Geir Ringstad
At our hospital, patients with idiopathic normal pressure hydrocephalus (iNPH) are selected for shunt surgery based on overnight intracranial pressure (ICP) monitoring, as 9 out of 10 patients with pulsatile ICP above threshold levels have proven to benefit from the treatment.1 Furthermore, the ICP test has high specificity for ruling out patients who would not respond to shunting and may therefore spare them complications with shunts. Even though ICP monitoring carries a low complication risk (1%), a noninvasive diagnostic alternative with a similar performance level would, in the end, be preferable.
The aqueductal stroke volume (ASV) from phase-contrast MRI (PC-MRI) has been proposed as such an alternative,2 and the theoretic link between pulsatile aqueductal CSF flow and ICP pulsatility has been established.3 However, no previous study had compared ASV with any invasive measure from the intracranial compartment. In short, we found no signs of clinical utility for ASV. ASV widely varied between patients, and we found no association between ASV and clinical or ICP parameters. Due to a high shunt response rate, and only 1 nonresponder, sensitivity and specificity for ASV could not be assessed.
Our study initiated a most-welcome debate,4,5 and the AJNR should be thanked for publishing “negative” results, which is mandatory to avoid publication bias. At the time, our main explanation for ASV’s lack of utility was the fact that ICP pulsatility fluctuates significantly over time, and the short duration of a PC-MRI acquisition could hardly be considered representative for average long-term intracranial pulsatility. Moreover, ASV seemed to be confounded by ventricular volume and aqueduct area. Later, it was also reported that respiration is a major determinant of CSF flow,6 which is not controlled for at cardiac-gated PC-MRI. Our optimism for using ASV as a clinical tool has therefore been somewhat curbed.
Other game-changers for our approach to iNPH imaging have been the studies of rodents demonstrating that amyloid-β and tau are cleared from the brain through paravascular pathways, denoted as the glia-lymphatic or “glymphatic” system and mediated by the water channel aquaporin 4 (AQP4).7 Tissue specimens from patients with iNPH and Alzheimer disease overlap significantly with regard to aggregations of amyloid-β,8 and the perivascular level of AQP4 is reduced in iNPH.9 As MRI contrast agents have properties that are suitable for making their entry into and being cleared by the glymphatic system, we have moved on to study glymphatic brain clearance in iNPH by utilizing gadobutrol as a CSF tracer, followed by multiple T1 acquisitions (gMRI).