Reduced reward responsiveness in treatment resistant depression of middle-aged adults: Association with carotid artery stiffness and tetrahydrobiopterin

Abstract

Nearly one third of the population diagnosed with major depressive disorder (MDD) fail to respond to two or more antidepressant drugs of adequate dose and duration. This necessitates identification of confounding psychological and physiological factors that could contribute to treatment resistant depression (TRD). The present longitudinal study investigated the influence of behavioural inhibition system (BIS) and behavioural approach system (BAS) in treatment resistance. Further, the association of depression severity with physiological factors contributing to arterial stiffness was also investigated. Baseline data was acquired from 101 middle-aged (36-56 years) patients on immediate diagnosis with MDD using DSM-V criteria. Follow ups were conducted at 06 months and 12 months during treatment. Psychological assessment battery at baseline and follow ups comprised of Hamilton depression rating (HAM-D) for depression severity, WHODAS-2 and BIS-BAS score. Atherosclerosis and central arterial stiffness were measured by intima-media thickness of internal carotid artery and brachial-ankle pulse wave velocity. Physiological factors influencing central vascular function viz., body-mass index, estimated glomerular filtration rate, HbA1c, central systolic and diastolic blood pressure, heart rate and tetrahydrobiopterin were also investigated. Our results show lower reward responsiveness (BAS-RR) and higher BIS scores in TRD patients along with differentially higher intima-media thickness of left internal carotid artery. Higher depression severity at all stages of the study was correlated with lower tetrahydrobiopterin and BAS-RR scores. We, therefore, suggest that vascular depression resulting due to increased intima-media thickness of left carotid artery and lower tetrahydrobiopterin could be contributing factors for treatment resistance in middle-aged MDD patients.

Fig 1. Study design and timelines for follow ups.

Fig 2. Patient demographic and clinical characteristics (Mean ±SD) of newly diagnosed MDD patients (n = 101) at baseline.

Fig 3. Parametric mean ±SD scores of responders and non-responders to anti-depressant treatment at various study check points.

£indicates p < 0.01 and € indicates p < 0.05 between responders and non-responders as determined using t-test.

Fig 4

Representative images of intima media thickness in a) healthy volunteers b) Treatment non-responders and c) Treatment responders. Graph denotes Mean ±SD of d) Brachial-ankle pulse wave velocity e) Intima-media thickness of left carotid artery f) Intima-media thickness of right carotid artery.

Fig 5. Individual and joint effect of treatment response and study time on various measures.

The tests were conducted using repeated measures ANOVA. The p-values were adjusted for multiple comparison using Bonferroni test.

Fig 6. Multivariable mixed effect model assessing the individual and joint effect of treatment response and study time on various measures.

The mixed effect models were adjusted for age, sex, education, and socio-economic status. The confidence intervals and p-values were adjusted for multiple comparison using Bonferroni test.

Fig 7. Adjusted predicted values of parameters viz.

HAM-D, BAS-RR, BAS-D, BAS-F, BIS, WHODAS, SBP, DBP, HR, W/H, BMI, EGFR, Hb1Ac, BaPWV, IMT-L, IMT-R and THB at different study time.

Fig 8. Heat map showing interdependence of study variables on a linear -1 to +1 colour scale with dark purple depicting maximum negative association and bright yellow depicting maximum positive association.

Numbers on the x and y axis from 0 to 16 denote HAM-D, BAS-RR, BAS-D, BAS–F, BIS, WHODAS, SBP, DBP, HR, W:H, BMI, eGFR, Hb1Ac, baPWV, IMT-L, IMT-R and THB respectively.