Ex vivo orbital volumetry using stereology and CT imaging: A comparison with manual planimetry.

OBJECTIVES

To evaluate the applicability of stereology and planimetry in orbital volume measurements using computed tomography (CT) and to compare the results between the two measurements.

METHODS

Experimental study using sheep craniums for CT imaging. Water filling measurements were performed, as the validation technique. Quantification techniques were also evaluated in five human subjects. To examine the proportion of agreement among measurements, we tested intra- and inter-observer agreement.

RESULTS

For stereology customization, a 1/8 systematic sampling scheme was considered as optimal; this resulted in a low coefficient of error (2.59 %) and low measurement time (1.9 mins). In sheep craniums, mean volume measured by water displacement, planimetry and stereology was 17.81 ± 0.59 cm, 17.87 ± 0.68 cm and 17.54 ± 0.49 cm, respectively. Total volumes, obtained by stereology, were highly correlated with the water-filling method (r=0.893; p = 0.001) and a paired t-test showed significant difference between methods (t=3.047; p = 0.014). Planimetry results displayed a high correlation with the water-filling method (r=0.957; p ≈ 0.001) but no statistically significant difference was found (p = 0.154). Mean difference using planimetry and stereology was 0.332 ± 0.322 cm. In human subjects, using stereology, the estimated volume ranged between 18.57 cm and 19.27 cm, and the mean orbital volume was 19.05 ± 0.50 cm with CE=3.75 ± 0.16 %. Mean measure time was 2.1 ± 0.1 mins.

CONCLUSIONS

Stereological measurements were superior to manual planimetry in terms of user effort and time spent. Stereology sampling of 1/8 was successfully applied in human subjects and yielded a strong correlation with manual planimetry.

KEY POINTS

• Stereology can be applied to measure the orbital volume using computed tomography. • Stereological measurements display high correlation with gold standard planimetry and combine low coefficient of error (2.59%) with low measurement time (1.9 min). • Stereology is superior in terms of user effort and time spent.