Cerebrospinal fluid markers before and after shunting in patients with secondary and idiopathic normal pressure hydrocephalus

Thirty-five consecutive patients diagnosed with NPH at the Hydrocephalus Research Unit, Sahlgrenska University Hospital, Goteborg, Sweden, and who had a repeat CSF examination were included (table 1). These are the same patients as reported in a previous paper [18]. All patients received a ventriculo-peritoneal shunt: a Sophy™ Programmable Pressure Valve (Sophysa, Orsay Cedex, France) or a Delta Valve (Medtronic PS Medical, Santa Barbara, USA). The Göteborg University Ethics Committee approved the study and verbal informed consent was obtained from all patients.

Diagnostic criteria for NPH have been described previously [8] and were a) gait disturbance; b) mental deterioration, or urinary incontinence, or both; c) enlarged ventricles on computerised tomography (CT) or MRI with Evan's index > 0.30; d) a lumbar CSF pressure < 20 cm H₂O; e) ventricular filling and block of convexity flow on radionuclide cisternography (RC). The CSF tap test [19] and regional cerebral blood flow (rCBF) measurement was performed on some patients with signs of other disorders, such as vascular lesions. Only patients with improvement at the CSF tap test and a characteristic pattern on rCBF (n = 11) [2] were included. The probable cause of the NPH state was assessed and patients were diagnosed with idiopathic etiology (INPH) if no known cause could be identified. Three months postoperatively, a re-examination was performed using the same protocol except RC. If a patient had not improved clinically and the ventricular system had not decreased in size at the re-examination, the shunt was tested by radionuclide shuntography and was surgically corrected if malfunctioning. Using the clinical measures described below, patients were divided into two groups for data analysis: those improved by shunt surgery or those unimproved by surgery.

Patients were evaluated before and three months after surgery according to a previously published study protocol [13, 20, 21] comprising a semi quantitative evaluation of twelve symptoms and signs measuring gait, balance, mental and bladder performance. Mini-Mental State Examination (MMSE) was used to assess global psychometric performance. Simple reaction time was calculated as the median of 10 tests using simultaneous sound and light stimuli. The other psychometric tests used were the identical forms test (perceptual speed and accuracy) and Bingley's test (learning and memory) [11]. Bingley's visual recognition test was performed by presenting a picture of 12 drawings of familiar objects over a 30 second period. Recognition was tested immediately. Each test was done twice and the mean calculated. The impairment of wakefulness was assessed as presence of somnolence-sopor-coma disorder (SSCD) based on the original coarse division of SSCD into a mild, a moderate and a severe form by Lindquist and Malmgren [22]. Clinical indices reflecting global, psychometric, balance, gait and continence performance were calculated as previously described [20] (Table 2; see Statistics).

Lumbar puncture (LP) was performed in the L3/L4 or L4/L5 interspace with the patient in a recumbent position on the morning before, and three months after shunt surgery. The CSF pressure was measured and the first 12 ml of CSF were collected in ice-chilled tubes, aliquoted, centrifuged and frozen at -80° until analyzed.

The CSF concentrations of the major monoamine metabolites, homovanillic acid (HVA), 5-hydroxy-indoleacetic acid (5-HIAA) and 4-hydroxy-3-methoxyphenylglycol (HMPG) [23], sulphatide [24], neuropeptide Y (NPY) [25], vasoactive intestinal peptide (VIP) [14], tau protein [26], ganglioside GD3 (GD3) [27], gamma amino butyric acid (GABA) [13] and neurofilament protein (NFL) [28] were determined as previously described. Albumin in serum and CSF was determined by nephelometry. The CSF/serum albumin ratio [29] was used as a measure of the blood-brain barrier (BBB) integrity [29].

Nonparametric Mann-Whitney U test was used for between group comparisons. Spearman correlation coefficients were calculated in correlation analyses. Changes between pre- and postoperative values were analyzed by Wilcoxon signed rank sum test. The clinical indices were constructed by calculating the mean of the variables included in each index (Table 2). Scores were standardized to enable comparisons across indices [20]: preoperative variables were z-standardized (mean of 0 and a standard deviation of 1.0) and postoperative variables were standardized using the mean and SD of the preoperative variables. The difference between the pre- and postoperative values was calculated for each index, and the mean of the differences (MoD) was calculated as the overall result after shunt surgery. Patients showing MoD > 0.05 were considered improved, which corresponded to the clinical impression in each case. A p-value < 0.05 was considered significant.

Clinical assessments of patient groups are given in Table 2. Preoperatively, the INPH patients were significantly better than the SNPH patients for MMSE, Bingley's test, gait performance and wakefulness but were significantly worse for continence.

After shunt surgery, 28 patients (80%) were improved (MoD 0.12 to 2.4), 5 (14%) were unchanged (MoD -0.04 to 0.04) and 2 (6%) deteriorated (MoD -0.19 and -0.45, data not shown). Fifteen of the SNPH patients were improved, whereas 2 deteriorated (MoD -0.19 to -0.45). Thirteen (72%) of the INPH patients improved, whereas 5 (28%) were unchanged (MoD -0.03 to 0.03, data not shown). The overall improvement (MoD) for all patients was 0.53 ± 0.60 (mean ± SD) without any significant gender differences (data not shown). Secondary NPH patients (MoD = 0.75 ± 0.72) showed more overall improvement than INPH patients, but the difference was not significant (MoD = 0.33 ± 0.37), Table 2. Specifically, eight tests were significantly improved in the SNPH patient compared to five in the NPH patients.

Preoperatively in both groups, the only pathological findings were elevated neurofilament protein (NFL) and elevated albumin (Table 3). SNPH patients also had an elevated albumin ratio. Apart from this, the only difference seen between SNPH and INPH patients was in NFL, which was higher in secondary cases (4884 ± 5855 vs 1352 ± 2989 ng/L, p = 0.01, Table 3). Preoperatively, there were no differences in any CSF biomarker between patients that improved and those that did not improve after surgery. Patients with cerebrovascular etiology (n = 4) versus those without, had higher sulphatide (287.5 ± 85.4 vs 197.2 ± 97.2 nmol/l; n.s.) and albumin ratios (12.3 ± 2.9 vs 9.9 ± 8.6; p = 0.09, data not shown).

After surgery, significant increases were seen in both SNPH and INPH groups for albumin, albumin ratio, NPY, VIP, GD3. On the other hand, HVA, HMPG and 5-HIAA and sulphatide remained unchanged (Table 3). In addition, GABA was significantly increased in SNPH patients and tau in the INPH patients. These postoperative changes were significant also in the improved subgroup of patients but not in the unimproved subgroup. However, this could be due to fewer patients in the unimproved group (% change similar in both groups). NFL was numerically but non-significantly reduced in all patients, except for the unimproved group where there was a small increase.