Normal pressure hydrocephalus: cerebral hemodynamic, metabolism measurement, discharge score, and long-term outcome

Background: Regional CBF study has been reported effective in the selection of patient with NPH. However, controversial outcome had been reported. We sought to determine if the combination of rCBF measurement, cerebrovascular reactivity, and regional metabolism were positive predictors of shunt responsiveness in NPH syndrome. Methods: Twenty-eight patients with clinical diagnosis of NPH were enrolled to study their rCBF in CSWM before and after the ACT challenge test, the regional CSWM metabolism by MRSI, and the clinical grading by the CSRIH defined by the Ministry of Health and Welfare of Japan in 1996. All the patients received VP shunting procedure by the same neurosurgical team. The pre- and postoperative clinical conditions were recorded. A patient was considered as "responder" when the patient's CSRIH total score decreased by one or more points. Patients have been followed for a median duration of 40.6 months (range, 28-67 months) with Karnofsky performance scale. Results: Twenty-three responders had significant improvement after VP shunting in clinical grading; 5 nonresponders were stationary after VP shunting. During the 3 years of follow-up, 5 of the 28 patients died, the other 6 were lost to follow-up (including telephone contact), and 3 had progressive deterioration. The prechallenge rCBF decreased in all the 28 subjects. In the 23 responders, the rCBF after challenge were greater than 20 mL/min per 100 g (P = .008), had a significantly better CRC in the anterior CSWM than the nonresponders (1.40 vs 1.06), and had normal NAA/Cre ratio in the anterior, middle, and posterior CSWM in MRSI study. In those nonresponders, the NAA/Cre ratio was less than 0.8 in at least 2 regions of CSWM, and in 23 patients with symptoms other than ataxia (dementia, incontinence), the NAA/Cre ratio was less than 1.5 at frontal CSWM area. Discharge CSRIH scale was well correlated with CRC (P < .03), the average ACT challenge CBF (P < .005), and the average rCBF (P < .02). There was a statistically significant correlation between discharge CSRIH scale and follow-up performance at 3 months (P = .017), 2 years (P = .018), and 3 years (P = .038). Conclusion: Measurement of cerebrovascular hemodynamic and regional metabolism can be a good predictor of outcome after shunting in patients with NPH. Magnetic resonance spectroscopic imaging at frontal CSWM has good correlation with clinical symptoms. After VP shunting procedure, the discharge CSRIH scale is a good predictor of long-term outcome of patients with NPH.