191 Hydrocortisone as an intervention (RCT) 6 The secondary outcomes of this study were sleep, quality of life, and hunger feeling. Hydrocortisone did not reduce parental distress or improve sleep problems or quality of life of patients. Additionally, the average hunger score and fasting hunger score increased during hydrocortisone compared to placebo. The effect of glucocorticoids on hunger is not completely unraveled, and our findings may be explained by the fact that glucocorticoids act differently on appetite than on other side-effects, for example by altering excretion of appetite-regulating hormones, such as leptin.32 Besides a different mechanism, a bias in reporting the hunger score could play a role since this proved to be difficult for parents, resulting in fewer patients to evaluate. An interesting observation in our data is that the delta scores of the first, ‘dexamethasone only’ course are remarkably higher than the subsequent delta scores in the RCT (figures 3 and 4, tables 1 and 4). This may be caused by regression to the mean, however other explanations may be possible. The decrease in side-effects during the RCT may be attributed to a placebo (by proxy) effect.29,30,33 Expectancies, which are an important learning mechanism and may steer placebo-effects,34 may have played a role in our study, since both parents and children were informed about the potential positive effect of hydrocortisone. Furthermore, a participation effect or classical conditioning may have occurred: by adding an oral suspension to standard treatment, patients can be triggered to show physiological responses to additional medication.29,35-37 However, since we did not include a third observational arm with treatment as usual, a direct comparison between the intervention with hydrocortisone or placebo and no intervention (natural course) cannot be made. Clinical implications and future directions The question remains, should we use hydrocortisone in clinical practice? Our study suggests that hydrocortisone has the same effect as placebo on the outcome. Therefore, hydrocortisone is not advised as standard of care for children with ALL who experience dexamethasone-induced neurobehavioural problems. The current study was not designed including a third ‘treatment as usual’ arm, therefore we cannot show that both hydrocortisone and placebo improve side-effects compared to a non-intervention setting. Based on the observations in our study, it would be interesting to explore the possibilities of nocebo- and placebo-effects in the respective prevention and treatment of dexamethasone-induced side-effects. A recent expert consensus paper regarding placebo- and nocebo-effects in adults stresses the importance of making optimal use of placeboeffects to achieve better treatment outcomes.38 Studying the effect of hydrocortisone and open-label placebo, which has been proven effective in children with functional abdominal pain or attention deficit hyperactivity disorder (ADHD), would be very interesting.37,39 Besides further research on the placebo-effect, we propose to create awareness about possible nocebo-effects of dexamethasone in clinical practice.
RkJQdWJsaXNoZXIy MTk4NDMw