J. Paediatr. Child Health (1 996) 32, 508-51 1

Effect of phototherapy on thyroid stimulatory hormone and free thyroxine levels

KL TAN,l Y CHIRINO-BARCELO,’ TC AW2 and R JOSEPH’

3 Department of Neonatology and *Department of Laboratory Medicine, National University Hospital, Singapore

Objective: To study the effect of phototherapy for neonatal hyperbilirubinaemia on thyroid function as neonatal thyroid screening is sometimes performed during exposure to phototherapy. Methodology: Infants with non-haemolytic hyperbilirubinaemia were sequentially allocated to fibre-optic phototherapy, conventional daylight phototherapy, or a combination of both. Bilirubin concentration was monitored 12 hourly by capillary blood sampling; venous blood was sampled for thyroid stimulating hormone (TSH) and free thyroxine (ffd determinations, at start of exposure, at 24 h, end of exposure and 1 day later. Comparable unexposed infants served as controls. Results: All 123 study infants and 25 controls remained well during the study. Bilirubin levels declined during phototherapy, being most rapid in the combination group. The TSH and ff4 values at start of exposure were 3.86f0.41 mU/L (mean f SEM) and 33.20-f.l.l6prnol/L, respectively, in the fibre-optic group, 3.62+0.38mU/Land 37.22*1.76pmol/Lin thedaylight group, and 4.40f0.48mU/Land 29.91 11,13pmol/Lin thecombined group, compared with 5.77+0.40mU/Land 34.46*1.68pmol/L in the control group. The TSH and fT4 values declined with increasing age in the phototherapy and control groups with end of exposurevaluesof 2.90i0.28 mU/Land 27.71 f0 .71 pmol/L, 2.77k0.31 mU/Land 33.52*1.22pmol/L, and 3.44*0.30mU/L and 27.54+0.88pmol/L, respectively, compared with 4.21 rt0.61 mU/L and 27.19+2.33pmol/L (at 72 h) in the control group. The pattern of TSH and tT4 decline in the exposed and control groups was similar, being related to increasing age. Conclusions: The validity of neonatal thyroid screening is not affected by fibre-optic or conventional phototherapy or by both combined.

Key words: free thyroxine, phototherapy, thyroid stimulatory hormone.

Screening for neonatal thyroid status is now routine in many centres. In a minority of cases an inconclusive result would necessitate a repeat thyroid screen; some of the infants requiring this repeat screening test might be undergoing phototherapy because of hyperbilirubinaemia. The validity of neonatal thyroid screening during exposure to phototherapy in an infant with already suspect thyroid status needs to be addressed. Photo- therapy is known to be associated with some side effects: increased fluid loss,’ temperature change2 and effect on ribo- flavin status3r4 have been documented so far, one in Italian: and the other an abstract6 Neither study can be accepted as con- clusive mainly because of lack of controls.

The present study, which included both fibre-optic and con- ventional phototherapy, evaluated the effect of phototherapy, whether localized or generalized, on thyroid function.

METHODS

In the National University Hospital, infants with hyperbiliru- binaemia were exposed routinely to continuous phototherapy

Correspondence KL Tan, Department of Neonatology, National University Hospital, 5 Lower Kent Ridge Road, Singapore 119074, Republic of Singapore

KL Tan, FRCPE, FRACP Y Chtrino-Barcelo, MD. TC Aw, MRCP. MMED, FRCPA R Joseph, MBBS. MMED

Accepted for publication 12 June 1996.

when their bilirubin levels were >255prnol/L (15mg/dL), or >222pmol/L (13mg/dL) in the first 48 h of life. The present study involved healthy full-term infants with no complication other than non-haemolytic hyperbilirubinaemia as previously defined;7 they were assigned sequentially to one of three forms of phototherapy:

(i) the fibre-optic ‘Biliblanket’ device (Ohmeda Critical Care, Columbia, MD, USA), using optic fibres to transmit light from a halogen light source onto a flat mat in direct contact with the infant’s back;

(ii) our standard phototherapy setup using seven overhead fluorescent lamps (Philips TLD 18W/54); and

