# Usage of Hounsfield unit to differentiate idiopathic condylar resorption: a preliminary study

## TL;DR

This study explores using Hounsfield units from CT scans to distinguish between two jawbone conditions, ICR and TMJOA, with mixed results.

## Contribution

The study introduces the potential use of Hounsfield unit distributions to differentiate between idiopathic condylar resorption and temporomandibular joint osteoarthritis.

## Key findings

- HU distributions helped distinguish control subjects from ICR and TMJOA groups.
- ICR condyles showed a tendency for an additional HU peak at 1350–1500 HU.
- No significant HU differences were found between ICR and TMJOA groups.

## Abstract

Background: The Hounsfield unit (HU) is a quantitative scale used to
describe radiodensity in computed tomography (CT) scans. Since idiopathic
condylar resorption (ICR) and temporomandibular joint osteoarthritis (TMJOA)
involve destruction of bone and cartilage in the mandibular condyle, we
hypothesized that HU values might be used to differentiate between the two
conditions. This study aimed to evaluate the usefulness of HU values in the
differential diagnosis of ICR and TMJOA. Methods: Twelve TMJOA and 9 ICR
patients, and 11 healthy subjects were recruited as the TMJOA, ICR, and control
groups, respectively. CT scans were performed, and HU values were measured in the
region of interest (ROI) with 5 mm thickness along the Z-axis from superior
condylar surfaces. HU distributions were then analyzed for each ROI.
Results: Control and TMJOA patients were significantly older than those
in the ICR group. Median HU values of the mandibular condyle did not differ
significantly among the three groups. All groups showed a unimodal HU
distribution peaking at 250–450 HU, while ICR condyles exhibited a tendency to
have an additional peak at 1350–1500 HU. Compared to the control group, the HU
distribution of the TMJOA and ICR condyles was significantly lower at 250–450
HU. After age adjustment, significant intergroup differences in the voxel ratio
were noted at each HU level at 250–300, 300–350, 400–450, 1400–1450, and
1800–1850 HU. However, no significant differences in HU values were observed
between the ICR and TMJOA groups. Conclusions: HU values and
distributions of the mandibular condyle may be used to differentiate between the
control group and the ICR and TMJOA groups. Further studies with a sufficient
sample size are needed to confirm whether HU values and distribution could become
important indicators for distinguishing between the TMJOA and ICR condyles.