MRI Assessment of Pituitary Iron and Volume in Thalassemia, and Relation to Hypothalamic-Pituitary-Gonadal Axis Function (HPG): A Feasibility Study.

Improvements in transfusion and chelation treatment have led to increased survival of patients with thalassemia (THAL), however, HPG axis dysfunction continues as the most prevalent complication of iron-overload in THAL. The critical target organ for HPG axis dysfunction in thalassemia is the anterior pituitary gland. In a pilot study on iron-induced endocrinopathies in chronically transfused thalassemia and sickle cell disease (SCD) patients, we developed MRI methods to quantitatively assess pituitary iron by measuring the transversal relaxation rate (R2), the anterior and posterior volume, and anterior pituitary height. We examined the hypothesis that iron will be elevated in the pituitary of heavily transfused thalassemia patients, but gland volume and height will correlate more closely with function. We scanned 13 subjects (controls 8, HbE-thalassemia and thalassemia major 4, iron-overloaded SCD 1) with a total of 17 measurements in a 1.5 Tesla imager of Children’s Hospital Oakland (Philips Gyroscan®) using a quadrature head coil. Two of 4 thalassemia patients were on hormone replacement treatment. The relaxation rate R2 of the anterior pituitary gland was determined from 2 or 3 sagittal slices (3 mm slice thickness, 0.3 mm gap, manually traced ROI covering more than 60% of the gland within each slice) by a turbo-spin echo sequence acquiring 8 evenly spaced echoes (TE = 15, 30,..., 120 ms, TR = 2500 ms). Anterior and posterior pituitary volumes were analyzed from sagittal slices by manual tracing (3D FFE, 1 mm³ isotropic voxel, TE = 4.6 ms, TR = 9.8 ms, a = 15°). Anterior pituitary height was obtained from routine T1 weighted sagittal multi-slice spin echo images (slice thickness 3 mm, gap 0.3 mm). Liver iron concentration (LIC) by SQUID biosusceptometry was used for total body iron estimation. A reproducibility of 2.6 % (COV) was determined in a normal subject for a mean R2 = 11.8 s⁻¹. The relaxation rates for the control group R2 = 10.8 ± 0.8 s⁻¹ and the thalassemia group R2 = 14.3 ± 1.4 s⁻¹ were significantly different (p (U-test) < 0.001) with no overlap. A weak Spearman rank correlation was found with LIC (R_S = 0.71, p < 0.04), but not with plasma ferritin. Anterior pituitary height H and volume V were related by V [mm³] = 95 · H [mm] - 187 (R² = 0.80). For the anterior pituitary, a significant correlation was found between R2 and volume (R_S = −0.59, p < 0.04). Although volumes and heights did not significantly differentiate between the patients and controls, the two THAL patients with abnormal pituitary function had lower anterior pituitary volume and height compared with the two others with normal function. The one iron-overloaded SCD patient had normal R2 and pituitary function; volume and height were in the same range as controls. In summary, the higher transverse relaxation rates indicate elevated iron concentrations in the anterior pituitary gland in all of our thalassemia patients. Pituitary volumes or heights alone may reflect pituitary function, but abnormal volume/height was only seen in patients with high R2. Additional studies in patients with thalassemia and SCD are required to confirm these preliminary observations.