(iii) a combination of both. The irradiance of the fibre-optic mat at optimal intensity and without the sheath, measured at the centre and at the four corners averaged 775pW/cm2 in the 440-500nm range, 437 pW/cm2 in the 425-475 nm range, 342 pW/cm2 in the 440-480nm range, and 867pW/cm2 in the 400-480 nm range; that of the standard overhead daylight lamps were216pW/cm2,213pW/cm2, 110pW/cm2and 420pW/cm2, respectively; the irradiance in the ‘combined’ group would be the sum of the irradiances in the previous two groups. The measurements were made by the IL 1400A radiometer/ photometer (International Light Corp, Newburyport, MA, USA) as previously de~cr ibed.~

In the fibre-optic group, the infants were swaddled with the fibre-optic mat set at maximal light intensity and without its sheath, placed in direct contact with their bare backs, to obtain optimum effect; no eye pads were required. In the other two setups. the infants were exposed completely unclothed with

Phototherapy and TSH, free thyroxine levels 509

eyes covered; care was taken in the third group to promptly place the infant back on the fibre-optic mat whenever (s)he slid off during exposure.

Phototherapy was started in the morning and was continuous in all three groups interrupted only for feeding, nursing and blood sampling; capillary sampling was performed at start and at 12 hourly intervals, with the lights switched off, and the infant removed from the cot. Where the bilirubin concentration ex- ceeded the initial pre-phototherapy level during exposure on two consecutive estimations, direct acting bilirubin was deter- mined; if minimal (<10pmol/L-0.6 mg/dL), phototherapy was deemed to have failed, and the infant then transferred to high intensity phototheraphy as previously described8mg Direct acting bilirubin was also determined in random samples. Exposure was continued until the bilirubin concentration had declined to <185pmol/L (11 mg/dL) on two consecutive determinations, the minimum duration being 24 h. The bilirubin level was deter- mined as cessation of exposure. Daily monitoring of bilirubin levels post-phototherapy was performed until a declining or stationary trend was observed, the minimum period being 2 days. Phototheraphy was repeated in those with rebound bilirubin values exceeding the initial pre-phototheraphy values, following the same guidelines.

Venous blood for thyroid stimulating hormone (TSH) and free thyroxine (fT4) determination was obtained at the start of photo- theraphy, 24 h later, at termination of exposure, and 1 day after. Fluid intake was increased during exposure to offset the in- creased fluid loss during phototheraphy. An additional smaller group of infants comparable in gestational age and birthweight, with bilirubin levels <170pmol/L(10.0mg/dL) and not exposed to phototheraphy served as controls. The infants were younger than the study group infants; there was no reason for them to stay longer in hospital. Blood was sampled for TSH and fT4 determination at the same time as the infants of the study groups at start and 24 h with the third sample at 72 h; it was not possible to match the timing of the third sample of the study groups which was only taken at cessation of exposure. The temperature of all the infants was maintained within the normal range throughout the study period. Informed consent was obtained.

Table 1 Subject data

The capillary tubes containing blood samples were put in labelled red coloured straws and placed in a light-proof box; bilirubin determination was done under standard conditions using the A 0 bilirubinometer (American Optical, Buffalo, NY, USA) regularly calibrated against known standards. Direct acting bilirubin was determined as previously described.'O The venous blood was centrifuged and the serum stored at -2O'C until TSH and fT4 determination by microparticle enzyme immuno- assay using the IMX system (Abbott Laboratories, Diagnostics Division, Abbott Park, IL, USA)."

The data were statistically evaluated using the Student's t- test. The study was approved by the Director of Medical Affairs of the National University Hospital.

RESULTS

Altogether 123 full-term healthy infants were studied: three groups each of 41 infants were exposed to fibre-optic, conventional daylight and combined phototheraphy respectively; 25 infants formed the control group (Table 1). The birthweight, gestational age, postnatal age and bilirubin concentration of the study groups were comparable; the bilirubin concentration was lower, and the postnatal age less in the control infants. The infants on fibre-optic phototheraphy were comfortable from commence- ment of exposure. The other infants were also comfortable after the initial period of restlessness in the unclothed state. All the infants remained well during and after the study period.

Phototheraphy was effective in all the groups, being most effective in the combined, and least in the fibre-optic group (Fig. l) , the latter also experiencing the longest duration of exposure; the determination of the TSH and fT4 at cessation of exposure in this group was therefore delayed compared with the other two groups. Only two failures of phototheraphy occurred in the fibre-optic group, and none in the other two groups. The failures were transferred to high intensity phototherapye with good response; they were removed from the study.

