Transcutaneous pressure-adjustable valves and magnetic resonance imaging: an ex vivo examination of the Codman-Medos programmable valve and the Sophy adjustable pressure valve

Abstract

Objective: We investigated the compatibility of magnetically adjustable cerebrospinal fluid valves with clinical magnetic resonance imaging. Torque acting on the valves, subjective sensations of valve-carrying volunteers, extension of artifacts on acquired images, changes in valve pressure setting, and accurate valve function after repeated exposure to the magnetic field were tested.

Methods: Two externally adjustable differential pressure valves, i.e., the Codman-Medos programmable valve (Medos S.A., Le Locle, Switzerland) (n = 5) and the Sophy programmable pressure valve (Sophysa, Orsay, France) (Model SP3, n = 4; Model SU8, n = 3; and Model SM8, n = 2) were exposed to the magnetic fields of 1.5-T clinical scanners. Ferromagnetic properties were investigated according to an established protocol. Subjective sensations during positioning and scanning and image artifacts were investigated using standard clinical imaging protocols. Changes in opening pressure setting during repeated magnetic field exposure were examined using valves affixed to a dummy.

Results: Deflection forces measured 117 dynes in the Medos valve, and 2439 (Model SP3), 2172 (Model SU8) and 1914 (Model SM8) dynes in the Sophy valves. Torque during positioning and during imaging was reported for the Sophy valves. Distortion of the magnetic field with artifacts on acquired images ranged 6 x 6 x 12 cm around the Sophy valve and 4 x 4 x 4 cm around the Medos valves. Artifact extension increased with longer TE times on T2-weighted images and when using gradient-echo sequences. The pressure setting of the Medos valves did not change in 31 of 88 tests. Below 170 mm H2O, the maximal disadjustment was 60 mm H2O in each direction. Minor changes of the pressure setting were observed at 50 and 30 mm H2O. In 11 of 15 tests at 200 mm H2O, the setting after magnetic field exposure was below 30 mm H2O. One Medos valve could no longer be programmed after being exposed four times to the static magnetic field. Sophy valves remained at or changed to "high" in 68 of 81 tests and to "low" in 9. All Sophy valves exhibited paramagnetic behavior after the tests. All remained programmable. Observed changes always occurred within the safety area of the magnet.

Conclusion: Subjective disturbances resulting from paramagnetic valve behavior are absent in Medos valves and are minor in Sophy valves. Image artifacts require careful planning of valve position. Artifacts observed in magnetic resonance imaging are less disturbing than those observed in computed tomography. Medos valves are more stable regarding disadjustment than are Sophy valves. Radiological control of valve setting after exposure to the magnetic field is mandatory in both. The 0.5-mT safety line encircling the area that patients with pacemakers should not enter is a useful safety borderline for patients with pressure-adjustable valves. Failure of the programming mechanism of one Medos valve after several exposures to the magnetic field requires clarification.