Introduction

Given their proximity to vital nuclei, brainstem neoplasms present challenges in terms of therapy. Furthermore, management differs between low grade and high-grade neoplasms. While reviewing a case of pontine hemangioblastoma it was observed that the ipsilateral dentate nucleus^1-3 was atrophic on SWI images. This led us to question if the differences in SWI signal within the dentate nuclei as a marker of iron content can be used to differentiate high grade and low-grade brain stem tumors.

Methods

This was an IRB approved, single institution, retrospective study. Primordial was searched for adult patients with pathologic biopsy proven primary or secondary brainstem neoplasms who underwent MR imaging with an SWI sequence from July 2018-June 2019. There were 11 patients (6 females) ranging from 25 to 69-years-old. There were 5 HG (WHO III-IV) and LG tumors (WHO I-II). Age and gender matched controls were selected from a large pool of normal studies done for suspected stroke. The dentate nuclei were identified and manually demarcated on SWI images. Weighted signal intensities were obtained by multiplying signal intensities of dentate nuclei at each slice by the ratio of area of dentate at each slice over total volume of dentate nuclei. Further, average SWI signal intensities were computed by adding weighted signal intensities of each slice (Eqn 1, 2). ROIs were also placed within the normal appearing white matter (NAWM) of the centrum semiovale on both sides for normalization. The average SWI signal intensities for both the dentate nuclei were divided by the NAWM values from respective side. This strategy of weighting and normalization allowed us to compare subjects across scanners/ vendors/ field strengths. Similarly, weighted signal intensity were also obtained for controls.
$$ Ave SWI = ∑_k Wght SWI_k (k= 1 to n).................(Eqn 1)$$
where Ave SWI is average SWI signal intensity, Wght SWI_k is weighted SWI signal intensity from the kth slice, n is number of slices where dentate nuclei will appear
$$ Wght SWI_k = SI_k × (Ar_k/ Vol)-----------------------(Eqn 2)$$
Where SI_k is signal intensity of dentate nuclei from the kth slice, Ar_k is the area of dentate nuclei at the kth slice, Vol is the volume of dentate nuclei

Results

Statistical analysis was performed on SPSS v25. No statistically significant difference was found between the dentate nucleus signal intensity between left and right side for the LG tumor group. For the HG tumor group, a significant difference was found in the dentate nucleus SWI signal (p=0.024), which was increased to the side of the tumor. A similar nonsignificant trend was observed in the LG tumor group (p=0.15). Evaluation of volumes showed no significant difference in the HG group (p=0.451), although there was a trend toward lower volumes of the dentate nucleus ipsilateral to the side of the LG tumor (p=0.10). No significant differences were found in the signal intensities of dentate nuclei of controls.

Discussion/ Conclusion

Significant differences in SWI values within the dentate nuclei of high grade tumors, which was increased ipsilateral to the tumor, and may reflect reduced iron content. A trend towards ipsilateral dentate volume reduction as well as increased SWI values was seen in low grade tumors. Although highly speculative it could potentially reflect a yet unexplored pathway for dentate iron homeostasis, possibly intraneuronal. The early destruction of axons caused by high grade tumors vs displacement/ mass effect by low grade tumors may contribute to changes in dentate volume and iron deposition. Future work with the use of advanced susceptibility imaging techniques such as QSM and STI may help to validate these findings.