The combined group had slightly but not significantly higher TSH values than the other two groups which had similar values at start of exposure. The fT4 level of the combined group was

Fibre-optic White light Combined Control

No. subjects (M:F) 41 (25:16) 41 (27:14) 41 (24:17) 25 (16:9) Bodyweight (g) 31 68f69 301 Of68 3111f70 31 60f95 Gestation (weeks) 38.7f0.2 38.4k0.3 38.3f0.4 39.16f0.30 Age (days) 3.91 f0.19 3.76rt0.16 3.55f0.14 2.86 f0.18 Haemoglobin (g/dL)

Start 180.7f2.7 188.8f3.9 177.0f3.2 - End 164.0f2.9 174.8f3.4 164.9f3.8 -

Start 0.552f0.008 0.578f0.013 0.540f0.010 - End 0.503f0.009 0.535f0.011 0.487f0.011 -

Start 256.5f3.2 253.9f3.7 264.6f4.5 113.3f6.1 Duration of phototherapy 93.80f 5.33 60.02f3.77 46.03f1.65 -

Haematocrit

Bilirubin (pmol/L)

TSH (mU/L)

ff, (pmol/L) Start 3.86f0.41 3.62 f0.38 4.40 f 0.48 5.77 f0.62

Start 33.20f 1.1 6 37.22 f 1.76 29.91 f1.13 34.46f 1.68

Values expressed as meanfSEM.

510 KL Tan et a/.

comparable to that of the fibre-optic group, but significantly lower than that of the daylight lamp group (P<O.OOl). at the start of the study; the infants were younger. The TSH values of the control group were significantly higher (P<O.OOl) than those of the fibre-optic and daylight groups, but not that of the combined group; the ff4 level was comparable to those of the study groups. In all study groups the TSH and fT4 values declined gradually over the whole period of exposure; this trend continued after cessation of phototheraphy except for the fibre-optlc group where a slight rise in TSH level was observed. The same pattern of decline with increasing age was observed in the control group (Fig. 1).

DISCUSSION

Phototherapy is commonly used for the treatment of severe hyperbilirubinaemia during the neonatal period, with such

3001

'"1 SEM of SB too small to be charted

P. I -.

0 40 9

" i n

n

F

1 ::phototherapy

0 Flberoptlc 0 Daylight lampa Q Combinatloi 1 M e a n t SEM 0 Control

" . II 0 24 40 72 96 1

Duratlon of phototherapy (h) Portphoto (day)

exposure not infrequently occurring when a repeat thyroid screening test is required because of inconclusive results in the initial screening. That such exposure might interfere with the thyroid function of an infant with a possible suspect thyroid status is therefore a cause for concern. This is especially so in many centres where neonatal thyroid screening is a routine procedure, and where phototheraphy is frequently used. As thyroid function changes rapidly immediately after delivery, the possibility of phototherapy affecting thyroid status at this period of rapid change is a valid one, and should be considered especially as it has been known to affect other aspects of the infant's well-being. Furthermore, the association of hypothyroid- ism and neonatal jaundice is well established. The validity of thyroid screening during phototherapy therefore needs to be examined. Hence the justification of this study.

Unlike other studies,6n7 this study not only included controls, but also evaluated the effect of fibre-optic phototherapy during which the infants' eyes *ere not exposed to phototherapy; with conventional phototherapy, stray light might still occasionally enter the eyes in spite of eye pads. The present study would thus be able to evaluate the effect of phototherapy on thyroid function when confined solely to the skin as well as when occasional exposure of the eyes also occurred.

There was some variation in the TSH and fT, values in the study and control groups; however the values were within the normal range observed in healthy infants in this age group. This study demonstrated that phototherapy whether in the conven- tional, fibre-optic or combination mode apparently did not sig- nificantly affect the thyroid status of the infant. The decline in the TSH and fT4 values in the phototherapy groups was very similar during the period of study, even though the combination group had significantly higher ff4 values at commencement of ex- posure due probably to the slightly younger age of the infants, although this was not observed in the control group; the control group with younger infants and higher TSH value also demon- strated the same trend. This suggested that the decline in both values was due to the increasing postnatal age of the infants, not to light exposure; in fact this is the normal trend seen in healthy infants in the first week of life.12-13 Thyroid function can be affected by external stimuli (e.g. cold exposure12); however, in this study the temperature was maintained within normal limits in all infants. Apparently, skin exposure to light did not affect thyroid status, as the present study demonstrated. Light entering the eye might be more potent in influencing thyroid function since it has a role in influencing circadian rhythm. The present study demonstrated lack of any such influence on thyroid function during and after conventional phototherapy, a most reassuring observation. However, most of the infants, even those exposed to daylight or combination phototherapy, were qui- escent and asleep under the eye pads during exposure after the initial phase of adaptation to the 'unswaddled' state. Photo- therapy should therefore not affect the accuracy of screening procedures for congenital hypothyroidism even when a repeat assessment is required during exposure to phototherapy. Indeed plasma growth hormone and cortisol are also not affected during conventional phototherapy, as is also the case with cholesterol and trigly~eridesl~ and hepatic function.16

