Fiberoptic Phototherapy vs Conventional Phototherapy

A Clinical Trial of Paul C.

Holtrop, MD; Kimberly Madison RN, MSN;

conducted a randomized, controlled trial to compare fiberoptic phototherapy with conventional phototherapy in healthy jaundiced newborns with birth weights greater than 2500 g. Twelve patients received fiberoptic phototherapy and 14 patients received conventional phototherapy. There were no significant differences between the groups with respect to birth weight, gestational age, feeding method, presence of hemolytic disease, hematocrit, reticulocyte count, or initial serum bilirubin level. Measured irradiance at 425 to 475 nm for conventional phototherapy was greater than that of fiberoptic phototherapy (9.2\m=+-\0.9\g=m\W/cm2 per nanometer vs 8.2\m=+-\1.2\g=m\W/cm2 per nanometer). Both types of phototherapy lowered the level of serum bilirubin after 18 hours of therapy (fiberoptic group, from 231 \m=+-\29to 210\m=+-\24\g=m\mol/L; conventional group, from 231\m=+-\21 to 188\m=+-\26\g=m\mol/L), but the mean serum bilirubin level was lower after 18 hours of therapy in the conventional phototherapy group (188\m=+-\26vs 210\m=+-\24\g=m\mol/L).There were no side effects in either group of newborns. Both methods of phototherapy decreased the serum bilirubin level, but conventional phototherapy did so more effectively, probably because of its greater irradiance. \s=b\ We

(AJDC. 1992;146:235-237)

phototherapy has Conventional tages: the newborn's eyes newborn confined is

usually

a

number of disadvan¬

must be covered,l and the to an incubator. Both ma¬

may inhibit parent-child bonding,2 and eye patching has caused airway obstruction.3 In addition, conventional phototherapy is delivered as a horizontal band of light energy, incident on the newborn's torso and limbs. As such, the irradiant energy is maximal at the body surface nearest to the light source, but decreases significantly when measured at the lateral surfaces of the neuvers

body. A recently developed technique of phototherapy uses light from a fiberoptic source that is fanned out on a cum¬ merbund. The cummerbund is wrapped around the Accepted

for publication October 7, 1991. Department of Pediatrics, Division of Newborn Medicine, William Beaumont Hospital, Royal Oak, Mich. Presented in part at the Society for Pediatric Research Annual Meeting, Anaheim, Calif, May 8, 1990. Reprint requests to the Department of Pediatrics, Division of Newborn Medicine, William Beaumont Hospital, 3601 W Thirteen Mile Rd, Royal Oak, MI 48073-6769 (Dr Holtrop). From the

M.

Jeffrey Maisels, MB,

BCh

newborn's torso.

Eye patching and the use of an incuba¬ apparently unnecessary, and the light transmitted is nearly always perpendicular to the newborn's skin. Two recent studies suggest that this technique of photo¬ therapy is probably as effective as conventional photo¬ therapy.4,5 We compared these two methods with regard tor

to

are

efficacy,

side effects, and convenience. PATIENTS AND METHODS

This

study was approved by the Human Investigation Com¬ mittee of the William Beaumont Hospital, Royal Oak, Mich, and consent for inclusion was obtained from one or both parents be¬ fore randomization into the study or control group. Patients were enrolled from June 1989 through July 1990. Criteria for inclusion in the study were the following: (1) healthy newborn with birth weight greater than 2500 g, (2) age greater than 24 hours, (3) no Rh blood group incompatibility, and (4) phototherapy ordered by the child's physician.

