Striatal networks for tinnitus treatment targeting

Abstract

Neuromodulation treatment effect size for bothersome tinnitus may be larger and more predictable by adopting a target selection approach guided by personalized striatal networks or functional connectivity maps. Several corticostriatal mechanisms are likely to play a role in tinnitus, including the dorsal/ventral striatum and the putamen. We examined whether significant tinnitus treatment response by deep brain stimulation (DBS) of the caudate nucleus may be related to striatal network increased functional connectivity with tinnitus networks that involve the auditory cortex or ventral cerebellum. The first study was a cross-sectional 2-by-2 factorial design (tinnitus, no tinnitus; hearing loss, normal hearing, n = 68) to define cohort level abnormal functional connectivity maps using high-field 7.0 T resting-state fMRI. The second study was a pilot case-control series (n = 2) to examine whether tinnitus modulation response to caudate tail subdivision stimulation would be contingent on individual level striatal connectivity map relationships with tinnitus networks. Resting-state fMRI identified five caudate subdivisions with abnormal cohort level functional connectivity maps. Of those, two connectivity maps exhibited increased connectivity with tinnitus networks-dorsal caudate head with Heschl's gyrus and caudate tail with the ventral cerebellum. DBS of the caudate tail in the case-series responder resulted in dramatic reductions in tinnitus severity and loudness, in contrast to the nonresponder who showed no tinnitus modulation. The individual level connectivity map of the responder was in alignment with the cohort expectation connectivity map, where the caudate tail exhibited increased connectivity with tinnitus networks, whereas the nonresponder individual level connectivity map did not.

Keywords: caudate nucleus; hearing loss; neuromodulation; striatal networks; tinnitus.

FIGURE 1

Cohort level tinnitus connectivity maps (7 T). Targets that show an effect of tinnitus on functional connectivity are color coded with respect to their corresponding caudate seed: CHd, dorsal caudate head (pink); CT, caudate tail (green); CHp, posterior caudate head (cyan); CBdl, dorsal lateral caudate body (red); CHv, ventral caudate head (yellow). ANCOVA with age as a covariate thresholded at FWE = 0.05. Bounding box color of bar graphs corresponds to caudate seed color. Target cluster descriptions in Table 2. A, anterior; D, dorsal; L, lateral; M, medial; P, posterior; V, ventral

FIGURE 2

Striatal networks associated with hearing loss (7 T). Targets that show an effect of hearing loss on functional connectivity are color coded with respect to their corresponding caudate seed: CHv, ventral caudate head (yellow); CHa, anterior caudate head (blue). Target cluster descriptions in Table 3. Conventions as in Figure 1

FIGURE 3

Striatal networks associated with tinnitus severity (7 T). Targets that show correlations between functional connectivity strength and tinnitus severity (TFI) are color coded with respect to their corresponding caudate seed: CT, caudate tail (green); CHa, anterior caudate head (in blue); CHp, posterior caudate head (in cyan). Target cluster descriptions in Table 4. Conventions as in Figure 1

FIGURE 4

Individual level tinnitus connectivity maps (3 T). Caudate neuromodulation responder U01‐10 exhibits increased connectivity between left CT and tinnitus networks of cerebellar/temporal regions, in alignment with the cohort level expectation map (Figure 1). Nonresponder U01‐12 exhibits increased connectivity between left CT and frontal or brainstem regions bilaterally, without tinnitus network involvement. Rectangular box denotes DBS lead location. Color bar, case–control t‐statistic; DBS, deep brain stimulation; A, anterior; D, dorsal; L, lateral; M, medial; P, posterior; V, ventral