This observational study reported that mothers who completed a series of steroid injections within 48 hours of premature delivery were less likely to have a child who later developed cerebral palsy. However, the number of mothers involved was small and the overall benefit was difficult to evaluate. This is a promising approach, but clearly one requiring additional evaluation.

It is well established that white matter changes (changes to the covering of nerve cell fibers) surrounding the cerebral ventricles in premature infants (called periventricular leucomalacia or PVL) are associated with the later development of cerebral palsy. This is especially true if bleeding has occurred into the area of PVL. Opinions differ as to the relative roles of maternal inflammation and of decreased oxygen (hypoxia) in producing this condition subsequent cerebral palsy. Some argue that both must be present for the development of cerebral palsy.

Several exciting approaches to preventing cerebral palsy based on either the inflammatory or the hypoxic theories are in trials of one sort or another. A study from Australia attempted to measure the benefit of administering adrenal steroids to mothers threatening to deliver their babies prematurely. These agents are now commonly used to suppress inflammation in a number of disorders including multiple sclerosis and optic neuritis (inflammation of the optic nerve) which are also disorders of white matter in the brain. The specific medication used is dexamethazone, a long acting steroid commonly used to treat brain inflammation.

Kent, et al¹ reviewed the outcome of 220 children born before 30 weeks of gestation (40 weeks is normal). The membranes covering the fetus and the umbilical cords of these children had been carefully examined and any steroid treatment prior to birth had been noted. These children were divided into 3 groups on the basis of the degree and location of inflammation:

      1. Those with no apparent inflammation.
      2. Those with inflammation of the fetal membranes only.
      3. Those with inflammation of the infants umbilical cord (considered to be the most severe exposure).

PVL was evaluated with cerebral ultrasound, and children were examined for cerebral palsy at one and three years of age. Unfortunately, many children were not available for follow-up. Most of the mothers had at least some steroid treatment before the child’s birth, but only 72% received the full treatment. Twelve infants died. There was no relationship between steroid use and survival.

Two thirds of the children had no evidence of inflammation. Thirty one infants in this group had been given no steroids or incomplete steroid coverage and were available for follow-up. Only one developed cerebral palsy (3%). Among the 77 with complete steroid coverage, 5 developed cerebral palsy (7%). Due to the small size of the groups, this difference did not reach statistical significance, meaning that it could have occurred by chance.

One third of the infants had evidence of inflammation either in the membranes or the umbilical cord. Lower gestational age was associated with evidence of inflammation. Nine children in this group whose mothers did not have the complete steroid treatment were available for follow-up study. Three had cerebral palsy (33%). Among the 49 who received the full steroid treatment, only 5 developed cerebral palsy (10%). This was statistically significant despite the small numbers.

Comment:

The authors refer to previous observations that the use of steroids in mothers delivering prematurely may decrease the risk of cerebral palsy. They concede that other studies have suggested that dexamethazone is not the best choice of steroid for this purpose. Their study is purely observational (see Research Fact Sheet- Terms Used in Research, May 2005) but it was carefully done and adds some weight to the argument in favor of steroid use.

It should be noted that infants with no pathological evidence of infection developed cerebral palsy. This suggests that either cerebral palsy can occur in the absence of infection or that the methods they used were insufficient. Many would argue that evidence of inflammation in the mother’s blood would have been a better marker of infection than examination of the fetal membranes and umbilical cord. However, inflammation can be caused by factors other than infection and suggests that these may be common.

If only the infants available for follow-up are considered, 4 developed cerebral palsy among the 40 who had either no steroids or an incomplete course (10%). Among the 126 who had the full course, 10 developed cerebral palsy (8%). This is a rather small difference and similar to reported incidence rates (see Research Fact Sheet- Ultrasound and the Prediction of Cerebral Palsy (CP), October 2004 and, Research Fact Sheet- Infection in the Newborn as a Cause of Cerebral Palsy, December 2004).Thus, the findings suggest that steroid treatment of mothers of premature infants may be helpful for preventing cerebral palsy; but the study results are clearly suggestive rather than conclusive.

¹Kent A, Lomas F, Hurrion E, Dahlstrom JE. Journal of Pediatrics and Child Health. 2005 Apr; 41(4)186-90.

© UCP Research & Educational Foundation, June 2005