Integrated IVIM- and DCE-MRI analysis reveals differential modulation of CSF transport along olfactory bulb and optic nerve pathways by taVNS

Cerebrospinal fluid (CSF) transport through the subarachnoid space (SAS) is essential for the brain’s waste clearance. While tracer-based methods such as dynamic contrast-enhanced MRI (DCE-MRI) have advanced the study of CSF pathways, their dependence on the injection site and limited repeatability highlight the need for complementary approaches. Intravoxel incoherent motion (IVIM)-MRI enables noninvasive and repeatable assessment of CSF dynamics, offering distinct biophysical insights. Recent interest in transcutaneous auricular vagus nerve stimulation (taVNS) suggests potential to influence CSF transport, yet its effects on intracranial pathways remain poorly understood. Here, we used IVIM- and DCE-MRI to characterize taVNS-associated alterations in CSF transport in Sprague–Dawley rats. Customized MRI-compatible electrodes were designed to fit the MRI cradle, enabling taVNS delivery during imaging under isoflurane anesthesia. IVIM-MRI was used for within-subject comparison of baseline versus post-taVNS states, while DCE-MRI was used for group-wise comparison between sham stimulation and taVNS groups. To quantify taVNS-associated changes in CSF transport, we analyzed the initial signal decay at low b-values ( D * ) derived from IVIM-MRI and CSF-related metrics derived from DCE-MRI, including the area under curve (AUC), arrival time (AT) and initial slope. Statistical analysis revealed significant changes in D * magnitude across intracranial CSF pathways, including SAS along the cisterna magna, perivascular space in the circle of Willis, SAS adjacent to the olfactory bulb (OB) and SAS adjacent to the dural venous sinus. For DCE-MRI, taVNS-associated changes in the initial slope across intracranial CSF pathways showed a strong linear correlation with changes in D * magnitude ( r = 0.958, p < 0.01). Moreover, taVNS altered the differences in AUC and AT between the OB and optic nerve (ON) pathways compared with sham ( p < 0.05 for both). Integrated analysis of IVIM- and DCE-MRI showed that taVNS altered intracranial CSF transport, enhancing CSF outflow toward the OB pathway while suppressing the ON pathway. This differential modulation of CSF transport along the OB and ON pathways suggests that taVNS may selectively engage distinct anatomical pathways rather than globally enhancing CSF transport. These findings highlight the need for integrated analytical approaches to better characterize the complex dynamics of CSF transport associated with neuromodulatory interventions.