Background: Observations and measurements of olfactory structures in humans have been difficult and not of common neuroradiological interest. Because of our interest in olfaction, we have studied the presence, size, and function of these structures in normal subjects and in patients with smell loss.
Methods: Magnetic resonance imaging studies of brain were performed in 220 consecutive patients in our medical center for a variety of clinical neurological investigations. Magnetic resonance imaging studies were performed in each subject including high-resolution coronal T2-weighted fast-spin echo images in the orbitofrontal region. Measurements of olfactory bulb diameter, olfactory sulcal depth, and morphology of the olfactory grooves were performed.
Results: Olfactory bulbs were present bilaterally in each patient studied. Olfactory bulbs appeared duplicated in 11 patients and triplicated in one (5.4% of the total group). Whereas olfactory sulcal depth was similar in all patients, olfactory bulb diameter in patients with duplicate or triplicate bulbs was significantly smaller than those in subjects with single bilateral olfactory bulbs. One patient with congenital hyposmia and olfactory bulb duplication had significant impairment in olfactory acuity. None of the other subjects complained of smell loss.
Conclusions: Olfactory bulbs with a duplicated or triplicated appearance and associated changes in olfactory groove morphology can be present in patients examined with orbital magnetic resonance imaging, and are not uncommon. Although the mechanism(s) for this finding is unclear, it may relate to neurodevelopmental and genetic factors.
From the *Department of Radiology, George Washington University Medical Center, †George Washington University School of Medicine; and ‡The Taste and Smell Clinic, Washington, DC.
Received for publication April 4, 2012; accepted April 20, 2012.
Reprints: Lucien M. Levy, MD, PhD, Department of Radiology, George Washington University Medical Center, 901 23rd St, NW, Washington, DC 20037 (e-mail: firstname.lastname@example.org).
The authors report no conflicts of interest.