After a pilot study of five patients to familiarize the nursing staff with the fiberoptic phototherapy system, the newborns were randomly assigned (using a computer-generated random table) to receive either conventional phototherapy or fiberoptic phototherapy. The newborns were stratified by feeding method (breast or bottle) and ABO blood group incompatibility. Conventional phototherapy was administered according to our usual nursery protocol using standard equipment. The pho¬ totherapy system consisted of four white (Sylvania Daylight, 20 W, F20T12/D, Sylvania GTE Products Corp, Hillsboro, NH) and four blue (General Electric, 20 W, F20T12-B, General Electric, Cleveland, Ohio) lamps (Olympic Bili-lite, Olympic Medical Corp, Seattle, Wash), placed 35 cm above the newborn, who was placed naked, except for diaper and eye patches, in an incuba¬ tor and removed only for feedings. We measured irradiance at the newborn's head, chest, and abdomen (Olympic Bilimeter, Mark II, Olympic Medical Corp) and expressed the result as the mean of the measurements. The coefficient of variation of the bilimeter measurements (for 20 measurements at the same loca¬ tion under one light source) is 0.0003%. In the fiberoptic phototherapy group (Wallaby Phototherapy System, Fiberoptic Medical Products Ine, Allentown, Pa), the cummerbund was wrapped snugly around the newborn's torso. The first 10 newborns in the fiberoptic phototherapy group were cared for in incubators, and the remaining two in a crib. All newborns wore eye patches. (Although eye patching is proba¬ bly unnecessary with fiberoptic phototherapy, some light es¬ capes beneath the cummerbund. Pending additional clinical ex¬ perience, we thought that eye patching was prudent.) Irradiance was measured on the surface of the fiberoptic cummerbund, and is expressed as a mean of several measurements taken at differ¬ ent areas over the surface of the cummerbund.

Downloaded From: http://archpedi.jamanetwork.com/ by a University of California - San Diego User on 06/01/2015

The decision to initiate phototherapy was made by the child's pediatrician. Total serum bilirubin concentration was measured just before the initiation of phototherapy (baseline), 6 hours af¬ ter beginning phototherapy, and then every 12 hours until pho¬ totherapy was discontinued. A direct serum bilirubin measure¬ ment, direct Coomb's test, maternal and newborn blood typing, reticulocyte count, and complete blood cell count with periph¬ eral smear were performed at, or shortly after, entry into the study. Newborns were removed from the study if their serum bilirubin level rose by more than 9 µ /L per hour after pho¬ totherapy was initiated.

size of 34 µ /L as a clinically important difference in serum bi¬ lirubin levels at the end of phototherapy between the fiberoptic and the conventional group. Using a .05 (5% chance of a type I error), ß= .1 (10% chance of a type II error), and an estimated SD of 24 µ /L for the level of serum bilirubin after 18 hours, the number of patients required in each group was 12. We also measured the irradiance of a fiberoptic panel at var¬ ious distances from the surface. Ten measurements, from vari¬ ous areas on the surface of the cummerbund, were taken 0, 1, 2, 3, and 4 cm from the surface. =

The child's temperature, while in the incubator, was measured

continuously with a skin probe and was taken axillary every 4 hours with a glass mercury thermometer. All study newborns

cared for in the nursery. Serum bilirubin levels were measured with direct spectrophotometry (Bilirubinometer, Advanced Instruments, Needham were

Heights, Mass, or Paramax, Irvine, Calif). Results were analyzed using paired and unpaired f tests for continuous variables and Fisher's Exact Test for categorical variables. Data are reported as

and SDs. To estimate the sample size, we used a software program (True Epistat, Epistat Services, Richardson, Tex) and chose an effect

means

Table 1.—Clinical and

Laboratory Characteristics of Study Groups* Conventional

Characteristic

Fiberoptic Phototherapy Phototherapy

Gestational age, wk Birth weight, g

Sex, M:F

(n = 12)

37.6±2.9

38.7±1.9 3520 + 547

.31

9:3

.30

3255±525 8:6

.22

13:1

11:1

.70

10:4

9:3

.60

62.5±21

66.5±18

.61

5:9

5:7

.56

2:12

3:8

.37

9.2 ±0.9 0.55±0.7

8.2 ±1.2

.039

0.57±0.6

.37

421 ±543 349±533 Reticulocyte count, 10VL 7.2±2.0 7.5±1.4 Apgar score at 1 min 5 8.7±0.6 score at min 8.6±0.8 Apgar Mean axillary temperature 37.1 ±0.2 during phototherapy, °C 37.1 ±0.2 'Continuous data are reported as means ± SDs.

