Fetal and neonatal MRI findings in CMV infected fetuses and newborns have already been extensively described [5, 6]. Malformations of cortical development, calcifications, periventricular white matter lesions and cysts, ventriculomegaly and cerebellar hypoplasia occur according to GA at the time of infection [4]. Despite this, any in vivo MRI depiction of olfactory system damage has been never reported so far.
Nevertheless, previous neuropathologic study on CMV infected human fetal brains from 23 to 28 weeks of GA reported damage in the olfactory bulbs consisting in disseminated cytomegalic cells, inflammation, necrosis and neuronal and radial glial cell loss. Supporting this evidence, pronounced olfactory deficits were described in mouse model of CMV infection [8] occurring long before the auditory deficits. Additionally, studies on murine CMV showed that the placental CMV inoculation of embryos leads to olfactory bulbs infection. The virus enters via the apical cilia of olfactory sensory neurons (OSN) located in the nasal olfactory epithelium. The OSN project to olfactory bulbs (OB) where they form synapses with mitral/tufted cells whom axons directly connect with neurons of primary olfactory cortex. Additionally, CMV secondary spread systemically to blood through the myeloid cells that infiltrate the olfactory epithelium, become infected and then migrate in the superficial cervical lymph nodes. In murine models OB infection was detected until 16 weeks after birth showing a longlasting persistence over time [8]. Interestingly the olfactory infection is a common, conserved route of mammalian herpesvirus entry to host [8, 9].
Olfaction develops antenatally before audition and vision. Nevertheless, the assessment of olfactory function is challenging as no specific tests on in newborns are available. However, the early detection of olfactory deficits might be relevant, given its function in fetuses and newborns in learning of maternal odors, guiding feeding and social behaviors and maintaining a strong parent–infant bound [11].
The radial glial cells surrounding the periventricular germinal epithelium and the progenitor cells within the germinal areas (the Ventricular Zone (VZ) and the Subventricular Zone (SVZ)) are of utmost importance for fetal brain development. Previous neuropathological studies in human fetal brains showed that the radial glial cells surrounding the periventricular germinal epithelium are the main cellular target of CMV infection. Additionally, the presence of CMV-infected cells in the SVZ and in the VZ, and in cortical plate and subplate too, was also depicted. It was shown that CMV infects cells exhibiting the phenotypic characteristics of neural stem cells/progenitors [7, 8]. These findings may explain the severe cellular loss in the VZ and SVZ occurring in CMV infected brain human fetus [7] associated to the impairment of brain development. Interestingly, a connection between the OB and the periventricular area have been suggested in human fetal brain. In support of this, pathological studies in fetal brain demonstrated the presence of a rostral migratory stream of neuroblasts coming from the SVZ and connecting the anterior horn of the lateral ventricle and the OB. Additionally, an extension of the lateral ventricle reaching the olfactory bulb and probably closes during fetal development, was described in human brains [12, 13, 14]. Therefore, in human fetal brain a connection between the periventricular area, where CMV infects cells of the VZ and SVZ, and the OB is probably present and may contribute to the spread of CMV infection within the fetal CNS.
The MRI in vivo involvement of OB in CMV infected newborns hasn’t been described so far. A previous study showed the pattern of physiological MRI appearance of OB from birth to adult age [15]: in newborns at 15 days of median age the OB are depictable as two hypointense oval structures with a less hypointense central areas in T2- weighted images. The central part was interpreted as an area of axons and synaptic networks connected to primary olfactory area and to ganglionic eminence still with immature myelination and rich in extracellular matrix; this area undergoes progressive myelination, similar to cerebral white matter, reduction of the extracellular matrix and is no more detectable in children older than 2 years of age. In the patient described the central portion of the OB showed noticeable abnormal T2 hyperintensity, while the surrounding peripheral nerve portion was spared (Fig. 1). It can be assumed that the central part is highly susceptible to damage due to its immaturity. As no specific tests for olfactory functions on newborns are available, MRI still remain a unique test for the assessment of central olfactory system in newborns. The lack of confirmation of olfactory dysfunction is the main limitation in the case presented and it may be argued that the appearance of OB described is due to immaturity. Nevertheless, Fig. 1 shows the physiological appearance of OB in a newborn without CMV infection and studied at the same corrected GA, where any severe signal alteration in OB is present, therefore it is unlikely that the OB MRI appearance would be due to immaturity. Interestingly, any changes weren’t observed in the other cranial nerves, in particular in the optic nerves that are a true neocerebral extension, such as the olfactory nerves, suggesting a specific involvement of OB by CMV infection.