Editor—To investigate the effect of antenatal steroids on thymic size, we reviewed the chest x ray pictures taken in the first 36 hours of life in two groups, each of 25 infants. The first group of infants was born in 1992–3; none of their mothers had received antenatal steroids. The second group was born in 1995–6; all of these mothers had received two doses of dexamethasone before delivery. The groups were matched for sex, birthweight (± 100 g), and gestation (±1 week). There was no significant difference in the causes of premature delivery; the most common stated cause of premature delivery was amnionitis (six from group 1, seven from group 2), although in most cases the cause of premature delivery was unclear (12 from group 1, 14 from group 2).

In group 1 we observed thymic shadows in 16 of the chest radiographs. Mothers of group 1 infants with no thymic shadows had hypertension (4 mothers) or suspected amnionitis (5 mothers). In group 2 no radiograph showed a thymic shadow. Using χ² analysis, these figures gave a probability value of p < 0.01 for the null hypothesis that antenatal steroids do not cause thymic atrophy in the fetus. Does the neonatal thymus grow after such suppression? In six of the group 2 infants we reviewed successive chest radiographs over the first 4 weeks of life. Three infants showed a steady enlargement of the thymic shadow in relation to the cardiac outline and thoracic cavity on successive radiographs. In all infants from both groups the total white cell counts and lymphocyte counts were within the normal ranges throughout their follow up.

Thymic stroma is particularly sensitive to endogenous or exogenous steroids which induce rapid apoptosis. The volume of the thymus in older children is reduced following stressful stimuli such as infection or chemotherapy.1-4 Birth stresses in term, but not preterm infants have been noted to cause this regression on a chest radiograph.4 This series of radiographs suggested that the use of antenatal steroids is a clinically significant cause of reversible thymic atrophy in the premature infant, in whom the thymus is usually particularly large and active.

We observed no sequelae to the atrophy induced by antenatal steroid. This is not surprising in view of the observation that very small thymic size in some patients with Catch 22 syndrome seems to have no relevance to their long term immunological outcome. In our series radiographic evidence of a change in thymic size was not accompanied by any measurable change in the total lymphocyte count; more subtle changes in T cell phenotype function and dynamics need to be examined. Thymic regeneration has been observed following the use of steroids in older infants.4 Studies using ultrasound scanning and formal T cell phenotyping are currently under way and may help determine what regulates the rate of thymic recovery. Our observation merits careful examination in neonates, as in older children and adults thymic size has been shown to influence the rate of recovery from insults including chemotherapy and radiotherapy.5