.75

Race, white:other breast:bottle

Feeding, Age phototherapy initiated, h

ABO

incompatibility:no incompatibility Positive Coombs':negative Coombs'

Irradiance, µW/cm2 per nanometer

Hematocrit concentration

significant differences in clinical and laboratory characteristics between the two groups at en¬ try into the study (Table 1). The measured irradiance, however, was significantly greater in the conventional group (9.2±0.9 µ / 2 per nanometer vs phototherapy 8.2±1.2 µ\\7 2 per nanometer, P=.039). The mean se¬ rum bilirubin levels during and after phototherapy are shown in Table 2. After 18 hours of phototherapy, the se¬ rum bilirubin level decreased significantly in both groups (P=.008 for the fiberoptic group, and =.00001 for the conventional group), but was significantly lower in the conventional phototherapy group than in the fiberoptic phototherapy group. The mean decrement per hour in total

were no

serum

treatment

(n 14) =

RESULTS There

.69 .55

.69

Table 2.—Serum Bilirubin Levels

bilirubin level for the first 18 hours of

was

greater in the conventional phototherapy

group than in the fiberoptic phototherapy group. The differences in serum bilirubin levels between the two groups remained statistically significant at 30 hours but not at 42 hours. However, only 11 (42%) of 26 newborns required phototherapy at 42 hours. Two newborns, one from each group, were removed from the study before a 6-hour serum bilirubin level was measured. On receiving the baseline serum bilirubin lev¬ els (344 and 284 µ /L, reported after phototherapy was initiated), their physicians chose to treat these newborns with two banks of conventional phototherapy. At 18 hours of treatment, two newborns in the fiberoptic group were changed to conventional phototherapy; one at the parents' request and one because the light bulb failed in the fiberoptic phototherapy system. Serum bilirubin lev¬ els in these newborns are included only to 18 hours. The Figure shows the measured irradiance as a function of distance from the fiberoptic panel. Even a distance of 1 cm from the surface of the panel significantly reduced the irradiance. The cummerbund produced no abrasions and there were no rashes, temperature abnormalities, or other side effects in either group of newborns. Initially, the nurses preferred to administer conventional phototherapy.

During Phototherapy Treatment*

Total Serum

Bilirubin, µ /L 30 h

42 h

Mean Decrement per Hour During First 18° ht

226±10

203±14

0.44±0.48

(n=4)

(n 4)

-"->

Baseline

Fiberoptic therapy

231 ±29

(n 12) =

Conventional

phototherapy

231 ±21

(n 14) =

(fiberoptic vs conventional) are reported as means ± SDs.

.94

6 h

233±32

(n 12) =

215±24

(n 14) =

.14

18 h

210±24

(n 12) =

=

200±17

186±29

(n 13)

(n 8)

(n 7)

.035

.025

.33

188±26 =

=

1.0±0.48

=

.0077

*Data

of

tlndicates mean decrement per hour of total serum bilirubin (as a percentage of initial total serum bilirubin level) du ring the first 18 hours

phototherapy.

Downloaded From: http://archpedi.jamanetwork.com/ by a University of California - San Diego User on 06/01/2015

greater than that in their study, which "exceeded 7 µW/ cm2/nm".

Effect of distance from the

fiberoptic phototherapy.

measurements.

