gms | German Medical Science

Objective: The implantation of gravitational valves has increased over last few years. Advantages with respect to outcome and reduction of overdrainage-associated complications were shown in several studies, especially for children. More and more of these patients will undergo MRI examinations. The purpose of this study is to show extent of magnetic field artefacts and their clinical relevance, allowing for more specific and adapted preoperative planning for the location of valves to preserve MRI as a diagnostic option for intracranial diseases.

Methods: From 2005 to 2010, a gravitational valve was implanted in 60 children. Surgical indications were mostly posthemorrhagic, congenital or postoperative (6) hydrocephalus after tumor resection. Standard-weighted SE and TSE sequences (T1, T2, FLAIR, 3.0 Tesla) were analyzed from 35 children during follow-up with reference to extent of magnetic field artefacts and clinical relevance at different levels of infratentorial fossa, cerebrum and tumor sites. Additionally, medical reports were analyzed for patient data including type and localization of the valve.

Results: Aesculap proSA- (7) and proGAV-valves (28) were implanted. The implantation site was exclusively retro- or supraauricular. Median age was 5.1 yrs (0–15 years) at time of implantation and 6.9 yrs at time of MRI examination. T1- weighted images were available in 100%, T2 images in 31 (89%), FLAIR images in 14 (40%) cases. In T1-weighted images, artefacts occurred at the cerebrum in 97%; mostly the temporal lobe was affected and further artefacts were observed for the petrous bone (54%), cerebellum (49%) and cerebellopontine angle (14%). The brainstem, the fourth ventricle or the tumor site itself were not affected. T2 and FLAIR images showed similar results with one exception of massive artefacts in infratentorial fossa. There was no statistically significant difference between analyzed sequences. Magnetic field artefacts did not cover any important intracranial structure.

Conclusions: Our study shows MRI artefacts caused by gravitational valves implanted in the retro- or supraauricular position regularly affect imaging of intracranial structures. However, in this cohort important structures were not covered and artefacts can be assessed as clinically non-relevant accordingly. Referring to our results, no disadvantages regarding the implantation of gravitational valves in these positions could be observed. Nevertheless, in some patients, e.g. neurofibromatosis, the option of valve implantation in a clavicular or thoracic location should be considered.