Phototherapy is usually used for only a relatively short duration. Even with fibre-optic phototherapy, where exposure is relatively longer, the average duration was only about 94 h. In an era where screening for congenital hypothyroidism is the norm

Fig. 1 Effect of phototherapy on bilirubin concentrations, ff, and TSH values.

in most neonatal centres, the concern that external influences might affect the screening results has to be addressed. A

Phototherapy and TSH, free thyroxine levels 51 1

substantial minority of infants would most likely be screened for thyroid status during exposure to phototherapy; this study is thus reassuring in that it demonstrates that such exposure would not affect the validity of the screening process. Such concerns can now be laid to rest.

ACKNOWLEDGEMENTS

We thank the Director of Medical Affairs of the National University Hospital, Singapore, for approval for this study, the nursing and laboratory personnel for their enthusiastic co-operation, and LH Ang for assistance.

REFERENCES

Oh W, Karecki H. Phototherapy and insensible water loss in the newborn infant. Am. J. Dis. Child. 1972; 124: 230-3. WU PYK, Wong WH, Hodgman JE, Levan N. Changes in blood flow in the skin and muscle with phototherapy. Pediatr. Res. 1974; 8: 257-61. Gromisch DS. Lopez R, Cole HS et el. Light (phototherapy) induced riboflavin deficiency in the neonate. J. Pediatr. 1977; 90: 118-21. Tan KL, Chow MT, Karim SMM. Effect of phototheraphy on neonatal riboflavin status. J. Pediatr. 1978; 93: 494-7. Bosco M. Bacolla A, Castelli G et a/. Fototerapi per iperbilirubinemie

6

7

8

9

10

11

12

13

14

15

16

neonatali e valutazione del T3. T, e TSH. Min. Pedietr. 1984; 36:

Dacou-Voutetakis C, Papadopoulos G. Anagnotakis D. Effect of pro- longed illumination (Phototherapy) on concentration of TSH and T, in human neonates. Pediatr. Res. (Abstr.) 1979; 13: 1185. Tan KL. Comparison of the efficacy of fiberoptic and conventional phototherapy for neonatal hyperbilirubinemia. J. Pediatr. 1994; 125:

Tan KL The pattern qf bilirubin response to phototherapy for neonatal hyperbilirubinemia. Pediatr. Res. 1982; 16: 670-4. Tan KL, Lim GC, Boey Kw. Efficacy of 'high intensity' blue light and 'standard' daylight phototherapy for non-haemolytic hyperbilirubin- aemia. Acts Paediatr. 1992; 81: 870-4. Tan KL, Boey KW. Efficacy of phototherapy on nonhaemolytic hyperbilirubinaemia. BMJ 1986; 293: 1361 -4. Joseph R. Aw TC, Tan KL. Free thyroxine as a supplement to thyrotropin in cord screening for hypothyroidism. Ann. Aced. Med. Sing. 1993; 22: 549-52. Fisher DA, Dussault JH, Sack J, Chopra IJ. Ontogenesis of hypo- thalamic-pituitary-thyroid function and metabolism in man, sheep and rat. Rec. Prog. Horm. Res. 1977; 33: 59-1 07. Fisher DA, Odell WD. Acute release of thyrotropin in the newborn. J. Clin. Invest. 1969; 48: 1670-7. Muhlendahl KEV, Ballowicz L. Growth hormone and cortisol in neonates during phototherapy. 2, Kinderheilk. 1975; 119: 53-8. Hadjigeorgiou E, Triliouri D, Trichopoulou A el a/. Influence of phototherapy on serum lipids of jaundiced newborn infants. Pediatr. Res. 1978; 12: 690-4. Tan KL, Jacob E, Chua KS, Woon W. Phototherapy and neonatal liver function. Biol. Neonate. 1979; 36: 128-33.

61 1-1 5.

607-1 2.