Bars

fiberoptic panel represent the

on irradiance means and SDs

with of 10

There was some difficulty in maintaining a snug fit of the fiberoptic cummerbund against the skin. Subsequently, fiberoptic phototherapy gained some acceptance, espe¬ cially when it was administered with the newborn in a crib. Several parents expressed satisfaction at being able to hold their child during phototherapy. COMMENT Both types of phototherapy decreased the serum bilir¬ ubin level, but conventional phototherapy did so more effectively than fiberoptic phototherapy, most likely be¬ cause of the greater irradiance delivered. The irradiance of conventional phototherapy as measured at the new¬ born's skin was already greater than that of fiberoptic phototherapy when measured at the surface of the panel. This difference would probably be even greater if the ir¬ radiance of fiberoptic phototherapy could be measured at the newborn's skin. It was unfortunate, although not planned, that we actually compared two different dosages of phototherapy. Our intent was to compare the fiberop¬ tic phototherapy system with conventional phototherapy as we currently use it. It is possible that increasing the ir¬ radiance of fiberoptic phototherapy would make it as ef¬ fective as our conventional phototherapy. In a nonrandomized trial of newborns with nonhemolytic jaundice, Rosenfeld et al4 found that the serum bilirubin level was significantly lower after 48 hours of fi¬ beroptic phototherapy than with conventional photother¬ apy, although the serum bilirubin levels after 12, 24, 36, and 60 hours of phototherapy were similar between the two groups. The irradiance of our conventional photo¬ therapy units (9.2 µ\ / 2 per nanometer) was probably

Gale et al5 found no difference between conventional and fiberoptic phototherapy. In their study, the cummer¬ bund was not wrapped around the newborns; they sim¬ ply laid on the fiberoptic pad. In our study and others, the newborns had mild eleva¬ tions of their serum bilirubin levels. The decision to use phototherapy was made by the child's pediatrician, although the need for phototherapy in most of these newborns is questionable,6,7 and its use may have adverse effects on maternal behavior and attitudes in the first few months of life.8 Fiberoptic phototherapy has some advantages over conventional phototherapy. Mothers can hold their new¬ borns, and eye patching is probably unnecessary. How¬ ever, in our sample, conventional phototherapy lowered the serum bilirubin level more effectively than fiberoptic phototherapy, probably because of its greater irradiance. Fiberoptic phototherapy has not been studied in new¬ borns with severe hyperbilirubinemia (serum bilirubin level greater than 342 µ /L). Although there is no rea¬ son to anticipate that it would not be effective at these levels, such studies should be done. The fiberoptic phototherapy systems were Medical Products Ine, Allentown, Pa.

References 1. Messner KH, Maisels

to the newborn

1978;17:178-182.

primate

MJ, Leure duPree AE. Phototoxicity

retina. Invest

Mary Lynn Chu, MD,

Vis Sci.

ized, controlled application of the Wallaby Phototherapy Sys-

compared with standard phototherapy. J Perinatol. 1990;10:239-242. 6. Newman TB, Maisels MJ. Does hyperbilirubinemia dam-

tem

age the brain of

healthy

full-term infants? Clin Perinatol.

1990;17:331-358. 7. Osborn LM, Lenarsky C, Oakes RC, Reiff Ml. Photother-

apy in full-term infants with hemolytic disease secondary to ABO incompatibility. Pediatrics. 1984;74:371-374. 8. Kemper KO, Forsyth B, McCarthy P. Jaundice, terminating breast feeding, and the vulnerable child. Pediatrics.

1989;84:773-778.

Journals

ARCHIVES OF NEUROLOGY

Younger Than

Ophthalmol

2. Fetus and Newborn Committee, Canadian Pediatric Society. Use of phototherapy for neonatal hyperbilirubinemia. Can Med Assoc J. 1986;134:1237-1245. 3. Al-Salihi FL, Curran YP. Airway obstruction by displaced eye mask during phototherapy. AJDC. 1975;129:1362. 4. Rosenfeld W, Twist P, Concepcion L. A new device for phototherapy treatment of jaundiced infants. J Perinatol. 1990;10:243-248. 5. Gale R, Dranitzki Z, Dollberg S, Stevenson DK. A random-

In Other AMA

Headaches in Children

supplied by Fiberoptic

7 Years

of Age

Shlomo Shinnar, MD, PhD

(Arch Neurol. 1992;49:79-80)

Downloaded From: http://archpedi.jamanetwork.com/ by a University of California - San Diego User on 06/01/2015