Enhanced Olfactory Cortex Connectivity in a Patient With PTSD With Olfactory Hallucinations

A 53-year-old, white, non-Hispanic woman self-referred to a clinical trial of an experimental medication for PTSD. She complained of daily intrusive memories, nightmares, and flashbacks of a traumatic event, as well as avoidance, anhedonia, and difficulty concentrating. She had no history of psychiatric treatment. Psychiatric assessment using the SCID-I² revealed diagnoses of PTSD and major depressive disorder, and a clinical rating assessment³ revealed PTSD symptoms within the “extreme” category. The patient did not meet diagnostic criteria for any personality disorder or psychotic disorder. She was medically cleared for participation in the trial, with only mild hypertension that was treated with metoprolol.

The patient’s PTSD stemmed from witnessing her mother’s murder at age 5 years. Although the patient had no memory of the event, she experienced intense flashbacks that coincided with olfactory hallucinations. These episodes included experiencing “the overwhelming scent of blood” and lasted up to 30 minutes. These episodes were triggered by seeing someone who reminded the patient of her mother’s murderer. These particular olfactory hallucinations disappeared at age 40 years; however, the patient continued to experience additional olfactory hallucinations related to subsequent traumas (including extensive sexual abuse), specifically the smell of “stale urine” and tobacco. She denied having a generally enhanced sense of smell.

We used resting-state functional connectivity MRI (CONN software⁴) to study the connectivity between the olfactory cortex and areas related to emotion (inferior frontal gyri [iFG], medial frontal gyri, superior frontal gyri, anterior cingulate cortex [ACC], amygdala, and insula). As controls, we used data previously collected from 12 matched patients with depression (all white women, aged 27–51 years; six controls had comorbid anxiety disorders). We found that our patient had markedly stronger functional connectivity between the olfactory cortex and the insula, ACC, and iFG, but not in an unrelated region (supplementary motor area) compared with controls. Assuming a Gaussian distribution, we calculated the probability of obtaining these results as 0.02 for the ACC and 0.08 for the insula and iFG. To study white matter tracts, we performed diffusion tensor imaging tractography on the patient and controls (TrackVis software⁵). We observed a normal number of streamlines and normal fractional anisotropy in the olfactory cortex (Figure 1).

We postulate that a traumatic event at age 5 years affected the patient’s brain development, resulting in enhanced connectivity between the olfactory cortex and brain regions that integrate emotion-related information, such as the insula, iFG, and ACC. It is possible that the development of a second olfactory hallucination associated with subsequent trauma was related to an already hyperconnected olfactory cortex. Although additional subjects are needed to reach statistically significant conclusions, we did not find any overt white matter differences between the patient and matched controls, which suggests that the event-related changes may be functional and not anatomical. Further study is clearly needed to determine the relationship between olfactory hallucinations and PTSD.

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the U.S. Department of Veterans Affairs or the U.S. government.