Saxena · Archives of general psychiatry 2002 · non-randomized controlled cohort study · n=82

Differential cerebral metabolic changes with paroxetine treatment of obsessive-compulsive disorder vs major depression.

Cited 204 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized controlled cohort study with an untreated control group

PubMed 11879163 · doi:10.1001/archpsyc.59.3.250 · record verified 2026-08-29

What was done

Researchers obtained [18F]fluorodeoxyglucose positron emission tomography brain scans before and after 8 to 12 weeks of paroxetine hydrochloride treatment in 25 subjects with obsessive-compulsive disorder (OCD), 25 with major depressive disorder (MDD), and 16 with concurrent OCD and MDD. A control group of 16 subjects was scanned 10 to 12 weeks apart without treatment. Treatment response was defined as more than a 25% decline in OCD symptom severity, more than a 50% decline in MDD severity, and a clinical global impression of much improved.

What was found

Regional metabolic changes differed significantly across diagnostic groups receiving identical paroxetine dosing and duration. Subjects with OCD alone showed significant metabolic decreases in the right caudate nucleus, right ventrolateral prefrontal cortex (VLPFC), bilateral orbitofrontal cortex, and thalamus that did not occur in any other group. Both the MDD and concurrent OCD plus MDD groups exhibited metabolic decreases in the left VLPFC and increases in the right striatum. Treatment response was associated with decreased striatal metabolism in nondepressed OCD patients but increased striatal activity in patients with concurrent OCD plus MDD. The abstract reports directional changes without numerical values or effect sizes.

Why it matters

This study shows that functional neuroanatomical responses to serotonin reuptake inhibitors vary based on underlying diagnostic pathophysiology rather than operating through a single uniform brain mechanism.

Limits

The trial lacked randomization and a placebo control, using only untreated subjects for comparison. Subgroup sample sizes were small, ranging from 16 to 25 individuals. The abstract reports no numerical metabolic rates, variance measures, or confidence intervals.

Cited by