This study reviewed evidence for cognitive impairments in trauma-exposed children with and without PTSD. Twenty-seven studies were eligible for meta-analysis, totalling 1526 participants, including 412 trauma-exposed children (PTSD unknown), 300 children with PTSD (PTSD+), 323 children without PTSD (PTSD-), and 491 trauma-naive controls. Eligible studies mostly investigated familial-maltreatment trauma (k=22). Trauma-exposed children (PTSD unknown) performed more poorly overall than controls (d=-0.57). Cognitive deficits were seen in PTSD+ compared to controls, including a large effect size (ES) for general intelligence (d=-0.88), moderate ESs for language/verbal (d=-0.65), visuospatial (d=-0.53), information processing (d=-0.62), learning and memory (d=-0.67), and executive skills (d=-0.52). PTSD+ showed poorer general intelligence (d=-0.28) and visuospatial skills (d=-0.42) compared to PTSD-, whilst PTSD- showed poorer executive function (d=-0.23) and learning and memory (d=-0.61) compared to controls. In conclusion, trauma-exposed children showed cognitive deficits compared to controls, although greatest deficits were associated with PTSD diagnosis.