Background: Post-traumatic stress disorder (PTSD) has inconsistently been associated with lower levels of cortisol.
Aims: To compare basal cortisol levels in adults with current PTSD and in people without psychiatric disorder.
Method: Systematic review and meta-analysis. Standardised mean differences (SMD) in basal cortisol levels were calculated and random-effects models using inverse variance weighting were applied.
Results: Across 37 studies, 828 people with PTSD and 800 controls did not differ in cortisol levels (pooled SMD=-0.12, 95% CI=-0.32 to 0.080). Subgroup analyses revealed that studies assessing plasma or serum showed significantly lower levels in people with PTSD than in controls not exposed to trauma. Lower levels were also found in people with PTSD when females were included, in studies on physical or sexual abuse, and in afternoon samples.
Conclusions: Low cortisol levels in PTSD are only found under certain conditions. Future research should elucidate whether low cortisol is related to gender or abuse and depends on the measurement methods used.