Salivary cortisol in depressed patients versus control persons: a systematic review and meta-analysis.
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of observational case-control studies
PubMed 20447770 · doi:10.1016/j.psyneuen.2010.04.001
What was done
The authors conducted a PRISMA-compliant systematic review, sequential cumulative meta-analysis, and meta-regression of case-control studies identified through comprehensive database searches to determine whether salivary cortisol levels differ between depressed patients and control individuals.
What was found
Twenty case-control studies with 1,354 depressed patients and 1,052 controls were identified. Random-effects meta-analysis showed salivary cortisol was higher in depressed patients by an average of 2.58 nmol/l (95% CI: 0.95 to 4.21, p=0.002) in the morning and 0.27 nmol/l (95% CI: 0.03 to 0.51, p=0.03) in the evening. In a fixed-effects model, the overall mean difference was 0.58 nmol/l. Morning reference intervals overlapped widely (0-29 nmol/l in depressed patients vs 1-23 nmol/l in controls). Study sequential cumulative meta-analysis suggested random error for the difference. Meta-regression indicated higher intra-assay coefficients of variation (p=0.07) and higher mean age (p=0.08) were associated with larger morning differences, whereas gender and depression severity were not.
Why it matters
Despite small statistical differences in salivary cortisol, the substantial overlap in reference ranges indicates salivary cortisol cannot reliably discriminate between individuals with and without depression.
Limits
Included studies were limited to case-control designs. Analyses of morning cortisol exhibited substantial heterogeneity, funnel plots suggested publication bias, assay kit variability influenced effect sizes, and cumulative analysis indicated the modest difference may stem from random error.
Cited by
- supports Elevated end-of-day or late-day cortisol spikes are associated with depression and anxiety.