37 Risk factors: a systematic review 2 ranges to compare differences.25,40-42,46,48,52 Regarding sex, four out of five studies (of which two high quality) did not find a significant difference between boys and girls.40,46,49,52 Only one lower quality study found an effect on one of their measured domains: listlessness. Girls seem to be at risk for listlessness; however no effect on all other domains (attention/ hyperactivity, emotional liability, and depressed mood) was found.48 All analyses regarding age and sex were univariate, no multivariate analyses were conducted. Overall, sex seems no risk factor for APRs, but certain age groups might be at risk for specific APRs. The evidence that younger children (0–6 years old) are at risk for behavioral problems is stronger, than the evidence that teenagers are at risk for psychosis. The latter needs to be confirmed in higher quality studies. Treatment factors (type of steroid, steroid dose, and cumulative dose) Six out of eight studies did not find more APRs when comparing dexamethasone to prednisone treatment, including four higher quality articles.4,25,37,38,43,51 Although the majority reports that steroid type is not a risk factor, evidence is not undisputed: two high quality studies did report more APRs during dexamethasone treatment.46,52 Steroid dose was investigated in four studies (one of higher quality).25,47-49 Three report no increased risk of APRs with increasing dose, one low quality study reports an effect on one of their measures domains (listlessness), but not on all other APR domains.48 Steroid dose seems no risk factor based on current evidence, which is overall of low quality. Only one study evaluated the risk of cumulative steroid dose and did not find an increased risk on APRs with a higher dose of prednisone nor dexamethasone.25 All studies on the risk of APRs by steroid type and dose were univariate, no multivariate analyses were used. Parental factors We did not identify any studies describing steroid-induced APRs and parental factors with our search. Medical history With our search, we did not find any studies describing medical history as a risk factor for steroid-induced APRs. Genetic predisposition Five articles studied the influence of genetic variation on steroid induced APRs,4,39,40,44,45 of which Eipel et al. described the same patient cohort twice.39,40 This was the largest patient cohort, consisting of 346 patients. The other studies included 37, 47, and 36 participants, respectively.4,44,45
RkJQdWJsaXNoZXIy MTk4NDMw