Synopsis

The limited resolution of 3Tesla MRI often leads to missed lesions or ambiguities in the tumour environment for patients with a pituitary macro-adenoma. In this study we developed a robust high-resolution 7Tesla MRI-protocol of the optico-chiasmatic system and evaluated its clinical value. The 7T MR-images reveal tiny lesions in the optic nerve or chiasm which are not visible at 3T. These lesions could explain the vision loss for 3 of the 7 evaluated patients and gave the physician new treatment possibilities. Overall, this study shows the great clinical opportunities of 7Tesla MRI for patients with pituitary macro-adenoma or other neuro-ophthalmic conditions.

Introduction

Patients with a macro-adenomas of the pituitary gland often suffer from visual problems, such as visual field losses, because the tumour compresses the optic chiasm.¹ Surgical removal of the tumour often restores most of the visual function.

3T MRI is used as the main imaging technique in the diagnosis, surgical planning and post-surgery follow-up of patients with pituitary adenomas. MRI is used preoperatively to assess the morphology of the tumour and optic chiasm, while the post-operative scans are used to detect any residual adenomas, as well as larger lesions involving the optic chiasm.

However, for some questions rising in the process of diagnosis and treatment, the spatial resolution/signal-to-noise of the 3T MRI technique is inadequate.² In preoperative planning it is sometimes hard to judge whether a lesion touches the chiasm or not. Furthermore, in a number of patients the vision problems reoccur years after successful surgery, while no abnormalities can be seen on the 3T MRI.

Although higher spatial resolution and signal-to-noise are possible at 7T than at 3T, there are also increased sensitivities to motion and magnetic susceptibility artefacts. The presence of air-tissue and air-bone transitions in the area of the optic chiasm are potentially problematic.³ Therefore, the aim of the study was to determine in a clinical setting whether the advantages of the higher field outweighed the disadvantages in a true patient population.

Purpose

To develop a robust high-resolution protocol for imaging the optic chiasm and optic nerves on a 7T scanner, and determine to what degree it has additional value in the diagnosis and treatment of patients with pituitary macro-adenomas compared to conventional 3T imaging.

Methods and materials

Images were acquired on a Philips 3 T and Philips 7T Achieva systems using a 32-channel head coil. The standard clinical protocol was run on the 3T system, and protocols based on the same sequences were optimized on the 7T (3D T1-weighted gradient-echo: TE/TR/flip-angle:2.1ms/4.9ms/7°, resolution:(0.7mm)³, FOV:246x246x100mm³, time: 4:50min; three orthogonal 2D T2-weighted spin echo: TE/TR:61ms/3403ms, resolution:0.45x0.45x1.0mm³, FOV: 140x180x24mm³, time:7:43min. For the transverse scan SPAIR fat-suppression was added, increasing the TR to 4000ms and scan time to 8:16min). Third order shimming was performed on a localized volume of interest before image acquisition. A total of 7 patients were scanned on both systems.

Results

The scan protocol developed on the 7T scanner proved to be robust for each of the patients scanned, with image quality superior to the 3T:no magnetic susceptibility artefacts were seen which would interfere with clinical diagnosis.

In one patient, a small lesion in the optic nerve was seen on the 7T T2-weighted images due to the higher resolution, which was not visible on the 3T images (Figure 1(a) and (b)). In another patient the ability to acquire much thinner slices (1 mm vs. 3 mm) with adequate signal-to-noise showed that the tumour was actually in physical contact with the chiasm (Figure 2(a)), whereas in the partial-volume averaged 3 mm slice acquired at 3T this was not apparent. In a third patient, a tiny lesion in a heavily deformed chiasm was seen at 7T which was not visible at 3T (Figure 3(c)).

Discussion

For three of the seven patients studied so far, the higher resolution of the 7Tesla MRI provided a direct clinical benefit, representing a significant percentage of patients studied. In one of the preoperatively scanned patients, the 7Tesla MRI clearly showed a tumour touching the chiasm, which was not apparent from the 3T Images. This information was key for the neurosurgeon to plan the best approach for the intervention. It could potentially also allow for a better prediction of the post-operative visual function.

In general, 3T MR images do not have sufficient resolution to reveal a potential cause for the visual field defects patients develop after a successful treatment. This was also the case for the patients from this study. For two of those patients, however, the 7T MRI showed a tiny lesion in either the optic nerve or optic chiasm. The location of the lesions corresponded to the observed visual field defects. This was an important confirmation for the ophthalmologist to develop a treatment strategy.

Conclusion

The increased resolution of 7Tesla MRI reveals tiny lesions in the optico-chiasmatic system which are not visible at 3T. This provides direct clinical opportunities for the treatment of patients with pituitary tumours and other neuro-ophthalmic conditions.

Acknowledgements

References

1. Ogra, S. et al. Visual acuity and pattern of visual field loss at presentation in pituitary adenoma. J Clin Neurosci 21, 735–740 (2014).

2. Pereira, A. M., Biermasz, N. R. Treatment of nonfunctioning pituitary adenomas: What were the contributions of the last 10years? A critical view. Annales d'Endocrinologie 73, 111–116 (2012).

3. de Rotte, A. A. J. et al. High resolution pituitary gland MRI at 7.0 tesla: a clinical evaluation in Cushing's disease. Eur Radiol 26, 271–277 (2016).