Evaluation of diffusivity in pituitary adenoma: 3D turbo field echo with diffusion-sensitized driven-equilibrium preparation

Abstract

Objective: Diffusivity of pituitary adenoma has not been investigated fully. The purpose of this study was to evaluate the feasibility of turbo field echo with diffusion-sensitized driven-equilibrium (DSDE-TFE) preparation for pituitary adenoma in the sella turcica and unaffected anterior lobe of the pituitary gland.

Methods: This retrospective study included 23 adult patients with pituitary adenomas. Among them, 6 each were prolactin-producing adenomas and growth hormone-producing adenomas (GH) and the remaining 11 were non-functioning adenomas (NON). The apparent diffusion coefficients (ADCs) were measured in the pituitary adenoma and in the unaffected pituitary gland using coronal reformatted plane.

Results: All pituitary adenomas were clearly visualized on DSDE-TFE and ADC maps without obvious geometrical distortion. There were no statistically significant differences in ADC of the all pituitary adenoma (1.50 ± 0.61 × 10(-3) mm(2) s(-1)) and the unaffected anterior lobe of the pituitary gland (1.49 ± 0.37 × 10(-3) mm(2) s(-1), p = 0.99). The ADC in prolactin-producing adenomas (2.04 ± 0.76 × 10(-3) mm(2) s(-1)) was significantly higher than that in GH (1.26 ± 0.47 × 10(-3) mm(2) s(-1); p < 0.05) and NON (1.33 ± 0.42 × 10(-3) mm(2) s(-1); p = 0.04). There was no statistically significant difference between GH and NON (p = 0.97). The intraclass correlation coefficient for ADC was 0.985 in adenomas and 0.635 in unaffected glands.

Conclusion: With its insensitivity to field inhomogeneity and high spatial resolution, DSDE-TFE proved a feasible method for evaluating the diffusivity in the pituitary gland and adenoma.

Advances in knowledge: DSDE-TFE could enable us to assess ADC of pituitary adenoma in the sella turcica with high resolution and few susceptibility artefacts.

Figure 1.

Diagram of three-dimensional turbo field echo with diffusion-sensitized driven-equilibrium sequence used in this study. *Pre-pulse gradients for eddy current compensation. RF, radiofrequency.

Figure 2.

A 23-year-old female with prolactin-producing microadenoma (0.28 cm³). T₂ weighted image (a) shows a hyperintense mass (arrowhead) in the right side of the pituitary fossa. Post-contrast T₁ weighted image (b) shows the hypointense mass (arrowhead). Apparent diffusion coefficient map (c) shows hyperintensity in the mass (2.54 × 10⁻³ mm² s⁻¹; arrowhead). Circles indicate the regions of interest in the adenoma, the unaffected anterior lobe of the pituitary gland and adjacent normal-appearing left temporal grey matter.

Figure 3.

A 63-year-old female with growth hormone-producing microadenoma (0.37 cm³). T₂ weighted image (a) shows a hypointense mass (arrowhead) in the right side of the pituitary fossa. Post-contrast T₁ weighted image (b) shows a hypointense mass (arrowhead). Apparent diffusion coefficient map (c) shows hypointensity in the mass (0.64 × 10⁻³ mm² s⁻¹; arrowhead).

Figure 4.

Comparison of pituitary adenoma and the anterior lobe of the pituitary gland. Apparent diffusion coefficient (ADC) derived from turbo field echo with diffusion-sensitized driven-equilibrium (a) shows no difference in the pituitary adenoma (1.50 ± 0.61 × 10⁻³ mm² s⁻¹) and in the unaffected anterior lobe of the pituitary gland (1.49 ± 0.37 × 10⁻³ mm² s⁻¹). The T₂ weighted images (rT2) (b) of the pituitary adenoma (1.17 ± 0.23) is higher than that of the anterior lobe of the unaffected pituitary gland (1.05 ± 0.12; p = 0.03). The T₁ weighted images (rT1) (c) of the pituitary adenoma (1.12 ± 0.11) is lower than that of the unaffected anterior lobe of the pituitary gland (1.25 ± 0.14; p < 0.01). n.s., not significant.

Figure 5.

Comparison among tumour subtypes. Apparent diffusion coefficient (ADC) derived from turbo field echo with diffusion-sensitized driven-equilibrium (a) in prolactin-producing tumours (PRL) (2.04 ± 0.76 × 10⁻³ mm² s⁻¹) is higher than that in growth hormone-producing tumours (GH) (1.26 ± 0.47 × 10⁻³ mm² s⁻¹; p < 0.05) and non-functioning tumours (NON) (1.33 ± 0.42 × 10⁻³ mm² s⁻¹; p = 0.04). There is no statistically significant difference between GH and NON (p = 0.97). The T₂ weighted images (rT2) (b) in PRL (1.42 ± 0.24) is higher than that in GH (0.94 ± 0.06; p < 0.0001) and NON (1.16 ± 0.11; p < 0.05). The rT2 in NON is also significantly higher than that in GH (p = 0.02). There are no statistically significant differences in T₁ weighted images (rT1) (c) among tumour subtypes; NON (1.17 ± 0.09), GH (1.10 ± 0.14) and PRL (1.06 ± 0.04; p > 0.05). n.s., not significant.