The study found people with major depression had a smaller hippocampus - largely accounted for by the high percentage of participants with recurrent depression.
Major depression is a serious mood disorder that affects around 1 in 6 people during their lifetime.
When it occurs, persistent feelings of sadness, frustration, loss or anger disrupt everyday life and can endure for weeks, months or even years.
The hippocampus - whose name comes from the Ancient Greek word for "seahorse" because of its shape - is an area of the brain that, among other things, is associated with forming new memories.
The ENIGMA study researchers, including a group from the Brain and Mind Research Institute (BMRI) at the University of Sidney in Australia, suggest their findings highlight a need to treat depression when it first occurs - especially in adolescents and young adults.
For the global study - which brings together 15 data sets from Europe, the USA and Australia - the team analyzed magnetic resonance imaging (MRI) brain scans of nearly 9,000 participants: 1,728 with major depression and 7,199 healthy individuals.
They also had access to clinical records of the participants with depression.
The researchers report their findings in the journal Molecular Psychiatry.
Smaller hippocampus largely accounted for by recurrent depression
The study has two main findings. The first - which confirms earlier clinical work at the BMRI - is that people with major depression have a smaller hippocampus.
The second finding is that the first finding is largely accounted for by people with recurrent depression - they represented 65% of the major depression participants.
Recurrent depression is a form of major depression where the depressive episodes come back regularly, interspersed with periods of no depression.
Another interesting finding is that people whose major depression started before they reached the age of 21 also had a smaller hippocampus. The researchers suggest this is consistent with the idea that many of these youngsters go on to have recurrent depression.
However, participants who had not experienced more than one episode of major depression - 34% of those with major depression - did not have a smaller hippocampus than the healthy subjects.
Jim Lagopoulos, an associate professor at BMRI, says these findings reveal new information about our brain structures and the mechanisms that might underlie depression. He adds:
"Despite intensive research aimed at identifying brain structures linked to depression in recent decades, our understanding of what causes depression is still rudimentary."
He says one reason we know so little about this is the lack of studies with sufficiently large numbers of participants. Another reason is the disease varies widely, as do the treatments, and there are also complex interactions between some of the clinical characteristics and brain structure.
Support for 'neurotrophic hypothesis of depression'
Co-author Ian Hickie, professor and co-director of BMRI, says the clinical implications of the findings are that we probably need to treat first episodes of depression effectively, "particularly in teenagers and young adults, to prevent the brain changes that accompany recurrent depression."
He says there is also a clear need for studies that can track changes in hippocampus size over time in people with depression. Results from such studies would help to clarify the question of cause and effect, "whether hippocampal abnormalities result from prolonged duration of chronic stress, or represent a vulnerability factor for depression, or both."
Prof. Lagopoulos also suggests the study lends support to the "neurotrophic hypothesis of depression," the idea that people with chronic depression have certain differences in brain biology - such as sustained higher levels of glucocorticoid - that shrink the brain.
Meanwhile, Medical News Today recently learned about new research that suggests brain inflammation links chronic pain with depression. The study is the first to discover brain inflammation caused by chronic nerve pain can affect signaling in regions associated with mood and motivation.