Fig. 4: Peripheral and central kynurenic acid (KYNA) levels in bipolar disorder (BD)—ongoing depressive symptoms.
From: Peripheral and central levels of kynurenic acid in bipolar disorder subjects and healthy controls
a Plasma KYNA levels in healthy controls (HCs; median = 37.7 nM, IQR = 18.6) and BD subjects (median = 37.0 nM, IQR = 23.2) stratified on total MADRS score. In age and sex adjusted analyses, BD subjects with MADRS score > 4 (median = 33.0 nM, IQR = 14.2) displayed significantly decreased plasma KYNA levels compared to HCs (P = 0.046) although the decrease did not reach significance when comparing with BD subjects and MADRS score < 5 (median = 38.0 nM, IQR = 23.6; P = 0.070). b Cerebrospinal fluid (CSF) KYNA levels in HCs and BD subjects stratified on total MADRS score. Median (IQR) CSF KYNA levels in HCs: 1.56 nM (1.03), BD subjects with MADRS score < 5: 1.91 nM (1.54), and BD subjects with MADRS score > 4: 1.58 (0.93). In age and sex adjusted analyses, no significant difference in CSF KYNA levels could be observed between BD subjects with MADRS score > 4 and HCs (P = 0.34), while this group displayed decreased CSF KYNA levels compared to remaining BD subjects without ongoing depressive symptoms (P = 0.041). However, as BD subjects displaying depressive symptoms more seldom had a history of psychosis (36 vs. 57%) the difference in CSF KYNA concentration between the two BD groups defined by MADRS score was largely explained by distribution of psychosis (β = −0.62; P = 0.041 unadjusted vs. β = −0.54; P = 0.081 adjusted for psychosis). All P-values are two-sided and derived from logistic regression models with age and sex as covariates (and in the comparison of CSF KYNA levels between BD subjects also controlling for a history of psychosis, see above)
