demonst ra ted tha t the Ohmeda Biliblan-

ket device could be used routinely for low bir th weight in-

fants; it was apparent ly inadequate for te rm infants, in

whom exposure was too long and failures occurred too of-

ten. A larger ma t might improve efficacy. I f efficacy can be

improved, fiberoptic phototherapy, because of its conve-

nience, ease of use, freedom from obstruction, a n d easy ac-

cessibility to the infant, might become a highly acceptable

mode of t r ea tmen t for hyperbi l i rubinemia.

My thanks to M/S Ohmeda for providing the bilirubin blankets for this study, the Chief Executive Officer of National University Hospital for permission to mount this study, the medical and nurs- ing personnel for their enthusiastic support, and Mr. L. E. Lee and Ms. P. Y. Chua for technical support.

A D D E N D U M

MANUFACTURER'S NOTE; The effectiveness of phototherapy for the treatment of hyperbilirubinemia in newborn infants is a com- bination of irradiance times body surface area, with the use of a therapeutic wavelength of light. When the fiber-optic system was used for preterm infants, with an average irradiance of 19.01/zW/

6 1 2 Tan The Journal of Pediatrics October 1994

cm 2 per nanometer, there was no significant difference compared with the fluorescence system; however, with larger, term infants, the fiber-optic device became less effective. The term infant was treated with the same low irradiance as the preterm infant; the re- sult is a reduced dose of phototherapy. The level of irradiance used by the fiber-optic device was not its maximal setting. The Ohmeda BiliBlanket has three levels of irradiance: low, 15 #W/cm 2 per na- nometer; medium, 25 tLW/cm 2 per nanometer; and high, 35/zW/ cm2 per nanometer. Each level has a standard deviation of 25%, so the infants treated in this study received fiber-0ptic phototherapy at a low or medium irradiahce setting.

David C. Lain, PhD Ohmeda--Specialty Products Division

Columbia, MD 21046

R E F E R E N C E S

1. Rosenfeld W, Twist P, Concepcion L. A new device for pho- totherapy treatment of jaundiced infants. J Perinatol 1990; 10:243-7.

2. Holtrop PC, Madison K, Maisels MJ. A clinical trial of fiberoptic phototherapy vs conventional phototherapy. Am J Dis Child 1992;146:235-7.

3. Holtrop PC, Ruedisueli K, Maisels JF. Double versus single phototherapy in low birth weight infants. Pediatrics 1992; 90:674-7.

4. Tan KL, Boey KW. Efficacy of phototherapy in neonatal hy- perbilirubinemia. BMJ 1986;293:1361-4.

5. Ballard JL, Khoury JC, Wediq K, Wang L, Eilers-Walsman BL, Lipp R. New Ballard score expanded to include extremely pi-ernature infants. J PED~AT~ 1991;119:417-23.

6. Tan KL. The pattern of bilirubin response to phototherapy for neonatal hyperbilirubinemia. Pediatr Res 1982;16:670-4.

7. Tan KL. Efficacy of fluorescent daylight, blue and green lamps in the management of nonhemolytic hyperbilirubinemia. J PE- I~IAT R 1989;1t4:132-7.

8. Vogl TP. Phototherapy of neonatal hyperbilirubinemia: bili- rubin in Unexposed areas of the skin. J PEDIATR 1974;85:707- 11.

9. Tan KL. Phototherapy for neonatal jaundice. Clin Perinatol 1991;18:423-9.

10. Tan KL, Lira GC, Boey KW. Efficacy of high intensity blue light and standard daylight for non-haemolytic hyperbilirubi- naemia. Acta Paediatr 1992;81:870-4.

11. Tan KL, Lira GC; Boey KW. Phototherapy for ABO hae- m01ytic hyperbilirubinaemia. Biol Neonate 1992;61:358-65.

AVAILABILITY OF JOURNAL BACK ISSUES

As a service to our subscribers, copies of back issues of THE JOURNAL OF PEDIATRICS for the pre- ceding 5 years are maintained and are available for purchase from the publisher, Mosby-Year Book, Inc., at a cost of $9.50 per issue. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby- Year Book, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis MO 63146- 3318, or call (800)453-4351 or (314)453-4351 for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues are available from UniVersity Microfilms International, 300 N. Zeeb Rd., Ann Arbor, MI 48106, (313)761-4700.